Informatics in Control, Automation and Robotics

By unwinding the assumptions that underlie the standard complete linkage method, the size of a hierarchical sequence reverts back from n levels to \(\frac{n \cdot (n - 1)}{2} + 1\) levels, and the time complexity to construct cluster sets becomes \(O(n^{4})\). To resolve this problem, distance graphs are used to find meaningful levels of an \(\frac{n \cdot (n-1)}{2} + 1\)-level hierarchical sequence prior to performing a cluster analysis. By doing so, it is possible to construct only the cluster sets for meaningful levels and reduce the time complexity from \(O(n^{4})\) to \(O(ln^2)\). Increasing the dimensionality of the data points helps reveal inherent structure in noisy data, which is necessary for finding meaningful levels. The means is theoretically validated. Empirical results from three experiments show that the means does not impose structure on a data set, that it is easy to use, and that it can identify cluster sets that have real world meaning.

Considerable research has been performed in applying reconfiguration scenarios to real-time systems at run-time. In fact, a reconfiguration scenario is a software operation that allows the addition, removal and update of real-time OS tasks which can share resources and are generally obliged to meet corresponding deadlines according to user requirements. Although, applying such scenarios has several advantageous consequences behind, it can have a severe impact on the real-time aspect within the system. The proposed solution is a protocol called Reconfigurable Priority Ceiling Protocol (denoted by RPCP). This protocol avoids deadlocks after any reconfiguration scenario and changes the priorities of tasks in order to reduce their response and blocking times to meet their deadlines. This protocol requires the use of two virtual processors in order to guarantee the non-interruption of execution during any reconfiguration step. A tool is developed to encode this protocol and is applied to a case study.

This paper describes two methods to determine the homogeneous transformation of a projector with respect to the robot hand. Since the projector itself has no exteroceptive capabilities, a camera is rigidly attached to the robot base or placed in the environment to detect the projected pattern. The camera’s extrinsic calibration parameters can be simultaneously solved, which is shown by the second method. Self-calibration implies that any kind of calibration tool may be omitted. For calibration, the robot hand has to make at least two movements around nonparallel rotational axes. At each robot configuration, correspondences between the camera and the projector are established to recover the transformation between them, up to an unknown scale factor. The system is described by the common known formulations \(\mathbf{AX }=\mathbf{XB }\) and \(\mathbf{AX }=\mathbf{ZB }\). Both can be arranged in a linear form with respect to the unknown extrinsic parameters and scale factors, and solved in least square sense. Further optimization allows to refine all intrinsic and extrinsic parameters.

The chapter presents a combinatorial optimization method for the low-power adaptive scheduling problem on variable speed processors and reconfigurable architectures. It deals with synchronous and flexible real-time OS tasks. A reconfiguration scenario is assumed to be a run-time software intervention which act on the system state to allow the addition-removal-update of OS tasks and consequently adapt the system to its environment under functional and extra-functional requirements. A reconfiguration can change the system behavior where temporal properties are violated or the energy consumption overcomes its limit and pushes the system to a non feasible state. A configuration scenario can be a result of the addition-removal-update of tasks in the system. The difficulty is how to recover the violated temporal parameters of the system’s tasks after any reconfiguration. We use a DVS processor which is with a variable speed to support run-time solutions to re-obtain the system’s feasibility. The challenge is to choose the suitable scaling factors for the processor speed to ensure the best compromise between the execution time and the energy consumption where all constraints are respected. We reformulate the problem and propose a combinatorial optimization method based on integer programming and heuristics to solve the problem. We compensate each approach by a mechanism which is based on the deadline adjustment of the tasks to satisfy the feasibility conditions when the available speeds of the processor do not filfull the needs. This mechanism make the system more reliable and flexible to respond appropriately to its environment.

To control robotic arms mounted on a flexible endoscope in Natural Orifice Transluminal Endoscopic Surgery (NOTES) procedure, Cable-Conduit Mechanisms (CCMs) are often used. Although the CCMs offer simplicity, safety, and easy transmission, nonlinear friction and backlash-like hysteresis between the cable and the conduit introduce some difficulties in the motion control of the NOTES system. It is challenging to achieve the precise position of robotic arms and force feedback information when the slave manipulator is inside the humans body. This paper presents the dynamic transmission characteristics of CCMs and control strategies to compensate for achieving precise position tracking of the robotic arms. The cable-conduit tension and position transmission are analysed and discussed for both sliding and presliding regimes. Unlike current approaches in the literature, position transmission of the CCM is modelled by an approximation of backlash-like hysteresis profile for both loading and unloading phases. In addition, nonlinear adaptive control algorithm is also used to enhance the tracking performance for a pair of CCMs regardless of the change of cable-conduit configuration during the operation. The backlash-like hysteresis parameters are online estimated under an assumption of presence of output feedback and unknown bound of nonlinear parameters. To validate the proposed approach, a prototype of single-DOF-flexible robotic system, which consists of a motion control device, a telesurgical workstation, and a slave manipulator, is also developed. The proposed compensation scheme is experimentally validated using the designed system. The results show that the proposed control scheme improves the tracking performances significantly regardless of the change of endoscope configuration.

