竇立謙,楊?闖,王丹丹,陳?濤,秦新立
Dou Liqian,Yang Chuang,Wang Dandan,Chen Tao,Qin Xinli
?
基于狀態(tài)觀測(cè)器的多無(wú)人機(jī)編隊(duì)跟蹤控制
竇立謙,楊?闖,王丹丹,陳?濤,秦新立
(天津大學(xué)電氣自動(dòng)化與信息工程學(xué)院,天津 300072)
針對(duì)一組欠驅(qū)動(dòng)四旋翼無(wú)人機(jī)在編隊(duì)飛行中僅有部分無(wú)人機(jī)可以直接獲取領(lǐng)機(jī)狀態(tài)信息的問(wèn)題,提出了一種基于狀態(tài)觀測(cè)器的分布式有限時(shí)間編隊(duì)跟蹤控制策略.根據(jù)四旋翼無(wú)人機(jī)系統(tǒng)嚴(yán)格反饋的結(jié)構(gòu)特點(diǎn),將四旋翼無(wú)人機(jī)的動(dòng)力學(xué)模型劃分為位置子系統(tǒng)和姿態(tài)子系統(tǒng),然后分別進(jìn)行位置控制器和姿態(tài)控制器的設(shè)計(jì).首先,考慮到在分布式的編隊(duì)控制策略下,并非所有的無(wú)人機(jī)都能直接與領(lǐng)機(jī)進(jìn)行通信并獲取領(lǐng)機(jī)的狀態(tài)信息.對(duì)每架四旋翼無(wú)人機(jī)分別設(shè)計(jì)分布式有限時(shí)間狀態(tài)觀測(cè)器估計(jì)自身與領(lǐng)機(jī)的相對(duì)狀態(tài)信息,穩(wěn)定性分析表明所設(shè)計(jì)的狀態(tài)觀測(cè)器的觀測(cè)誤差能夠在有限時(shí)間內(nèi)趨近于零;其次,在四旋翼無(wú)人機(jī)狀態(tài)觀測(cè)器觀測(cè)結(jié)果的基礎(chǔ)上設(shè)計(jì)了有限時(shí)間位置控制器,穩(wěn)定性分析表明位置控制器能夠在有限時(shí)間內(nèi)實(shí)現(xiàn)對(duì)領(lǐng)機(jī)位置的穩(wěn)定跟蹤;然后根據(jù)位置環(huán)控制量解算出期望的姿態(tài)角,基于滑模控制方法設(shè)計(jì)了姿態(tài)控制器,穩(wěn)定性分析表明各架無(wú)人機(jī)的姿態(tài)角能在有限時(shí)間內(nèi)跟蹤上期望的姿態(tài)角;最后,從仿真結(jié)果中能夠看出所設(shè)計(jì)的狀態(tài)觀測(cè)器的觀測(cè)誤差能夠在有限時(shí)間內(nèi)趨近于零,即每架無(wú)人機(jī)的狀態(tài)觀測(cè)器能夠在有限時(shí)間內(nèi)觀測(cè)到自身與領(lǐng)機(jī)的相對(duì)狀態(tài)信息.從無(wú)人機(jī)飛行軌跡中能夠看出各架無(wú)人機(jī)能夠在有限時(shí)間內(nèi)形成并保持期望的隊(duì)形.
四旋翼無(wú)人機(jī);編隊(duì)跟蹤控制;狀態(tài)觀測(cè)器;分布式;有限時(shí)間穩(wěn)定
四旋翼無(wú)人機(jī)因其具有結(jié)構(gòu)簡(jiǎn)單、機(jī)身質(zhì)量輕、靈活性強(qiáng)而且能夠垂直起降和定點(diǎn)懸停等優(yōu)點(diǎn),在環(huán)境監(jiān)測(cè)、目標(biāo)搜索、事故救援等領(lǐng)域得到廣泛應(yīng)用[1].多無(wú)人機(jī)編隊(duì)通過(guò)多架無(wú)人機(jī)的協(xié)同飛行使系統(tǒng)的綜合效能大大提高.隨著任務(wù)日趨復(fù)雜,無(wú)人機(jī)數(shù)量逐漸增多,對(duì)編隊(duì)控制提出更高的要求[2].無(wú)人機(jī)編隊(duì)控制是多無(wú)人機(jī)協(xié)同控制的重要研究方向.無(wú)論在軍事領(lǐng)域,還是在民用領(lǐng)域,多機(jī)編隊(duì)都有巨大的應(yīng)用前景,獲得國(guó)內(nèi)外學(xué)者廣泛的關(guān)注.
