FBCRI Based Real-time Path Planning for Unmanned Aerial Vehicles in Unknown Environments with Uncertainty

FBCRI Based Real-time Path Planning for Unmanned Aerial Vehicles in Unknown Environments with Uncertainty

  • 摘要: This paper presents an FBCRI (feedback based compositional rule of inference) based novel path planning method to satisfy the requirements of real-time navigation, smoothness optimization and probabilistic obstacle avoidance. With local path-searching behaviors in regional ranges and global goal-seeking behaviors in holistic ranges, the method infers behavior weights using fuzzy reasoning embedded with feedback, and then coordinates the behaviors to generate new reference waypoints. In view of the deterministic decisions and the uncertain states of a UAV (unmanned air vehicle), chance constraints are adopted to probabilistically guarantee the UAV's safety at a required level. Simulation results in representative scenes prove that the method is able to rapidly generate convergent paths in obstacle-rich environments, as well as highly improve the path quality with respect to smoothness and probabilistic safety.

     

    Abstract: This paper presents an FBCRI (feedback based compositional rule of inference) based novel path planning method to satisfy the requirements of real-time navigation, smoothness optimization and probabilistic obstacle avoidance. With local path-searching behaviors in regional ranges and global goal-seeking behaviors in holistic ranges, the method infers behavior weights using fuzzy reasoning embedded with feedback, and then coordinates the behaviors to generate new reference waypoints. In view of the deterministic decisions and the uncertain states of a UAV (unmanned air vehicle), chance constraints are adopted to probabilistically guarantee the UAV's safety at a required level. Simulation results in representative scenes prove that the method is able to rapidly generate convergent paths in obstacle-rich environments, as well as highly improve the path quality with respect to smoothness and probabilistic safety.

     

/

返回文章
返回