We demonstrate an approach for collision- and oscillationfree navigation of multiple robots or virtual agents amongst each other. Each entity acts independently and uses only both the position and velocity of nearby entities to predict their future trajectories in order to avoid collisions. Entities take into account that the other entities are responding to them likewise to prevent oscillations. Categories and Subject Descriptors I.2.9 [Artificial Intelligence]: Robotics; I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence General Terms Algorithms, experimentation Keywords Planning, multi-robot systems
Jamie Snape, Stephen J. Guy, Jur P. van den Berg