Trajectory Planning for Lane Change with Intelligent Vehicles Using Fuzzy Logic and a Dynamic Programming and Quadratic Programming Algorithm
Jiahao Li,Shengqin Li,Juncheng Wang
DOI: https://doi.org/10.3390/electronics13234732
IF: 2.9
2024-11-30
Electronics
Abstract:With the increasing demand for autonomous driving, ensuring safe and efficient lane-changing behavior in multi-lane traffic scenarios has become a key challenge. This paper proposes an algorithm for active lane-changing decision-making and trajectory planning designed for intelligent vehicles in such environments. The lane-changing intent is evaluated using fuzzy logic, followed by an assessment of lane-changing feasibility based on a lane utility evaluation function. A hierarchical model for path and speed planning is established. Path clusters are generated using quintic polynomials. With a multi-objective cost function designed to ensure collision safety, smoothness, road boundaries, and trajectory continuity, dynamic programming (DP) and quadratic programming (QP) are employed to obtain the trajectory with the minimum cost among the trajectory set fitted by fifth-order polynomials, which is the optimal lane-changing trajectory. For speed planning, obstacles are projected onto the S–T coordinate system, which is a coordinate system with time as the horizontal axis and the distance(s) of the planned path as the vertical axis, and multi-objective cost functions for speed, acceleration, and speed continuity are designed. The speed curve is optimized using DP followed by QP under given constraints. Simulation results show that the proposed algorithm makes safe and effective lane-changing decisions based on traffic conditions, vehicle distances, and speeds. The model generates smooth and stable paths while ensuring the safe and efficient execution of lane changes. This process meets real-time requirements and verifies the reliability of the algorithm.
engineering, electrical & electronic,computer science, information systems,physics, applied