The paper presents an algorithm for real-time motion control of a mobile manipulator in unstructured environments. The mobile manipulator consists of a velocity controlled mobile platform and a torque controlled manipulator. Therefore, a combination of torque and velocity control is used. For the obstacle avoidance two different principles are used: virtual repulsive velocity and action-reaction principle. The proposed control method has been verified on real system, composed of a mobile platform and a four DOFs manipulator arms. The results have been compared to the manipulator with a fixed base.

This paper deals with the problem of the dynamic stress optimisation for manipulator robots. The maximum dynamic stress search is a challenging problem, insofar the dynamic stresses change in a very complex manner and they are strongly influenced by the payload, internal parameters of the robot, and instantaneous values of the articular variables (positions, velocities, and accelerations). This problem is solved using a hybrid simplex-genetic algorithm. This algorithm allows solving the dilemma of exploration vs. exploitation in the search space. A simulation study shows the greater performance of the proposed approach as compared to a classical approach of the configuration space inverse sweeping.

This work proposes control structures that efficiently combine force control with vision servo control of robot manipulators. Impedance controllers are considered which are based both on visual servoing and on physical or fictitious force feedback, the force and visual information being combined in the image space. Force and visual servo controllers included in extended hybrid control structures are also considered. The combination of both force and vision based control allows the tasks range of the robot to be extended to partially structured environments. The proposed controllers, implemented on an industrial SCARA-type robot, are tested in tasks involving physical and virtual contact with the environment.

In this paper we present a globally convergent and ultimately exponentially convergent tracking controller for an X4 Flyer rotor craft. The desired or reference trajectory is restricted in one way: its scalar thrust must be uni signed and bounded away from zero. The proposed controller is a two phase one. Phase I converges the attitude (orientation) error and thrust error to zero globally and ultimately exponentially. Phase II is a position tracking and relative yaw angle tracking controller, which is not globally convergent by itself. By carefully postponing a switch from Phase I to Phase II, we are able to assure that Phase II will ultimately drive the position error and relative yaw angle error to zero exponentially. In problem-statement/theorem/proof format, we define the system, the control objective, the proposed controller, and a rigorous proof that the controller maintains boundedness of all closed loop signals and provides global convergence of the tracking error to zero and also ultimately exponential convergence of the tracking error to zero.

In this paper the mechanical design for a new 7-dofs leg is investigated as a walking module for a humanoid robot. In particular, a dynamic simulation is deduced by means of a Newton-Euler formulation and implemented numerically in order to compute the needed input actuator torques. A simulation is carried out for a similar built leg prototype that is used as a walking module for WABIAN-RIV (WAseda BIped humANoid robot-Revised IV). Experimental tests are carried out on this existing leg in order to validate the proposed formulation for a similar application. Thus, the validated formulation has been used in order to design the actuators for a new leg prototype.

In this paper two well known and two new methods, and corresponding algorithms, of trajectory planning with a constrained end-effector path tracking error are presented and compared. They are: an exact path following method, the Taylor's algorithm of following a straight line with prescribed accuracy, a local optimization method, and a method of local trajectory shortening. It appears that the last two methods provide path following with a prescribed accuracy while keeping a planned trajectory as short as possible. Presented algorithms can extend robot programming languages with a tool of trajectory planning.

When modeling the dynamics of robotic systems containing electric motors, the force generated by the motor is generally considered only as an applied torque or force that is independent of mechanical state variables such as velocity. Due to the electromechanical coupling effects in the motors, this approach leads engineers working on a robotic system to designing faulty controllers. In this paper, we propose a dynamics analysis model in which DC motor dynamics are embedded into a mechanical dynamics model such that the electromechanical coupling effects are included in the overall model. A model for the DC motor is developed based on its equivalent circuit model and incorporated into the generalized recursive dynamics formula previously developed by our group. The resulting dynamic numerical simulation program provides an effective and realistic approach for analyzing the electromechanical dynamics of robotic systems driven by DC motors. The developed numerical simulation tool is evaluated by applying to an industrial robot and a flexible antenna system driven by DC motors for a satellite.

This paper presents the basis of a mathematical model for simulation of planar flexible-link manipulators, taking into consideration the effect of higher stiffness zones at the link tips. The proposed formulation is a variation of the finite segment multi-body dynamics approach. The formulation employs a consistent mass matrix in order to provide better approximation than the traditional lumped masses often encountered in the finite segment approach. The formulation is implemented into a computational code and tested through three examples; cantilever beam, rotating beam and three-link manipulator. In these examples, the length of the rigid tips at both sides of each link ranges from 0% to 6.25% of the whole link length. The zones of higher stiffness at the link tips are treated as short rigid zones. The effect of the rigid zones is averaged along with some portions of the flexible links, thereby allowing further simplification of the dynamic equations of motion. The simulation results demonstrate the effectiveness of the proposed modeling technique and show the importance of not ignoring the effect of the rigid tips.

