Collision Detection and Prediction Using a Mutual Configuration State Approach

A configuration state approach is presented that simplifies the mutual collision analysis of objects with known shapes that move along known paths. Accurate and fast prediction of contact situations in games such as robot soccer enables improved anticipatory and corrective actions of the state estimation, control and planning processes. An overview is given how advantage is taken of the configuration state approach in order to deal with collision state correction and collision avoidance in the robot soccer control system MI20 developed at our university.