We examine the properties of an integrated, intelligent controller for multifingered manipulators. The discussion addresses the dynamic constraints on computation and resource allocation in systems which must continually adjust objectives and the set of behaviors employed to attain them. We present a means of specifying general force domain tasks and include techniques for reflexive, sensor-based control and for optimal combinatoric grasp synthesis as alternative grasp formation mechanisms. Examples are presented to illustrate the acquisition of information, the incremental refinement of models, the use of uncertain information in reflexive grasp formation, and the use of complete geometric information in optimal grasping.

This paper proposes an efficient algorithm for computing finger forces involved in a three-dimensional objects grasp. Effective finger force computation is necessary for the successful manipulation on an object by a multifingered robot hand. Based on previous works, the stability forces are computed as a solution of an optimization problem. This optimization problem is mapped into a linear quadratic problem under inequality constraints. We propose a new approach for this problem: the problem is solved as a minimal distance calculation problem in the forces space. The results obtained by simulation demonstrate the efficiency and the numerical stability of the method. This method is used with the LMS mechanical hand as a component of the global control strategy dedicated to the object manipulation.