The paper presents the investigation of collectives of term weighting methods for natural language call routing. The database consists of user utterances recorded in English language from caller interactions with commercial automated agents. Utterances from this database are labelled by experts and divided into 20 classes. Seven different unsupervised and supervised term weighting methods were tested and compared with each other for classification with k-NN. Also a novel feature extraction method based on terms belonging to classes was applied. After that different combinations of term weighting methods were formed as collectives and used for meta-classification with rule induction. The numerical experiments have shown that the combination of two best term weighting methods (Term Relevance Ratio and Confident Weights) increases classification effectiveness in comparison with the best individual term weighting method significantly.

Research over the past decade has demonstrated the capability of the electrocardiographic (ECG) signal to be used as a biometric trait, through which the identity of an individual can be recognized. Given its universality, intrinsic aliveness detection, continuous availability, and inherent hidden nature, the ECG is an interesting biometric modality enabling the development of novel applications, where non-intrusive and continuous authentication are critical factors. Examples include personal computers, the gaming industry, and the auto industry, especially for car sharing programs and fleet management solutions. Nonetheless, from a theoretical point of view, there are still some challenges to overcome in bringing ECG biometrics to mass markets. In particular, the issues of uniqueness (related to inter-subject variability) and permanence (related to intra-subject variability) are still largely unanswered. This work focuses on the uniqueness issue, evaluating the performance of our ECG biometric system over a database encompassing 618 subjects. Additionally, we performed tests with subsets of this population. The results cement the ECG as a viable trait to be used for identity recognition, having obtained and Equal Error Rate of \(9.01\,\%\) and an Error of Identification of \(15.64\,\%\) for the entire test population.

Wheeled robots on rough terrain are needed to effectively change wheel control strategies since optimal slip and maximum traction levels differ depending on soil types such as sandy soil, grassy soil or firm soil. In a view point of wheel control, this paper focuses on a prediction method of optimal control parameters such as optimal slip ratio and traction coefficient acting on wheels to maximize traction or minimize energy consumption. In this paper, optimal control parameter (OCP) models based on surface reaction index (SRI) are experimentally derived using characteristic data from wheel-soil interaction through indoor experiments by a testbed for analysis of wheel-soil interaction on three types of soil; grass, gravel and sand. For estimating surface reaction index (SRI), actual traction coefficient, including information of motion resistance, is observed by a state estimator which is constructed from longitudinal wheeled robot dynamics. The actual traction coefficient and slip ratio on wheels are employed to estimate surface reaction index (SRI) by a numerical method on the basis of derived optimal models. The proposed algorithm is verified through outdoor driving experiments of a wheeled robot on various types of soil.

The problem of two-player pursuit-evasion games with unmanned aerial vehicles (UAVs) in a three-dimensional environment is tackled. A game-theoretical framework is presented, enabling the solution of dynamic games in discrete time. Depending on the cardinality of the action sets, the time complexity of solving such games could rise tremendously. Therefore, a tradeoff between available actions and computational time of the solution has to be found. It was shown that the chosen action space allows manoeuvres with sufficient accuracy, assuring the convergence of the games, while the computational time of the algorithm satisfies the real-time specifications. The UAVs taking part in the pursuit-evasion games are two identical quad-rotor systems with the same dynamical constraints, while the evaders’ absolute velocity is smaller than the pursuers’. The approach was simulated on an embedded computer and successfully tested for real-time applicability. Hence, the implementation and real-time execution on a physical UAV system is feasible.

The Climbing Robot CREA is developed to climb up flat concrete walls. Due to its big size and weight the robot uses the suction system to generate necessary adhesive force. This suction system consists of eleven chambers which are thermodynamically connected to one common reservoir. The robot also uses the wheel-based locomotion which introduces chalenging control dilema when integrating with suction system. This paper addresses these difficulties by introducing new control scheme that is able to reach a satisfactory trade-off between contradictory criteria. An exponentially stable controller is developed for each chamber that engages automatically with wall and generates desired adhesive force with lowest possible friction and influence on other chambers.

To save time and money while designing new products, industry needs tools to design, test and validate the product using virtual prototypes. These virtual prototypes must enable to test the product at all Product Life-cycle Management (PLM) stages. Many operations in PLM involve human manipulation of product components in cluttered environment (product assembly, disassembly or maintenance). Virtual Reality (VR) enables real operators to perform these tests with virtual prototypes. This work introduces a novel path planning architecture allowing collaboration between a VR user and an automatic path planning system. It is based on an original environment model including semantic, topological and geometric information, and an automatic path planning process split in two phases: coarse (semantic and topological information) and fine (semantic and geometric information) planning. The collaboration between VR user and automatic path planner is made of 3 main aspects. First, the VR user is guided along a pre-computed path through a haptic device whereas he VR user can go away from the proposed path to explore possible better ways. Second the authority of automatic planning system is balanced to let the user free to explore alternatives (geometric layer). Third the intents of VR user are predicted (on topological layer) to be integrated in the re-planning process. Experiments are provided to illustrate the multi-layer representation of the environment, the path planning process, the control sharing and the intent prediction.

