Simulation, Modeling, and Programming for Autonomous Robots

Typical dynamic robotic simulators model the rigid body dynamics of robots using ordinary differential equations (ODEs). Such software libraries have traditionally focused on simulating the rigid body dynamics robustly, quickly, and accurately toward obtaining consistent dynamics performance between simulation and

in situ

. However, simulation practitioners have generally yet to investigate maintaining

temporal

consistency within the simulation: given that simulations run at variable rates, how does the roboticist ensure the robot’s control software (controller, planners, and other user-level processes) runs at the same rate that it would run in the physical world? This paper describes an intersection of research between Robotics and Real-Time Operating Systems that investigates mechanisms for addressing this problem.

Simulation in robotics is often a love-hate relationship: while simulators do save us a lot of time and effort compared to regular deployment of complex software architectures on complex hardware, simulators are also known to evade many of the real issues that robots need to manage when they enter the real world. Because humans are the paragon of dynamic, unpredictable, complex, real world entities, simulation of human-robot interactions may look condemn to fail, or, in the best case, to be mostly useless. This collective article reports on five independent applications of the MORSE simulator in the field of human-robot interaction: It appears that simulation is already useful, if not essential, to successfully carry out research in the field of HRI, and sometimes in scenarios we do not anticipate.

In this paper we present a dynamic simulator for intervention autonomous underwater vehicles. Prototyping and testing of such robots is often tedious and costly, and realistic simulation can greatly help validating several aspects of the project. In order to benefit from existing software, the presented system is integrated with ROS, through the Gazebo dynamic simulator, and the underwater image rendering UWSim. The whole approach allows realistic rendering of dynamic multi-robot simulation, with contact physics, buoyancy, hydrodynamic damping and low-level PID control. This paper details the modeling choices that are done and exposes how to build its own AUV model. Integration with other ROS programs is exposed, and a simulation shows an example of behavior during a black box recovery mission.

The DARPA Virtual Robotics Challenge (VRC) [1] was a cloud-based robotic simulation competition. Teams competed by writing control software for a humanoid robot to perform disaster response tasks in real-time simulation. Simulating the physics and sensors of a humanoid robot in real-time presented challenges related to the trade-off between simulation accuracy and computational time. The Projected Gauss-Seidel (PGS) iterative solver was chosen for its performance and robustness, but it lacks the accuracy and the fidelity required for reliable simulation of task-level behaviors. This paper presents the modeling decisions and algorithmic improvements made to the Open Dynamics Engine (ODE) physics solver that improved PGS accuracy and fidelity without sacrificing its real-time simulation performance in the VRC. These improvements allowed for stable simulation regardless of user input during the VRC, and supported reliable contact dynamics during VRC tasks without violating the near real-time requirement.

The Model-based development of robotics applications relies on the definition of models of the controls that abstract the computation and communication platform under the synchronous assumption. Computation, scheduling and communication delays can affect the performance of the controls in way that are possibly significant, and an early evaluation allows to select the best control compensation or the best HW/SW implementation platform. In this paper we show a case study of the application of the open T-Res framework, an environment for the co-simulation of controls and real-time scheduling, on a quadcopter model example, highlighting the possible tradeoffs in the selection of the scheduling strategy and priority assignment.

Many robotic projects use simulation as a faster and easier way to develop, evaluate and validate software components compared with on-board real world settings. In the human-robot interaction field, some recent works have attempted to integrate humans in the simulation loop. In this paper we investigate how such kind of robotic simulation software can be used to provide a dynamic and interactive environment to both collect a multimodal situated dialogue corpus and to perform an efficient reinforcement learning-based dialogue management optimisation procedure. Our proposition is illustrated by a preliminary experiment involving real users in a Pick-Place-Carry task for which encouraging results are obtained.

In this work we describe a simulation environment for an autonomous all-terrain mobile robot. To allow for extensive test and verification of the high-level perception, planning, and trajectory control modules, the low-level control systems, the sensors, and the vehicle dynamics have been modeled and simulated by means of the V-Rep 3D simulator. We discuss the overall, i.e., high and low-level, software architecture and we present some validation experiments in which the behavior of the real system is compared with the corresponding simulations.

