It is with much regret that I've to resign from the editorship of Robotica after 23 years of service because of ill-health and advanced age. I hope the new editor Greg Chirikjian will make further progress in expanding the scope of the journal and responding to the challenges of our times.

Organic transistors have been developed to a state where, combined with pressure sensors, they may prepare the way for the development of an artificial skin that could be utilised by robots and in automation applications.

In this paper, we have developed and implemented a system that combines autonomous obstacle avoidance with force-reflective tele-operation. In this system, a tele-operated mobile robot is controlled by a local two-degrees-of-freedom force-reflective joystick that a human operator holds while he is monitoring the screen. The force-reflective joystick transforms the relation between a mobile robot and the environment to the operator as a virtual force. A virtual force is generated in the form of a new collision vector and reflected to the operator, which makes the tele-operation of a mobile robot safe from collision in an uncertain and obstacle-cluttered remote environment. A mobile robot controlled by a local operator usually takes pictures of remote environments and sends the images back to the operator over the Internet. Because of limitations of communication bandwidth and the narrow view-angles of the camera, the operator cannot observe shadow regions and curved spaces. To overcome this problem, a new form of virtual force is generated along the collision vector according to both distance and approaching velocity between an obstacle and the mobile robot, which is obtained from ultrasonic sensors. This virtual force is transferred back to a master (two degrees of freedom joystick) over the Internet to enable a human operator to feel the geometrical relation between the mobile robot and the obstacle. It is demonstrated by experiments that this haptic reflection improves the performance of a tele-operated mobile robot significantly.

The properties of the software part of a control system implemented in walking robots are described. The paper presents also the navigation method elaborated for a family of hexapods. The general structure of the programming system is given together with the functional description of the system modules. The event-based action scheme of the central part of the system responsible for data distribution and high level navigation is addressed in detail.

We developed a new type of a human-sized BWR (biped walking robot) driven by the closed-chain type of a joint actuator. Each leg of the BWR is composed of three pitch joints and one roll joint. In all, a 12 degree-of-freedom robot, including four arm joints, was developed. The BWR was designed to walk autonomously; it is actuated by small 90W DC motors/drivers and is has DC batteries and controllers. A new type of the joint actuator for the BWR is composed of the four-bar-link mechanism driven by a ball screw which has high strength and high gear ratio despite its light weight.

In this paper, analyses on the four-bar-link mechanism applied to the joint actuator and on the structure of the BWR are presented. Through walking experiments of the BWR, the superior trajectory-tracking ability of the proposed joint actuator is validated.

The paper addresses kinematic and geometrical aspects of the Orthoglide, a three-DOF parallel mechanism. This machine consists of three fixed linear joints, which are mounted orthogonally, three identical legs and a mobile platform, which moves in the Cartesian x-y-z space with fixed orientation. New solutions to solve inverse/direct kinematics are proposed, and we perform a detailed workspace and singularity analysis, taking into account specific joint limit constraints.

Tools that make it possible to measure the performances of manipulators are essential in many technical applications, for instance, when the optimal path to accomplish a task has to be chosen, or when different manipulator architectures have to be compared. This paper proposes some indices that fully describe the passive dynamic performances of manipulators with two degrees of freedom (dof). The pro- posed indices make it possible to compare the passive dynamic performances of different manipulator architectures, which can perform the same tasks, and can be used to build diagrams which highlight the effects of variations in the manipulator geometry on the manipulator dynamics. These features make them easy to be used in a design context. Finally, some applications of the proposed indices will be presented and discussed.

The paper presents a control oriented dynamic formulation for constrained robotic systems. We refer to this formulation as control oriented since it can be directly applied to design a model-based tracking control strategy. It enables one to obtain dynamic models for systems subjected to material constraints and non-material referred to as program constraints. The dynamic model for a constrained robot, where equations of program constraints specify tasks and motion requirements, is used as a program motion planner for a controller design. The formulation provides a theoretical framework, which is a basis for the study of robot performance of constrained tasks. Examples of developments of control oriented dynamic models are presented, and their applications for control are demonstrated and discussed.

