Abstract: A collision-free, coordinated motion planning is presented for multiple robots in a shared workspace. The proposed method consists of two phases. In the first phase, a collision-free path with respect to stationary obstacles for each robot is planned according to task requirements without considering the conflicts among the robots by employing the probabilistic roadmap (PRM) method. In the second phase, the robot path is described by continuous configuration sequences, the state space of the whole system is constructed, and the investigated problem is transformed into a continuous path search problem in a high dimensional state space. On this basis, a collision-avoidance strategy for multiple robots, a dynamic modification method of motion sequence priorities, and an improved A^* algorithm are proposed, and collision-free, coordinated motion planning for multiple robots are implemented. The feasibility and effectiveness of the proposed approach are demonstrated with 2 simulation cases. The results show that the proposed method can rapidly, effectively obtain the coordinated collision-free paths for multiple robots.