Scholarlink

<p>An efficient path tracking algorithm is essential for achieving intelligent agricultural automation. However, the narrow navigable ridges of tea plantations result in variable row spacing between tea plants and complex turning radii. These characteristics render traditional algorithms insufficiently responsive to changes in path curvature and impose higher demands on path-following precision. To tackle this issue, this paper proposes an adaptive adjustment-based path tracking control method for mobile robots in tea plantations. Firstly, a look-ahead pose prediction model is introduced to improve the efficiency of the robot’s look-ahead predictions. Secondly, an evaluation function integrating lateral and heading error weights and a normal acceleration formula are incorporated to dynamically optimize angular and linear velocities. Finally, adaptive path tracking control is achieved through the dynamic adjustment of the tea plantation mobile robot’s movement parameters. This paper presents a simulation scenario based on a real tea plantation and performs path tracking simulations within this scenario. The experimental results demonstrate that, compared to the traditional pure pursuit algorithm with a fixed look-ahead distance, the proposed algorithm reduces the overall average lateral error by between 14.38 and 26.13% across different test paths. Comparative results indicate that, although the MPC algorithm minimizes global error slightly better, the proposed algorithm achieves lower mean error in regions with high curvature. Additionally, the average single-frame computation time is approximately 2.5% of what the MPC algorithm requires. In robustness tests under simulated muddy road surfaces and slip disturbances, the algorithm maintained an overall average lateral error of 0.032&#xa0;m, comparable to the MPC algorithm, while reducing errors by 8.31–20.99% relative to the pure pursuit algorithm. These results show that the proposed method effectively improves the accuracy of path tracking and the stability of the system in complex, confined spaces while keeping computational overhead low. This makes it a viable engineering solution for automated operations of robots in tea plantations.</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

Path tracking control method for mobile robots in tea plantations based on adaptive adjustment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月3日

查看全文
  • Jin Li,
  • Jinjun Cai,
  • Shuping Dong,
  • Zhihao Lai,
  • Chuhong Ou,
  • Lei Ye,
  • Zhipeng Guo

摘要

An efficient path tracking algorithm is essential for achieving intelligent agricultural automation. However, the narrow navigable ridges of tea plantations result in variable row spacing between tea plants and complex turning radii. These characteristics render traditional algorithms insufficiently responsive to changes in path curvature and impose higher demands on path-following precision. To tackle this issue, this paper proposes an adaptive adjustment-based path tracking control method for mobile robots in tea plantations. Firstly, a look-ahead pose prediction model is introduced to improve the efficiency of the robot’s look-ahead predictions. Secondly, an evaluation function integrating lateral and heading error weights and a normal acceleration formula are incorporated to dynamically optimize angular and linear velocities. Finally, adaptive path tracking control is achieved through the dynamic adjustment of the tea plantation mobile robot’s movement parameters. This paper presents a simulation scenario based on a real tea plantation and performs path tracking simulations within this scenario. The experimental results demonstrate that, compared to the traditional pure pursuit algorithm with a fixed look-ahead distance, the proposed algorithm reduces the overall average lateral error by between 14.38 and 26.13% across different test paths. Comparative results indicate that, although the MPC algorithm minimizes global error slightly better, the proposed algorithm achieves lower mean error in regions with high curvature. Additionally, the average single-frame computation time is approximately 2.5% of what the MPC algorithm requires. In robustness tests under simulated muddy road surfaces and slip disturbances, the algorithm maintained an overall average lateral error of 0.032 m, comparable to the MPC algorithm, while reducing errors by 8.31–20.99% relative to the pure pursuit algorithm. These results show that the proposed method effectively improves the accuracy of path tracking and the stability of the system in complex, confined spaces while keeping computational overhead low. This makes it a viable engineering solution for automated operations of robots in tea plantations.