Human Preferences and Robot Constraints Aware Shared Control for Smooth Follower Motion Execution

TL;DR

This paper introduces a shared control framework for robot teleoperation that integrates operator preferences and robot constraints, enabling smoother transitions and movements during object grasping tasks.

Contribution

It presents an alternating shared control approach that considers operator input and robot manipulability for improved grasping and movement execution.

Findings

Smoother follower movements during mode switching.

Effective incorporation of operator preferences in grasping pose generation.

Enhanced teleoperation experience with reduced operator load.

Abstract

With the continuous advancement of robot teleoperation technology, shared control is used to reduce the physical and mental load of the operator in teleoperation system. This paper proposes an alternating shared control framework for object grasping that considers both operator's preferences through their manual manipulation and the constraints of the follower robot. The switching between manual mode and automatic mode enables the operator to intervene the task according to their wishes. The generation of the grasping pose takes into account the current state of the operator's hand pose, as well as the manipulability of the robot. The object grasping experiment indicates that the use of the proposed grasping pose selection strategy leads to smoother follower movements when switching from manual mode to automatic mode.