Static And Dynamic Collision Avoidance For Multi-robots In Crowded Environment

Résumé: A team of multiple mobile robots that work in parallel offers a number of advantages over single robot systems. Multiple robots have the potential to finish a given task faster than a single robot and they are able to perform some tasks that are outside the scope of a single-robot system. As a primary problem, motion-level conflict resolution requires that the robots avoid collisions not only with static obstacles but also with other robots. If a robot regards its neighboring robots as static obstacles, the conflict among the robots is inevitable in some cases. It becomes worse in crowded multi-robot environments. Therefore, the use of multiple robots in the same workspace requires the necessity to coordinate between them. Coordination among the robots may be of two types, centralized and decentralized. In the centralized approach, a single robot acts as the coordinator which monitors the movement of robots and hence the goal accomplishment is centered round the coordinator. In the decentralized strategy, there is no single coordinator in the environment. Each robot coordinates its own movement and ensures that it does not collide with any other robot while goal accomplishment. In this project, a new hybrid multi-robots navigation strategy in crowded environment (while avoiding static and dynamic obstacles) will be proposed. Thus, the possibility of collision between the pre-planned trajectories will be calculated by a centralized coordinator. Hence, the time to reach the possible collision points by each robot is calculated according to their speed. Consequently, the robots concerned by the collision have to take decision to avoid the collision by applying the technique of priority between robots.

Publié dans la revue: