Pareto efficiency in synthesizing shared autonomy policies with temporal logic constraints

TL;DR

This paper develops a two-stage policy synthesis method for shared autonomy systems that balances system performance and human workload, using Pareto efficiency and multi-objective optimization with temporal logic constraints.

Contribution

It introduces a novel approach combining Pareto efficiency and Tchebychev scalarization for synthesizing shared autonomy policies with temporal logic specifications.

Findings

The method effectively balances human effort and system performance.

It provides a diverse set of Pareto optimal policies.

The approach improves coverage of optimal solutions compared to linear scalarization.

Abstract

In systems in which control authority is shared by an autonomous controller and a human operator, it is important to find solutions that achieve a desirable system performance with a reasonable workload for the human operator. We formulate a shared autonomy system capable of capturing the interaction and switching control between an autonomous controller and a human operator, as well as the evolution of the operator's cognitive state during control execution. To trade-off human's effort and the performance level, e.g., measured by the probability of satisfying the underlying temporal logic specification, a two-stage policy synthesis algorithm is proposed for generating Pareto efficient coordination and control policies with respect to user specified weights. We integrate the Tchebychev scalarization method for multi-objective optimization methods to obtain a better coverage of the set of Pareto efficient solutions than linear scalarization methods.