四旋翼無(wú)人機(jī)是一個(gè)具有6自由度和4個(gè)控制輸入的欠驅(qū)動(dòng)系統(tǒng),具有非線性、強(qiáng)耦合、多變量等特點(diǎn),控制問(wèn)題復(fù)雜.早期的編隊(duì)控制主要采用集中式控制方法,其特點(diǎn)是精度高、便于控制,但要依賴(lài)于中央控制單元的計(jì)算能力和全局通信能力.
目前,應(yīng)用比較成熟的多無(wú)人機(jī)編隊(duì)控制方法主要有:leader-follower法[3]、基于行為法[4]和虛擬結(jié)構(gòu)法[5].文獻(xiàn)[6]提出了一種基于leader-follower法的編隊(duì)控制算法.文獻(xiàn)[7]研究了基于行為法的多無(wú)人機(jī)編隊(duì)控制問(wèn)題,但文中假設(shè)每架無(wú)人機(jī)之間可以相互通信以共享狀態(tài)信息.文獻(xiàn)[8]設(shè)計(jì)了一種小型四旋翼無(wú)人機(jī)群起飛后自主形成正多邊形編隊(duì)的分布式運(yùn)動(dòng)控制方法,在四旋翼無(wú)人機(jī)串級(jí)控制系統(tǒng)框架下,同時(shí)采用平均一致性算法和有領(lǐng)導(dǎo)一致性算法,但文中將四旋翼模型簡(jiǎn)化為一階系統(tǒng),控制誤差較大.文獻(xiàn)[9]設(shè)計(jì)了一種分布式協(xié)同控制算法,使用二階四旋翼無(wú)人機(jī)模型,但是期望隊(duì)形無(wú)法在有限時(shí)間內(nèi)形成.
本文針對(duì)通信受限情況下,多機(jī)編隊(duì)中領(lǐng)機(jī)參考狀態(tài)不能被所有無(wú)人機(jī)直接獲取的情況,設(shè)計(jì)了一種基于狀態(tài)觀測(cè)器的分布式有限時(shí)間編隊(duì)控制策略.對(duì)每一架無(wú)人機(jī)設(shè)計(jì)狀態(tài)觀測(cè)器,可以在有限時(shí)間內(nèi)觀測(cè)自身與領(lǐng)機(jī)的狀態(tài)差.利用狀態(tài)觀測(cè)器的觀測(cè)值設(shè)計(jì)編隊(duì)控制器實(shí)現(xiàn)有限時(shí)間內(nèi)編隊(duì)隊(duì)形的形成與保持.通過(guò)穩(wěn)定性分析來(lái)驗(yàn)證分布式有限時(shí)間控制器的有效性,并通過(guò)仿真來(lái)驗(yàn)證該控制策略的性能.
圖1?無(wú)人機(jī)結(jié)構(gòu)和坐標(biāo)系
本文控制目標(biāo)是為每一架無(wú)人機(jī)設(shè)計(jì)一種基于分布式有限時(shí)間狀態(tài)觀測(cè)器的編隊(duì)跟蹤控制器,使多架無(wú)人機(jī)形成并保持固定的編隊(duì)隊(duì)形飛行,具體為
圖2?第架四旋翼無(wú)人機(jī)控制結(jié)構(gòu)
把模型(1)中的位置子系統(tǒng)寫(xiě)成
其中
對(duì)時(shí)間求導(dǎo)得
基于觀測(cè)器的觀測(cè)值,設(shè)計(jì)位置控制器的形式為
選擇李雅普諾夫函數(shù)的形式為
期望偏航角為
旋翼提供的升力為
下面設(shè)計(jì)姿態(tài)跟蹤控制器,定義姿態(tài)跟蹤誤差為
對(duì)其進(jìn)行求導(dǎo)得
將模型(1)中的姿態(tài)子系統(tǒng)代入式(35)得
分兩步證明系統(tǒng)(38)有限時(shí)間穩(wěn)定,首先證明漸近穩(wěn)定,選取李雅普諾夫函數(shù)的形式為
通過(guò)文獻(xiàn)[15]中引理2.1可知系統(tǒng)(38)全局有限時(shí)間穩(wěn)定.
為驗(yàn)證所提控制策略的正確性和有效性,以4架無(wú)人機(jī)組成編隊(duì)并使編隊(duì)中心沿著虛擬領(lǐng)機(jī)飛行進(jìn)行仿真分析.圖3給出了各無(wú)人機(jī)間的通信拓?fù)浣Y(jié)構(gòu),其中0代表虛擬領(lǐng)機(jī),1~4分別代表4架四旋翼從機(jī).
圖3?編隊(duì)通信拓?fù)?/p>
無(wú)人機(jī)初始位置、速度及角速度參數(shù)如下:
圖4?觀測(cè)器狀態(tài)值與每架無(wú)人機(jī)和領(lǐng)機(jī)速度差的差值曲線
圖5?觀測(cè)器狀態(tài)值與每架無(wú)人機(jī)和領(lǐng)機(jī)位置差的差值曲線
圖6?x-y平面及立體空間編隊(duì)飛行軌跡
本文針對(duì)分布式下多架四旋翼無(wú)人機(jī)的編隊(duì)控制問(wèn)題,提出了一種基于狀態(tài)觀測(cè)器的分布式有限時(shí)間編隊(duì)跟蹤控制策略.分布式情況下,虛擬領(lǐng)機(jī)的狀態(tài)信息只有部分從機(jī)可以直接獲得.利用狀態(tài)觀測(cè)器來(lái)觀測(cè)自身與領(lǐng)機(jī)的狀態(tài)差,狀態(tài)觀測(cè)器能夠在有限時(shí)間內(nèi)估計(jì)出自身與領(lǐng)機(jī)的相對(duì)狀態(tài).根據(jù)觀測(cè)器的觀測(cè)結(jié)果設(shè)計(jì)的編隊(duì)控制器,能夠在有限時(shí)間內(nèi)實(shí)現(xiàn)多架無(wú)人機(jī)形成并保持期望的隊(duì)形.基于滑模方法設(shè)計(jì)的姿態(tài)控制器能夠使無(wú)人機(jī)姿態(tài)在有限時(shí)間內(nèi)跟蹤上期望的姿態(tài)角.最后,仿真結(jié)果驗(yàn)證了所設(shè)計(jì)觀測(cè)器的觀測(cè)誤差能夠在有限時(shí)間內(nèi)趨于零,所設(shè)計(jì)的控制器能夠保證多架無(wú)人機(jī)快速形成并保持期望的編隊(duì)隊(duì)形.
[1] Kendoul F,Lara D,F(xiàn)antoni I,et al. Nonlinear control for systems with bounded inputs:Real-time embedded control applied to UAVs[C]// IEEE Conference on Decision and Control. San Diego,CA,USA 2006:5888-5893.
[2] 宗?群,王丹丹,邵士凱,等. 多無(wú)人機(jī)協(xié)同編隊(duì)飛行控制研究現(xiàn)狀及發(fā)展[J]. 哈爾濱工業(yè)大學(xué)學(xué)報(bào),2017,49(3):1-14. Zong Qun,Wang Dandan,Shao Shikai,et al. Research status and development of multi UAV coordinated formation flight control[J]. Journal of Harbin Institute of Technology,2017,49(3):1-14(in Chinese).
[3] Wang D,Zong Q,Tian B,et al. Neural network disturbance observer-based distributed finite-time formation tracking control for multiple unmanned helicopters[J]. ISA Transactions,2018,73:208-226.
[4] Turpin M,Michael N,Kumar V. Decentralized formation control with variable shapes for aerial robots[C]// IEEE International Conference on Robotics and Automation. Saint Paul,MN,USA,2012:23-30.
[5] Lalish E,Morgansen K A,Tsukamaki T. Formation tracking control using virtual structures and deconfliction[C]//IEEE Conference on Decision and Control. San Diego,CA,USA,2006:5699-5705.
