Trust and self-confidence based autonomy allocation for robotic systems

Abstract

In this paper, we investigate shared autonomous and manual control of robotic systems. A trust and self-confidence based autonomy allocation scheme is proposed based on objective and unbiased measures of human-robot collaboration (HRC) systems. We also provide analytical tools to evaluate the long-term effects of the proposed allocation scheme on the overall robot performance and human workload. Experiments are designed to test and show that the proposed trust and self-confidence based allocation scheme can capture human autonomy allocation pattern. Furthermore, we develop a pattern correction algorithm using nonlinear model predictive control (NMPC) to help the human gradually adapt to a modified autonomy allocation pattern for better overall performance. The effectiveness of this scheme is shown in simulations.