Bioinspired robots are useful tools to study complex biomechanical processes of animal locomotion. Key movements and kinematic parameters are under the control of experimenters, which is impossible to perform when experimenting with living animals. The primary challenge to test biological hypotheses is designing realistic robots taking inspiration from swimming snakes. Yet, underlying biomechanics of undulatory swimming i.e., anguilliform swimming, remains poorly understood. Many of underwater snakebots are made of rigid segments that form a broken-line system, unlike the skeleto-muscular systems of living snakes include more than 200 vertebrae, conferring an extreme fluidity. This paper introduces a novel design based on hybrid continuum cable driven robot (HCDR) developed through interaction with biologists and roboticists. This Biology-Push design significantly increases the fluidity and freedom of the robot's motion. Thus, improved mimic snake's locomotion is given using cable-driven to represent linkages between muscles and vertebrae providing good fluidity. In addition, the association of rigid and flexible parts allows a homogeneous distribution of actuators and masses to design autonomous swimming snake robot. Combining literature data and kinematic of swimming snakes’ analyses, we implemented a kinematic model to control prototype’ motion in both a plane and a volume. Finally, a comparative study between the device kinematics and the snake's movements is carried out.

Making collaborative robots physically interact with human requires to solve some technical issues of safety. This work presents a new methodological approach to define a mechanical model of force-deformation response for biologically inspired safety mechanisms. A nonlinear Hunt-Crossley based parametric model is proposed; its parameters are designed by an optimal approach balancing safety criteria with a metrological performance criterion. A particular attention is paid on the safety bounds so that they are not overly conservatives. The resulting optimal parameters show an excellent minimization of the objective function while ensuring the safety conditions.

This paper presents the challenges in the development of a new middleware for collaborative robotics and also for the new needs of interaction between robots and humans. The authors introduce first the main characteristics of a middleware; three types of middleware are proposed in the literature: Message Oriented Middleware, Middleware based on Remote Procedure Calls and also middleware based on the Object Request Broker. Thus a state of art of the most common middleware is given; ROS, RT-middleware and Orocos are detailed. Strong points and weak points are highlighted to justify the need to develop a new middleware. Based on this statement, the specifications of a new middleware are given with an important issue: develop a middleware that will meet complex research issues and also industrial issues for the future of robotics. In order to meet these issues, we propose the foundations of our approach. The middleware will be real-time, transferable, maintainable and multi-manufacturers. These requirements will also be able to guaranty the safety interaction of humans and machines in a dynamic and collaborative environment. Finally the efficiency of the proposed approach is demonstrated by using two different robots: a 6 dof industrial robot and a new 16 dof dextrous robot hand.

The optimization of the geometric parameters of a mechanism according to some desired objectives is an important aspect of design. The paper presents the multi-objective optimization of a new 6DOF haptic device, the Delthaptic. The design requirements are classified into constraints or objectives according to their priority and their assumed influence. It allows to simply formulate the optimization problem by using a Penalty method. An optimal solution is chosen into the resulting Pareto front and geometric parameters are fixed.

The majority of recent researches have been focused on developing compliant joint for rotary motion. Few authors contributed to the problematic of safety in pure linear motion, i.e. prismatic joint. The contribution of this work is to present a new design capable of achieving, passively, a nonlinear elastic behavior for prismatic joint implementation, the so-called Prismatic Compliant Joint (PCJ). This new device is based on a six-bar mechanism equipped with a linear spring. Hence, this structure generates the desired nonlinear stiffness behavior under a specified external force. The elastic characteristic will comply with force safety criteria of physical Human/Robot interaction (pHRI); a Hunt-Crossley model based one. In order to fit the PCJ response curve to the established safety measures, an optimization based on genetic algorithm method tunes PCJ intrinsic parameters subject to the chosen constraints.

