Advances in Service and Industrial Robotics

In general, i.e. when the kinematic construction of a redundant robot arm does not meet special requirements, the inverse kinematic task has only differential solutions that do not exist in the kinematically singular points and suffer from large angular velocity components in the vicinity of the singularities. Recently it has been pointed out that these problems are caused by the use of some generalized matrix inverses, and they can be evaded by the application of a quasi-differential approach that transforms the task into a Fixed Point Problem and solves it with a convergent iteration without the use of any matrix inversion. In this paper it is shown that further possibilities are available in the quasi-differential approach that differ from each other in the use of the elements of the Null Space of the Jacobian. It is concluded that for maintaining the continuity of the solutions the use of the elements of this null space are practically important. This conclusion is illustrated via simulations for an irregularly extended PUMA-type robot arm, that is an 8 Degree of Freedom system.

Schoenflies-motion generators (SMGs) are 4-degrees-of-freedom (dof) manipulators whose end effector can perform translations along three independent directions and rotations around one fixed direction (Schoenflies motions). Such motions constitute the 4-dimensional (4-D) Schoenflies subgroup of the 6-D displacement group. The most known SMGs are the serial robots named SCARA. Pick-and-place tasks are typical industrial applications that can be accomplished by SMGs. Over the SCARA, lower-mobility parallel manipulators (PMs) have been proposed as SMGs. Here, a novel type of SMG with parallel architecture is presented together with its kinematics analysis. The proposed SMG has a single-loop not-overconstrained architecture, actuators on or near the base and decoupled kinematics.

The selection of the Jacobian matrix and the performance measure affects the results in optimization problems such as the placement of a task in the workspace of the robot. In this paper, a comparative study of three performance measures that are produced by alternative determinations of the Jacobian matrix is presented. As it is shown in the illustrative example, the selection of the Jacobian affects the values of the considered performance measures and thus in most of the cases.

This paper considers the kinematic analysis of a 2 dof closed-loop chain for the night-sky observation known as the Canterbury tracker. Closed-loop mechanisms are characterized by a higher number of joints than dofs, so kinematic constraints act to make dependent their configurations. For some configurations of the joints, it is possible that one or more constraints become redundant, giving instantaneously extra dofs to the mechanism. Thus, conventional input-output singularities do not represent the unique possibility of having controllability issues. In this paper, the screw theory is exploited to determine singular configurations of the proposed trackers. According to the results of this analysis, the geometric parameters of the mechanism are chosen to make it suitable for solar tracking purpose, taking into account the existence of these singular configurations to avoid tracking failures.

This paper describes a novel docking mechanism for space applications. Generally speaking, docking mechanisms have two main objectives. On the one hand, they have to recover the linear and angular errors between a servicer and a customer spacecraft. On the other hand, they have to dissipate the kinetic energy associated with the relative motion between the bodies. The proposed docking mechanism consists of an active part mounted on the servicer spacecraft and a passive one linked to the target. The active part is equipped with a retractable rod fitted to a two degrees of freedom rotational joint. The rod is pointed toward a spherically suspended socket mounted on the target using electrical actuators exploiting optical feedback. The multibody simulations used to test the feasibility of the system are briefly presented. Finally, some alternative mechanical implementations of the pointing system will be described.

Parallel kinematic manipulators are for sure the best choice for those applications requiring high stiffness and dynamic performances. This paper presents the mechatronic design of a low cost 6-DoFs parallel kinematic manipulator, which is the scaled version of the one that will be installed into the wind tunnel of Politecnico di Milano for tests related to the H2020 Life50+ project. The aim of the prototype is to allow the development of control algorithms and design test procedures without stopping the activities of the wind tunnel. For this reason the robot is designed to be as close as possible to the one that will be used for the wind tunnel tests.

Due to their parallel nature, behaviour-based control architectures can strongly benefit from an implementation on FPGAs. The problem is, to find out the real benefit of such an implementation, since general calculations are difficult, due to the heterogeneity of such systems. In this paper, we present early results of the integration of a behaviour-based inverse kinematics solver, based on the iB2C-architecture, on a common FPGA. It is shown, that this implementation is feasible and that the resulting performance is satisfying. Further benefits are evaluated.

A novel parallel robot stemming from the 3-SRU (spherical-revolute-universal) under - actuated joints topology is presented in this paper. The conceptual design here proposed takes advantage of a reconfigurable universal joint obtained by locking, one at a time, different rotations of a spherical pair by means of an automatized system. Such local reconfiguration causes a slight modification of the robot legs mobility which is enough to provide the end-effector with different kinds of motion. The first part of the paper is dedicated at formally demonstrating the motion capabilities offered by the 3-URU kinematics; in the second part, a mechanical solution for the realization of the reconfigurable joint is shown.

This paper deals with the mechatronic design and control of a 3-RPS (Revolute-Prismatic-Spherical) parallel manipulator, which moving platform has three d.o.f.s, one translation and two rotations. Each of the three legs connects the fixed with the moving platform through a revolute and a spherical joint, respectively, while each prismatic joint consists of a DC electromechanical actuator. A prototype was built and experimentally tested by devoting particular attention to the control of the electric drives, as based on the technique of the Path-Step-Diagrams, which is commonly used for controlling pneumatic cylinders.

The simulation of the normal and oblique impact of a rigid rod with a granular material has been studied. This model solved the inconsistency of the dynamic force at the beginning of the impact. Two different impact angles have been analyzed: the normal impact and the impact with an angle $$\theta =45^\circ $$. The penetration depth, the normal velocity of the tip during impact and the normal contact force during the impact have been analyzed. For the normal impact the dynamic force reaches a maximum during the impact, while the static and the normal forces show continuous increase. For the oblique impact the dynamic and the total forces show a maximum during the impact, while the static force is increasing throughout the impact.

Mechatronic tracking of a target by two or more guns simultaneously offers the possibility to eliminate the laser range finder used for measurement of the target distance. A simple exchange of data between autonomous mechatronic agents in an effort to fulfill the common goal is proposed. Qualitative and quantitative configuration requirements are investigated in order to achieve the required accuracy.

This article presents an approach to collision avoidance of industrial robots. It is based on artificial potential fields or force fields generated by virtual charges which are placed on the obstacles. The obstacles are detected by range image sensors. To avoid fault detections e.g. as a result of occlusions and distortions two cameras are used to supervise the robot work space. An algorithm is worked out which eliminates the distortions in the camera images. After this, a method is developed to calculate a 3D image of the environment. For fast computation of the artificial forces the obstacles are classified into geometric primitives. The virtual charges are placed in 3D on the objects. From the charge positions and their potential value an artificial force can be calculated which acts on the end-effector. The force corrects the primary planed robot path to avoid collisions with the obstacles.

The paper deals with a new concept for autonomous path following for a certain class of tasks, where the robot and the environment (object) are tightly coupled and its constraints completely define the motion needed to perform a task. The main idea is that after initializing a motion of such a close-chain system, the controller preserves the motion until the task is completed, without any path planning. One of the important features of the robot system necessary for the proposed concept is that the robot arm has to be compliant. Our main contribution is a proof-of-concept demonstration of path following control. For that, we have implemented the control on a KUKA LWR 4 robot arm, performing the table leveling task by turning the handle of a crank mechanism. The experiments have shown that the robot can perform the desired motion even when the constraints change during the motion, e.g. the admissible motion path.

The paper presents the construction and operating mode of a platform for controlling mobile robots. A number of, at least, three independent mobile robots, who are to be entrusted with the supervision of the zones, will report via a 4G transmission, with a certain frequency of a central server the current position in the geographical coordinates. Latitude and longitude will be accompanied by photos received from a high-resolution camera and these images will be processed with a certain target for identifying an object with certain features. The system will be built around a hardware architecture open-source (for example Atmel, Raspberry PI). Another target of this paper is to present an example where we are going to examine an image processing and text extraction system that will compare an implementation done using the OpenImaj SWT algorithm and the Tesseract OCR compared to a implementation done in Matlab based on MSER. These services can then be called through a restful Web Service by any application, mobile or otherwise to extract text from images.

