Risk-aware lane-change trajectory planning with rollover prevention for autonomous light trucks on curved roads.

Lane-change trajectory planning of autonomous light trucks is closely related to safety, driving stability and transportation efficiency, especially in complex road and traffic conditions. However, the collision risk assessment and the impact of roll dynamics, which are considered as critical safety factors, are not well investigated in most of the previously proposed planning strategies. In this study, a novel lane-change trajectory planning framework for autonomous light trucks on curved roads is proposed to improve the vehicle safety, vehicle stability and transportation efficiency. To begin with, curvilinear Gaussian collision risk model considering longitudinal motion, lateral motion and short-term motion trend of surrounding vehicles is utilized to accurately establish the risk map of the traffic environment. Then, collision risk, lateral and roll dynamics are considered in the planning stage to screen out the safe and dynamic-stable trajectory candidates. Furthermore, the technique for order preference by similarity to ideal situation (TOPSIS) approach is employed to handle the multi-criteria decision-making problem by scoring and ranking the performance of the trajectory candidates and select the optimal one. Finally, compared with artificial potential field method and the planner without dynamic constraints, the experimental results validate the effectiveness and performance of the proposed planning framework through MATLAB/Simulink-TruckSim joint platform. • Curvilinear Gaussian collision risk model is utilized to establish the risk distribution map of the traffic environment. • Collision risk, lateral and roll dynamics are considered in the planning stage to optimize the trajectory. • TOPSIS algorithm is employed to solve the multi-objective optimization problem for optimal trajectory selection. [ABSTRACT FROM AUTHOR]