A policy-blending formalism for shared control

TLDR

Shared‑control teleoperation uses a robot to assist users by predicting intent and helping accomplish tasks, thereby making operation easier and more seamless. This work introduces a policy‑blending formalism for shared control, illustrating how existing techniques instantiate it and analyzing intent prediction and arbitration. The authors formulate intent prediction with inverse reinforcement learning under simplifying assumptions, evaluate it on teleoperated robotic manipulator data, and model arbitration as a control‑theoretic problem aligned with user preferences. A user study shows that effective arbitration must be contextual, depending on the robot’s and user’s confidence and individual traits, and highlights challenges such as adaptation, legibility, and the closed loop between prediction and behavior.

Abstract

In shared control teleoperation, the robot assists the user in accomplishing the desired task, making teleoperation easier and more seamless. Rather than simply executing the user’s input, which is hindered by the inadequacies of the interface, the robot attempts to predict the user’s intent, and assists in accomplishing it. In this work, we are interested in the scientific underpinnings of assistance: we propose an intuitive formalism that captures assistance as policy blending, illustrate how some of the existing techniques for shared control instantiate it, and provide a principled analysis of its main components: prediction of user intent and its arbitration with the user input. We define the prediction problem, with foundations in inverse reinforcement learning, discuss simplifying assumptions that make it tractable, and test these on data from users teleoperating a robotic manipulator. We define the arbitration problem from a control-theoretic perspective, and turn our attention to what users consider good arbitration. We conduct a user study that analyzes the effect of different factors on the performance of assistance, indicating that arbitration should be contextual: it should depend on the robot’s confidence in itself and in the user, and even the particulars of the user. Based on the study, we discuss challenges and opportunities that a robot sharing the control with the user might face: adaptation to the context and the user, legibility of behavior, and the closed loop between prediction and user behavior.