【摘要】 针对固定参数模型预测控制(model predictive control, MPC)在路径跟踪控制器中跟踪误差大、难以满足精准农业作业需求的情况，以及传统模型预测控制中时域参数固定的局限，提出一种时域参数自适应调整的控制策略。建立拖拉机动力学模型，在MPC算法的基础上，引入改进粒子群优化算法，对时域参数进行自适应调整；搭建MPC轨迹跟踪仿真框架，验证控制器的可行性。仿真结果表明：相比于固定时域MPC控制器，所提出的自适应时域MPC控制器的轨迹跟踪，横向误差绝对均值可降低22%～28%,提高了跟踪精度。更多还原

【Abstract】 In the field of precision agriculture, the accuracy of tractor path tracking directly affects the quality and efficiency of agricultural operations. However, the path tracking controller based on the fixed-parameter model predictive control(MPC) faces large tracking errors, which makes it difficult to meet the high-precision requirements of precision agricultural operations. A key factor leading to this issue lies in the limitations of the fixed time-domain parameters in the traditional MPC. These fixed parameters cannot flexibly adapt to the changes in the tractor’s operating environment, speed, and path conditions, resulting in a decline in control performance and an increase in tracking errors in actual operations.To address the problem, this paper proposes a control strategy with adaptive adjustment of time-domain parameters. To lay a solid foundation for the subsequent control strategy design, a tractor dynamics model is built. This model fully considers the dynamic characteristics of the tractor during movement, including factors such as tire-ground interaction, vehicle inertia, and steering system response, so as to accurately reflect the actual movements of the tractor.On the basis of the MPC algorithm, this paper introduces an improved particle swarm optimization(PSO) algorithm to realize the adaptive adjustment of time-domain parameters. The traditional PSO algorithm has certain defects, such as easily falling into local optimum and slow convergence speed in the later stage. The improved PSO algorithm optimizes the inertia weight and learning factors, which enhances its global search ability and convergence speed. By using this improved algorithm, the time-domain parameters in the MPC can be adjusted in real-time according to the current tracking error, the tractor’s running speed, and the curvature of the planned path. When the tracking error is large, the time-domain parameters can be adjusted to improve the responsiveness of the controller; when the tractor is moving along a smooth path with small errors, the parameters can be optimized to ensure the stability of the system.To verify the feasibility and effectiveness of the proposed controller, an MPC trajectory tracking simulation framework is built. In the simulation environment, different working conditions are set, including different path types(straight lines, curves with different curvatures), different tractor speeds, and different ground adhesion conditions, to comprehensively test the performance of the controller. The fixed time-domain MPC controller is used as a comparison group in the simulation experiment.The simulation results show that compared with the fixed time-domain MPC controller, the trajectory tracking performance of the adaptive time-domain MPC controller proposed in this paper is markedly improved. Specifically, the absolute mean value of the lateral error is reduced by 22% to 28%, which effectively improves the tracking accuracy. This indicates that the adaptive adjustment strategy of time-domain parameters overcome the limitations of fixed parameters, make the controller better adapt to various complex working conditions, and provide a more reliable technical support for precision agricultural operations.更多还原