In this paper, we present the status of ongoing research aimed at tackling the issues of programming robots for small-size productions where fast set-up times, quick changeovers and easy adjustments are essential. We use a probabilistic approach where uncertainties are taken into account, making the deterministic requirements of an assembly process less strict. Concretely, actions from an action library are modelled through parameters, simulation is used to facilitate learning of uncertainty-tolerant actions, and a Domain-Specific Language (DSL) is used to convert the abstractly specified actions into corresponding executable actions. The approach is tested on an application example from industry.

Field experiments with a team of heterogeneous robots require human and hardware resources which cannot be implemented in a straightforward manner. Therefore, simulation environments are viewed by the robotic community as a powerful tool that can be used as an intermediate step to evaluate and validate the developments prior to their integration in real robots. This paper evaluates a novel multi-robot heterogeneous cooperative perception framework based on monocular measurements under the MORSE robotic simulation environment. The simulations are performed in an outdoor environment using a team of Micro Aerial Vehicles (MAV) and an Unmanned Ground Vehicle (UGV) performing distributed cooperative perception based on monocular measurements. The goal is to estimate the 3D target position.

Today Virtual Reality (VR) simulation technology is a well-known field of virtual training and engineering and widely applied in research and in the industry. Multi-domain VR simulation systems cover multiple technical and visual aspects not limited to a single task or domain. While current systems mostly neglect the rendering component and provide purely functional graphics and simple virtual environments, we present the concepts of eRobotics and matching system structures to combine complex simulations and realistic virtual environments in a holistic VR simulation system. These environments not only provide attractive visual presentations, they also help to realize close-to-reality testing of virtual prototypes and positively affect the accuracy and performance of simulated components like optical sensors.

In this paper, an inverse dynamics approach by means of adaptive coupled oscillators is used in the modelling and control of a lower limb orthosis applied at the knee and ankle joint level. This design is aimed at providing assistance and rehabilitative measures to humans with lower limb disorders and as such presents a platform for which their mobility performance can be improved. Adaptive oscillators are known to have the capability of learning high level parameters of sinusoidal, quasi-sinusoidal or non-sinusoidal signals (amplitude, frequency and offset). However, the later signal (non-sinusoidal) considered in this paper requires a number of oscillators in parallel to replicate the moving joint regarding filtering via adaptive oscillator. The dynamic model for the knee and ankle are considered to take the form of a damped pendulum model connected by two revolute joints. This maps the input torque of both joints (knee and ankle) to their output trajectories, hence integrating the different forces at the joint level of the different joints. The coupling effect is achieved by the use of coupled Adaptive Frequency Oscillator (AFO) for the estimation of the joint trajectories. Tracking performance for the knee-ankle orthosis is studied for non-sinusoidal reference trajectories, having a global coupling between the joints. The results obtained using SCILAB show a good performance of the controller trajectory tracking capabilities even in the presence of external disturbances.

We present an approach to analytically construct a robotic team, i.e., team members and deployment order, that achieves a specific task with quantified probability of success. We assume that each robot is Markovian, and that robots interact with each other via communication only. Our approach is based on probabilistic model checking (PMC). We first construct a set of Discrete Time Markov Chains (DTMCs) that each capture a specific “projection” of the behavior of an individual robot. Next, given a specific team, we construct the DTMC for its behavior by combining the projection DTMCs appropriately. Finally, we use PMC to evaluate the performance of the team. This procedure is repeated for multiple teams, the best one is selected. In practice, the projection DTMCs are constructed by observing the behavior of individual robots a finite number of times, which introduces an error in our results. We present an approach – based on sampling using the Dirichlet distribution – to quantify this error. We prove the correctness of our approach formally, and also validate it empirically on a mine detection task by a team of communicating Kilobots.