The paper studies the control problem of a class of hyper-redundant robots with uncertain components by using the output control. To avoid the complexity of distributed controllers, in this paper is proposed a lumped output controller. A weighted technique is used to generate a boundary output. A PD controller is proposed and an algorithm for determining the controller gains is discussed. Lyapunov techniques are used to prove the asymptotically stability of the control system. Numerical simulations and prove the efficiency of the method.

Industrial manipulators are usually equipped with a set of basic sensors (encoders and current sensors) that could not be sufficient to implement more advanced service and control algorithms, like the Payload Check, the Impedance Control or the Cartesian Soft Servo. A software layer between the real sensors and the algorithms is developed in order to provide a new set of “measures”. Such a layer, called Virtual Sensor, exploits the information provided by the real sensors to compute new virtual measures, to be used in the algorithms layer to implement new services. The Virtual Sensor is introduced in a standard COMAU controller and its effectiveness is tested in a service algorithm, called Payload Check, able to detect a wrong declaration of the robot payload. Good performances have been obtained for a wide range of COMAU manipulators.

This work presents a novel control approach for a tendon-driven soft robotic system. The soft robotic system composed of a silicon continuum, tendons and antagonistic actuation yields a highly complex mechanical model. As the high complexity is not feasible here, a linear time invariant system is approximated instead for the controller design. A fractional order $$\text {PD}^{\alpha }$$ controller is applied to meet performance and the high robustness requirements due to the neglected nonlinear dynamics. Simulation and experimental data confirm a superior performance of the FO controller while exhibiting a higher robustness to model mismatches and better disturbance rejection properties.

This paper presents the open software package FloBaRoID (FLOating BAse RObot dynamical IDentification), which aims to provide a package implementing all necessary methods to identify robot dynamics parameters starting from a kinematic model. The package features data acquisition and preprocessing, extraction of identifiable base dynamics parameters, and finding physically consistent dynamics parameters for stable control and simulation. The paper details each of these steps and exemplifies the software usage with experimental results for the 7-DOF robot Kuka LWR 4+.

Usually, industrial robots are controlled by means of separate position loops for each axis, and each position loop is closed by a classical discrete-time PID, with integer-order derivative and integral. It is well known that the application of fractional-order derivatives can improve the dynamic position-tracking performance of a closed-loop mechatronic device. Nevertheless, the discrete-time implementation of fractional-order control algorithms presents several issues, which determine differences with respect to continuous-time simulations. In the present paper, the practical discrete-time implementation of a robotic axis controller combining fractional-order and integer-order derivative terms is discussed, and experimental results are shown.

The exact solution of practical trajectory optimization problems are usually not feasible in real time. The use of approximate solutions deduced form a nominal optimal solution is a potential alternative. This requires evaluation of parametric sensitivities of the nominal solution in real time. This is addressed in this paper. Improvements to existing methods are introduced that not only increase their efficiency but also ensure the constraint satisfaction of the approximation. The proposed methods are applied to a practical trajectory planning problem for a planar manipulator where the initial state is given but the final state is unknown prior to task execution.

Usually, programming robot systems is expensive and complex and only profitable for companies with big lot sizes. Hence these systems do currently not play a big role in small or medium sized enterprises. This work extends the programming paradigm published in [1] that is based on the playback programming method, so that also sensor information can be used to generate more complex robot programs by means of playback programming in addition to the already existing functionality. This is achieved by developing a concept for sensor-based loops and branches that fits well to the programming paradigm. Finally, the enhanced programming system is evaluated in a user study with experts and non-experts with respect to its intuitiveness. The user study is divided into a part that tests the user interface and a part that evaluates the system as a whole, so that possible weak spots of the system or the user interface can be detected and can be taken into account in further work with this programming system.

We study the effects of variable latencies and noise-effects in vision based navigation. Based on the observations, we adapt a new robust estimation solution that is simple to integrate into a path-following controller and shown to provide a smoothed, high-bandwidth feedback for real-time control of a mobile robot. The strong dependency of steering oscillations originating from the noises and inaccuracies of the robot’s pose estimates is highlighted. The system is capable of positioning the mobile manipulator’s gripper in the vicinity of a target only by navigation of its nonholonomic mobile base.

This paper presents a novel low cost strategy for the geometric calibration of industrial robots. The nominal geometric parameters of the robot are therefore enhanced by 30 calibration parameters such as length errors, axes misalignment, gear backlash and joint offsets. Additionally the elastic deflection between motor and arm positions due to finite gear stiffness is included. For the calibration procedure, a laser pointer is mounted on the end-effector of the robot pointing towards a measurement plane. This point is measured by a camera and the beam is controlled to three defined evaluation points by movements of the robot closing a kinematic loop. The procedure is repeated for 83 different optimized poses of the robot. The closed kinematic loops are used to calculate the unknown error parameters utilizing the Levenberg-Marquardt algorithm. For verification purposes, the accuracy of the EE is measured at 100 random positions. A comparison of the proposed method to a laser-tracker calibration is presented. Both methods result in a similar accuracy range.

This paper presents a method to generate joint trajectories for a redundant manipulator. The control system of the manipulator determines the joint references so that the goal pose can be reached without any collisions, in real-time. The control system checks weather any part of the manipulator is at risk of colliding with itself or with any obstacles. If there is a risk of collision, then the collision server computes the exact points where the collision is about to happen and calculates the shortest distance between the colliding objects. The joint trajectories of the manipulator are modified so that collisions will be avoided while at the same time, the trajectory of the end-effector maintains its initial trajectory if possible. Experimental results are given for a 7 DOF redundant manipulator to demonstrate the capability of the collision avoidance control system.

Cobotix is one of the latest paradigm changes emerged in the era of modern industrial robotics at the focal point of the so-called Industry 4.0 concept. Safe physical human-robot contact is the core enabling technology and therefore, it has special importance in research and also in education of robotics engineers. This paper introduces a low-cost, yet complete experimental and educational purpose compliant robot that allows for studying the main principles of physical human-robot interaction. The setup consists of a NOVINT Falcon 3-axis haptic device with parallel mechanism and a 3-axis force sensor, which allows for sensing the interaction forces between the robot and the environment. The controller is implemented on a Real-time Linux platform. The resulted software is open source and shared with the robotics community.

This paper studies push recovery of humans while walking which is an important research topic in biomechanics and robotics bringing systems to their limits. The main goal of the presented research is to determine which passive elements can best support given push recovery motions. The research is based on a multi-body system model of human walking in the sagittal plane, which has active torques in all joints and parameterized spring damper torques in parallel in the lower limb joints. In addition, we use recorded data of human push recovery and the solution of least squares optimal control problems. The results show that a significant reduction of the active torques required for the motion can be achieved including the spring-damper elements. This approach can be applied to component design of humanoid robots.

The development of sensitive manipulators has decisively enlarged the range of robot applications in recent years. This robot class enables direct interaction between humans and robots, even in an industrial environment. High demands from the industry are placed on robot systems in general, such as robustness, precision, high and constant speed, etc., of which collaborative robots are not excepted. In this paper important parameters of sensitive robots are listed and well-known systems from different manufacturers are compared by extensive measurements. Peculiarities are identified, analysed, and presented and valuable suggestions for programmers and applicators are derived.

In this paper a prototype of an automatic machine for the herbaceous grafting of tomatoes is presented. The herbaceous grafting is important in order to reduce the use of pesticides. For this reason, its advantages are both on the environment and on the crops quality. This innovative robotic system, electro-pneumatically controlled, was developed by the Department of Mechanical and Aerospace Engineering (DIMEAS) of the Politecnico di Torino in collaboration with the Department of Agricultural Forest and Food Sciences (DISAFA) of the Università degli Studi di Torino. The aim of this device is to implement a fully automatic machine capable of picking, choosing, cutting and joining the two plants involved in order to obtain the grafted one.

In this paper, a falling motion planning for humanoid robots is proposed as inspired from human falling in order to decrease the impact force significantly. Firstly, motion data of human falling are obtained from a motion track system. Then, two falling motion primitives, namely, knee bending with stretching and hip bending with stretching are identified from human falling tests. A motion strategy is used to combine the falling motion primitives in motion strategy for humanoid falling in order to minimize the impact effects. Simulations and experiments with a humanoid robot have verified the proposed falling motion strategy.

