A Real-Time Multirobot System Obstacle Avoidance and Autonomous Motion Control Among Moving Obstacles With Continuously Time-Varying Positions and Orientations

摘要

Most interesting environments are in general not known precisely and/or time varying and robots capable of maneuvering among moving obstacles will be capable of accomplishing a much larger and a more versatile class of tasks. In this paper we present a method to address the interesting and difficult problem of real-time planning and efficient motion control of a multirobot system to dynamically avoid mobile robots and obstacle(s) as these move. The method follows upon our previous work which employs a notion of complex potential fields representation, and involves the use of state variables which preserve the Newman Boundary condition of such complex potential fields while allowing the mobile robots to autonomously and dynamically avoid each other as well as other moving obstacles.