农机化研究

针对传统人工割草作业存在效率低、劳动强度大，而遥控作业模式仍无法减少作业人员等问题，以丘陵果园履带割草机为执行机构，设计了基于RTK的导航控制系统。通过构建车体滑移转向模型，根据转向半径实时计算两侧电机转速，设计了基于模糊PID的智能导航控制算法，并结合JY901S九轴陀螺仪与卡尔曼滤波技术，实现多源传感器数据融合；基于果园区块边界特征自动生成折返形作业路径，减少漏割和复割作业；通过RTK定位模块实时获取割草机的位置和航向信息，并依托卡尔曼滤波优化的航向反馈，输出最优航向角估计值，由导航控制系统实时调节两侧电机转速，快速调整割草机进入规划路径。试验结果表明：该导航控制系统具有控制精度高、轨迹偏差小等特点，可在7.6 s内收敛至0.03 m稳态误差范围内；丘陵果园地形试验中，在初始横向偏差为0.5 m、航向偏差为60°、作业速度为1.0 m/s的条件下，割草机作业平均横向偏差不超过0.072 m,可有效适应坡地环境。研究可为丘陵果园智能化作业装备开发提供实践参考。

Traditional manual mowing suffers from low efficiency and high labor intensity, while remote-controlled operation modes still fail to reduce workforce requirements, focused on a crawler mower for hilly orchards as the execution mechanism, designing and implementing an RTK-based navigation control system with experimental validation. A skid-steer vehicle model was established to calculate real-time motor speeds on both sides based on steering radius. Simultaneously, an intelligent navigation control algorithm based on fuzzy PID was designed, integrating the JY901S 9-axis gyroscope and Kalman filtering technology to achieve multi-source sensor data fusion. Zigzag operation paths were automatically generated based on orchard boundary features to minimize missed or repeated cutting areas. Real-time position and heading information of the mower was acquired through an RTK positioning module, and heading feedback optimized via Kalman filtering outputed the optimal heading angle estimation. The navigation control system dynamically adjusted motor speeds on both sides to rapidly align the mower with planned paths. Experimental results demonstrated that the navigation system exhibited high control precision and minimal trajectory deviation. The system converged to a steady-state error within 0.03 m in 7.6 s. Under field tests in hilly orchard terrain with an initial lateral deviation of 0.5 m, heading deviation of 60°, and operating speed of 1.0 m/s, the average lateral deviation during mowing was less than 0.072 m, effectively adapting to sloped terrain. This study can provide practical references for developing intelligent equipment in hilly orchard environments.