Khái niệm cốt lõi
This paper proposes a novel optimal control framework that couples the predicted trajectories of surrounding vehicles with collision avoidance constraints to generate collision-free optimal trajectories for autonomous vehicles in highway traffic.
Tóm tắt
The paper presents an approach for optimal trajectory planning of autonomous vehicles in highway traffic scenarios that involve interactions with surrounding vehicles. The key highlights are:
The authors propose a novel optimal control framework that integrates the predicted trajectories of surrounding vehicles into the collision avoidance constraints of the trajectory planning problem.
The trajectory planning is formulated as an optimal control problem, which is solved using Pontryagin's Minimum Principle (PMP). This involves introducing necessary conditions and jump conditions to handle the state constraints related to collision avoidance.
The predicted trajectories of surrounding vehicles are obtained using a modified high-order Markov chain model, which can capture the dynamic behavior of the vehicles.
The effectiveness of the proposed approach is demonstrated through simulations of various highway traffic scenarios, including cases where the surrounding vehicles exhibit different driving behaviors (constant speed, deceleration, acceleration).
The results show that the trajectory planner can generate collision-free optimal trajectories for the ego vehicle by effectively coupling the predicted trajectories of the surrounding vehicles with the collision avoidance constraints.