Tracing algorithm and control strategy for crawler rice combine harvester auxiliary navigation system

Abstract

Crawler combine harvesters are major equipment for rice harvesting. Multiple operating parameters and harvesting path need to adjust in real-time to balance harvesting quality and operational efficiency, which is highly demanding and labour-intensive. Application navigation system is one of the effective ways to solve the problems, but navigation path tracking errors are significant due to track sinking and slippage in paddy field. Therefore, tracing algorithm and control strategy for crawler rice combine harvester auxiliary navigation system was developed. Firstly, a circular arc-tangent line tracking model was constructed based on the crawler combine harvester kinematic characteristics in the paddy field to calculate the steering angle. Then, hydraulic steering actuator controller was designed based on fuzzy control methods with particle swarm algorithm. Finally, a least-squares support-vector machine regression-based steering feature identification method was proposed to construct a functional relationship between the control variable and the actual yaw rate, correcting the control errors due to crawler sinking and slippage. Simulations show that the rise time for path correction, steady-state adjustment time, maximum overshoot, and average steady-state error of crawler combine harvester was 7.5 s, 14.7 s, 0.148 m, and 0.064 m, respectively. Field tests show that the average harvest width was 2.02 m, and the average deviation and harvest width rate was 0.18 m, and 91.9%, respectively, under different vehicle speeds. • Crawler combine harvester path tracking algorithm and control strategy was proposed. • Circular arc-tangent model, PSO and LS-SVM regression were used to improving paddy field adaptability. • Field experiment results indicates that rice harvesting navigation average deviation was ≤0.18 m.