This paper presents first an experimental platform developed for hexapod locomotion analysis and experiments climbing common types of obstacles. The platform contains two elements: a model of a six legged mobile robot used for real tests on different obstacle categories and a software simulator interface which allows study of the robot stability in gravitational field, assuring in the same time the control of the mobile robot during real experiments. For the hexapod robot this paper presents briefly the leg design and some constructive characteristics. The control part is a combination of Matlab, a micro-controller based development board and dedicated servomotor controller board. Secondly, this paper presents some strategies implemented for climbing stairs and experiments achieved on straight stairs and on spiral stairs. The locomotion sequences strategies for climbing stairs were designed in order to overcome them using the following two constraints: maintain the robot’s body parallel with respect to the ground and so, maintain the maximum height of the robot during locomotion.

Robot arms with redundant degrees of freedom are well known to match the flexibility of motion inherent to the human hand. This is typically useful for most industrial robots where one of the design requirements is to execute work operations with more natural kinds of movements. Such work tasks are performed subject to different constraints, while the end- effector tracks a given path in task space. In the general case task constraints can be mapped into admissible ranges of variation of the joint variables. This paper elaborates a method for kinematic path control in sliding mode employing vector space methods. The sliding mode control makes use of the redundant degrees of freedom of a robot arm to satisfy the constraints of the joint variables during the motion of the end- effector along a prescribed path in task space. In fact the proposed method allows the robot arm to reconfigure similarly to a human hand movement, while retaining the assigned end- effector movement. The method is computationally efficient and it is suitable for kinematic control when the number of the degrees of freedom of the robot arm is greater than the dimension of the task space.

Modern service robots are a combination of a mobile platform and a robotic manipulator. One of the main and most difficult tasks in front of these robots is the object transportation between two points. They should be able to detect any desired object. Then move the platform as close as possible to the desired object. Afterwards the manipulator arm should position the gripper near it. The last step is to grasp and transport the object. This paper presents a novel approach for solving the manipulator arm positioning problem. The presented method combines computer vision and Iterative Learning Control techniques in order to compensate any imprecisions of the robot kinematics and dynamics. This results in an efficient solution, which succeeds in precise positioning near the desired object even when there is a very little knowledge of those mathematics models. It is a robust method, which auto adapts to mechanical wear during normal operations, not severe damages or imprecise factory assembly. The method is then validated on a physical robotic manipulator.

Balancing and reacting to strong and unexpected pushes is a critical requirement for humanoid robots. We recently designed a capture point based approach which interfaces with a momentum-based torque controller and we implemented and validated it on the iCub humanoid robot. In this work we implement a Receding Horizon control, also known as Model Predictive Control, to add the possibility to predict the future evolution of the robot, especially the constraints switching given by the hybrid nature of the system. We prove that the proposed MPC extension makes the step-recovery controller more robust and reliable when executing the recovery strategy. Experiments in simulation show the results of the proposed approach.

Recent research suggests intrinsically safe robots, such as through soft limbs or artificial skins, to enable close-quarter human-robot collaboration. Intrinsically safe robots allow for risk-minimized instead of collision-free path planning. Risk-minimized path planning can integrate non-binary knowledge—including obstacle probabilities, robot speed, or data age—into the choice of a robot path. In this contribution, we propose a novel approach to probabilistic obstacle detection on color images that is specifically suited for use in real-time risk-minimized path planning. Our approach enhances an existing neural network for object detection by incorporating spatial coherence via a second neural network and an optimization step inspired by simulated annealing. Finally, a bias towards false-positive obstacle detection allows us to avoid the Sleeping Person Problem for online learning. In our experiments, we show that a GPGPU implementation of our approach can process Full HD images at a soft real-time rate of 15 Hz. We conclude that our probabilistic obstacle detection is fit for use in real-time risk-minimized path planning.

The paper describes a computational solution with cloud implementing, extending robot-vision capabilities of real-time multiple articulated objects recognition for on-the-fly robot grasping. Articulated objects are recognized by matching the unknown object’s skeleton computed from the input image in a cloud virtual machine (VM) with a set of learned skeleton signatures. This High Performance Computing (HPC) process represents a powerful capability for qualitative shape matching because it unambiguously synthesizes and helps estimating the topology of the object and its shape. The skeleton-based matching process is performed as an application-driven robotic service in a private cloud, ten times faster than the robot controller is able to do it and nearly twice faster than two PC-type robot terminals for multiple parts moving on conveyor belts. The parameters of the virtualization process and experimental results which confirm the solution are presented.

The goal of this research work was to create a prototype of a robotic 3D vision system for cranial prostheses automatic inspection. The proposed device is able to automatically scan the entire prosthesis under study by means of an automatic tilt-rotational mechanism and the embedded software allows to generate an automatic inspection report by comparing the specimen with a reference digital sample. The output of the procedure gives information about the geometrical conformity of the new fabricated prosthesis and automatically accept or discard the product. The presented prototype is the first one in literature to have such an automated quality control and rejection procedure in 3D vision and it is also the first robotic vision system applied to cranial prosthesis manufacturing.

This paper presents an approach for depth sensor based detection of target objects and automated generation of trajectories for a robot-aided object scanning process. In general quality management concerning purchased or manufactured products is of central importance. Inspection and verification processes can be very time consuming or require high personnel expenditures which encourages the utilization of automated quality assessment methods. The presented strategy combines the detection of known objects inside the workspace of a robot based on an Iterative Closest Point (ICP) algorithm with trajectory planning and G code generation. For this purpose the workspace around the target object is divided into sub-volumes to identify a set of 3D support points that are the basis of the final scanning path. This method shows promising results in simulation as well as in first tests on a real system.

In this paper a real-time collision avoidance algorithm, based on the method of artificial potentials an intended for collaborative robotics applications was studied. Within this work, the movements of a person are detected and acquired by a vision system and a dummy, developed to interact with a robot in a simulated workspace, replicates these actions. Ellipsoids are then defined to entirely include several parts of the dummy and the end-effector of the robot. The minimum distance between the ellipsoids of the dummy and the one of the end-effector is the input of the collision avoidance algorithm. The results of tests are presented to show the effectiveness of the algorithm. Finally, the influence of the velocity of the obstacle on the capability of the algorithm of ensuring safe collision avoidance is analyzed.

The paper presents the design of a prototype of HeritageBot Platform for demo purposes with features of low-cost construction and user-oriented performance. The design requirements are presented for application in monitoring and working within frames for conservation of goods of Cultural Heritage. The design of the robotic platform is conceived with modular structure with a module for locomotion and a module for small flight. A prototype is presented with its construction peculiarities and preliminary test results.

Transportation tasks within warehouses are nowadays more and more solved using of fleets of autonomous robots. A fleet allows coordinating the different robots in such a way that it balance the load caused by transportation tasks. This allows the robot fleet to be a cost-efficient solution for moderate and changing loads compared to fixed conveyor belts.To allow such a flexible load balancing it is necessary to estimate the time it may take to perform a certain transportation. This is of interest if different transportation tasks can be assigned to an individual robot and the order may have an impact on the time spent to perform a transportation task.In this paper, we will present a method which can learn to estimate the time spent on certain transportation tasks. The method is evaluated per its prediction accuracy on a different set of data which were obtained from the deployment of a robotic fleet in an industrial environment.

Robots often use a topological graph to perform their navigation. To perform this navigation efficiently the traversal time along the edges of the graph needs to be properly estimated.In this paper, we show an approach which estimates the traversal time along edges using only the information of the traversal time from one vertex to any other vertex in the graph. The approach does not need any detailed information which edges were actually traversed. Instead, it is assumed that the robot moves the fastest path in the graph.To address the problem of noise measurements the approach uses a probabilistic model to estimate the traversal time. This paper we show how the probabilistic model can be simplified to allow to solve the estimation problem efficiently.Finally, we show an evaluation of the approach on different sets of generated graphs and traversals showing that the approach estimates the of the traversal times for the shortest path correctly.

The purpose of this paper is to present a 3 DoF underactuated mechanism with one flexible component. It is called FLEGX (FLEXible LEG) and it would be the first step in the design of a jumping humanoid robot with flexible limbs. An early system-level design validation of the FLEGX mechanical configuration was performed using the software MSC.Nastran$$^{\circledR }$$ and MSC.Adams$$^{\circledR }$$-Matlab/Simulink$$^{\circledR }$$ integrated environment.