As the complexity of robots deployed in the real world increases, the use of formal specifications in the development of safety-critical robot systems is becoming increasingly important. A formal specification gives confidence in the correctness, completeness, and accuracy of a system design. In this paper, we present a formal specification of a redundant control architecture for a mobile robot in the form of a model. The model is created using the Architecture Analysis and Design Language (AADL). This formal language allows the model to be analysed to prove system properties of interest. In this case, we are interested in proving the response time of the robot to external obstacles and to internal errors. We present the model and the results of these analyses with the goal of proving that the architecture is sufficiently safe for use in a safe robot wheelchair.

In this article, we present the process of modeling control algorithms as means to increase reliability of software components. The approach to developing Embedded Control Software (ECS) is tailored to Component-Based Software Development (CBSD). Such tailoring allows to re-use the ECS development process tools in a development process for robotics software. Model-to-text transformation of the ECS design tool is extended to model-to-component transformation suitable for CBSD frameworks. The development process and tools are demonstrated by a use case.

This paper gives an overview of reflections about more generic robotic architectures models and their associated tools. The objective of our work is not to define a new robot software but rather to specify common robotic requirements for future component-based models. These models could be used as a common-base by the robotic sub-communities whatever the purpose their different robots have been designed for, whatever the targeted hardware, the chosen frameworks or the host operating systems. Even if we are not yet strongly familiar with the specificities of the Model-Driven Architecture (MDA) and with the Domain-Specific Language (DSL), we are self-convinced by the powerful benefits that these two fields could bring to robotics and to robotic architecture models. In this paper, we discuss about the characteristics a robotic architecture model should own to be efficiently designed by software model engineers and easily but efficiently used by robot engineers.

We present

gusimplewhiteboard

, a software architecture analogous to

ROS:services

and

ROS: messages

, that enables the construction and extremely efficient inter-process relaying of message-types as

C++11

objects, All

gusimplewhiteboard

objects reside in shared memory. Moreover, our principle is to use idempotent message communication, in direct contrast to previously released platforms for robotic-module communication, that are based on an event-driven subscriber model that queues and multi-threads. We combine this with compiled, time-triggered, logic-labeled finite state machines (

llfsms

) the are executed concurrently, but scheduled sequentially, in an extremely efficient manner, removing all race conditions and requirements for explicit synchronisation. Together, these tools enable effective robotic behaviour design, where arrangements of

llfsms

can be organised as hierarchies of machines and submachines, enabling composition of very complex systems. They have proven to be very powerful for Model-Driven Development, capable of simulation, validation, and formal verification.

The design, simulation and programming of robotics systems is challenging as expertise from multiple domains needs to be integrated conceptually and technically. Domain-specific modeling promises an efficient and flexible concept for developing robotics applications that copes with this challenge. It allows to raise the level of abstraction through the use of specific concepts that are closer to the respective domain concerns and easier to understand and validate. Furthermore, it focuses on increasing the level of automation, e.g. through code generation, to bridge the gap between the modeling and the implementation levels and to improve the efficiency and quality of the software development process. Within this contribution, we survey the literature available on domain-specific (modeling) languages in robotics required to realize a state-of-the-art real-world example from the RoboCup@Work competition. We classify 41 publications in the field as reference for potential DSL users. Furthermore, we analyze these contributions from a DSL-engineering viewpoint and discuss quantitative and qualitative aspects such as the methods and tools used for DSL implementation as well as their documentation status and platform integration. Finally, we conclude with some recommendations for discussion in the robotics programming and simulation community based on the insights gained with this survey.

Safety is a key challenge in robotics, in particular for mobile robots operating in an open and unpredictable environment. To address the safety challenge, various software-based approaches have been proposed, but none of them provide a clearly specified and isolated safety layer. In this paper, we propose that safety-critical concerns regarding the robot software be explicitly declared separately from the main program, in terms of externally observable properties of the software. Concretely, we use a Domain-Specific Language (DSL) to declaratively specify a set of safety-related rules that the software must obey, as well as corresponding corrective actions that trigger when rules are violated. Our prototype DSL is integrated with ROS, is shown to be capable of specifying safety-related constraints, and is experimentally demonstrated to enforce safety behaviour in existing robot software. We believe our approach could be extended to other fields to similarly simplify safety certification.

This paper describes the development of an electric car prototype, aimed at autonomous, energy-efficient driving. Starting with an urban electric car, we describe the mechanical and mechatronics add-ons required to automate its driving. In addition, a variety of exteroceptive and proprioceptive sensors have been installed in order to obtain accurate measurements for datasets aimed at characterizing dynamic models of the vehicle, including the complex problem of wheel-soil slippage. Current and voltage are also monitored at key points of the electric power circuits in order to obtain an accurate model of power consumption, with the goal of allowing predictive path planners to trace routes as a trade-off between path length and overall power consumption. In order to handle the required variety of sensors involved in the vehicle, a MOOS-based software architecture has been developed based on distributed nodes that communicate over an onboard local area network. We provide experimental results describing the current stage of development of this platform, where a number of datasets have been already grabbed successfully and initial work on dynamics modeling is being carried on.

Robot programming is an interdisciplinary and knowledge-intensive task. All too often, knowledge of the different robotics domains remains implicit. Although, this is slowly changing with the rising interest in explicit knowledge representations through domain-specific languages (DSL), very little is known about the DSL design and development processes themselves. To this end, we present and discuss the reverse-engineered process from the development of our Grasp Domain Definition Language (GDDL), a declarative DSL for the explicit specification of grasping problems. An important finding is that the process comprises similar building blocks as existing software development processes, like the Unified Process.

The following paper presents the ROS-I interface developed to control Comau manipulators. Initially, the Comau controller allowed users to command a real robot thanks to motion primitives formulated through a Comau motion planning library. Now, either a ROS or a non ROS -compliant platform can move either a real or a virtual Comau robot using any motion planning library. Comau modules have been wrapped within ROS and a virtual model of a Comau robot has been created. The manufacturer controller has been innovatively used to drive both the real and the simulated automata.

We introduce Robot Unit Testing (RUT) as a methodology to bring modern testing methods into robotics. Through RUT the range of robotics software that can be automatically tested is extended beyond current practice. A robotics simulator is used to bridge the gap between well automated tests that only check a robot’s software and time consuming, inherently manual tests on robots of alloy and circuits. An in-depth realization of RUT is shown, which is based on the Robot Operating System (ROS) framework and the Gazebo simulator due to their prominence in robotics research and inherent suitability for the RUT methodology.

An increasing number of applications require the integration of heterogeneous hardware and software components. Due to the high levels of complexity that such integrations demand, several solution have been proposed in the state of art of software engineering. This paper introduces the IMI2S framework: a distributed computing software platform aimed to cope with such levels of complexity by simplifying the functional decomposition of the problems through the implementation of highly decoupled, efficient and portable software. We will present the design issues addressed in the development of the IMI2S framework. We will show through two case studies its flexibility and its general efficacy.

Autonomous robot fleets are complex systems that require the interaction and communication between heterogeneous hardware and software. Despite many years of work in robotics, there is still a lack of established software architecture and middleware, in particular for large scale multi-robots systems. Many research teams are still writing specific hardware orientated software that is very tied to a robot. This vision makes sharing modules or extending existing code difficult. A robotic middleware should be designed to abstract the low-level hardware architecture, facilitate communication and integration of new software. In this paper, we present and compare seven existing middlewares capable of being used in multi-robot systems. We also present two dedicated cloud based multi-robots platforms. After this analysis, we discuss why a cloud of robots and not a cloud for robots is more suitable in a fleet context.

Service robots become increasingly capable and deliver a broader spectrum of services which all require a wide range of perceptual capabilities. These capabilities must cope with dynamically changing requirements which make the design and implementation of a robot perception architecture a complex and tedious exercise which is prone to error. We suggest to specify the integral parts of robot perception architectures using explicit models, which allows to easily configure, modify, and validate them. The paper presents the domain-specific language RPSL, some examples of its application, the current state of implementation and some validation experiments.

Integrating robotic systems into our everyday life needs that we prove that they will not endanger people, i.e. that they will behave correctly with respect to some safety rules. In this paper, we propose a validation toolchain based on a Domain Specific Language. This DSL allows to model the software architecture of a robot using a component-based approach. From these models, we provide tools to generate deployable components, as well as a two-step validation phase. This validation first performs a real-time analysis of the component architecture, leading to an evaluation of the software architecture schedulability. Then we can check the validity of some behavioral property on the components.

We consider the problem of automating the verification of distributed control software relying on publish-subscribe middleware. In this scenario, the main challenge is that software correctness depends intrinsically on correct usage of middleware components, but structured models of such components might not be available for analysis, e.g., because they are too large and complex to be described precisely in a cost-effective way. To overcome this problem, we propose to identify abstract models of middleware as finite-state automata, and then to perform verification on the combined middleware and control software models. Both steps are carried out in a computer-assisted way using state-of-the-art techniques in automata-based identification and verification. Our main contribution is to show that the combination of identification and verification is feasible and useful when considering typical issues that arise in the implementation of distributed control software.

This paper describes a software infrastructure for developing and composing task and motion planners. The functionality of motion planners is well defined and they provide a basic primitive on top of which it is possible to develop planners for addressing higher level tasks. It is more challenging, however, to identify a common interface for task planners, given the variety of challenges that they can be used for. The proposed software platform follows a hierarchical, object-oriented structure and identifies key abstractions that help in integrating new task planners with popular sampling-based motion planners. Examples of use cases that can be implemented within this common software framework include robotics applications such as planning among dynamic obstacles, object manipulation and rearrangement, as well as decentralized motion coordination. The described platform has been used to plan for a Baxter robot rearranging similar objects in an environment in an efficient way.

This paper describes a Component-Based Meta-Model (WCOMM) and framework (FraCC) as part of a complete Model-Driven Software Development process and toolchain: C-Forge. The approach given in the design of WCOMM and FraCC is presented highlighting the differences with other similar approaches. To illustrate the use of C-Forge, the development of a control architecture for the robots in project MISSION is presented.

Learning maps from sensor data has been addressed since more than two decades by Simultaneous Localization and Mapping (SLAM) systems. Modern state-of-the-art SLAM approaches exhibit excellent performances and are able to cope with environments having the scale of a city. Usually these methods are entailed for on-line operation, requiring the data to be acquired in a single run, which is not always easy to obtain. To gather a single consistent map of a large environment we therefore integrate data acquired in multiple runs. A possible solution to this problem consists in merging different submaps. The literature proposes several approaches for map merging, however very few of them are able to operate with local maps affected by inconsistencies. These methods seek to find the global arrangement of a set of rigid bodies, that maximizes some overlapping criterion. In this paper, we present an off-line technique for merging maps affected by residual errors into a single consistent global map. Our method can be applied in combination with existing map merging approaches, since it requires an initial guess to operate. However, once this initial guess is provided, our method is able to substantially lessen the residual error in the final map. We validated our approach on both real world and simulated datasets to refine solutions of traditional map merging approaches.

The focus of this paper is on the effect of muscle force optimization algorithms on the human lower limb stiffness estimation. By using a forward dynamic neuromusculoskeletal model coupled with a muscle short-range stiffness model we computed the human joint stiffness of the lower limb during running. The joint stiffness values are calculated using two different muscle force optimization procedures, namely: Toque-based and Torque/Kinematic-based algorithm. A comparison between the processed EMG signal and the corresponding estimated muscle forces with the two optimization algorithms is provided. We found that the two stiffness estimates are strongly influenced by the adopted algorithm. We observed different magnitude and timing of both the estimated muscle forces and joint stiffness time profile with respect to each gait phase, as function of the optimization algorithm used.