This paper presents a new design of finger proposed by the ROBIOSS team of the PPRIME Institute: it is a fully actuated bio-inspired four-degree-of-freedom (DOF) finger driven by four actuators. It has been developed with the aim to replicate fine manipulation with fingertips with a high degree of interaction with the environment. This paper proposes to realize a robotic hand for inside-hand fine manipulation and adaptive grasping. The robotic hand is equipped with fingers whose design is based on a human anatomical finger model. Thus, several fingers can be assembled for building a human-sized dexterous hand with an anthropomorphic look. The modular design offers the ability to choose the number of fingers to be used as well as to adjust finger placement based on the manipulation task requirement. The tendon-based actuation presents a routing of the tendons that minimizes friction, kinematic, and static coupling between different finger axes in the transmission from motors to joints. Unlike many existing robotic hands, including our first anthropomorphic hand, we address the difficulties by decoupling joint motions with a new solution for the universal joint at the base of the finger. The results obtained demonstrate an excellent dynamic behavior and accuracy of the finger motion. Finally, the new finger design led to the development of a fully actuated mechanical hand with four fingers and with 16 DOF: the ROBIOSS hand. The hand was embedded on an industrial robot. A manipulation task that uses simultaneously abduction-adduction motion and flexion-extension motion of the finger demonstrates the potential of the hand for accurate manipulation.

The aim of this work is to improve the safety and control robustness of robots interacting physically with humans by enhancing their contact perception. More specifically in the abdominal area, respiratory motions are clearly influencing the contact dynamics but have not yet been accounted for in the estimation of the contact impedance. We propose here a combined mechanical-respiratory model of impedance along with its on-line identification. Numerical and practical experiments with living subjects validate the identification process and show the relevance in accuracy of accounting for the respiratory beats.

In this paper, dynamic simulation methods for reliability evaluation of common industry-based control system architectures are investigated. Control system design often employs complex reliability structures in the forms of several levels of software and hardware redundancies, hot and cold standby systems. This is required in order to achieve certain plant availability and safety functions. Control system maintenance requires expert knowledge due to the complexity of troubleshooting steps involved with a hardware or software failures of a large system. Hence, it is crucial to understand the effect of recovery time on reliability and on overall availability in a critical control system. Dynamic Fault Tree Analysis (DFTA), Markov Chains and Reliability Block Diagrams (RBD) are presented and a block library is introduced for addressing the aforementioned modelling problems. In order to be able to evaluate dynamic fault trees and Markov Chains, Monte Carlo simulation has been used. An industry-based case study is presented, where critical failures of a redundant Programmable Logic Controller (PLC) system are identified by a Failure Mode and Effect Analysis (FMEA). The bottom up process of modelling control system reliability is discussed.

This paper discusses an optimal torque distribution for a redundant 3-RRR spherical parallel manipulator (SPM). The SPM has parallel singular configurations within its workspace. The singularity leads to an amplification of the actuated joint torques which violates the actuator torque limits. The use of a redundant actuator solves the effect of the singularity. The quadratic sum of the actuated joint torques is minimized in order to obtain an optimal torque distribution. Experiments carried out show the advantage of introducing redundancy of actuators to improve the behavior of the SPM.

This paper deals with a spherical parallel manipulator developed for a master device, used in a teleoperation system for surgical application. The aim is to improve the calculation of the direct kinematic model (DKM) even when the parallel manipulator is in a singular configuration by using an extra sensor for a passive joint. A new simple approach for the DKM calculation of the haptic device, in order to meet requirement of real time applications, is developed. The measure of a passive joint's angle reduces the calculation time and improves the accuracy of the direct model particularly in the vicinity of the singular positions. Experiments were conducted on tele-operation system despite of the presence of singularities in the master workspace. The obtained results demonstrate the feasibility of the approach.

The current objective of this robotics research study is mainly to develop a cheap and not cumbersome systems for special applications. A PROMIS (Pprime RObot for Minimally Invasive Surgery) system is designed with this philosophy. The developed system is composed of master, slave and control system for collaborative operation between the surgeon and the robot. A parallel robot haptic device is used as master to teleoperate a serial robot with three degrees of freedom (dof). The advantage of this method is the possibility of coupling master and slave with non-matching kinematics through a kinematic transformation. The developed control system architecture is presented in this paper. The control architecture was designed to control the teleoperation system where the master and slave robots do not have homothetic kinematics. Finally, the viability of the developed control is demonstrated by the experimental results.

In tele-operation applications, especially in surgery, haptic devices need to exhibit a high degree of rigidity and accuracy. This paper deals with the development of a spherical parallel manipulator (SPM) that enables the satisfaction of those characteristics. The parallel architecture enables the use of this kind of robot as master in a tele-operation system. Moreover, the SPM has a center of rotation that makes it a natural candidate and more adapted to minimally invasive surgery application. However, the parallel manipulator presents a poor behavior in singular configuration.