[6] Gu Y,Seanor B,Campa G,et al. Design and flight testing evaluation of formation control laws[J]. IEEE Transactions on Control Systems Technology,2006,14(6):1105-1112.
[7] Kim S. Behavior-based decentralized control for multi-UAV formation flight[J]. Indian Journal of Genetics & Plant Breeding,2011,74(4):409-413.
[8] 邢關(guān)生,杜春燕,宗?群,等. 基于一致性的小型四旋翼機(jī)群自主編隊(duì)分布式運(yùn)動(dòng)規(guī)劃[J]. 控制與決策,2014,29(11):2081-2084.
Xing Guansheng,Du Chunyan,Zong Qun,et al. Consensus-based distributed motion planning for autonomous formation of miniature quadrotor groups[J]. Control and Decision,2014,29(11):2081-2084(in Chinese).
[9] Ghommam J,Luque-Vega L F,Castillo-Toledo B,et al. Three-dimensional distributed tracking control for multiple quadrotor helicopters[J]. Journal of the Franklin Institute,2016,353(10):2344-2372.
[10] Zuo Z. Trajectory tracking control design with command-filtered compensation for a quadrotor[J]. IET Control Theory & Applications,2013,4(11):2343-2355.
[11] Andreasson M,Dimarogonas D V,Johansson K H. Undamped nonlinear consensus using integral Lyapunov functions[C]// IEEE American Control Conference. Montreal,QC,Canada,2012:6644-6649.
[12] Paden B E,Sastry S S. A calculus for computing Filippov’s differential inclusion with application to the variable structure control of robot manipulators[C]// IEEE Conference on Decision and Control. Athens,Greece,1987:578-582.
[13] Orlov Y. Finite time stability and robust control synthesis of uncertain switched systems[J]. SIAM Journal on Control & Optimization,2004,43(4):1253-1271.
[14] Khalil H K. Nonlinear Systems[M]. 3rd ed. Upper Saddle River,NJ:Prentice-Hall,Inc,2002.
[15] Du H,Zhu W,Wen G,et al. Finite-time formation control for a group of quadrotor aircraft[J]. Aerospace Science & Technology,2017,69:609-616.
State Observer-Based Formation Tracking Control for Multiple Quadrotors
Dou Liqian,Yang Chuang,Wang Dandan,Chen Tao,Qin Xinli
(School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China)
Abstract:A distributed finite-time formation tracking control strategy based on a state observer is proposed for a group of under-actuated quadrotors,where only a part of the quadrotors can directly obtain the state information of the leader quadrotor. According to the strict feedback architecture of the quadrotor system,the dynamic model of the quadrotor is divided into the position and attitude subsystems,from which the position and attitude controllers are designed,respectively. First,given that not all quadrotors can communicate with the leader and obtain its state information under the condition of using a distributed formation control strategy,the distributed finite-time state observer is designed for each quadrotor to estimate its relative state information and that of the leader quadrotor. The stability analysis shows that the error of the state observer can reach zero in finite time. Second,a finite-time position controller is designed based on the observed results of the state observer. The stability analysis shows that the position controller can achieve stable tracking of the position of the leader quadrotor in finite time. Based on this finding,the desired attitude angle can be calculated according to the values of the position controller,and the attitude controller can be designed using the sliding mode control method. The stability analysis shows that the desired attitude angle of each quadrotor can be tracked in finite time. Finally,the simulations results show that the error of the designed state observer can reach zero in finite time,indicating that the state observer of each quadrotor can obtain the relative state information between itself and the leader quadrotor in finite time. The flight trajectory of each quadrotor can be used to generate and maintain the desired formation in finite time.
quadrotor;formation tracking control;state observer;distributed;finite-time stable
TK448.21
A
0493-2137(2019)01-0090-08
2018-03-19;
2018-08-17.
竇立謙(1976—??),男,博士,副教授,douliqian@tju.edu.cn.
王丹丹,dandanwang0910@163.com.
國(guó)家自然科學(xué)基金資助項(xiàng)目(61873340,61773279).
the National Natural Science Foundation of China(No. 61873340,No. 61773279).
10.11784/tdxbz201803057
(責(zé)任編輯:孫立華)