The idea of component-defined visualization is introduced and benefits for different challenges in robotics software development are discussed – including system maintenance, component integration, and identification of critical behavior or malfunction. Design considerations for integration in state-of-the-art robotic software frameworks are presented – with an open source implementation for the

Finroc

framework as a proof-of-concept. Its use in two very different autonomous systems is illustrated. Experiments with these systems indicate that the proposed approach has in fact relevant advantages.

Though significant progress on autonomous navigation has been made, the natural world offers interesting examples of navigational techniques that are worth exploring and understanding. The cognitive mechanism of mental rotation has been revealed in numerous cognitive and neuroscientific experiments; its reason for existence and evolution, however, has yet to be thoroughly understood. It is speculated that this mechanism may assist primates in navigation. This paper explores how mental rotation can be used in navigation by developing an autonomous robotic navigation algorithm that draws inspiration from the mechanism. This algorithm was tested on a robot tasked with navigating to a specified goal location contained within the agent’s initial view. The testing suggests that mental rotation can be used as an asset in navigation.

Research on robot systems either integrating a large number of capabilities in a single architecture or displaying outstanding performance in a single domain achieved considerable progress over the last years. Results are typically validated through experimental evaluation or demonstrated live, e.g., at robotics competitions. While common robot hardware, simulation and programming platforms yield an improved basis, many of the described experiments still cannot be reproduced easily by interested researchers to confirm the reported findings. We consider this a critical challenge for experimental robotics. Hence, we address this problem with a novel process which facilitates the reproduction of robotics experiments. We identify major obstacles to experiment replication and introduce an integrated approach that allows (i) aggregation and discovery of required research artifacts, (ii) automated software build and deployment, as well as (iii) experiment description, repeatable execution and evaluation.We explain the usage of the introduced process along an exemplary robotics experiment and discuss our approach in the context of current ecosystems for robot programming and simulation.

This paper presents an evolution method used to modify the morphology of humanoids to make them more efficient in a specific direction of walking. Starting from the NAO’s model used in the 3D Simulation Soccer League, the walking specializations are based on 5 to 8 parameters that are being evolved. A black-box optimization process is run and guided by a decision-making function that defines the outcome of the humanoid evolution process. The simulation results lead to four optimized morphological profiles, each of them specialized for either forward, or lateral, or diagonal walk, or in-place turn respectively. These results could be used to build heterogeneous humanoids inside a team of soccer players.

This paper examines the process involved in separately packing goods of several kinds into a container. During such a process, the volume of each kind of good (“block”) must be optimized to minimize container size. For this optimization, densest packing procedures can be used to determine the feasibility of packing by a robot. This paper analyzes packing procedures generated automatically using partial order representation of feasible procedures, which determines stable and feasible procedures by considering that process planning affects the stability, contact forces, and torques of the target packing pattern and its transient piles.

This paper analyzes and discusses the problem of optimizing the size of a team of robots for multi-robot exploration. We are concerned with the number of robots for a given exploration task that minimizes both exploration time and cost. Minimizing time means that the exploration should be done as fast as possible. Minimizing cost means that the number of robots and their energy consumption should be as low as possible. To solve this problem, we report in this paper, on a series of exploration simulations based on ROS and MORSE using a cluster of computers. The simulated code is exactly the same as that which would run on the actual robots. Such a simulation infrastructure is crucial to “quickly” execute experiments with different parameters such as the number of robots or their initial positions.

In this paper, we suggest

gripper quality metrics

that indicate the performance of a gripper given an object CAD model and a task description. Those, we argue, can be used in the design and selection of an appropriate gripper when the task is known. We present three different gripper metrics that to some degree build on existing grasp quality metrics and demonstrate these on a selection of parallel jaw grippers. We furthermore demonstrate the performance of the metrics in three different industrial task contexts.

Before autonomous robotics can be used for dangerous or critical missions, performance guarantees should be made available. This paper overviews a software system for the verification of behavior-based controllers in context of chosen hardware and environmental models. Robotic controllers are automatically translated to a process algebra. The system comprising both the robot and the environment are then evaluated by VIPARS, a verification software module in development, and compared to specific performance criteria. The user is returned a probability that the performance criteria will hold in the uncertainty of real-world conditions. Experimental results demonstrate accurate verification for a mission related to the search for a biohazard.

