This paper presents original concepts to represent and deal with kinematic constraints imposed on mobile robots. This is based on formulations derived originally for the dynamics of multibody systems and applied to robotic manipulator arms. This representation has several advantages such as its ability to deal with kinematically excited modes, derive a physically meaningful state-space representation, and find corresponding constraining forces. The paper shows how to apply these concepts to different types of mobile robots such as those modelled as unicycle or bicycle.

Natural and manmade-shared control system examples are presented. Analogies and differences among these systems are discussed. Then, basic requirements for successful cooperation in shared-control systems are identified. Finally, an ambitious general framework for human-machine shared control is proposed. Only intelligent machines that would otherwise function autonomously are considered.

This paper presents kinematic and dynamic analysis of a holonomic vehicle with continuously-variable transmission. Four ball wheels, independently actuated by DC motors, enable for moving the vehicle in any direction within the plane and rotating it around its center. The angle between the two beams holding the balls can be changed to alter the gear ratio and other dynamical characteristics of the vehicle. This feature is exploited in augmenting the vehicle stability, optimizing output power, selecting an appropriate gear ratio, and in impedance matching. A simple adaptive friction-compensation-based controller is proposed to handle the complex friction properties.

A intelligent control architecture for semi-autonomous systems is presented. This has been motivated by the need for an architecture to differentiate between users of different perceptual and cognitive capabilities when interacting and using machines with automatic features. A good example for that is a robotic wheelchair. The proposed architecture evolves around three variables: degree of autonomy to reflect the user capabilities, user's level of confidence in commanding the machine, and strength of conflict between the user's command and the machine's autonomous response. The analogy between this architecture and horseback riding is demonstrated.

Much research work has been dedicated in the last years to the control of manipulators (rigid, flexible, or with flexible joints) with stationary bases. To enlarge the workpace of manipulators, to avoid obstacles or singularities, and to open new application areas a new generation of robots has been created. This class contains the mobile manipulator which is basically a manipulator mounted on a vehicle. In this case not only the kinematics but also the dynamics of the manipulator is affected by the base motion and coupling forces. Since it is usually difficult, if not impossible, to model these, a novel control method is proposed in this paper. The main idea is to design a regular manipulator motion controller that considers the dynamic coupling effects as external disturbances. Here, robustness is assured by estimating and compensating for the coupling effects together with any other disturbances rather than computing them.

This paper deals with environmental modeling of constrained industrial robots. The considered constraints arise during the contact of a robot with an external object in its environment. Such a situation happens mainly if a robot is desired to apply forces and/or torques on an external surface for machining purposes, or if it is used to mate parts with limited clearances. A novel method that is based on task-space formulations and on the modes-of-motion approach is developed to represent constraints and constrained tasks. It is shown that this methodology can handle constrained motions generally without having any of the substantial limitations of other proposed methods. Special attention is paid in this paper to surface forces, a general reduction method is given where its results acknowledge the existence of these forces. Constrained systems under surface forces maybe not solvable: the sufficient and necessary condition for the existence of a solution is given and then illustrated by an example.>

Diiririg marly industrial robot appli- cations, the end effector coiiies into contact with clxtcmial objects. The analysis and control of tlir constrained iiiotioii of tlie end effector reprcscmt an important class of robotic. problcmis. In this papc'r, we present a general irictliod to derive the state space form for the constrailled systcmi in tlw presence of generally rriodclccl friction forces. Trllis method is based 011 Jourdain's priiiriplc. for wii- strained systems where the total virtual power of the constraint forces axid torques is zero. As a rc- stilt, general and efficient explicit expressions for tlie acceleration statc vector and for tlw wiiitwct forcm are available.