Ni, Xinrui (Tsinghua University), Li, Boyu (Tsinghua University), Cui, Yunkuan (Tsinghua University), Han, Xu (Tsinghua University), Yao, Danya (Tsinghua University), PEI, Xin (Tsinghua University), Qi, Yuliang (Hebei Expressway Group Co., Ltd. Jingxiong Branch), Jin, Shuqing (Hebei Expressway Group Co., Ltd. Jingxiong Branch), Liu, Yinna (Hebei Expressway Group Co., Ltd. Jingxiong Branch)

A Driving Corridor Based Lane-Changing Trajectory Planning Approach for Automated Vehicles

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Lane-change maneuvers in dynamic environments present significant challenges to the advancement of autonomous driving, particularly with respect to safety, comfort, and efficiency. To address these challenges, this paper presents a lane-change trajectory planning approach based on a dynamic safety corridor. To characterize the feasible spatio-temporal space, the lateral space is divided into three regions, and the longitudinal safety corridor is constructed based on the time at which the vehicle enters each region. Meanwhile, the Time-to-Collision metric is integrated into the safety corridor framework to reduce dynamic risks. To deal with the complexity of the integrated trajectory planning model, we divide the process into two stages. Since the construction of the spatio-temporal corridor depends on the lateral trajectory, we perform the first stage to generate a trajectory set via a sampling-based method and obtain the best lateral trajectory through a comprehensive evaluation metric. In the second stage, by incorporating the corridor constraints, an optimization-based method is applied to generate the optimal longitudinal trajectory. A model predictive control framework is employed to address prediction uncertainties in the dynamic environment. Finally, simulation results demonstrate that the proposed approach ensures safety and achieves high trajectory quality in terms of comfort and efficiency, outperforming the baseline approach. Real-world lane-changing scenarios extracted from the NGSIM dataset are employed to further validate the practicality and effectiveness of the proposed method.