Skip to main content

Cooperative Merging Speed Planning: A Vehicle-Dynamics-Free Method

by Zejiang Wang, Adian S Cook, Yunli Shao, Guanhao Xu, Jianfei Chen
Publication Type
Conference Paper
Book Title
2023 IEEE Intelligent Vehicles Symposium (IV)
Publication Date
Page Numbers
1 to 8
Publisher Location
New Jersey, United States of America
Conference Name
IEEE Intelligent Vehicles Symposium (IV)
Conference Location
Anchorage, Alaska, United States of America
Conference Sponsor
Conference Date

Various cooperative merging control strategies at on-ramp have been proposed in the last decade. Approximated vehicle longitudinal motion models, e.g., kinematics model, have been broadly adopted for controller synthesis because of their simplicity. However, what appears problematic is that the models used for controller validation remain, in many cases, the same as the ones used for controller design. Indeed, actual vehicle dynamics contain rich behaviors that the simplified models cannot fully cover. In this paper, we first demonstrate that the actual vehicle speed can be dissimilar to the reference from a speed planner once vehicle dynamics is considered. Then, we propose two data-driven speed generators agnostic to vehicle dynamics. SUMO/Simulink joint simulations demonstrate that the proposed reference speed planners can successfully merge vehicles with distinct dynamics characteristics by following the desired sequence, speed, and intervehicle distance at the merging point while avoiding collisions.