Local implicit surface based virtual fixtures in point cloud virtual environment for teleoperation

It is a critical task to teleoperate a robot in a partially known or unstructured environment without any assistance. In this paper, local implicit surface based virtual fixtures are generated real time in a point cloud augmented virtual environment for operation guidance. The point-set implicit local surface method is modified to accomplish local forbidden region virtual fixtures. The robot-centered potential force field model is applied for the guidance virtual fixture generation. The resultant force generated from both forbidden region virtual fixtures and guidance virtual fixtures are fed back in real time to the human operator through a haptic device. With assistance from the guidance force, human operators can implement obstacle avoidance tasks efficiently. The experiment results show that the proposed method is effective for robot teleoperation.