Driven by the emergence of autonomous/semi-autonomous driving technologies, the mixed situation of autonomous vehicles and human drivers is of considerable significance. Toward this end, it is necessary to better understand human driving characteristics so as to predict the actions of the other cars. In this regard, we develop a basic framework for modeling driver behaviors in view of human prediction ability. Through the game theoretic estimation of the counterpart’s behaviors and the corresponding time-evolution of unsafe collision areas, we compute an objective collision model. In turn, we design a human-like predictive perception model on collision with an adjacent vehicle based on the objective collision model and the driver’s subjective level of safety assurance. Since drivers have different safety requirements, the subjective estimate on the collision was designed as a region in which has less safety than the driver’s own safety requirement in the objective probabilistic collision prediction. The region that is subjectively perceived based on the driver’s own safety standard is regarded as a deterministic unsafe region for the driver. That is to say, the subjective perception acts as a collision area with the collision probability of 1 so that the driver should avoid while driving. In our subsequent work, we will address the issue of controller design to avoid the subjective collision estimation.
- Dynamic Systems and Control Division
Development of a Predictive Collision Risk Estimation Scheme for Mixed Traffic
- Views Icon Views
- Share Icon Share
- Search Site
Yoo, JH, & Langari, R. "Development of a Predictive Collision Risk Estimation Scheme for Mixed Traffic." Proceedings of the ASME 2014 Dynamic Systems and Control Conference. Volume 1: Active Control of Aerospace Structure; Motion Control; Aerospace Control; Assistive Robotic Systems; Bio-Inspired Systems; Biomedical/Bioengineering Applications; Building Energy Systems; Condition Based Monitoring; Control Design for Drilling Automation; Control of Ground Vehicles, Manipulators, Mechatronic Systems; Controls for Manufacturing; Distributed Control; Dynamic Modeling for Vehicle Systems; Dynamics and Control of Mobile and Locomotion Robots; Electrochemical Energy Systems. San Antonio, Texas, USA. October 22–24, 2014. V001T10A005. ASME. https://doi.org/10.1115/DSCC2014-6144
Download citation file: