论文部分内容阅读
目前,多旋翼无人机控制器设计问题中存在着大量的依靠经验的调参工作。为了使调参简单而又可靠,本文基于控制器参数与控制系统性能响应存在的对应关系,提出了自动调参思想。在满足控制器各项性能指标的前提下,利用粒子群算法(Particle swarm optimization,PSO)提炼出优化目标和约束条件。对被控对象进行建模并搭建非线性模型。然后,利用工程实践方法估算出参数范围,并利用粒子群快速优化特点自动寻找在约束条件下符合性能指标的控制器参数。最后,通过Matlab/Simulink对模型进行仿真验证。仿真结果分析表明,PSD可快速准确地对飞行控制进行自动调参。
At present, there are a large number of reference tasks based on experience in the design of multi-rotor UAV controllers. In order to make the reference simple and reliable, based on the correspondence between the controller parameters and the performance response of the control system, this paper proposes the idea of automatic reference. On the premise of meeting each performance index of controller, the optimization objectives and constraints were extracted by Particle Swarm Optimization (PSO). The accused object is modeled and a nonlinear model is built. Then, the engineering practice is used to estimate the parameter range, and the particle swarm optimization is used to automatically find the controller parameters which meet the performance index under the constraints. Finally, the model is verified by Matlab / Simulink. Simulation results show that PSD can quickly and accurately adjust the flight control automatically.