Detection and tracking of moving objects is an essential problem in situational awareness context and hence crucial for many robotic applications. Here we propose a method for the detection of moving objects with a 3D laser range sensor and a variation of the method for tracking multiple detected objects. The detection procedure starts with the ground extraction using random sample consensus approach for model parameter estimation. The resulting point cloud is then downsampled using voxel grid approach and filtered using a radius outlier rejection method. Within the approach, we have utilized a procedure for building short-term maps of the environment by using the octree data structure. This data structure enables an efficient comparison of the current scan and the short-term local map, thus detecting dynamic parts of scene. The ego-motion of the mobile platform is compensated using the available odometry information, which is rather imperfect, and hence is refined using the iterative closest point registration technique. Furthermore, due to sensor characteristics, the iterative closest point is carried out in 2D between the short-term map and the current, where the non-ground filtered scans are projected onto 2D. The tracking task is based on the joint probabilistic data association filter and Kalman filtering with variable process and measurement noise which take into account velocity and position of the tracked objects. Since this data association approach assumes a constant and known number of objects, we have utilized a specific entropy based track management. The experiments performed using Velodyne HDL-32E laser sensor mounted on top of a mobile platform demonstrate the suitability and efficiency of the proposed method.

Managing the bandwidth requirements of a team of robots operating cooperatively is an ubiquitous and commonly overlooked problem, despite being a crucial issue in the successful deployment of robotic teams. As the team’s size grows, its bandwidth requirements can easily rise to unsustainable levels. On the other hand, general-purpose compression techniques are commonly used to transmit data through constrained communication channels, and may offer a solution to this problem. In this paper, we study the possibility of using general-purpose compression techniques to improve the efficiency of inter-robot communication, firstly by comparing the performance of various compression techniques in the context a of multi-robot simultaneous localization and mapping (SLAM) scenarios using simplified occupancy grids, and secondly by performing tests with one of the compression techniques on real-world data.

The use of traditional position controllers in robots working in close proximity to people can lead to risks that result from the unexpected physical human-robot interactions (pHRI). To obtain effective tracking during normal operations, and retain the ability to recover from position errors in a smooth and damped manner, following contacts with external objects/agents, Proxy-based Sliding Mode Control (PSMC) has been proposed. While the efficacy of this controller in fully actuated manipulators has been studied, its use in underactuated systems has not yet been considered. This paper introduces PSMC for a class of underactuated systems. Specifically it considers the control of flexible joint manipulators with passive elastic elements in series with the motors. The Proxy-based Sliding Mode Control is developed to address the stability requirements of this type of dynamic system, while a torque controller to regulate the output torque of the actuation units is designed using Feedback Linearization and Linear Quadratic optimal control approaches. The performance of the proposed scheme is demonstrated in dynamic simulation of an anthropomorphic compliant robot arm.

A contactless torque/speed sensing module with digital signal processing circuit is designed and studied. To enhance more sensitivity with respect to the exerted torque, the cross-section of the mechanical structure is designed to be orange-slice-like and two links are connected to the both sides of the flexible orange-slice-like body. To realize this novel torque sensor applied to high-speed shafts, a couple of photo interrupters are equipped. In addition, there are numerous reflective strips attached to the links to cooperate with photo interrupters. Once the flexible body is deformed by an external torque, a relative angle between the two links is induced. Therefore, the two pulses sequences, out of the two corresponding photo detectors, would present a certain degree of time delay. By conversion of this time delay between the two pulse sequences, the exerted torque can be quantified. A set of digital signal processing circuit, which mainly consists of counter ICs, is incorporated to convert the time delay and time period (i.e., inverse of the shaft speed) into digital data in terms of torque and rotational speed of shaft so that almost no signal interference is involved. One of merits of the proposed torque sensor is: real-time measurement on torque applied becomes feasible even if the shaft is rotating at high speed. Another advantage of the fully-digital signal processing circuit is: no need to conduct A/D conversion and free of noise, cross-talk and EMI.

A passive vibrissa (whisker) is modeled as an elastic bending rod that interacts with a rigid obstacle in the plane. Aim is to determine the obstacle’s profile by one quasi-static sweep along the obstacle. To this end, the non-linear differential equations emerging from Bernoulli’s rod theory are solved analytically followed by numerical evaluation. This generates in a first step the support reactions, which represent the only observables an animal solely relies on. In a second step, these observables (possibly made noisy) are used for a reconstruction algorithm in solving initial-value problems which yield a series of contact points (discrete profile contour).

In this paper, a trajectory tracking control problem for non-linear systems is considered. For this purpose, a time-varying sliding mode controller will be applied to a linear time-varying approximate model of the nonlinear system around the desired trajectory. The authors proposed the simple design procedure of the pole placement controller for linear time-varying systems, by which the time-varying closed loop system becomes equivalent to some linear time-invariant system with desired constant eigenvalues. This implies that, by applying the conventional sliding mode controller to this equivalent time-invariant system, a time-varying sliding mode controller for a trajectory tracking problem of nonlinear systems is obtained. To show the validity of this method, a simulation results of the practical problem will be presented.