Game Theoretic Modeling of Driver and Vehicle Interactions for Verification and Validation of Autonomous Vehicle Control Systems

Abstract

Abstract-Autonomous driving has been the subject of increased interest in recent years both in industry and in academia. Serious efforts are being pursued to address legal, technical, and logistical problems and make autonomous cars a viable option for everyday transportation. One significant challenge is the time and effort required for the verification and validation of the decision and control algorithms employed in these vehicles to ensure a safe and comfortable driving experience. Hundreds of thousands of miles of driving tests are required to achieve a well calibrated control system that is capable of operating an autonomous vehicle in an uncertain traffic environment where interactions among multiple drivers and vehicles occur simultaneously. Traffic simulators where these interactions can be modeled and represented with reasonable fidelity can help to decrease the time and effort necessary for the development of the autonomous driving control algorithms by providing a venue where acceptable initial control calibrations can be achieved quickly and safely before actual road tests. In this paper, we present a game theoretic traffic model that can be used to: 1) test and compare various autonomous vehicle decision and control systems and 2) calibrate the parameters of an existing control system. We demonstrate two example case studies, where, in the first case, we test and quantitatively compare two autonomous vehicle control systems in terms of their safety and performance, and, in the second case, we optimize the parameters of an autonomous vehicle control system, utilizing the proposed traffic model and simulation environment.

The Smart Black Box: A Value-Driven High-Bandwidth Automotive Event Data Recorder

“…Above, η, ζ ≥ 0 are weighting parameters that can be varied to examine solution sensitivity or represent user preferences. The optimization problem (15) may not be easy to solve because the dimension |B k | is typically large. In what follows we introduce a simple but effective value filtering method so that the continuity term W B k (D) can be dropped.…”

Decision making in dynamic and interactive environments based on cognitive hierarchy theory, Bayesian inference, and predictive control

“…The level-k framework has been exploited for modeling human-human, human-machine, and machine-machine 1 interactions in automotive systems [3], [16], [19], aerospace systems [13], [14], as well as in cyber-physical security [15]. It has been revealed in [16] that an ego agent using a decision strategy corresponding to a level-k model with some fixed k may behave poorly when interacting with another agent using a decision strategy of level-σ with σ = k − 1.…”

Liability Design for Autonomous Vehicles and Human-Driven Vehicles: A Hierarchical Game-Theoretic Approach

“…A majority of existing studies have focused on designing AVs' driving algorithms in various scenarios to ensure traffic efficiency, safety, or ethics (Reece and Shafer, 1993;Yoo and Langari, 2013;Kim and Langari, 2014;Yu et al, 2018;Wang et al, 2015;Talebpour et al, 2015;Wang et al, 2016;Gong et al, 2016;Gong and Du, 2018;Li et al, 2018;Huang et al, 2019a,c,b;Smith, 2019), but ignoring human drivers' behavioral adaptation to AVs as humans are exposed to more and more traffic encounters with AVs. Human drivers may have a weaker incentive to exercise "due care" when faced with AVs.…”