Scholarlink

<p>In path planning with the traditional artificial potential field method (T-APF), intelligent vehicle are prone to falling into local minima when navigating complex multi-obstacle environments, resulting in target inaccessibility. Additionally, this method has inherent limitations in dynamic obstacle avoidance. To overcome these drawbacks, this paper proposes an adaptive fuzzy potential field method (SA-APF) with dynamic constraints. First, a four-wheel steering kinematics model is established, and an endpoint pre-aiming strategy based on subdivided repulsive fields is designed to avoid the target unreachability caused by local minima. Then, dynamic obstacle potential fields are integrated with fuzzy control feedback on repulsion and corner to enable autonomous obstacle avoidance and speed adaptation. An MPC (Model Predictive Control) controller incorporating upper-layer state references is also designed to improve tracking accuracy. MATLAB-Carsim co-simulations and real-vehicle tests confirm SA-APF resolves local minima, boosts target accessibility, shortens paths by 10.14% with smaller curvature, and restricts lateral tracking error within 0.35&#xa0;m, validating its feasibility.The above verifies the rationality of the improvement method and provides a theoretical basis for further strengthening the reliability of active obstacle avoidance of intelligent vehicles at low speeds.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent Vehicle Park Path Planning Based on Improved Artificial Potential Field Method

  • 期刊论文

  • 发表日期：

    2026年3月17日

查看全文
  • Ke Liu,
  • Qiuyue Du,
  • Weiliang Zhan,
  • Hongzhao Dong,
  • Yuxin Zhang,
  • Yi Gong

摘要

In path planning with the traditional artificial potential field method (T-APF), intelligent vehicle are prone to falling into local minima when navigating complex multi-obstacle environments, resulting in target inaccessibility. Additionally, this method has inherent limitations in dynamic obstacle avoidance. To overcome these drawbacks, this paper proposes an adaptive fuzzy potential field method (SA-APF) with dynamic constraints. First, a four-wheel steering kinematics model is established, and an endpoint pre-aiming strategy based on subdivided repulsive fields is designed to avoid the target unreachability caused by local minima. Then, dynamic obstacle potential fields are integrated with fuzzy control feedback on repulsion and corner to enable autonomous obstacle avoidance and speed adaptation. An MPC (Model Predictive Control) controller incorporating upper-layer state references is also designed to improve tracking accuracy. MATLAB-Carsim co-simulations and real-vehicle tests confirm SA-APF resolves local minima, boosts target accessibility, shortens paths by 10.14% with smaller curvature, and restricts lateral tracking error within 0.35 m, validating its feasibility.The above verifies the rationality of the improvement method and provides a theoretical basis for further strengthening the reliability of active obstacle avoidance of intelligent vehicles at low speeds.