An important concept proposed in the early stage of robot path planning field is the shrinking of the robot to a point and meanwhile expanding of the obstacles in the workspace as a set of new obstacles. The resulting grown obstacles are called the Configuration Space (Cspace) obstacles. The find-path problem is then transformed into that of finding a collision free path for a point robot among the Cspace obstacles. However, the research experiences obtained so far have shown that the calculation of the Cspace obstacles is very hard work when the following situations occur: 1. both the robot and obstacles are not polygons and 2. the robot is allowed to rotate. This situation is even worse when the robot and obstacles are three dimensional (3D) objects with various shapes. Obviously a direct path planning approach without the calculation of the Cspace obstacles is strongly needed. This paper presents such a new real-time robot path planning approach which, to the best of our knowledge, is the first one in the robotic community. The fundamental ideas are the utilization of inequality and optimization technique. Simulation results have been presented to show its merits.

The 3-PS structure features one rigid body (platform) connected to another rigid body (base) by means of three kinematic chains (limbs) of type PS (P and S stand for prismatic pair and spherical pair, respectively). All the 3-degree-of-freedom parallel manipulators with three connectivity-5 limbs, each one constituted of one passive (i.e. not actuated) prismatic pair, one passive spherical pair and one actuated kinematic pair of any type, become 3-PS structures when the actuated pairs are locked. Direct kinematics of this class of manipulators is tied to the properties of the 3-PS structure. In particular, the direct position analysis is tied to the assembly modes of the 3-PS structure; whereas the determination of the singularities of the direct instantaneous problem is tied to the determination of the singular geometries of the 3-PS structure, where instantaneous relative motions between platform and base are possible. The solution of these two problems is necessary both for designing the manipulators and for controlling them during motion. This paper deal with the determination of the singular geometries of the 3-PS structure.

Instantaneous kinematics and singularity analysis of a class of three-legged, 6-DOF parallel manipulators are addressed in this paper. A generic method of derivation of reciprocal screw and consequently, the instantaneous kinematics model is presented. The advantage of this formulation is that the instantaneous kinematics model possesses well-defined geometric meaning and algebraic structure. Singularity analysis is performed under three categories, namely forward, inverse and combined singularities. A new concept of Passive Joint Plane is introduced to correlate the physical structure of the manipulator and these geometric conditions. In the inverse kinematic analysis, a new approach is introduced. At each leg end point a characteristic parallel- epiped is defined whose sides are the linear velocity components from three main joints of the leg. An inverse singularity occurs when the volume of this parallelepiped becomes zero. Examples are demonstrated using RRRS and RPRS-type parallel manipulators.

A survey of the field of control for flexible multi-link robots is presented. This research area has drawn great attention during the last two decades, and seems to be somewhat less “attractive” now, due to the many satisfactory results already obtained, but also because of the complex nature of the remaining open problems. Thus it seems that the time has come to try to deliver a sort of “state of the art” on this subject, although an exhaustive one is out of scope here, because of the great amount of publications. Instead, we survey the most salient progresses – in our opinion – approximately during the last decade, that are representative of the essential different ideas in the field. We proceed along with the exposition of material coming from about 119 included references. We do not pretend to deeply present each of the methods quoted hereafter; however, our goal is to briefly introduce most of the existing methods and to refer the interested reader to more detailed presentations for each scheme. To begin with, a now well-established classification of the flexible arms control goals is given. It is followed by a presentation of different control strategies, indicating in each case whether the approach deals with the one-link case, which can be successfully treated via linear models, or with the multi-link case which necessitates nonlinear, more complex, models. Some possible issues for future research are given in conclusion.

To fully utilize the information from the sensors, this paper proposes a new sensor-fusion technique where the data sets for the previous moments are properly transformed and fused into the current data sets to enable an accurate measurement. Exploration of an unknown environment is an important task for the new generation of mobile service robots. The mobile robots may navigate by means of a number of monitoring systems such as the sonar-sensing system or the visual-sensing system. Note that in the conventional fusion schemes, the measurement is dependent on the current data sets only. Therefore, more of sensors are required to measure a certain physical parameter or to improve the accuracy of the measurement. However, in this approach, instead of adding more sensors to the system, the temporal sequence of the data sets are stored and utilized for the accurate measurement. The theoretical basis is illustrated by examples and the effectiveness is proved through the simulations and experiments. The newly proposed, STSF (Space and Time Sensor Fusion) scheme is applied to the navigation of a mobile robot in an unstructured environment, as well as in structured environment, and the experimental results show the performance of the system.

In this paper, wall-climbing robots with a single suction cup are studied. The robots mainly consist of three parts: A vacuum pump, a sealing mechanism and a driving mechanism. Basic conditions that the robots can adhere to and move reliably on a vertical surface are first established, then the sealing mechanism with an air spring and regulating springs is analyzed, and the synthetic rigidity formula for the sealing loop is obtained. Finally, two application examples are given: One for the ultrasonic inspection of cylindrical stainless steel nuclear storage tanks, and the other for cleaning high-rise buildings.

Vision-aided flexible link robot positoning using the Camera Space Manipulation (CSM) method is developed. The primary motivation for this work is to use an autonomous vision-aided robotic system to pick-up and accurately move a flexible object that it encounters. The work consists of analytical and experimental investigation of the performance of CSM for a kinematic model of the PUMA manipulator with a flexible structure at the wrist which accounts for the gravitation. Trade-offs between camera view parameters and axial deflection model parameters were investigated. View parameter reestimation and maneuvering resulted a very accurate placement of the end-effector at the target.

This paper presents an adaptive impedance control scheme for an $n$-link constrained rigid robot manipulator without using the regressor. In addition, inversion of the estimated inertia matrix is also avoided and the new design is free from end-point acceleration measurements. The dynamics of the robot manipulator is assumed that all of the matrices in robot model are unavailable. Since these matrices are time-varying and their variation bounds are not given, traditional adaptive or robust designs do not apply. The function approximation technique is used here to represent uncertainties in some finite linear combinations of the orthogonal basis. The dynamics of the output tracking can thus be proved to be a stable first order filter driven by function approximation errors. Using the Lyapunov stability theory, a set of update laws is derived to give closed loop stability with proper tracking performance. A 2 DOF planar robot with environment constraint is used in the computer simulations to test the efficacy of the proposed scheme.

A biped walking robot, named EP-WAR3 (Electro-Pneumatic-Walking-Robot), has been designed, built and tested at LARM (Laboratory of Robotics and Mechatronics) in Cassino. EP-WAR3 is provided with a suitable binary pneumatic actuation in order to be controlled through a common PLC (Programmable-Logic-Controller) as an event-based system in an on/off environment. The walking stability of the biped robot is obtained by using suction-cups, which are installed on the underside of each foot. EP-WAR3 is able to walk along a straight line with two different step sizes, turn right and left, and to climb and to descend stairs. A suitable motion analysis and programming technique of the PLC controller is proposed in order to obtain suitable walking capabilities and flexibility of the robot.

This paper presents a method of autocalibrating a stereo system using the full parameterization of the fundamental matrix. The technique is an improvement of classical autocalibration methods that use corresponding points and it is aimed to reduce the number of processes required to achieve a satisfactory calibration. An analysis of the matching points used in the calibration process is carried out in order to select them based upon their quality or reliability. Using this technique, a significant reduction of the convergence errors in the calibration process is obtained.

The work presented here concerns the design and manufacturing possibility of a polymodular autonomous microrobot moving like an earthworm. The study consists of the development of a new locomotion actuator based on shape memory alloys. This actuator is formed by a flexible frame forced in post-buckling by its assembly on a rigid skeleton cage. This mechanical structure, forming a knot in the microrobot chain, uses its two post-buckling equilibrium positions to achieve first the support and second the local lengthening enabling the movement of the microrobot. Initially, we present the technical solution designed for the locomotion actuator and the mechanical characterization of its malleable structure. Then, we present the instrumentation by educated SMA of a malleable structure and the capability of microrobot locomotion task in a vertical tube.

In this paper, we demonstrate a multi-phase genetic programming (MPGP) approach to an autonomous robot learning task, where a sumo wrestling robot is required to execute specialized pushing maneuvers in response to different opponents' postures. The sumo robot used has a very simple, minimalist hardware configuration. This example differs from the earlier studies in evolutionary robotics in that the former is carried out on-line during the performance of a robot, whereas the latter is concerned with the evolution of a controller in a simulated environment based on extended genetic algorithms. As illustrated in several sumo maneuver learning experiments, strategic maneuvers with respect to some possible changes in the shape and size of an opponent can readily emerge from the on-line MPGP learning sessions.