Haptics-aided path planning and virtual fixture based dynamic kinesthetic boundary for bilateral teleoperation of VTOL aerial robots

摘要

In this paper, we present a novel semi-autonomous shared control framework based on an admittance configured haptic interface for bilateral teleoperation of VTOL aerial robots. A haptics-aided path planner and a virtual fixture based dynamic kinesthetic boundary (VFDKB) are proposed to better assist pilots to navigate through complex environments and perceive obstacle and path boundary. By combining the autonomous path planner and VFDKB, the novel framework maximizes autonomy and maintains flexibility to allow pilots to freely fly the robot with minimal effort for navigation and obstacle avoidance. The RRT based path planner uses the human operator's force input to dynamically determine the target position and heuristically bias the sampling process for better efficiency in finding a collision-free path. Cost functions are used to assess the difference between goal positions and generated paths for determining the replanning process and distance-optimal path. A virtual fixture based dynamic kinesthetic boundary prioritizes the haptic cue of more critical information between obstacles and path guidance virtual fixture (GVF) for pilot to achieve safe navigation. The theoretical analysis provides proof for guaranteed obstacle avoidance performance of the novel approach. Simulation and experimental study on an aerial robotic platform were conducted, and the outcomes confirm our claims.