基于B样条优化的改进纯追踪农机导航曲线跟踪方法

Improved Pure Pursuit Agricultural Machinery Navigation Curve Path Tracking Method Based on B-spline Optimization

摘要:

针对驾驶农机采样路径平滑性对跟踪效果的影响和曲线路径变曲率特性导致跟踪适应性较低的问题，提出了一种基于B样条路径优化的积分补偿模糊纯追踪控制方法。首先采用B样条插值方法平滑优化采样路径信息；再基于行驶规律分析农机曲率与作业速度对跟踪控制的相互作用关系，设计了模糊调整前视距离的纯追踪控制方法，该方法根据前期试验结果采用拖拉机速度和前视路径平均曲率模糊整定纯追踪前视距离，同时采用跟踪横向偏差设计前轮转角积分补偿，减小稳态误差。设计构建了曲线跟踪Simulink仿真模型并进行了仿真试验。仿真结果表明，改进方法的平均绝对横向偏差均值和最大绝对横向偏差均值相较于传统纯追踪控制方法降低4.8%和7.1%。进行了拖拉机田间正弦曲线跟踪试验，插值路径对比试验结果表明，路径插值后跟踪误差较插值前降低75.9％。控制算法对比试验结果表明，拖拉机以速度1.0、1.5、2.5 m/s跟踪不同幅值正弦曲线路径时，采用改进纯追踪方法平均绝对横向偏差均值分别降低36.80％、62.50％和61.03％，标准差均值分别降低27.8％、24.0％和46.3％。最后跟踪田间随机采样路径，当速度为2.5 m/s时，横向偏差标准差为0.06 m。本文方法有效提高了农机曲线路径跟踪精度，满足需要曲线作业的农业生产需求。

Abstract:

Aiming at the influence of the smoothness of the sampling path of driving agricultural machinery on the tracking effect and the low tracking adaptability caused by the variable curvature of the curve path, a fuzzy pure tracking control method with integral compensation based on B-spline path optimization was proposed. Firstly, B-spline interpolation method was used to smooth and optimize the sampling path information. Based on the driving law, the interaction between the curvature of agricultural machinery and the operating speed on the tracking control was analyzed, and the pure tracking control method of fuzzy adjusting the forward-looking distance was designed. According to the previous test results, the pure tracking forward looking distance was fuzzy adjusted by the tractor speed and the average curvature of the forward-looking path. At the same time, the tracking lateral deviation was used to design the front wheel angle integral compensation to reduce the steady state error. The Simulink simulation model of curve tracking was designed and constructed, and the simulation experiment was carried out. The simulation results showed that the average absolute and maximum absolute lateral deviation of the improved method were reduced by 4.8% and 7.1% compared with that of the traditional pure tracking control method. The sinusoidal curve tracking test of agricultural machinery in the field was carried out. The interpolation path comparison test results showed that the tracking error after path interpolation was reduced by 75.9% compared with that before interpolation. The control algorithm comparison test results showed that when the agricultural machinery tracked the sinusoidal path with different amplitudes at speeds of 1.0 m/s, 1.5 m/s and 2.5 m/s, the average absolute transverse deviation of the improved pure tracking method was reduced by 36.80%, 62.50% and 61.03%, and the average standard deviation was reduced by 27.8%, 24.0% and 46.3%, respectively. Finally, the random sampling path in the field was tracked. When the speed was 2.5 m/s, the horizontal deviation standard deviation was 0.06 m. This method effectively improved the accuracy of agricultural machinery curve path tracking, and met the agricultural production needs of curve operation.