正在加载图片...
工程科学学报,第39卷,第11期:1743-1752,2017年11月 Chinese Journal of Engineering,Vol.39,No.11:1743-1752,November 2017 D0l:10.13374/j.issn2095-9389.2017.11.018;http://journals..ustb.edu.cn 无人直升机自抗扰自适应轨迹跟踪混合控制 姜辰,王浩文四,李健珂,李亮君 清华大学航天航空学院,北京100084 ☒通信作者,E-mail:bobwang@mail.tsinghua.cdu.cn 摘要为满足无人直升机高精度轨迹跟踪的控制需求,并降低直升机动力学模型误差对飞行控制器飞行控制效果产生的 影响,提出自抗扰自适应直升机混合控制.该控制器的内环控制采用模型跟随自适应控制,通过使用动量反向传播算法 (MOBP)对该内环控制参数进行实时优化.通过使用自抗扰控制(ADRC)对直升机的水平速度进行控制.仿真结果表明,该 混合控制器能够实现直升机对预定轨迹的跟踪.相对PID和级联ADRC控制,该控制器具有更好的抗扰性和鲁棒性.通过在 200kg级的专业植保无人直升机XV2上搭载所提出的控制器,使其自主飞行轨迹跟踪控制的均方根误差在0.6m以内. 关键词无人直升机:自适应控制:自抗扰控制:飞行控制:神经网络 分类号V249.1 Trajectory-tracking hybrid controller based on ADRC and adaptive control for unmanned helicopters JIANG Chen,WANG Hao-wen,LI Jian-ke,LI Liang-jun School of Aerospace Engineering,Tsinghua University,Beijing 100084,China Corresponding author,E-mail:bobwang@mail.tsinghua.edu.cn ABSTRACT To enable unmanned helicopters to fly autonomously in precise paths and to reduce the influence of helicopter dynamic model error,this paper proposes a hybrid controller with active disturbance rejection control (ADRC)and adaptive control for trajecto- ry tracking.This paper proposes the model reference adaptive control strategy for the inner-loop controller.This paper uses the momentum back-propagation (MOBP)neural network algorithm to tune the parameters of the proposed inner-loop controller.This paper uses ADRC in the proposed controller for velocity control.The simulation results indicate that the proposed controller can achieve good trajectory tracking.Compared with the PID controller and cascade ADRC,the proposed hybrid controller is more robust and has better anti-disturbance capability.This paper uses the proposed hybrid controller for trajectory-tracking control of the XV2, which is an unmanned helicopter with a gross take-off weight level of 200 kg.With the help of the hybrid controller in our flight test, the root mean square error of the XV2 trajectory-tracking control is within 0.6m. KEY WORDS unmanned helicopter:adaptive control;active disturbance rejection control:flight control:neural network 近年来,无人机在植保领域得到越来越广泛的应 确跟踪的控制器对无人直升机的植保作业有重要 用.以保有量最多的YAMAHA RMAX为例0,飞控手 意义. 通过对直升机进行姿态遥控以完成航化喷洒作业回 在线性化的飞行动力学模型的基础上,多种线性 由于飞控手难以完成无人机超视距遥控飞行,实现无 控制算法被应用于直升机飞行控制,如PD控制回、线 人直升机高精度自主航迹对无人直升机在大地块植保 性二次调节控制田、鲁棒控制可等.针对直升机的非 作业领域的应用至关重要.因此设计能够进行轨迹精 线性动力学特性,一些研究将动态逆网、智能模糊控制 收稿日期:2017-01-18工程科学学报,第 39 卷,第 11 期: 1743--1752,2017 年 11 月 Chinese Journal of Engineering,Vol. 39,No. 11: 1743--1752,November 2017 DOI: 10. 13374 /j. issn2095--9389. 2017. 11. 018; http: / /journals. ustb. edu. cn 无人直升机自抗扰自适应轨迹跟踪混合控制 姜 辰,王浩文,李健珂,李亮君 清华大学航天航空学院,北京 100084  通信作者,E-mail: bobwang@ mail. tsinghua. edu. cn 摘 要 为满足无人直升机高精度轨迹跟踪的控制需求,并降低直升机动力学模型误差对飞行控制器飞行控制效果产生的 影响,提出自抗扰自适应直升机混合控制. 该控制器的内环控制采用模型跟随自适应控制,通过使用动量反向传播算法 ( MOBP) 对该内环控制参数进行实时优化. 通过使用自抗扰控制( ADRC) 对直升机的水平速度进行控制. 仿真结果表明,该 混合控制器能够实现直升机对预定轨迹的跟踪. 相对 PID 和级联 ADRC 控制,该控制器具有更好的抗扰性和鲁棒性. 通过在 200 kg 级的专业植保无人直升机 XV-2 上搭载所提出的控制器,使其自主飞行轨迹跟踪控制的均方根误差在 0. 6 m 以内. 关键词 无人直升机; 自适应控制; 自抗扰控制; 飞行控制; 神经网络 分类号 V249. 1 收稿日期: 2017--01--18 Trajectory-tracking hybrid controller based on ADRC and adaptive control for unmanned helicopters JIANG Chen,WANG Hao-wen ,LI Jian-ke,LI Liang-jun School of Aerospace Engineering,Tsinghua University,Beijing 100084,China  Corresponding author,E-mail: bobwang@ mail. tsinghua. edu. cn ABSTRACT To enable unmanned helicopters to fly autonomously in precise paths and to reduce the influence of helicopter dynamic model error,this paper proposes a hybrid controller with active disturbance rejection control ( ADRC) and adaptive control for trajecto￾ry tracking. This paper proposes the model reference adaptive control strategy for the inner-loop controller. This paper uses the momentum back-propagation ( MOBP) neural network algorithm to tune the parameters of the proposed inner-loop controller. This paper uses ADRC in the proposed controller for velocity control. The simulation results indicate that the proposed controller can achieve good trajectory tracking. Compared with the PID controller and cascade ADRC,the proposed hybrid controller is more robust and has better anti-disturbance capability. This paper uses the proposed hybrid controller for trajectory-tracking control of the XV-2, which is an unmanned helicopter with a gross take-off weight level of 200 kg. With the help of the hybrid controller in our flight test, the root mean square error of the XV-2 trajectory-tracking control is within 0. 6 m. KEY WORDS unmanned helicopter; adaptive control; active disturbance rejection control; flight control; neural network 近年来,无人机在植保领域得到越来越广泛的应 用. 以保有量最多的 YAMAHA RMAX 为例[1],飞控手 通过对直升机进行姿态遥控以完成航化喷洒作业[2]. 由于飞控手难以完成无人机超视距遥控飞行,实现无 人直升机高精度自主航迹对无人直升机在大地块植保 作业领域的应用至关重要. 因此设计能够进行轨迹精 确跟踪的控制器对无人直升机的植保作业有重要 意义. 在线性化的飞行动力学模型的基础上,多种线性 控制算法被应用于直升机飞行控制,如 PID 控制[3]、线 性二次调节控制[4]、鲁棒控制[5]等. 针对直升机的非 线性动力学特性,一些研究将动态逆[6]、智能模糊控制
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有