基于横/航向误差的模型预测农机导航轨迹跟踪控制研究

Lateral/Yaw Error Based Model Prediction of Agricultural Machinery Navigation Trajectory Tracking Control

摘要:

随着农业智能化与精准化的快速发展，农机导航路径跟踪技术成为提升作业效率与资源利用率的核心环节，为此本研究提出一种基于横/航向误差的模型预测控制（Model predictive control，MPC）方法，旨在提升农机导航轨迹跟踪的精度与适应性。基于运动学模型构建了低维度横/航向误差跟踪模型，考虑实际应用，加入延迟量状态方程，在状态空间下对模型进行线性化和离散化，推导预测模型，并转换为二次型优化模型进行求解。仿真结果表明，对于速度变化、不同横向误差初始条件、路径曲率突变等都保持了高精度的跟踪性能。将算法部署至实际系统进行验证试验，结果表明，导航轨迹跟踪横向误差维持±0.02 m以内，在2σ统计学意义下航向误差±0.5°以内，解决了传统MPC控制器因土壤条件时变导致跟踪性能下降的问题。研究结果为农机高精度自主导航提供了理论支持，对推动无人农场技术发展具有重要意义。

Abstract:

With the rapid development of agricultural intelligence and precision，agricultural machinery navigation path tracking technology has become the core link to improve operational efficiency and resource utilization. A model predictive control (MPC) method based on lateral and yaw error was proposed to improve the accuracy and adaptability of agricultural machinery navigation track tracking. A low-dimensional lateral error tracking model was constructed based on the kinematic model，and considering actual application，the delay state equation was also added，and based on the model，linearization and discretization were carried out in the state space，the prediction model was derived，and it was converted into a quadratic optimization model for solving. The simulation experiments showed that this algorithm maintained high-precision tracking performance in cases such as speed changes，different initial conditions of lateral errors，and sudden changes in path curvature，and the system maintained high-precision tracking performance. The algorithm was deployed into the actual system，and the above working conditions were verified respectively. The experiments showed that the lateral error of the algorithm was maintained within ±0.02 m，and the heading error was within ±0.5° in the statistical sense of 2σ， which solved the problem of the traditional MPC controller's tracking performance degradation caused by time-varying soil conditions. The research result can provide theoretical support for high-precision autonomous navigation of agricultural machinery，which was significant for promoting the development of unmanned farm technology.