2019/10/11 by Berat Mert Albaba, Albaba, Mert, Yıldıray Yıldız +1 · 1 citation
Psychology · Engineering · #Human-Automation Interaction and Safety #Air Traffic Management and Optimization #Evacuation and Crowd Dynamics
paper · pdf · doi:10.48550/arxiv.1910.05092
Predicting the outcomes of cyber-physical systems with multiple human\ninteractions is a challenging problem. This article reviews a game theoretical\napproach to address this issue, where reinforcement learning is employed to\npredict the time-extended interaction dynamics. We explain that the most\nattractive feature of the method is proposing a computationally feasible\napproach to simultaneously model multiple humans as decision makers, instead of\ndetermining the decision dynamics of the intelligent agent of interest and\nforcing the others to obey certain kinematic and dynamic constraints imposed by\nthe environment. We present two recent exploitations of the method to model 1)\nunmanned aircraft integration into the National Airspace System and 2) highway\ntraffic. We conclude the article by providing ongoing and future work about\nemploying, improving and validating the method. We also provide related open\nproblems and research opportunities.\n