Null space velocity control is essential for achieving good behaviour of a redundant manipulator. Using the dynamically consistent pseudo-inverse, the task and null space motion and forces are decoupled. The paper presents a globally stable null space velocity controller and the gradient projection technique in conjunction with the dynamically consistent pseudo-inverse. The physical meaning and influence of the compensation terms in null the space velocity controller are explained. The performance of the proposed null space controller is tested on 4. d.o.f planar redundant manipulator interacting with the environment.

The focus of this work is on sensor-based motion planning for a biped robot operating in a scene with unknown obstacles. Using on-line sensor information about its surroundings, the robot negotiates obstacles on its way. Depending on the obstacle's shape, size, and location, the decision-making system chooses an appropriate walking pattern. The robot then negotiates the obstacle and resumes stable motion. The walking pattern is chosen from a small number of precomputed patterns that together cover a reasonably wide range of possible situations. The overall control strategy is based on the concept of Zero Moment Point. Each precomputed pattern guarantees dynamically stable motion; its stability is obtained by adjusting the swing leg center of mass and hip position trajectories. The approach is fast enough for real-time implementation. Simulation experiments demonstrate stability of motion when negotiating various obstacles.

One of the important issues of walking machine active force control is a successful distribution of the body force to the feet to prevent leg slippage. In this paper, a new force distribution method, the Friction Constraint Method (FriCoM), is introduced. The force distribution during the walking of a typical quadruped crawl gait is analyzed by using the FriCoM. Computation results show that the distributed forces of the feet are continuous during the walking. This reflects the change of the force distribution during actual conditions. The comparison with a pseudo-inverse method shows that the FriCoM is more practical. The FriCom also requires less computation time than that by an incremental optimization method. Some problems, such as the singularity in the application of the FriCoM, are discussed. The FriCoM will be used in the active force control of a quadruped robot that is taken as a platform for the research on the study of terrain adaptation.

This paper deals with the stability of a rigid body under multiple contact forces. First, the problem is considered at the force planning level, and the stability of a force distribution is formulated. For this problem, the stiffness tensor is derived, and its basic properties are analyzed. Necessary and sufficient conditions for stability of a force distribution are established in an analytical form. These conditions, considered under unilateral frictional constraints, are studied on an illustrative example. Next, it is shown that stabilization of an unstable force distribution can be done by a simple control law. The stability conditions for this control law are formulated by transforming the stiffness tensor to the center of stiffness. Finally, conclusions on the contradiction between the Liapunov stability and the contact stability of the objects are drawn.

Often inspection and maintenance work involve a large number of highly dangerous manual operations, especially within industrial fields such as shipbuilding and construction. This paper deals with the autonomous climbing robot which uses the “caterpillar” concept to climb in complex 3D metallic-based structures. During its motion the robot generates in real-time the path and grasp planning in order to ensure stable self-support to avoid the environment obstacles, and to optimise the robot consumption during the inspection. The control and monitoring of the robot is achieved through an advanced Graphical User Interface to allow an effective and user friendly operation of the robot. The experiments confirm its advantages in executing the inspection operations.

Three standard types of obstacles are selected for the study: vertical step, isolated wall and ditch. When one of these obstacles is detected in the direction of motion, a quadruped reads only one or two parameters of obstacle and then calculates corner distances, which makes it possible to determine the locomotion sequences before it reaches the obstacle. Vertical step is fully defined by one parameter, step height. An isolated wall is defined by two parameters: wall thickness and vertical height. A ditch is defined by ditch width. The dimensions of the suggested model are used to calculate the crossing performance of a quadruped, for the purpose of comparison. All the crossing gaits are illustrated by simulation figures. Based on the results of this research, a more comprehensive study in the walking over a combination of these obstacles can be conducted and a four-legged walking machine which moves with mobility can be designed.

