基于转向非线性模糊补偿的水田履带拖拉机路径跟踪控制方法

Path Tracking Control for Crawler Tractors in Paddy Fields Based on Fuzzy Compensation for Nonlinear Steering Characteristics

摘要:

针对水田作业中履带拖拉机因差速转向非线性与履带滑移导致路径跟踪精度下降的问题,本文提出一种基于实时侧滑估计与模糊补偿的自动导航控制方法。建立了履带拖拉机运动学模型,通过转向系统参数辨识与中位标定,确立了转向控制参数与转向曲率的线性关系,并设计了自适应预瞄纯追踪算法。在此基础上,通过运动学分析进行了侧滑估计并构建了以实时滑移率与期望转向角为输入的模糊控制器,在线修正由履带侧滑与液压转向死区引起的决策偏差。Matlab/Simulink仿真结果表明,在跟踪S形与双扭路径时,相比传统纯追踪算法,该方法最大横向偏差降低54.1%、66.7%,平均绝对偏差减少34.4%、55.8%,表现出更优的跟踪精度与抗干扰能力。以1GLX-230型履带自走式旋耕机为平台的田间试验进一步验证了其有效性：连续作业行平均最大横向偏差为0.059 m,平均绝对偏差为0.025 m,且97.33%跟踪偏差被控制在±0.05 m范围内。该方法有效提升了水田复杂环境下履带拖拉机路径跟踪稳定性、对行精度与整体鲁棒性,满足精准农业自动导航作业要求。

Abstract:

Aiming at the problem of path tracking accuracy degradation in crawler tractors operating in paddy fields,caused by the nonlinearity of differential steering and track slippage,an automatic navigation control method was proposed based on real-time sideslip estimation and fuzzy compensation. A kinematic model of the crawler tractor was established. Through steering system parameter identification and neutral calibration,the linear relationship between steering control parameters and steering curvature was determined,and an adaptive preview pure pursuit algorithm was designed. Building upon this foundation,sideslip estimation was performed via kinematic analysis,and a fuzzy controller using real-time slip rate and desired steering angle as inputs was constructed to correct decision-making deviations online caused by track sideslip and hydraulic steering dead zone. Simulation results in Matlab/Simulink showed that,when tracking S-shaped and double-lane change paths,the proposed method reduced the maximum lateral deviation by 54.1% and 66.7% and the mean absolute deviation by 34.4% and 55.8% compared with that of the traditional pure pursuit algorithm,demonstrating superior tracking accuracy and anti-interference capability. Field tests conducted on a 1GLX-230 crawler self-propelled rotary tiller further verified its effectiveness: the average maximum lateral deviation across consecutive operation rows was 0.059 m,the average mean absolute deviation was 0.025 m,and 97.33% of the tracking deviations were controlled within the ±0.05 m range. This method effectively enhanced the path tracking stability,inter-row alignment accuracy,and overall robustness of crawler tractors in the complex paddy field environment,meeting the requirements for automatic navigation operations in precision agriculture.