This paper explores the 2D Audio Localization using only the Direction of Arrivals (DOAs) of a fixed acoustic source coming from an audio sensors network and proposes a new method for estimating the position of the source using a Gaussian Probability over DOA approach (G-DOA) in the 2D space. This new method was thought for Robotic purposes and introduces a new perspective of the Audio-Video synergy using Video Sensor Localization in the environment for extrinsic Audio Sensor Calibration. Our approach achieves more precise solutions using more sensors and shows better results compared to the analytic Weighted Least Square method (WLS-DOA). Test results using Microsoft Kinect as DOA-sensors within the ROS framework show that the algorithm is robust, modular and can be easily used for robot applications.

Traversing unstructured environments, (statically stable) legged robots could be applied effectively but, they face two main problems: the

high complexity

of the system and the

low speed of locomotion

. To address the complexity of the controller, we apply a control layer that abstracts the legged robot to an omni-directional moving mass. In this control scheme, we apply the gait generator as proposed by Estremera and de Santos. We present theory to determine the theoretically maximum achievable velocity of a quadruped and compare the (omni-directional) maximum velocity of the selected gait generator with this optimum to validate its performance. For our use case the theoretically maximum achievable velocity is 1

ms

− 1

; in simulations we achieve a velocity for straight movement of maximum 0.75

ms

− 1

. Normal turns with a radius larger than 0.45

m

are possible at a velocity of at least 0.1

ms

− 1

; the performance of crab turns is too unpredictable to be useful. The gait generator as proposed by Estremera and de Santos is partially capable of supporting omni-directional movement at satisfactory velocities.

Biomimetic Chemical Plume Tracing (CPT) problem is complex because it couples nonlinearity of biological systems with uncertainty of time-varying plume diffusion. A vision-based simulator is proposed to decouple these difficulties to facilitate multiple runs under controlled environment. This enables identification of efficient biological CPT algorithm. The simulator is used to simulate Embodiment Sensing (ES), i.e. sensing using physical attributes of animals. Wings and antennae of silk moth are used for ES, and evaluated for CPT using vision-based simulator. Results suggest (1) vision-based plume field mimics actual plume diffusion in terms intermittency, and (2) similar performance as that for surge-cast algorithm. The contribution is two-fold, (1) vision-based plume diffusion simulator decouples uncertainty of plume diffusion from nonlinearity of biological system to facilitate biomimetic CPT study, and (2) feasibility of using physical attributes of silk moth to achieve good CPT performance.

Robot perception subsystems typically form complex networks, with boxes representing computations and arrows presenting the exchanged data. Taking into account that data acquired from robot sensors may arrive with different frequencies, as well as that computations may by performed on different processor cores, a problem of handling of asynchronous data flows appears. Hence appropriate tools facilitating the implementation are highly demanded. In this article we propose a solution to the aforementioned problem, enabling the activation of a conditional behaviour of a given computational block, depending on the presence of data in its input buffers. Theoretical considerations led to the implementation of these mechanisms in a component-oriented framework for development of robot perception subsystems: DisCODe. Operation of the solution was verified on an exemplary perception subsystem using RGB-D camera.

In this paper, we present a healthcare sensor managing system that manages healthcare sensor devices in a distributed environment. It is an independent system, but can cooperate by wireless communication with client systems, such as robots and smart phones. We have designed five key concepts of our sensor managing system including plug and play, status managing, scheduling of requests, expandability, and compatibility. We have developed a sensor managing system based on our design concepts, and applied this system to a healthcare application. It consists of three parts: a healthcare robot system, a sensor manager system, and sensor device systems. It can be applied in various use case scenarios for heterogeneous devices, between single and multiple clients. To verify the efficiency of our system, we report functionality experiments focusing on each of the five key concepts.

