Towards game theoretic AV controllers: measuring pedestrian behaviour in Virtual Reality

Camara, Fanta and Dickinson, Patrick and Merat, Natasha and Fox, Charles (2019) Towards game theoretic AV controllers: measuring pedestrian behaviour in Virtual Reality. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019) Workshops, Nov 8 2019, Macau, China.

Documents
Towards game theoretic AV controllers: measuring pedestrian behaviour in Virtual Reality
Accepted Manuscript

Request a copy
[img] PDF
camara2019gamesVR_IROS_CVW.pdf - Whole Document
Restricted to Repository staff only

2MB
Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

Understanding pedestrian interaction is of great importance for autonomous vehicles (AVs). The present study investigates pedestrian behaviour during crossing scenarios with an autonomous vehicle using Virtual Reality. The self-driving car is driven by a game theoretic controller which adapts its driving style to pedestrian crossing behaviour. We found that subjects value collision avoidance about 8 times more than saving 0.02 seconds. A previous lab study found time saving to be more important than collision avoidance in a highly unrealistic board game style version of the game. The present result suggests that the VR simulation reproduces real world road-crossings better than the lab study and provides a reliable test-bed for the development of game theoretic models for AVs.

Keywords:autonomous vehicles, game theory, Cognitive architectures for action selection
Subjects:H Engineering > H230 Transport Engineering
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:37261
Deposited On:27 Sep 2019 08:47

Repository Staff Only: item control page