OPTIMAL MOTION PLANNING OF A MULTIPLE-ROBOT SYSTEM BASED ON DECOMPOSITION COORDINATION

In this paper, a solution to the problem of the optimal control of multiple-robot systems in the presence of obstacles is presented. All the robots are subject to state and control constraints. The proposed method is based on nonlinear programming and decomposition coordination. The optimal control problem is solved by satisfying the constraints on the states and the controls as well as those on the obstacles. In order to generalize the problem, N(R) robots with N(O) obstacles are considered. No distinction is made between the different types of robots. This allows the generalization of this problem to the mobile robots case with a known work space. In order to generate the optimal control that accounts for the presence of obstacles, the Euclidian distance between convex and compact objects is used. All robot segments and obstacles are considered as convex and compact sets in R3. An additional property is used to obtain a distance function C1. This property is essential in elaborating the decomposition-coordination algorithm. The proposed method is programmed on a VAX station and is used as a CACSD tool for the path planning of two modular assembly robots PAMIR developed at the Control and Automation Laboratory of the ESIEE.