Mobile autonomous robotics is on its way to industry. Beyond the useful logistics applications where the material for a manufacturing procedure is delivered on demand, the mobile manipulation of objects is still under development. The modern flexible production lines can be supported by autonomous industrial mobile manipulators (AIMM) acting as a cobot for numerous tasks. During the MBZIRC (Mohamed Bin Zayed International Robotics Challenge) an autonomous mobile manipulator was developed showing the potential for industrial applications.

The paper describes the main features of a team of co-operant climber robots designed for autonomous inspections inside ships and offshore platforms. The autonomous units are conceived for monitoring of the status of internal metallic surfaces, taking into account the peculiarities related to the shape of the typical metallic structures characterizing the marine field. The team consists of four units at least, pneumatically powered and controlled by Actuator-Sensor-interface technology (ASi). The number of the team units can be modified as a function of the inspection. Each unit is equipped with video-camera and ultrasonic thickness measurement device; other sensors and transducers can be added and handled, depending on the inspection requirements.

The paper reports a feasibility study of car suspensions equipped with an innovative bushing composed by an elastomer having magneto-rheological characteristics in order to improve the performances in terms of handling and comfort. The analysis focuses on the effects produced by the variations in stiffness due to the magneto-rheological effect on the elastomeric material of the bushing. This magneto- rheological effect is designed to act on the performances of the vehicle particularly during the transients of the dynamic behavior. The performances can be achieved by acting on the characteristics of the suspension bushing on the car body.The desired suspension performances, in terms of camber or steering angle, can be obtained by controlling the characteristics of the magneto-rheological bushings. The diagrams of these characteristics, obtained through an Elasto-kinematic analysis, synthetically show the changes of performances due to the variations in the magnetic field produced by the current applied in the inside coil of the bushing.The feasibility analysis could be particularly considered for industrial application of high performance vehicles.

In this paper, the dynamic stability analysis and control of an articulated heavy vehicle is investigated. First, lateral dynamics of an articulated heavy vehicle is described by a 3-DOF rigid body model. Then, it is considered as a linear system with steering input with delay. System states are the lateral and yaw motion of the tractor unit and the yaw motion of the semitrailer unit. Thus a less conservative stability criterion is developed for this system with time-varying delay. The key idea is constructing a new type of LKF containing a quadruple-integral term. Moreover, novel free-weighting matrices are introduced to increase the degree of freedom in the sufficient stability condition. The proposed delay-dependent stability measure is presented in the terms of linear matrix inequalities (LMIs). Numerical simulations are performed to show the performance of the proposed approach and demonstrate that the suggested scheme is remarkably less conservative compared to the available stability analysis approaches in the literature.

The main idea of the present work is to define the domain in which it is possible to adopt very simple models of vehicle dynamics for applications in the testing of Advanced Driver Assistance Systems (ADAS) in lieu of complex models. The aim is to reduce the computational burden, and consequently the computing time. In particular, in the paper, the performances of a very simple model of vehicle dynamics, the Single Track with linear tires, have been compared with those of a complex and complete model, with non-linear tires, included in a commercial software (IPG CarMaker). For sake of shortness, the comparison has been carried out focusing on the lateral dynamical behaviour, and consequently the testing of a Lane Keeping Assistant (LKA) system has been carried out. Of course both the vehicle dynamic models, and the ADAS system have been integrated in a common simulation environment (Simulink), and tested in the standard traffic scenarios defined in EuroNCAP test protocols.

A relatively new technology for the electric vehicles considers the use of brushless permanent magnet motors directly connected to the car wheels (in-wheel motors or hub motors). In order to evaluate the performance that can be obtained, a complete dynamic model of a four-wheel drive (4WD) electric vehicle equipped with four in-wheel motors is developed and a correspondent parametric simulator is implemented in Matlab/SimulinkTM. The simulator is also employed for designing, testing and comparing various control logics which reproduce the handling behavior of a real vehicle.

Autonomous or self-driving vehicles are becoming a consolidate reality that involves both industrial and academic fields also for its impact in social and governmental communities, well far from automotive engineering. The intent of the present paper is to design an automatic steering control for an autonomous vehicle equipped with steer-by-wire and drive-by-wire technologies. The steering action is calculated to let the vehicle follow a reference path which is stored in a Digital Map properly built to be available in real-time. A Proportional + Derivative (PD) control strategy is deigned based on the Parameter State Approach (PSA) and it is coupled with a Feedforward (FF) term for improving the path tracking control in cornering maneuvers. Some experimental results are shown to demonstrates the efficacy of the controller presented.

The characterization of the radiation pattern of low-frequency antennas is a challenging problem that requires an advanced strategy to reach the expected measurement accuracy. A micro Unmanned Aerial Vehicle (UAV) has been used as a radio frequency test-source to perform such measurements with an innovative technique, obtaining outstanding results. In particular, a Differential Global Navigation Satellite System (DGNSS) based positioning has been introduced through the use of an external GNSS (Global Navigation Satellite System) receiver to obtain the UAV position during the flight with a high level of accuracy. The system has been experimentally tested since a challenging topic is to define all the issues that can occur in the combined use of unmanned aerial systems and electromagnetic components. For this reason, it was investigated to assess sensors problems (such as interferences and positioning troubles with the external sensor) and to define some best practices in the use of such system for this particular field of application to reach the expected result. Finally, an experimental case to demonstrate the suitability of the proposed system is presented in this paper. The radiation pattern of a log-periodic antenna at 250 MHz has been measured with good results.

Agriculture and Remote Sensing (RS) have shared a long common story. Spectral properties of vegetation can be related to many phenol-/physiological parameters of crop. The recent technology advance has made available for users both low cost multispectral sensors and platforms (Remotely Piloted Aerial Systems, RPAS). In Precision Farming the current moment is crucial, since scientists have still not answered all the questions concerning performances of RS+RPAS systems, nor consistency of costs with those required by the low profit agricultural sector. We, firstly, try to lists the main tasks that are expected from RS+RPAS in agriculture (energy balance and thermal remote sensing excluded). Finally a discussion is opened about those critical aspects that, in our opinion, make the current adoption of RS+RPAS still unreliable, or not still proper, in agriculture.

A problem of finding an optimal size of a swarm of robots in a way of effective cooperation is not an easy task to solve. There are many factors, which influence the optimal size of the robotic swarm. Among major factors that have to be considered, belong communication, structure of environment and behavior of agents in the swarm. This paper presents a method for creating a decentralized self-adapting swarm of robots. The task is to set an optimal size of the swarm in a role of space exploration. Communication among robots is restricted to communication through the environment. The only way how agents communicate, is through artificial pheromone marks. This fact gives us an ability to create a decentralized algorithm for controlling and coordination of a robotic swarm, which is robust and efective.

In the design of an unmanned aerial vehicle (UAV), one crucial problem is the flight autonomy of the overall system. The powertrain of electric UAVs, namely: battery, electronic speed controllers, motors and propellers has been analysed to evaluate their effect on the efficiency of the system, and thus the autonomy. Current literature and datasheets on UAV performances, especially on electric motor and propeller, are usually scattered and poor. Two different test benches have been developed to characterize speed controllers, motors and propellers; experimental results have been used to tune motor-propeller numerical models. The proposed work allowed to shade some doubts about the reliability of the available data and validate a model-based design methodology.

Currently available 3D reconstruction and aerial mapping solutions based on the Structure-from-Motion (SfM) and bundle adjustment, require careful equipment selection/configuration (camera, optics and imaging parameters’ settings), accurate flight planning (number of images and their overlap percentage) and often user intervention during post processing. This paper studies the various factors affecting the quality of the final 3D reconstruction, their dependencies and trade-offs trying to push the whole 3D reconstruction and aerial mapping solutions to a more automated scheme with minimum user intervention. We address the design trade-offs between the camera and the flight plan parameters and their effect on the images’ quality and the number and quality of matched features consequently. Then we consider the camera calibration as a fundamental stage in the SfM framework that affects afterwards the depth estimation and point cloud densification steps. An automated photogrammetry calibration solution - based on the Caltech Camera Calibration Matlab Toolbox and Andreas Geiger corner detector - is implemented and tested, allowing for a user-free precise camera calibration process. Finally, a professional photogrammetry application software based on the SfM (PIX4Dmappper) is used to assess the use of additional data sets - like the commonly used GPS data – on the overall resulting 3D reconstruction quality in terms of model richness and accuracy. Two geo-localization data sets (standard GPS used by the UAV autopilot and an L1/L2 GPS corrected with PPK) were used with various stated uncertainty bounds processed with the same images’ data set. The comparison of the resulting 3D reconstructed models highlights the decisive importance of identifying the actual uncertainty bounds - combining the equipment nominal uncertainty with the system introduces factors like the camera capture/GPS synchronization- instead of using relaxed (larger) uncertainty bounds or tighter bounds.

