Abstract: According to the features of headland turning，agricultural machine headland turning control algorithm is presented.Applying simplified bicycle kinematics model，the path planning is produced by assembling three straight lines；using the dynamical look-ahead distance which is implemented by the BP neural network，an improved pure pursuit algorithm is applied in the headland turning.MATLAB/Simulink simulation results show that headland path planning algorithm and headland turning algorithm presented are simple，occupy small headland space and have a high tracking accuracy.The control method is feasible and practical.

摘要： 根据农业机械田间地头转向的特点，提出了一种基于改进纯追踪模型的农业机械地头转向控制方法。以简化二轮车作为运动学模型，使用直线组合进行地头转向路径规划，采用BP神经网络实现动态调节前视距离改进纯追踪算法，用该算法进行路径跟踪，实现地头转向动作。MATLAB/Simulink仿真结果表明，该基于改进纯追踪模型的农业机械地头转向控制方法路径规划简单、占用空间小、跟踪精度较高。