A new design philosophy for the transmission of robot manipulators is proposed and an example of a practical implementation is presented. The philosophy combines the advantages of conventional geared robots in terms of relocating the actuators away from the joints and the alternative direct-drive approach. The gimbal drive is an example of a non-linear transmission where there is negligible friction, no backlash or compliance and which provides a varying reduction ratio for gravity compensation and for some dynamic decoupling. The gimbal drive is implemented on the three degrees of freedom SPRINTA prototype. Static repeatability, as well as tracking error and dynamic repeatability for the industrial goalpost test are measured. The performance demonstrates the potential of such a type of robot.

An approach to planning time-optimal collision-free motions of robotic manipulators is presented. It is based on using a negative formulation of the Pontryagin Maximum Principle which handles efficiently various control and/or state constraints imposed on the manipulator motions, which arise naturally out of manipulator joint limits and obstacle avoidance. This approach becomes similar to that described by Weinreb and Bryson, as well as by Bryson and Ho if no state inequality constraints are imposed. In contrast to the penalty function method, the proposed algorithm does not require an initial admissible solution (i.e. an initial admissible trajectory) and finds manipulator trajectories with a smaller cost value than the penalty function approach. A computer example involving a planar redundant manipulator of three revolute kinematic pairs is included. The numerical results are compared with those obtained using an exterior penalty function method.

In this paper we consider the problem of coordinating multiple motion devices for welding. We focus on the problem of coordinating a three-axis positioning table and a six-axis manipulator. The problem is complex as there are nine axes involved and a number of permutations are possible which achieve the same movements of the weld torch. The system is redundant and the robot has singular configurations. As a result, manual programming of the robot system is rather difficult to complete.

Our approach to the coordination problem is based on a subdivision of tasks. The welding table is coordinated to align the weld point surface to be anti-parallel to the gravity direction. The six-axis robot is constrained to move the weld torch along the weld trajectory. Robot coordination is achieved by placing the positioning table in a good maneuverability position, i.e. far from its singular configurations and far from the motion limits of the six-axis arm and the motion limits of the track. While considering multiple criteria, including the welding orientation, a Genetic Algorithm was employed to globally optimize six relevant redundant degrees of the multiple robotic system for welding. The joint angles of the arm were generated by inverse kinematics. Our method of redundancy coordination is superior to pseudo-inverse techniques, for it is more global and accurate.

This paper introduces a new approach to robust path planning for mobile robots entirely based on information from environment perception sensors. This method avoids the use of odometry which leads to the accumulation of errors resulting from the robot's position computing. We proceed as follows: we create regions inside which the robot detects the same obstacle segments. A node graph represents all the regions and their links. Then a planning algorithm is used to find a path which joins a start to a goal region. The final stage consists in applying a robust robot motion control as regards the uncertainties of the environment model. This approach contributes to a control system for indoor robots which is environment referenced. The sensors we deal with are first a continuous laser or ultrasonic scanning system, then a discrete ultrasonic belt whose limits of use we show.

In Part 1 of this paper we have investigated numerically the quasiperiodic and frequency locked solutions of mathematical model of a robot with one degree of freedom. In this paper we extend our investigations to the region of transient chaos. The zones of chaotic transients are very broad and lie beyond the parameter range of engineering significance. Transiently chaotic zones exhibit a complex structure, fractally intertwined with tongues of regular pattern and cover a broad range of control parameter L. The crisis point for the onset of sustained chaos lies extremely far from the point of onset of transient chaos.

This paper attempts firstly to derive a mathematical model of the dynamics of a set of dual fingers with soft and deformable tips which grasps and manipulates a rigid object with some dexterity. To gain a physical insight into the problem, consideration is restricted to the case that the motion of the whole system is confined to a horizontal plane. Secondly on the basis of the derived model it is shown that the rotation of the object can be indirectly controlled by the change of positions of the center points of both contact areas on the object. Then, each of the center points of contact areas can be positioned by inclining the last link of each corresponding finger against the object. It is further shown that, when both forces of pressing the object becomes almost equal, the equation of motion of the object in terms of rotational angles assumes the form of a harmonic oscillator with a forcing term, which can be regulated coordinately by the relative angle between the two last links contacting with the object. It is also shown that dynamics of this system satisfy passivity. Finally, design problems of control for dynamic stable grasping and enhancing dexterity in manipulating things are discussed on the basis of passivity analysis.