Many human service robotics applications are looking to cloud robotics as a possible paradigm in which robots, seen as remote and autonomous agents, have the possibility to connect to a common network and share information and knowledge they gather from real world on a complex and powerful data infrastructure. Conversely, they (the robots) can also consume data collected by other agents or made available on accessible database and repositories. In this paper, we propose a possible management of different services, exploiting the possibilities offered by cloud robotics in a complex scenario (smart city, smart agriculture, search & rescue). A high-level cloud platform manages several unmanned robots (UAVs and UGVs) with the goal of providing support to the end users that require it via a remote connection (web, SSH, VPN, etc.). The robots are used in the operating area while forwarding real-time data and video streaming to the final users, connected to the same cloud platform, that can manage them by remote.

FREEDOM robot has been developed for exploring dangerous or inaccessible sites by human operators, either from the ground and/or from the air, in urban environment, due to planned or emergency response events. The system, composed by ground and aerial modules, it is based on the design concept of taking advantage of both systems sharing design philosophy and management. One of the main design issue is the possibility of extending the inspection capability by providing power supply from the ground module to the flying module. In this contribution, system basic features and the application to agriculture are proposed.

This research is focused on developing a robotic painting system for artistic and graphic applications by means of an anthropomorphic robot equipped with an airbrush. Firstly, we introduce a mathematical colour spray model, based on a radially symmetric Gaussian distribution of colour intensity within the spray cone. Then, we present an experimental characterization of colour intensity in a spot, by varying the distance between airbrush and target surface and the spraying time. The experimental results of this pilot study validate the paint intensity model and provide the basis for further investigations.

In this paper a procedure to reduce the motor size in articulated mechanism is developed. In particular, the paper presents a method that allows reducing the peak torque requirement of a single degree of freedom mechanism, through the use of an elastic element. Unlike other works in literature, in this work the choice of the trajectory primitive and the resulting inertial effects are taken into account. Both numerical and experimental results show that the inclusion of a single spring can sensibly decrease the peak torque requirements, thus allowing a cost-effective design modification.

In the last years, beginning from the mid of the eighties, many solutions have been designed for new technologies related to additive manufacturing, AM. The machines proposed have been a natural consequence of CNC machines already present in the market. This basically happened because the attention was focused mainly on production technology and not the machine as a whole. The constraints to which CNC machines are usually submitted, such as lathes or milling machines, are usually different from the requirements of AM machines. In order to follow a typical robotic approach this paper proposes the design of an AM machine based on a MIM, metal injection molding, technology by outlining the design process and by showing how it has been adapted to the specific case in order to consider the specific technology and thus proposing an original solution.

General purpose robots are established tools in a variety of industrial applications. An important goal in actual research is to transfer the advantages of these tools into more unstructured environments like househoulds or small and medium sized enterprises. One challenge in this field is to enable non-experts to use all the capabilities of a robot. This includes two aspects: Robots must be intuitive to program and robust to execute. The main contribution of this work is a novel programming approach, that concerns both aspects.Thus our system enables users to guide a robot kinesthetically through a task without prior knowledge. By observing resources in the workspace, the demonstrated task is encoded as a finite state machine (FSM). This FSM allows the reproduction of a task by the robot itself. Furthermore, our approach can integrate a deviation detection method to robustify task reproductions.

This paper is aimed at describing the authors experience on developing a practical robotic course for students of the last year of Master of Science in Mechanical Engineering at Politecnico di Milano. The course is born from the will to satisfy the request of a course where students are able to put into practice all concepts acquired during theoretical lessons. The aim of the course is to design a 3-DoF robot with a given architecture. The subject is intrinsically interdisciplinary, so students, organized in groups, are require to focus their attention on different aspects related to the design of a robotic device, from the modeling of structural components using CAD programs, to the dynamic analyses of the system and to the development of the control software.

The subject of this paper is the mechatronic design of a 4-DOFs hot-wire CNC cutting machine, which is mainly composed by a conventional Cartesian robot with 3-DOFs, plus one that corresponds to the rotation of a fork, in order to orient the hot-wire during the cutting. The proposed cutting machine is actuated through four stepper motors and controlled in such a way to obtain a CAD-CAM system, which is able to generate both planar and skew ruled surfaces, because of the 3-dimensional motion of the hot-wire that is installed on the fork (end-effector). A prototype was built and several experimental tests have allowed the validation of the proposed mechatronic design.

This paper presents the first prototype of a new concept nanogripper whose overall size has been reduced as much as permitted by a new fabrication process based on Nano Technology. The jaws lumen size is adequate to the mechanical manipulation of microorganisms colonies.

In this paper, a 3-finger gripper is developed for grasping fabrics that are folded or unfolded. Its design is based on the investigation of human fingers’ movements in grasping a piece of fabric that is laid on a table and it is conceptualized using the theory of cam-follower design. The concept, the design and the prototype of the gripper are presented, along with the analysis for specifying the characteristics and parameters of the mechanism. The proposed gripper has a camshaft with three cams (2-axial and 1-globoidal), guided by only one actuator, for moving the three fingers (2-linear motions and 1-rotational). A prototype of the gripper is implemented using 3D printing technique which keeps the total cost and weight very low. The prototype has been tested experimentally under several grasping tasks, where its efficiency is demonstrated.

Main factors influencing modern industrial processes are competition in a global market and more and more customized products. Thanks to new technologies, the idea of the “next day delivery” is becoming a standard for all kind of customers. This requires more flexible, automated and efficient processes. In fact, industrial robots play a fundamental role: they are becoming more and more “smarter”, faster, less expensive and “collaborative”. Therefore, it is clear how the link between the robots use and the integration of manipulation devices is extremely tight. In fact, a correct and smart manipulation allows more effective, efficient and sustainable industrial processes.The present paper presents an innovative industrial gripper, called “QuBu Gripper”. It shows all steps that have interested the design process and it is focused on the experimental results obtained. The main function of this gripper is to manipulate products belonging to the family “oil ring seals” within an automated robotic molding cell. The criteria that have guided the entire design process, until the implementation and industrialization of the first prototype, are flexibility, versatility, cost rationalization and performance improvement.

Nowadays, robotic systems are not an exclusive of the industry environment anymore. Indeed, in the last decades, they assumed an important role in other application fields such as medicine (e.g. robot-assisted surgery) and agriculture (e.g. fruit picking). To meet the requirements of these new fields of application, this transition has involved both an adaptation of old-technologies and the development of new ones.In the agricultural field, the objects to be manipulated are usually characterized by non-uniform shapes, dimensions and weights. Moreover, they are typically soft and fragile, so they need to be handled with care.To accomplish these requirements, several types of new grippers have been created or adapted, also thanks to recent developments in manufacturing technologies and materials. However, many of these innovative grippers are either at a prototyping stage or expensive.This applied-research work wants to investigate the innovative available low-cost grippers able to deal with objects of different dimensions, weights and surface conformation, which could be employed in productive sectors such as agriculture. In particular, the few information (i.e. maximum object diameter and weight, durability) given by producers are supplemented with data on gripper capability to handle statically and dynamically objects with the aforementioned characteristics. Such data have been obtained through several experimental tests (i.e. pick & place operations) and provide a set of gripper-characteristic-maps that allows evaluating the gripper adequacy for a given application.

In this paper the shape memory alloys are presented and their special characteristics as the possibility of shape recovery are illustrated. Then some possibilities about the control are analyzed and some applications developed at the University of L’Aquila, Italy, are presented and discussed.

