Motion information based avoidance control for 3-D multi-agent systems

This paper presents a closed-form cooperative avoidance control design for 3-dimensional rigid-body agents. New avoidance functions are designed to integrate the collision risk with the addition of motion and posture information. Time-varying sensing regions enable the conflict resolution actions only when there is a potential risk of collision, thus further reducing some unnecessary interference with other objectives. These characteristics lead to the less-conservative avoidance response, smoother maneuvers, and milder motion trajectories. The stability of the proposed method is analyzed and the corresponding proofs and guaranties for collision-free maneuvering, are provided. Performance comparisons are presented for two simulation scenarios, which illustrate the effectiveness and reliability of the new method.