Human-automated vehicle interaction
DRIVERS' following behavior
Despite numerous studies on general human-robot interactions, in the context of transportation, automated vehicle (AV)-human driver interaction is not a well-studied subject. These vehicles have fundamentally different decision-making logic compared with human drivers and the driving interactions between AVs and humans can potentially change traffic flow dynamics. Accordingly, through an experimental study, this paper investigated whether there is a difference between human-human and human-AV interactions on the road. This study focuses on car-following behavior and conducted several car-following experiments utilizing Texas A&M University’s automated Chevy Bolt. Utilizing NGSIM US-101 dataset, two scenarios for a platoon of three vehicles were considered. For both scenarios, the leader of the platoon follows a series of speed profiles extracted from the NGSIM dataset. The second vehicle in the platoon can be either another human-driven vehicle (Scenario A) or an AV (Scenario B). Data is collected from the third vehicle in the platoon to characterize the changes in driving behavior when following an AV. A data-driven and a model-based approach were used to identify possible changes in driving behavior from Scenario A to Scenario B. The findings suggested there is a statistically significant difference between human drivers’ behavior in these two scenarios and human drivers felt more comfortable following the AV. Simulation results also revealed the importance of capturing these changes in human behavior in microscopic simulation models of mixed driving environments.
Experimental setup for data collection from car-following behavior
Human drivers are more comfortable following an automated vehicle and choose smaller gaps
• Rahmati, Y., M., Khajeh-Hosseini, A., Talebpour, B., Swain, and C., Nelson. Influence of Autonomous Vehicles on Car-Following Behavior of Human Drivers. Accepted for presentation at the 98th Annual Meeting of the Transportation Research Board of National Academies, January 13–17, 2019.
CAVS REAR-END COLLISIONS
While safety is the ultimate goal of designing connected automated vehicles (CAVs), current measures of safety do not provide enough insight into the nature of CAV crashes. This study aims at developing a systematic approach to assess CAV safety, focusing particularly on rear-end crashes at intersections in mixed driving environments. The main hypothesis is that the major reason behind these crashes is the mismatch between human drivers' expectations and CAVs braking decisions. Accordingly, the present study develops two artificial intelligence (AI) techniques to learn and model the braking behavior of human drivers at intersections under two different driving conditions (free-flow and car-following) and compare the results to that of CAVs. Our initial results indicate that the proposed modeling frameworks not only can accurately classify different braking behaviors using vehicle trajectory data, but also reveals a mismatch between humans’ and CAVs’ braking decisions. According to these findings, an initial CAV deceleration profile is designed based on the observed behavior of human drivers while ensuring (1) compatibility with human expectation, and (2) safety of CAVs (see Figures).
Stopping behavior at stop-signs in free-follow and car-following modes
Generating safe stopping behavior for automated vehicles gaps
Rahmati, Y., A., Samimi Abianeh, M., Tabesh, A., Talebpour, and F., Sharifi. Driving to Safety: Who Is At Fault in CAVs Rear-End Collisions. Accepted for presentation at the 98th Annual Meeting of the Transportation Research Board of National Academies, January 13–17, 2019.
Collaborative Left Turn Maneuver
Connectivity and automation provide the opportunity to enhance safety and mitigate congestion in transportation systems. In fact, these technologies can enhance the efficiency of drivers/vehicles’ decision-making by managing and coordinating the interactions among human-driven and connected, automated vehicles. Such management and coordination can lead to developing a collaborative connected, automated driving environment. Game theory, as a methodology to model the outcome of the interactions among multiple players, is a perfect tool to characterize the interactions between these vehicles.
One of the most challenging maneuvers to model is drivers/vehicles’ tactical decisions at intersections. Focusing on unprotected left turn maneuvers, we aim at developing a game theory based framework to characterize driver behavior in unprotected left turn maneuvers in a connected, automated driving environment. A two-person non-zero-sum non-cooperative game under complete information is selected to model the underlying decision-making. NGSIM data is used to calibrate the payoff functions based on Maximum Likelihood Estimation. Validation results indicate that this framework can effectively capture vehicle interactions when performing conflicting turning movements while achieving a relatively high accuracy in predicting vehicles' real choice.
Game structure for unprotected left turn
Rahmati, Y., and A., Talebpour. Towards a Collaborative Connected, Automated Driving Environment: A Game Theory Based Decision Framework for Unprotected Left Turn Maneuvers, IEEE Intelligent Vehicle Symposium (IV), Redondo Beach, CA, June 11-14, 2017.