This paper describes the implementation of a robust dynamic hand gesture recognizer using a depth sensor. The recognizer uses only depth image information, and the hand position provided by a hand tracker library, in order to construct its feature vectors. The recognizer builds two types of feature vectors to increase accuracy; the frame feature vectors that describe a static hand, and the sequence feature vectors that describe a contiguous segment of frames. The recognizer also uses two statistical classifiers. The frame feature vectors are utilized by the frame classifier. The results of the classifier, then become part of the sequence feature vector, which in turn are utilized by the sequence classifier. The results show that the accuracy of the recognizer increases more than twice, when using both classifiers. The recognizer also does not make any assumption for when a gesture begins or when it ends. Instead it learns to differentiate between noise, and a real gesture. A humanoid robot, ROBIN, is used for validation of the approach for human-robot interaction.

This paper presents an analysis of humanoid robot ability to perform human upper body language regarding several typical human emotional states. A young actor has been involved as a model of affective states. Experiments on real robot have been carried out in order to find its dynamics. A criterion that allows us to detect robot possibility of performing emotions by movements is defined. Analysis of how to compensate robot shortcomings such as limited degrees of freedom and joint’s activation in performing emotions is given.

Recent research in robotics aims at combining the abilities of humans and robots through human-robot collaboration. Robots must overcome additional challenges to handle dynamic environments within shared workspaces. They especially must perceive objects and the working progress to synchronize with humans in shared tasks. Due to unpredictable human interaction, local information about objects detected by eye-in-hand cameras and stored within a world model falls in value as soon as respective objects get out of sight. Our contribution is an approach to making world models aware of human influences and thus allowing robots to decide, whether information is still valid. To this end, we annotate pieces of information with certainty values encoding how trustworthy they are. Certainty is adapted over time according to additional knowledge about human presence within the workspace, provided by a global sensor. Thus, we achieve human-awareness through fusion of local and global sensor data. Our concept is validated through a prototype implementation and experiments that regard certainty of objects in different scenarios of human presence.

Socially interactive robots need the same behaviors and capabilities of people to interact naturally with humans. One of these capabilities is the perception of nonverbal cues. The paper presents a biologically inspired perception system for a social robot. This system is based on the psychological theory of perceptual cycle. It is composed of two main parts: schema and exploration. The schema represents the bottom-up information processing, whereas the exploration represents the top-down information processing. The system has been implemented and evaluated on a humanoid robot. The experiments have shown promising results. Several interaction sessions were conducted with the robot. The robot was able to perceive the nonverbal cues of the interaction partner and behave accordingly.

The paper presents new mechanical design and control of a hyper-redundant, under actuated, soft robot arm with 20 DOFs with gripper. The robot mechanism is powered by the 9 servo-motors and the power transmission from the actuators to robot links and further to the end-effector is realized intra structurally by strings. Controllability and dexterity of the soft robotic arm is verified by model simulation before implementing control algorithms to the robot controller. For the purpose of simulation the algorithms of the inverse kinematics are realized. Mechanical prototype in its’ early phase of integration is shown in this paper, too. Control performances of the hyper redundant soft robot arm are evaluated by a simulation example.

In this paper, we investigate a design optimization of a planar cable robot included in a gait training machine called the Cable Driven Legs Trainer. Generally, design of cable-driven parallel manipulator is constrained by generating non-negative tension in all the cables in order to guarantee the feasibility of a robot’s pose, and moreover, cable interference problem should be avoided. The cable robot of the gait trainer is employed to move the lower limb with the purpose to mimic the normal walking motion. Therefore, the structure of this robot is optimized using a discrete approach such that the gait kinematics is carefully produced.

This paper presents a novel kinematic of haptic device with 4 d.o.f for medical use. The kinematic of the proposed device is based on delta robot architecture. The challenging task was to obtain the three orientations of the tool for each Cartesian location of the mobile platform. Gimbal joints were specifically used in this purpose. The kinematic and haptic models of the new device are presented. A 3d printed prototype was developed to validate the new kinematic and present a the first conceptual design.

In this paper, a novel hybrid hexapod legged robot R3HC is proposed as based on a collaboration between Universities of Huelva and University of Cassino. This robot has been designed and built for patrolling areas where flat surfaces are mixed with obstacles and/or small unevenness. Its main design features, which are based on previous experiences with the Cassino Hexapod robot series, are also discussed. Both hardware and software are aiming at low-cost and user-friendly features as also described in the preliminary tests.

The paper deals with a new design for a low-cost lower limb exoskeleton, which allows a simple construction by using lightweight materials, an easier wearing and an adaptability to various human legs dimensions. The designed exoskeleton allows 3 DoFs for hip, knee and ankle joints. The actuation of the exoskeleton joints is achieved by one rotational servomotor and 2 electric linear motors. The CAD model was simulated by considering the exoskeleton with or without the human leg weight applied on a scaled CAD model. A prototype of the developed exoskeleton was manufactured and tested at LARM lab at the Cassino University.

This project is aimed to provide a textile pneumatic actuator capable of generating a 180° rotary motion. The prototype is intended for the rehabilitation of the elbow joint. It boasts: the use of biocompatible materials, low weight, small dimensions, low hysteresis and uniform motion. The actuator shows good extension characteristics and a good stability. In the article, are presented: the project, the prototype, and the results of measurements, which show the actuator torque.

Robotic systems to restore, augment and support human capabilities hinder the natural interaction with the world. Different approaches based on physiological measurements such as brain activity, muscle contraction, kinematics, or eye movement, can be exploited to automatically and reliably detect the intention of the user to perform a movement. Once the intention of the user is detected or classified, it can trigger or control an exoskeleton supporting the target gesture. All these features together provide a personalized communication between the robot and the user making human-robot interaction natural and seamless. Thus, the acceptability and usability of the system is maximized. Several integrated robotic actuators driven by user’s intention are here described to demonstrate the potentiality of these technologies both for rehabilitation and assistance purposes.

The paper considers a two-legged model of lower limb exoskeleton with hybrid pneumatically assisted electric drive. The main interest of the work is dynamics and motion laws of the control system of human stepping patterns. The exoskeleton is designed to help patients with lost the mobility of the lower limbs, in particular athletes and astronauts at different stages of rehabilitation.

According to recent epidemiological surveys 2.5 millions of people affected by a Spinal Cord Injury (SCI) are experiencing a variable degree of disability.

The hearing-impaired person has high risks of accidents by the car or the motorcycle out of the door. In this study, the wearable hearing support system which can identify the direction of sound source by using the interaural time differences (ITDs) of sound pressure and inform the sound source direction to the hearing-impaired person via the vibration of vibrators on the shoulders is developed. This system could inform the direction of the front, side, even rear sound source to the hearing-impaired person dynamically.

Among the growing number of different exoskeletons, passive and quasi-passive solutions hold the upper-hand compared to powered solutions in price, accessibility, complexity, weight and user acceptance. This paper evaluates a modification of an originally passive ankle exoskeleton with an active clutch, making it quasi-passive. We developed an electric quasi-passive clutch to improve the performance of the original exoskeleton design, with the aim of mitigating the problems on the clutch engagement timing and user physiological variability. In order to evaluate the exoskeleton and the clutch operation, we performed a study where 7 users wore the exoskeleton and performed trial walks. Qualitative user feedback that focused on the device comfort, users perception of exoskeletons effect and smoothness of the clutch operation was collected, along with quantitative data on clutch operation during walking on a flat surface. Results show improved and more reliable exoskeleton clutch operation which was also expressed in qualitative user feedback.

In the past decade many studies on human motor control have investigated how humans are moving their arms. In robotics, these studies were usually used as a foundation for human-robot cooperation tasks. Nonetheless, the gap between human motor control and robot control remains challenging. In this paper we investigated, how human proprioceptive abilities could enhance performance of cooperative manipulative tasks, where humans and robots are autonomous agents coupled through physical interaction. In such setups, the robot movements are usually accurate but without the proprioceptive capabilities observed in humans. On the contrary, humans have well developed proprioceptive capabilities, but their movement accuracy is highly dependent on the speed of movement. In this paper we proposed an approach where we exploited the speed-accuracy trade-off model of a human together with the robotic partner. In this way the performance can be improved in a human-robot cooperative setup. The performance was analyzed on a task where a long object, i.e. a pipe, needs to be manipulated into a groove with different tolerances. We tested the accuracy and efficiency of performing the task. The results show that the proposed approach can successfully estimate human behavior and successfully perform the task.

