Regret-based Bayesian sequential decision-making for human-agent collaborative search tasks

We consider domain search tasks using autonomous multi-agent teams collaborating with a human operator. A Bayesian sequential decision-making strategy is first proposed for the optimal allocation of manual and autonomous sensing mode in the presence of sensing uncertainty. However, this type of optimal strategy is not always the style of human decision-making. In particular, regret has been shown to play a critical role in rational decision-making. Humans experience regret when they perceive that they are better off with another option and tend to make choices to avoid such experience. Furthermore, it has been shown that team members share a same mental model perform better than teams with a more accurate but less similar mental model. Therefore, to enable more effective human-agent collaboration (HAC) and better overall task performance, we embed regret analysis into the proposed decision-making framework to provide suboptimal but more human-like decisions. We compare simulation results of decision-making strategies with and without regret analysis and show that regret-based decision-making can integrate human tendency in risk-averse and risk-seeking for better HAC.