To control the robots navigation path tracking and achieve trackless precise material pushing, a local path planning method based on adaptive control algorithm of navigation variable weight is proposed. By allocating tracking weights to the lateral distance of the cattle pen and the course deviation of the target waypoint, the dependence on the global positioning is reduced, and the pushing operation of the robot with a fixed lateral distance is realized, and the following of the global path is considered. At the same time, the planner monitors the power of the drive motor and adjusts the angular velocity and the lateral distance by adjusting the holding distance to adapt to different dietary thicknesses. The experimental results show that when the global positioning error is within ±100 cm, the method proposed in this paper can maintain the accuracy of the cattle fence within the lateral range of 2 cm, and solve the adaptive path adjustment problem of the robot under extreme dietary load conditions, and effectively solve the problem of robot navigation control in the complex feeding channel scene.- 引用格式:何梦腾,潘浩文,邓洪兴,等.基于点云多尺度方向一致性的羊只体尺测量方法［J］.河北大学学报(自然科学版),2026,46(2):113-127.引用格式:颜铭宏,高德华,李茂强,等.基于导航变量权重自适应的推料机器人局部路径规划方法［J］.河北大学学报(自然科学版),2026,46(2):128-141.DOI:10.3969/j.issn.1000-1565.2026.02.002基于导航变量权重自适应的推料机器人局部路径规划方法颜铭宏¹,高德华¹,李茂强²,马永龙²,马驰¹,张文恺¹,田世玉¹,冉碗情²(1.宁夏大学 机械工程学院,宁夏 银川 750021;2.北方民族大学 机电工程学院,宁夏 银川 750021)摘 要:为了控制机器人对导航路径的跟踪,实现无轨精准推料,本文提出一种基于导航变量权重自适应控制算法的局部路径规划方法.通过对牛栏横向距离和目标航点航向偏差进行跟踪权重分配,降低对全局定位的依赖,实现机器人以固定横向距离进行推料作业,并遵循全局路径;同时,规划器监测驱动电机的功率,并通过调整保持距离以调节角速度与横向距离,达到适应不同日粮厚度的目的.实验结果表明,全局定位误差在±100 cm内时,本文提出的方法可实现对牛栏的横向距离保持精度在2 cm以内,解决了极端日粮载荷条件下机器人自适应路径调整问题,有效应对饲喂通道复杂场景下机器人导航控制问题.关键词:推料机器人;权重自适应;局部路径规划;PID算法中图分类号:TP242.3 文献标志码:A 文章编号:1000-1565(2026)02-0128-14DOI:10.3969/j.issn.1000-1565.2026.02.002Local path planning method for feed-pushing robot based on adaptive weights of navigation variablesYAN Minghong¹, GAO Dehua¹, LI Maoqiang², MA Yonglong², MA Chi¹, ZHANG Wenkai¹, TIAN Shiyu¹, RAN Wanqing²(1.College of Mechanical Engineering, Ningxia University, Yinchuan 750021, China; 2. College of Mechanical and Electrical Engineering, North Minzu University, Yinchuan 750021, China)Abstract: To control the robots navigation path tracking and achieve trackless precise material pushing, a local path planning method based on adaptive control algorithm of navigation variable weight is proposed. By allocating tracking weights to the lateral distance of the cattle pen and the course deviation of the target waypoint, the dependence on the global positioning is reduced, and the pushing operation of the robot with a fixed lateral distance is realized, and the following of the global path is considered. At the same time, the planner monitors the power of the drive motor and adjusts the angular velocity and the lateral distance by adjusting the holding distance to adapt to different dietary thicknesses. The experimental results show that when the global positioning error is within ±100 cm, the method proposed in this paper can maintain the accuracy of the cattle fence within the lateral range of 2 cm, and solve the adaptive path adjustment problem of the robot under extreme dietary load conditions, and effectively solve the problem of robot navigation control in the complex feeding channel scene.- 收稿日期:2025-12-01;修回日期:2026-01-06 基金项目:宁夏自然科学基金项目(2022AAC03258);宁夏回族自治区重点研发项目(2024BBF02029) 第一作者:颜铭宏(1998—),男,宁夏大学在读硕士研究生,主要从事智能化农牧装备方向研究.E-mail:790169572@qq.com 通信作者:高德华(1989—),男,宁夏大学副教授,主要从事智能化农牧装备方向研究.E-mail:dehuag@nxu.edu.cn 第2期颜铭宏等:基于导航变量权重自适应的推料机器人局部路径规划方法河北大学学报(自然科学版) 第46卷