Fast and Efficient Trajectory Generation for Ball Interception Using Motion Primitives

Ball-interception is an vital strategic operation for most soccer competitions. This paper proposes a trajectory generation method that combines trajectory prediction and motion primitives for a soccer-robot to perform ball-passing and interception operations during a RoboCup competition. Compared to existing algorithms, our method has better real-time replanning capability as well as better obstacle avoidance performance. Our method is thoroughly tested in both simulation and real-world environments, and the results show that our approach is both efficient and robust.