Kinesthetic teaching is a commonly employed method for programming robots using the

Programming by Demonstration

(PbD) paradigm. It is widely regarded as an intuitive approach to robot programming, which can be performed by shop-floor workers. Much research in this area has focused on pick-and-place tasks while demanding assembly tasks have received less attention so far. Nonetheless, in various contributions kinesthetic teaching is utilized to gain insight into human assembly strategies by deriving trajectories, mating forces, etc. To evaluate the discrepancies between kinesthetic teaching and manual assembly in the context of industrial assembly tasks, we conducted a user study with 78 participants featuring four different tasks. Our results confirm the ease of learning attributed to kinesthetic teaching but also suggest that trying to transfer human assembly strategies using this method may suffer from a substantial flaw.

This paper presents a motion optimization system for an industrial quality inspection process where a vision device coupled with a manipulator robot arm is able to perform quality and completeness inspection on a complex solid part. In order to be deployed in an industrial production plant, the proposed system has been engineered and integrated as a module of an offline simulator, called WorkCellSimulator, conceived to simulate robot tasks in industrial environments. The novelty of the paper concerns the introduction of time constraints into the motion planning algorithms. Then, these algorithms have been deeply integrated with artificial intelligence techniques in order to optimize the inspection cycle time. This integration makes the application suitable for time-constrained processes like, e.g., autonomous industrial painting or autonomous thermo-graphic detection of cracks in metallic and composite materials.

This paper presents a project-based laboratory for senior-level students in computer engineering that is based on the LEGO Mindstorms kits extended with a set of off-the-shelf microcontrollers and custom electronics. It is organized in an integrated set of projects, which individually cover a subset of typical issues and challenges involved in the development of a complete robotic system. The pedagogical goal is to equip students with an understanding of how engineering of complex projects is a multi-dimensional decision making process and with teamwork and self-learning skills.

Point cloud registration is an essential part for many robotics applications and this problem is usually addressed using some of the existing variants of the Iterative Closest Point (ICP) algorithm. In this paper we propose a novel variant of the ICP objective function which is minimized while searching for the registration. We show how this new function, which relies not only on the point distance, but also on the difference between surface normals or surface tangents, improves the registration process. Experiments are performed on synthetic data and real standard benchmark datasets, showing that our approach outperforms other state of the art techniques in terms of convergence speed and robustness.

In this paper, we introduce a healthcare robot system for older people and its experiments in private and public spaces. We designed a healthcare robot system and healthcare functionalities, and conducted a long-term study in a real environment. Our healthcare robot system consists of three parts: a kiosk type service robot platform, a healthcare software system with healthcare service modules, and a medical server system. 1) The kiosk type service robot platform is used for giving helpful information to older people through a touch screen. 2) The healthcare software system is designed to enable easy modification of healthcare service modules according to the purpose of the robot. 3) The medical server system stores health information of older people for managing their health conditions. For validating our software design and implementation in real environments, we deployed this healthcare robot system in private and public places of a retirement village. In these experiments, older people interacted with the robots and used healthcare functionalities for over 12 weeks. During the experiments, the robots sent records of the interactions to our medical server. The server provides this information to clinicians who are supervising the older peoples’ health status. When the experiments were completed, the participants completed questionnaires. The results showed that older people in private places used the healthcare service for checking their health conditions, and older people in public places like to use the entertainment services. We confirmed that our kiosk type robot can help older people as well.

Part of the acknowledgement for the paper starting on page 450 of this volume was inadvertently omitted. The full acknowledgement should read as follows:

Acknowledgments.

The research leading to these results has received funding from the European Community’s Seventh Framework Programme FP7/2007-2013 (Programme and Theme: ICT-2011.2.1, Cognitive Systems and Robotics) under grant agreement no. 600578, ACAT.

The research has furthermore received founding from the Danish Council for Strategic Research under the grant agreement no. 12-131860, CARMEN.

The research has also received funding from a project MB/WM/21/2013 realised by Białystok University of Technology.