In this paper, the kinematics and dynamics of a parallel manipulator with a new architecture supposed to be used as a moving mechanism in a flight simulator project is studied. This manipulator with three independent degrees of freedom consists of a moving platform connected to a based platform by means of three legs. Kinematic solutions for this manipulator at position, velocity and acceleration levels are obtained. Moreover, the dynamical equations of motion of the manipulator are determined using Newton-Euler's equations and applying the natural orthogonal complement (NOC) method. Using kinematics and dynamics and also performing simulation for different manoeuvres of moving platform, the motion and the actuator forces of the legs are obtained.

Machine vision-based sensing enables automatic hover stabilization of helicopters. The evaluation of image data, which is produced by a camera looking straight to the ground, results in a drift free autonomous on-board position measurement system. No additional information about the appearance of the scenery seen by the camera (e.g. landmarks) is needed. The technique being applied is a combination of the 4D-approach with two dimensional template tracking of a priori unknown features.

The Clifford algebra for the group of rigid body motions is described. Linear elements, that is points, lines and planes are identified as homogeneous elements in the algebra. In each case the action of the group of rigid motions on the linear elements is found. The relationships between these linear elements are found in terms of operations in the algebra. That is, incidence relations, the conditions for a point to lie on a line for example are found. Distance relations, like the distance between a point and a plane are found. Also the meet and join of linear elements, for example, the line determined by two planes and the plane defined by a line and a point, are found. Finally three examples of the use of the algebra are given: a computer graphics problem on the visibility of the apparent crossing of a pair of lines, an assembly problem concerning a double peg-in-hole assembly, and a problem from computer vision on finding epipolar lines in a stereo vision system.

This paper presents a new method of analyzing and calculating the uncertainty configurations of parallel manipulators. The main feature of the method that makes it attractive with respect to the methods available in the literature, is its ability to obtain simplified uncertainty configuration equations. It's very useful in real-time control. The application of the proposed method is illustrated in detail by two examples.

In this paper neural network (NN) control techniques for non-model based PD controlled robot manipulators are proposed. The main difference between the proposed technique and the existing feedback error learning (FEL) technique is that compensation of robot dynamics uncertainties is done outside the control loop by modifying the desired input trajectory. By using different NN training signals, two NN control schemes are developed. One is comparable to that in the FEL technique and another has to deal with the Jacobian of the PD controlled robot dynamic system. Performances of both controllers for various trajectories with different PD controller gains are examined and compared with that of the FEL controller. It is shown that the new control technique performed better and robust to PD controller gain variations.

Teleoperation, one of the oldest areas of robotics, has experienced considerable growth in the last two decades. Main causes for this trend are the need for increased safety levels for human operators and lower production costs. In this work, a three d.o.f. local manipulator (two d.o.f. for force and one d.o.f. for torque) is developed. This hand controller, intended for robot or mobile teleoperation systems, has force reflection in two axes and torque reflection in the third axis. using a robotic hand developed at INAUT as a remote device, laboratory experiments on each axis (one at a time) have shown good results. An impedance controller at the remote system allows one to carry out interactive tasks with the environment such as polishing, insertion and grinding, where it is necessary to control and accommodate the interaction forces and torques in order to avoid hazards for both the manipulated objects and the remote robot.

Although robots have some kinematic and dynamic constraints such as the limits of the position, velocity, acceleration, jerk, and torque, they should move as fast as possible to increase the productivity. Researches on the minimum-time trajectory planning and control based on the dynamic constraints assume the availability of full dynamics of robots. However, the dynamic equation of robot may not often be exactly known. In this case, the kinematic approach for the minimum-time trajectory planning is more meaningful. We also have to construct a controller to track precisely the minimum-time trajectory. But, finding a proper controller is also difficult if we do not know the explicit dynamic equations of a robot.

This paper describes an optimization of trajectory planning based on a kinematic approach using the evolution strategy (ES), as well as an optimization of a sliding mode tracking controller using ES for a robot without dynamic equations.