Non-Connected Vehicle Detection Using Connected Vehicles
-
2022-12-01
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report, 1/1/2018 to 12/31/2022
-
Corporate Publisher:
-
Abstract:Connected vehicle (CV) technologies are entering the realm of deployment. They have the potential to help drivers make safe, reliable, and informed decisions, and thereby to enhance network capacity and reduce congestion. During the transition to CV technologies, there will be mixed-flow traffic streams of CVs and non-CVs, i.e., human-driven vehicles (HDVs). While most studies have analyzed the CV car-following behavior in a pure CV environment, there is the need for a comprehensive CV car-following model in general mixed flow environment. Specifically, mixed-flow traffic introduces key challenges for CV operations due to potential lane changes by HDVs in adjacent lanes, which can cause stop-and-go waves and traffic oscillations. An understanding of the interactions between CVs and HDVs in the lane-change process can be leveraged to enhance the CV platoon operations. This study proposes a deep reinforcement learning-based proactive longitudinal control strategy (PLCS) for CVs to counteract disruptive HDV lane-change behaviors that can induce disturbances, and to preserve the smoothness of traffic flow in the CV platooning control process. In it, a Transformer-based lane-change traffic condition predictor is constructed to predict longitudinal trajectories of HDVs and whether an HDV will likely perform a disruptive lane change under the ambient traffic conditions. If no disruptive lane change is predicted, an extended intelligent driver model is activated for the CV to perform smooth car-following behavior under cooperative CV platooning control. If a disruptive lane change is predicted, a rainbow deep Q-Network (RDQN)-based lane-change preclusion model is proposed through which the CV can alter the lane-change traffic condition to preclude the HDV’s lane change. Results from numerical experiments suggest that a CV controlled by the PLCS is effective in reducing disruptive lane-change maneuvers by an HDV in its vicinity and can improve string stability performance in mixed-flow traffic. Further, the effectiveness of the PLCS is illustrated under different lane-change scenarios, CV control setups, and HDV driver types.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
File Type:
Supporting Files
-
No Additional Files
More +