This paper focuses on the structure and property of the singularity loci of the 3/6-Stewart-Gough platform for general orientations. Based on the singularity kinematics principle, a planar singularity-equivalent-mechanism is proposed, by which the complicated singularity analysis of that parallel mechanism is transformed into a simpler position analysis of the planar mechanism. All the possible positions of the planar mechanism form the singularity loci of the 3/6-Stewart-Gough manipulator. The result shows that the singularity equation become quite simple moreover the structure and property of the singularity loci are also identified and explained. For the most general orientations of the typical 3/6-Stewart-Gough platform, the singularity locus equation is a special irresolvable polynomial expression of degree three, which in infinite parallel principal sections includes a parabola, four pairs of intersecting straight lines and infinity of hyperbolas. This result is beneficial to analysis of the similar issue of other Stewart-Gough manipulators.

Non-overconstrained 3-dof spherical parallel manipulators of a structural type 3-RCC, 3-CCR, 3-CRC are introduced. The mechanism has three limbs that connect in parallel the moving platform to the fixed base. Each limb is an opened kinematic chain made of a sequence of one revolute pair R and two cylindrical pairs C. The orientation of the end-effector is obtained by actuating simultaneously the three limbs. A structural type analysis and synthesis, which is based on the algebraic properties of a Lie group of the displacement set, is employed to find the geometrical conditions for the assembly of these spherical parallel mechanisms and also the structurally singular configurations. Then an enumeration of the structural types is given and remarkable special cases of orientational mechanisms are also described, namely 3-HGR, 3-RGH and 3-HGH.

This paper presents a robust input shaping technique that significantly reduces (almost eliminates) the residual vibration of manipulation systems typified by a flexible-jointed robot manipulator. The technique consists of two stages. In the first stage, a ramp function is superimposed onto the main trajectory to be preshaped. In the second stage, the outcome of the first stage is convolved with a sequence of two impulses. The robustness of the technique to the uncertainties in the system natural frequency and damping ratio are quantified through simulation and experimental evaluation. Simulation and experimental results demonstrate that the technique is not only effective in reducing the residual vibrations, but also it is robust to the uncertainties of as ∓35% from the ideal value of the system natural frequency. Further, it has been found that the proposed input shaping technique is insensitive to the uncertainties in the damping ratio. The results allow us to suggest that the proposed technique is versatile and robust enough to apply it to the motion design of any flexible-jointed manipulation system making a point-to-point motion.

We present the positional abilities of a humanoid manipulator based on an improved kinematical model of the human arm. This was synthesized from electro-optical measurements of healthy female and male subjects. The model possesses three joints: inner shoulder joint, outer shoulder joint and elbow joint. The first functions as the human sternoclavicular joint, the second functions as the human glenohumeral joint, and the last replicates the human humeroulnar rotation. There are three links included, the forearm and the upper arm link which are of a constant length, and the shoulder link which is expandable. Mathematical interrelations between the joint coordinates are also taken into consideration. We determined the reachability of a humanoid arm, treated its orienting redundancy in the shoulder complex and the positional redundancy in the shoulder-elbow complexes, and discussed optimum configurations in executing different tasks. The results are important for the design and control of humanoid robots, in medicine and sports.

The system here described has the capability of generating range images that include robot motion. The system has two main modules, the motion and the image generator. Motion is modeled using a Bezier's curve method. To compute a range value corresponding to a pixel image, the robot position in the coordinated system is obtained from trajec-tory generation. In this way, distortion is produced in the image, or sequence of images, as a consequence of motion. The obtained range images represent scenes perceived by the robot from a specific location or during a specified dis-placement in a very “real” view.

In this paper, a simple and convenient method – Recursive Matrix method – is proposed for kinematic and dynamic analysis of all types of complex manipulators. After addressing the principle of the method, an example – a 3-DOF parallel manipulator with prismatic actuators – is demonstrated for the efficiency of the method in solving kinematic and dynamic problems of complex manipulators. With the inverse kinematic solutions, the inverse dynamic problem is solved with the virtual powers method. Matrix relations and graphs of the acting forces and powers for all actuators are analysis and determined. It is shown that the proposed method is an effective mean for kinematic and dynamic modelling of parallel mechanisms.

This paper presents a singularity analysis for a 3-DOF parallel manipulator with R-P-S (Revolute-Prismatic-Spherical) joint structure. All three types of singularities are investigated with most attention paid for direct kinematics singularities (DKS). The loci of inverse kinematics and combined singularities are identified using a new approach. The equation of DKS is defined first from the condition of existence of an instantaneous motion. The geometrical method is used to find the loci of trajectories corresponding to DKS-s. As a result of these investigations, an optimization procedure was proposed of a robot design in order to have an enlarged singularity free part of the working space. The construction of a singularity free path is discussed without changing the robot trajectory by selecting the appropriate inverse kinematics task solution.