Off-the-shelf electronic market is large, diverse and easily accessible by many. Credit card size computers (example: Raspberry Pi) or micro-controller boards (example: Arduino) can be used for learning how to code and how to control embedded systems. Nevertheless, there is a lack of off-the-shelf, open source devices that would enable us to learn about and make use of human signal processing. An example of such a device is an electromyograph (EMG). In this paper we investigated, if an EMG device could fulfill the aforementioned gap. EMG device we used for conducting our experiment was a five channel open source EMG Arduino shield. The performance of the device was evaluated on three healthy male subjects. They were instructed to perform basic finger movements which we classified and executed on the robotic hand. The EMG signal classification was performed using a Support Vector Machine (SVM) algorithm. In our experimental setup the average EMG signal classification accuracy was 78.29%. This we believe demonstrates there are EMG devices on the market today that provide access to cost effective prototyping and learning about EMG signals.

An important open challenge in robotic assistive exoskeletons is how to control them to maximize their physical benefits on users. We addressed this challenge on a back support exoskeleton in a preliminary user study, which evaluated three possible assistive strategies on a lifting task. One strategy modulated the assistance on the posture of the torso. The two additional direct strategies assisted proportionally to forearm electromyography and grip pressure on the fingertip, respectively.The experiments highlighted that the direct strategies modulate assistance more appropriately than the posture-based one. Additionally, the associated acquisition devices were not considered obtrusive for the lifting task. The insights from this exploratory study will guide further development of the control interface to operate the robotic back support.

This paper presents the design process of an exoskeleton for executing human fingers’ extension movement for the rehabilitation procedures and as an active orthosis purposes. The Fingers Extending eXoskeleton (FEX) is a serial, under-actuated mechanism capable of executing fingers’ extension. The proposed solution is easily adaptable to any finger length or position of the joints. FEX is based on the state-of-art FingerSpine serial system. Straightening force is transmitted from a DC motor to the exoskeleton structures with use of pulled tendons. In trial tests the device showed good usability and functionality. The final prototype is a result of almost half a year of the development process described in this paper.

This paper describes a model of a prosthesis of ankle able to control the pitch and the release of energy recovered during the gait; moreover, the model is able to provide to the user more power than the one recovered during the gait, in order to ensure a natural gait. The development of the model has been articulated in 3 steps: the study of the gait of able body people; the design of a mechanical system able to adjust the pitch of the ankle and at the same time able to store the mechanical energy; the development of an active control of the foot and of the energy recovery system. The conceived model is made of a four-bar linkage with a further fifth element. Two of those elements can modify the length: one is an active shock absorber for damping and for the energy recovery; the other is a linear actuator for the pitch adjusting. The first is equipped with two control flow valves; the second is made of a screw nut mechanism. The control system is based on a central control unit that detects signals from three sensors to determine the gait phase, to control the shock absorber and to adapt the actuator to the correct position.

The development of an active orthosis for inferior limb is presented. A procedure for the customization of the device is proposed. It is based on the design of few standardized parts and on two alternative ways to measure the relevant features of the user: the first making use of a instrumented template; the second making use of a 3D scanner. Then the complete manufacturing process for production of the structural carbon fiber components is presented.

This paper presents the structure and the main innovations of P.I.G.R.O. (Pneumatic Interactive Gait Rehabilitation Orthosis). It is an active exoskeleton electro-pneumatically controlled with 6 DoF (Degree of Freedom) in the sagittal plane. Robotic neurorehabilitation trainings are its main field of application. Some preliminary tests are carrying on with brain stroke and ictus patients.

This article shows some innovative artificial and textile pneumatic muscles designed at the Department of Mechanical and Aerospace Engineering (DIMEAS) of the Politecnico di Torino. The study was carried on using a specific test bench suitable to work with these textile pneumatic muscles. Various geometries were tested giving interesting information about their behavior analyzed with different materials and constructions. These prototypes are useful to be integrated into active clothing for robotic rehabilitation of upper limbs.

In this paper, a mechanical design and control method of a leg-exoskeleton are proposed for a new wheelchair for persons with problems in lower limbs. For the mechanical design, a pedal-cycling actuation method is proposed with crank-rocker mechanism driven by one motor with a simple mechanical structure to guarantee the user’s safety. Regarding the master-slave control, the user’s motion intention is detected by force sensors on pedals and it is used as the control message for the leg-exoskeleton motion and wheelchair motion. Experiments are discussed to show the characteristic force during the pedalling action. The experiment results give a good inspiration to optimize the control method.

Optimal rehabilitation results can be achieved only through an intense and constant physical therapy. The adoption of cyber-physical technologies in the field of healthcare can guarantee benefits in terms of patient care and health results. Compared to manual therapy, cyber-physical technologies have additional risks that should be reduced and controlled. The introduction of these technologies will also pose a legislative and ethical challenge. According to Biomedical Ethics, robots should act in the best interests of the humans and not to harm them and robotic research activities should look at the “Precautionary principle”. Another important ethical aspect is the respect for one’s autonomy. Technological acceptance is also to be considered: the point is if patients want to be taken care of by robots and if they agree to be diagnosed by robots. We must also think about the problems of an equal access to the new technologies, discrimination and stigmatization. The ethical question is: where does a medical professional have to stop with the artificial enhancement of humans?

Stroke patients are often affected by hemiparesis. In the rehabilitation of these patients the function of the hand is often neglected. Thus in this work we propose a robotic approach to the rehabilitation of the hand of a stroke patient in hospital and also at home. Some experimental results can be presented here especially for inpatients. Further experimental results on home-patients must be acquired through a telemedicine platform, designed for this application.

This work is devoted to monitor and trace a preliminary representation of the set of devices designed to take part to resuscitation process for patients subjected to a cardiac arrest. The resuscitation procedure is well defined and reviewed by the scientific community and different devices are present in literature and are adopted on patients. A standard definition of the characteristics for these device is not present and their efficacy is controversial in literature.

Robotic devices for rehabilitation purposes have been increasingly studied in the past two decades and are becoming more and more diffused, due to their effective support to the traditional therapy. They allow to automate in a repeatable manner the rehabilitative exercises and to quantify outcomes, giving important feedback to the therapist. This paper deals with the design, development and preliminary characterization of a robotic system, with an exoskeleton device, for assisted upper-limb rehabilitation, in which surface EMG measurements are used to implement a force-based active and resistive control. A prototype of the system has been realized, measurements of important parameters of the motion permitted to optimize the design and preliminary tests on the control strategy were carried out.

The assisted limb rehabilitation process is commonly associated with advanced control of the affected limb through robotic assistance and human interference. The robotic element is only expected to be able to reproduce the motion suitable for large variations of patient’s condition within a reasonable accuracy and stiffness. Therapist’s intervention of fine-control is in the format of planar elements (like pelvis linkage) or joints (like knee), which relate to trajectory or orientation adjustments. The rehabilitation process has to consider the patient’s ability, limit and motion constraint that form those two factors. The parameters for controlling these is associated with kinematic, that defines the behaviour and characteristic of the lower limb. The developed 3D Python simulation system allows for this fine-tuning in the form of slice analysis and interval analysis. The results show that Bezier could be successfully used in various development aspects of parallel robots. The Hybrid and Hexapod robot configurations in this study can then be linked to a Haptic controller that runs on Python’s Haptic engine.

Robotic technologies are progressively gaining considerable importance in motor rehabilitation. In this context, the development of non-invasive man-machine interfaces has a significant role. Among other physiological signals, surface EMG is of paramount importance. However, the detection of surface EMG signals in rehabilitation is currently based almost exclusively on a single or a few electrode pairs. In the last decade, some important limitations of this approach emerged. This work reviews the more recent knowledge about issues in the use of simple bipolar system when the objective is to estimate muscle activation and highlight the benefits provided by multi-channel surface EMG to muscle activity estimation.

The paper presents the need for upper limb rehabilitation and gives one possible solution with rehabilitation robotics. The research objectives, medical and technical requirements analysis for an upper limb rehabilitation robotic system are presented. Designing a usable upper limb rehabilitation robotic system must address the needs of its users.

In this work the energy cost of a human operator during the harvesting of saffron flowers is discussed. The amounts of energy cost was achieved for a traditional harvesting hand-made activity compared to a harvesting phase with a facilitator machine. In the paper the facilitator machine prototype is described with its mechanical performances. Oxygen uptake (VO2) and carbon dioxide consumption (VCO2) were taken in five male healthy volunteers using a progressive exercise test on a cycle ergometer to calculate the respiratory ratio (VCO2/VO2) corresponding to the anaerobic threshold (AT) which, just it exceed 1, indicates the fatigue onset. VO2 was also assessed when volunteers simulated hand-made saffron harvesting or assisted by the facilitator device, to establish if the VO2 reached or not the AT value/fatigue onset in the one or in the other harvesting modalities.

This paper presents a method for detecting the mechanical stiffness of micro-metric biological tissues by means of compliance tests performed with a MEMS-Technology based microgripper. Thanks to an actuating rotary comb drive working in cooperation with another sensing rotary comb drive, the system is able to recognize the tissue sample stiffness. Such characterization is possible thanks to a proper control system that is applied to the whole mechanical structure.

In the paper the robotic systems for restorative medicine including massage are considered. The features of robot training, the features of unusual environment – patient’s soft tissues and the features of non-invasive interaction of robot with soft tissues are emphasized. Nowadays there are specialized and research robots for massage only in the world but universal robotic systems for massage are absent. So it is necessary to have the devices which can adapt the serially produced non-medical robots for performing massage. These devices have to perform the functions of mechanics, electronics, computing and to interact with biological environment – patient’s soft tissues. So they can be named bio-mechatronic modules. The following bio-mechatronic modules are considered: the handle for manual spatial continuous robot training at soft tissues deforming, active force module for compensations of displacements of a patient at his breathing, module for program training of force points and spring compensator of force overload at unexpected obstacle. These bio-mechatronic modules directly interact with soft tissue by the tool imitating masseur’s hand and they take part in the following: admittance control for the robot training by demonstration; position-force control at reproduction; bio-diagnostics and bio-technical control of patient’s state; amortization of quick approach to untrained areas and as tool carrier.

The clinical study of postural control requires a disturbance to be imposed to the subject under evaluation. Among various kinds of disturbance, a mechanical stimulation, consisting in an impulsive force impressed to a certain point of the body, can be used. This paper describes the study of a device conceived to generate such a disturbance. The device is based on a commercial pneumatic actuator, equipped with appropriate force and motion feedback sensors, and properly controlled. The major item is to take into account the interaction between the device and the human, in order to individuate the optimal control technique to generate the desired force pattern. A mathematical model of the device and the human-machine interaction is presented and a sliding mode control technique is proposed. Finally, the results of simulations are reported and discussed.

A few years ago, we presented a new parallel robot kinematics (called “Dionis”) suitable for positioning an endoscopic tool above a patient, with a virtual center of rotation at the insertion point. A first prototype has been realized. This enables us to address a few specific mechanical design issues with the purpose of significantly increasing the stiffness of this design. FEM simulations presented here show promising results.

The posterior bundle of the medial collateral ligament of the elbow prevents the posterior dislocation of the elbow itself. In this paper a method, based on motion capture, to evaluate the biomechanical role of the bundles of the medial collateral ligament is presented. Anatomical sections of ligaments, and valgus and varus stresses were implemented on a cadaveric elbow. Clusters of markers are fixed on operational table, humerus and ulna. The motion of the ulna with respect to the humerus during manoeuvres is recorded and processed. Results of the measuring methodology allow to evaluate elbow stability. Three conditions are compared: elbow with intact MCL, anterior bundle section, anterior plus posterior bundle sections. A progressive increased dislocation of the elbow is quantified. The method is recognised as a valid instrument to asses bundle function in pre and post ligaments reconstruction surgery.

In cross-country sit-skiing all athletes compete in a sitting position, but some of them have the ability to control their trunk more than others. The trunk plays an essential role in two performance determinants: propulsion generation and balance maintenance. The aim of the study is to design a new testing device assessing athletes’ responses to these propulsion determinants. The new device was composed of a seat surrounded by a sensorized aluminum frame. To assess propulsion generation, two force sensors were mounted in the anterior and posterior side of the frame, while two force sensors were embedded in two ropes elongated from the top of the frame. To measure trunk control, the device was mounted on an electrically-driven sledge, which can be moved in anterior and posterior direction giving unpredictable balance perturbations to the athlete. In the paper a pilot test is presented. One athlete with low trunk control level was tested. The main results were: (1) no drawbacks related to discomfort or pain, (2) lower level of generated force without back support compared to the condition with back support, and (3) higher trunk range of motion and trunk angular velocity in response to unpredictable balance perturbations when the acceleration of the perturbations increased. These preliminary results suggest that the new device is suitable for testing sitting athletes. Future tests with a higher number of athletes with different levels of disability can give valuable new information, which can be used for evidence-based classification.

A deeper investigation of foot anatomy and a more accurate biomechanical model can be objects of interest to investigate daily activities, as to optimize orthopedic tools and bipedal robots. Several studies address the development of multi-segments kinematic foot models, but less to the dynamic analysis, because of instruments limitations. The aim of this work is the development of a two-segments foot model for biomechanical analysis. The model has been validated considering gait cycle of a healthy volunteer. The application of two adjacent force plates allowed the simultaneous recording of the ground reaction forces separately for the fore and hindfoot, as the pointing out of their involvement during stance. Ankle power generation at toe off in traditional model presents an overestimation around 50% of the total value in the two-segments model.

In this paper, we verify three different 6 degrees of freedom Kuka Agilus robots for application in neurosurgery. Application specific reachability maps are generated for robots with 707 mm (R700), 901 mm (R900), and 1101 mm (R1100) horizontal reach. The reachability of each robot reflects a working volume of a standard stereotactic frame which utilizes the center of arc principle. A working volume with 100% reachability yield has been identified for the R900 and R1100 robots when the robot is positioned sideways to the patient. The R700 robot doesn’t have a 100% reachability yield work volume. Robot configurations within the reachability map are further optimized given two dexterity performance indices: the condition number and a new fuzzy joint limit avoidance function. In the experiments, we have further evaluated the impact on robot work volume given robot orientation with respect to the patient. After reorienting the robot a significant increase in work volume with 100% reachability yield was obtained for all three robots.

In vitro experimental tests are of great importance in human joint biomechanics. These tests allow the study of the kinetostatic and dynamic behaviour of human joints: a limb specimen is connected to a rig, controlled loads are applied, and the joint displacements are measured. Fixation of the bone to the rig is a typical problem that raises in these applications. A new fixation device is presented in this paper that features several interesting and innovative characteristics: it consists of a passive parallel mechanism that allows adjustment of the pose of the bone with respect to the rig with six degrees of freedom; if needed, the bone can be removed from the rig and repositioned exactly at the same pose during the experimental tests; all six degrees of freedom of the fixation device can be locked by acting only on two screws, thus simplifying the fixation operation; the parallel structure of the mechanism guarantees a high fixation stiffness. The complete design is presented and discussed, together with the results from the workspace and stiffness analyses of the mechanism, that are particularly important for the considered applications.

This paper presents a performance evaluation of an automaton mechanism in a Blossoming Flower Clock that is preserved in Beijing palace museum. Even though the mechanical structure inside this clock is uncertain, the feasible mechanisms for these automata are synthesized through a systematically design procedure in a previous study. In this paper, the historical background of this clock is introduced first, and then the dimension design, kinematic and dynamic analysis are proposed to provide the numeral results for a performance characterisation. The proposed procedure can also be applied to other automata for reconstructing and analysing the mechanical performance.

This paper outlines the main design issues for an upper limb rehabilitation device. In particular, human motions have been measured and analyzed in order to identify a safe workspace required for a rehabilitation device. A preliminary design solution is proposed based on a cable-driven parallel architecture, which can provide the required operation workspace and significantly improve the safety of the rehabilitation procedure as compared with exoskeletons or traditional robotic devices.

An investigation on the influence of the design parameters of the tendon system on the behavior of an under-actuated finger is presented. The study was carried on by simulations with Working Model 2D™ and by an experimental apparatus. The results showed a good agreement between simulation results and experiments. A correct choice of the design of the tendon system parameters permits a correct working of an under-actuated mechanical finger driven by un-extendable tendons; hence this study can help in correctly designing the finger of under-actuated grasping devices. This both for industrial and agricultural grasping devices and for human hand prosthesis.