Scholarlink

<p>This study developed an omnidirectional following system (OFS) to enhance the versatility and adaptability of an agricultural follower robot in various human-robot collaborative agricultural tasks. By analysing the motion relationship between the operator and the follower robot, a virtual constraint model is established that allows the follower robot to adopt any relative position to follow the operator flexibly. With an omnidirectional operator-robot positioning model and omnidirectional following control model, the follower robot senses the operator’s movements and achieves omnidirectional following under preset positional constraints. In the three common paths such as straight-line, circular, and variable curvature paths, ten separate experiments with different relative orientations between the operator and robot were conducted respectively to verify the performance of the OFS. The results indicate that for straight-line paths, the longitudinal root mean square error (RMSE) ranges from 0.140&#xa0;m to 0.160&#xa0;m, while the lateral RMSE ranges from 0.061&#xa0;m to 0.102&#xa0;m. For circular paths, the longitudinal RMSE ranges from 0.059&#xa0;m to 0.135&#xa0;m, and the lateral RMSE ranges from 0.072&#xa0;m to 0.137&#xa0;m. In variable curvature paths, the longitudinal RMSE ranges from 0.158&#xa0;m to 0.299&#xa0;m, and the lateral RMSE ranges from 0.061&#xa0;m to 0.226&#xa0;m. Additionally, the study analyses the effects of different constraint parameters and path curvature on the following errors. The OFS extends the traditional fixed front-back following model by enabling the robot to follow at any relative position with the operator (e.g., right rear, left rear), which demonstrates broader application potential for human-robot collaboration in agricultural environments.</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

An omnidirectional following system based on virtual constraint model for agricultural follower robots

  • 期刊论文

  • 发表日期：

    2026年1月7日

查看全文
  • Hengda Li,
  • Pingyi Liu,
  • Haichao Li,
  • Liang Sun,
  • Xiaoliang Zhang,
  • Zhenghao Li

摘要

This study developed an omnidirectional following system (OFS) to enhance the versatility and adaptability of an agricultural follower robot in various human-robot collaborative agricultural tasks. By analysing the motion relationship between the operator and the follower robot, a virtual constraint model is established that allows the follower robot to adopt any relative position to follow the operator flexibly. With an omnidirectional operator-robot positioning model and omnidirectional following control model, the follower robot senses the operator’s movements and achieves omnidirectional following under preset positional constraints. In the three common paths such as straight-line, circular, and variable curvature paths, ten separate experiments with different relative orientations between the operator and robot were conducted respectively to verify the performance of the OFS. The results indicate that for straight-line paths, the longitudinal root mean square error (RMSE) ranges from 0.140 m to 0.160 m, while the lateral RMSE ranges from 0.061 m to 0.102 m. For circular paths, the longitudinal RMSE ranges from 0.059 m to 0.135 m, and the lateral RMSE ranges from 0.072 m to 0.137 m. In variable curvature paths, the longitudinal RMSE ranges from 0.158 m to 0.299 m, and the lateral RMSE ranges from 0.061 m to 0.226 m. Additionally, the study analyses the effects of different constraint parameters and path curvature on the following errors. The OFS extends the traditional fixed front-back following model by enabling the robot to follow at any relative position with the operator (e.g., right rear, left rear), which demonstrates broader application potential for human-robot collaboration in agricultural environments.