A Predictive Autonomous Decision Aid for Calibrating Human-Autonomy Reliance in Multi-Agent Task Assignment

TL;DR

This paper introduces a game-theoretic model and an adaptive decision aid using LSTM and Bayesian filtering to calibrate human reliance in multi-agent task assignment, enhancing team performance.

Contribution

It presents a novel adaptive decision aid that calibrates human reliance through interaction sequences, improving collaboration in multi-agent systems.

Findings

Significantly improves human-autonomy interaction performance.

Outperforms myopic decision aids without reliance understanding.

Effectively adapts to human reliance dynamics.

Abstract

In this work, we develop a game-theoretic modeling of the interaction between a human operator and an autonomous decision aid when they collaborate in a multi-agent task allocation setting. In this setting, we propose a decision aid that is designed to calibrate the operator's reliance on the aid through a sequence of interactions to improve overall human-autonomy team performance. The autonomous decision aid employs a long short-term memory (LSTM) neural network for human action prediction and a Bayesian parameter filtering method to improve future interactions, resulting in an aid that can adapt to the dynamics of human reliance. The proposed method is then tested against a large set of simulated human operators from the choice prediction competition (CPC18) data set, and shown to significantly improve human-autonomy interactions when compared to a myopic decision aid that only suggests predicted human actions without an understanding of reliance.