The work presented in this paper deals with the generation of trajectories for humanoid robots imitating human gaits captured with a motion capture system. Once the human motion is recorded, this one is modified to be adapted to the robot morphology. The proposed method could be used for human-like robots of different sizes and masses. The generated gaits are closed to the human's ones while respecting the robot balance and the floor contacts. First the human joint angles are computed from the markers coordinates and applied directly to the robot kinematics model. Then, from this non-corrected motion, the trajectories of both feet and of the Zero Moment Point (ZMP) are generated respecting the constraints of floor contact and balance control. From this data, an inverse kinematic algorithm is used to compute the joint angles of the robot according to the feet and ZMP trajectories. The results with the robot HRP-2 (AIST, Kawada Industries, Inc) and the small-sized humanoid HOAP-3 (Fujitsu Automation Ltd) are compared with the human motion.

In this paper, we describe the development of a nonlinear exponential control law for a multi-degree of freedom (MDOF) wheeled mobile robot. These vehicles have many advantages over the more classical 2-DOF mobile robots. Indeed, the MDOF vehicles can negotiate tight turns easily and are able to travel sideways. Nevertheless, the MDOF robots are over constrained systems, which are difficult to control. By this reason, MDOF robots are limited to simple applications, like following a guide-wire. In order to obtain an autonomous behavior, we propose the use of a nonlinear control law, which exponentially leads the robot to a desired position with a given orientation. A new set of error variables is defined in order to propose the control vector. This control vector makes the final configuration globally exponentially stable. Results of simulation, which illustrate the effectiveness of the control, are also included.

The work deals with the development of the control of an articulated mechanical hand with the aim of fine manipulation of objects. This hand possess four fingers with four degrees of freedom. The sixteen joints are actuated using cable transmission and sixteen DC servo actuators. Position and force controls as well as their performances are presented and developed through examples. For the force control, an original method based on learning is developed. This method, using neural networks, allows us estimate the contact force by modeling the behaviour of the elastic transmission made with cables.

A collision-free path planner for manipulators based on a local approach is proposed. The task is defined as a combination of two displacements: the first one brings the robot closer to the goal configuration; and the second one enables the robot to avoid the local minima. This method solves many of classical problems found in the potential field methods. However, a zig-zaging phenomenon appears in some heavy cluttered environments. To avoid this situation, a graph based on the local geometry of the environment is constructed and an A* search is performed in order to find a new dead-lock free position. Tests in heavy cluttered environments were successfully performed.

Presents a local path planner for mobile robots in an environment with obstacles, where the nonholonomic constraints are considered. This planner uses only the distance information between the robot and the obstacles, thus it is well adapted for a robot equipped with onboard sensors, such as ultrasonic sensors. The obstacles are mapped as linear constraints into the velocity space of the robot; since the obstacles constraints are linear, they form a convex subset which represents the velocities that the robot can use without collision with the objects. We call this convex subset the "feasible velocities polygon" (FVP). The planner is composed of two modules; the first one is based on an optimization problem, which is transformed into a minimal distance calculation problem, in the velocity space, between the FVP and a point, which represents the reference velocity obtained by considering the nonholonomic constraint, the second module, which is used when a dead-lock situation occurs, uses the FVP representation to find, locally, the best velocity to escape from the dead-lock. The major advantages of this method are the very short calculation time and a continuous stable behavior of the velocities. The results presented demonstrate the capabilities of the proposed method for solving the collision free path planning problem.

The authors describe an efficient and fast algorithm for finding the minimum distance between two convex polyhedrons in three-dimensional space. To obtain the minimal distance, the proposed computational scheme is based on a direct approach to minimize the distance function which produces a succession of optimal search directions along the boundary of the objects. This algorithm combines the gradient projection method (Rosen, 1960) and an additional optimal search direction when the gradient projection method leads to a zigzagging phenomenon. In this case, the additional optimal search direction accelerates significantly the convergence of the process. Extensive numerical experiments with convex polyhedra showed the performance of the algorithm when compared with previous approaches.>

A method to find the optimal placement of manipulators by using multiple kinematic criteria is presented. By adequately locating a manipulator in order to accomplish a specified task, the method allows improvement in several indices of kinematic criteria of the manipulator for different points in the path. In the formulation presented, these indices are incorporated in an objective function which allows them to carry out their global optimization. The method considers the constraints imposed by the reach and by the joint limits of the manipulator. The improvements that can be achieved for the performances of manipulators by applying the method are shown in an example.>