An approach to the control of multiple coordinated redundant robots working in an environment with obstacles is developed in which the extended task space technique is presented for establishing the dynamic equations of the coordinated redundant robots using the redundant degrees of freedom. This approach can be used to control the motion of an object while avoiding obstacles. Three subsystem error equations are generated: a position error subsystem, an internal force error subsystem, and a constraint error subsystem. The adaptive law is derived to estimate the unknown parameters of the coordinated redundant robots and the object in terms of the three error subsystem equations.>

The optimal torque control problem of a system of multiple coordinated redundant manipulators using joint redundancy is discussed. A local optimal control law and a global optimal control law, both minimizing torque loading at the joints of the multiple redundant manipulators, are presented. The Pontryagin minimum principle is used to solve the global optimal control problem. The local and global optimal control approaches can be used to control the motion of an object held by the redundant manipulators and the internal forces which do not contribute to the motion of the object. From the task space point of view, the errors in the object motion and in the internal forces converge asymptotically to zero. From the joint space point of view, the joint torques which are provided by the joint actuators of the multiple redundant manipulators are the global minimum under the global optimal controller or the local minimum under the local optimal controller.>

An approach to motion and force control of multiple coordinated robot arms based on an adaptive scheme is developed. The adaptation law uses Popov hyperstability theory to estimate online the uncertain parameters of multiple robot arms and payload. The approach can be used to control the motion of an object held by the arms, the contact forces between the object and the environment, and the internal forces that do not contribute to the motion and the contact forces. Three subsystem error equations are generated, i.e. a position error subsystem, a contact force error subsystem, and an internal force error subsystem. The unknown parameters of the multiple coordinated robot arms and the object are estimated in terms of the three error subsystem equations. It is shown that the proposed adaptive control scheme improves the position and internal and contact force tracking accuracy of a class of systems with an uncertain knowledge of the dynamic model.>