摘要

This paper introduces a novel constraint in the master joystick workspace, termed the dynamic kinesthetic boundary, that aids a pilot to navigate an aerial robotic vehicle through a cluttered environment. The proposed approach exploits spatial cues by projecting the remote environment into a hard boundary in the master device workspace that provides a pilot with a natural representation of approaching obstacles. The approach is distinguished from classical force feedback approaches by allowing normal operation of the vehicle in free flight and only imposing constraints when approaching and interacting with the environment. A key advantage is that contact with the environment constraint is immediately perceptible to a pilot, allowing them to make suitable adjustments to their inputs. Modulation of the velocity reference for the slave robot ensures obstacle avoidance while allowing a vehicle to approach as close as desired to an object, albeit at a slow speed. A comprehensive user study was performed to systematically test the proposed algorithm and comparisons to two existing state-of-the-art approaches are provided to demonstrate the relative performance of the proposed approach.

  • 出版日期2016-5