張維剛 張朋 韋昊 熊覺振
摘? ?要:在低附著路面情況下,針對(duì)現(xiàn)有以線性時(shí)變模型預(yù)測(cè)控制(LTVMPC)為基礎(chǔ)的無(wú)人駕駛汽車路徑跟蹤精確性和穩(wěn)定性問(wèn)題,提出一種改進(jìn)的控制算法. 以汽車動(dòng)力學(xué)理論為基礎(chǔ),將四輪輪胎側(cè)偏角和滑移率精確地表示為車輛狀態(tài)量的非線性函數(shù),在預(yù)測(cè)時(shí)域內(nèi)對(duì)車輛狀態(tài)方程線性化處理而求解雅可比矩陣時(shí),為降低系統(tǒng)維度,將輪速作為非狀態(tài)量,建立改進(jìn)的三自由度車輛模型,在二次規(guī)劃性能指標(biāo)中加入橫擺角速度跟蹤誤差項(xiàng)以提高路徑跟蹤性能,考慮質(zhì)心側(cè)偏角對(duì)跟蹤精度和車輛穩(wěn)定性的影響,修正參考橫擺角,建立改進(jìn)的LTVMPC. 在Carsim-Simulink聯(lián)合仿真平臺(tái)進(jìn)行低附著系數(shù)路面情況下的雙移線跟蹤仿真,結(jié)果表明改進(jìn)后的控制算法在保證實(shí)時(shí)性的前提下,提高了路徑跟蹤的精確性和車輛行駛的穩(wěn)定性.
關(guān)鍵詞:無(wú)人駕駛汽車;路徑跟蹤;線性時(shí)變模型預(yù)測(cè)控制;二次規(guī)劃
中圖分類號(hào):U467.1? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?文獻(xiàn)標(biāo)志碼:A
An Improved Path Tracking Control Algorithm
for Autonomous Vehicle Based on LTVMPC
ZHANG Weigang ZHANG Peng WEI Hao XIONG Juezhen
(State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,Hunan University,Changsha 410082,China)
Abstract:Under the case of low adhesion road,an improved control algorithm is proposed to solve the problem of the lack of accuracy and stability in the path tracking of autonomous vehicles based on Linear Time-Varying Model Predictive Control (LTVMPC). Based on vehicle dynamics,the slip angles and ratios of four tires are accurately expressed as the nonlinear function of vehicle state parameters. The Jacobian matrix is obtained by taking wheel speed as constant when linearizing the vehicle state equation in prediction horizon so as to reduce the dimension of the system,which aims to establish the improved 3-DOF vehicle model. The yaw rate tracking error is added into the performance index of quadratic programming to improve the path tracking performance and the influence of the slip angle of vehicle on the tracking accuracy and vehicle stability is considered to modify the reference yaw angle,which improves the overall performance of LTVMPC. The double lane change tracking simulation under the condition of low adhesion coefficient is performed on Carsim-Simulink co-simulation platform,and the results show that the improved control algorithm can increase the accuracy of path tracking and stability of vehicle while ensuring real-time performance.
Key words:autonomous vehicle;path tracking;Linear Time-Varying Model Predictive Control(LTVMPC);quadratic programming
路徑跟蹤是無(wú)人駕駛汽車自主行駛過(guò)程中的基本任務(wù)和關(guān)鍵環(huán)節(jié),其主要目標(biāo)是使車輛自動(dòng)沿著規(guī)劃路徑安全穩(wěn)定地行駛[1]. 當(dāng)車輛行駛在低附著路面上時(shí),由于此時(shí)車輛極易處于失穩(wěn)狀態(tài),路徑跟蹤精確性和車輛行駛穩(wěn)定性就面臨著很大的挑戰(zhàn).
為了增強(qiáng)路徑跟蹤控制效果,相關(guān)研究人員提出了許多算法. 模型預(yù)測(cè)控制(MPC)由于能夠有效處理系統(tǒng)約束、方便建立多輸入多輸出控制系統(tǒng)且具備前饋加反饋控制的優(yōu)點(diǎn),已經(jīng)成為其中最有效的方法之一[2]. MPC首先利用已有模型來(lái)預(yù)測(cè)系統(tǒng)未來(lái)的動(dòng)態(tài),并將系統(tǒng)當(dāng)前的狀態(tài)作為初始狀態(tài),以某項(xiàng)性能指標(biāo)達(dá)到最優(yōu)為目標(biāo),形成一個(gè)有限時(shí)域開環(huán)最優(yōu)化問(wèn)題,通過(guò)在線求解該問(wèn)題而獲得最優(yōu)控制量[3-4]. 由于在每個(gè)采樣時(shí)刻都求解二次規(guī)劃或非線性規(guī)劃問(wèn)題,當(dāng)預(yù)測(cè)模型過(guò)于復(fù)雜或考慮過(guò)多的非線性約束時(shí),計(jì)算量會(huì)過(guò)于龐大而限制其在車輛控制系統(tǒng)中的實(shí)際應(yīng)用[5-6]. 為提高算法的實(shí)時(shí)性,線性時(shí)變模型預(yù)測(cè)控制(LTVMPC)往往作為一種次優(yōu)的選擇[7].
LTVMPC在每個(gè)時(shí)間步長(zhǎng)內(nèi),將非線性車輛動(dòng)力學(xué)模型在當(dāng)前工作點(diǎn)連續(xù)線性化,且一般會(huì)將最優(yōu)化問(wèn)題轉(zhuǎn)化為二次規(guī)劃問(wèn)題進(jìn)行求解. 動(dòng)力學(xué)預(yù)測(cè)模型是LTVMPC的核心,模型的精度和復(fù)雜度直接決定了路徑跟蹤的精確性和實(shí)時(shí)性[8]. 文獻(xiàn)[9]使用三種不同精度的車輛模型來(lái)建立LTVMPC路徑跟蹤控制器,指出精度高的車輛模型可以有效提高跟蹤性能,但復(fù)雜的模型會(huì)相應(yīng)增加計(jì)算負(fù)擔(dān),當(dāng)模型的維度達(dá)到一定階次后,性能的提升幅度有限. 文獻(xiàn)[10]基于簡(jiǎn)化的三自由度車輛動(dòng)力學(xué)模型建立了典型的LTVMPC軌跡跟蹤控制器,但由于未考慮左右輪胎受力的差異及低附著工況下質(zhì)心側(cè)偏角和橫擺角速度對(duì)車輛穩(wěn)定性的影響,路徑跟蹤精確性和車輛行駛穩(wěn)定性的問(wèn)題依然存在.
文獻(xiàn)[11]研究了質(zhì)心側(cè)偏角對(duì)路徑跟蹤效果的影響,指出把參考橫擺角定義為期望路徑航向角,橫向誤差和橫擺角誤差將難以同時(shí)收斂趨于0,而利用質(zhì)心側(cè)偏角對(duì)參考橫擺角進(jìn)行補(bǔ)償能提高路徑跟蹤的精確性. 文獻(xiàn)[12-13]在LTVMPC性能指標(biāo)中加入橫擺角速度誤差,指出該項(xiàng)的加入能夠提高路徑跟蹤性能,但未對(duì)其影響作具體的闡述. 質(zhì)心側(cè)偏角與地面對(duì)輪胎的橫擺力矩和側(cè)向力變化范圍直接相關(guān),橫擺角速度表征了車輛的轉(zhuǎn)彎能力和動(dòng)態(tài)行為,通過(guò)對(duì)兩者實(shí)施相應(yīng)的動(dòng)力學(xué)約束能夠提升車輛的橫擺穩(wěn)定性[14]. 因此,在LTVMPC路徑跟蹤控制算法中,應(yīng)綜合考慮質(zhì)心側(cè)偏角和橫擺角速度的影響,以提高路徑跟蹤的精確性和車輛行駛的穩(wěn)定性.
基于以上分析,本文針對(duì)低附著路面情況下,基于LTVMPC的無(wú)人駕駛汽車路徑跟蹤控制精確性和穩(wěn)定性問(wèn)題,在精確建立四輪輪胎側(cè)偏角、滑移率和車輛狀態(tài)參數(shù)非線性關(guān)系的基礎(chǔ)上,考慮預(yù)測(cè)模型維度對(duì)算法實(shí)時(shí)性的影響,簡(jiǎn)化雅可比矩陣的求解,建立改進(jìn)的三自由度車輛模型. 在二次規(guī)劃性能指標(biāo)中加入橫擺角速度誤差項(xiàng),利用質(zhì)心側(cè)偏角修正參考橫擺角,并對(duì)橫擺角速度和質(zhì)心側(cè)偏角施加穩(wěn)定性約束,建立改進(jìn)的LTVMPC. 在Carsim-Simulink聯(lián)合仿真平臺(tái)進(jìn)行低附著路面情況下的雙移線跟蹤仿真,驗(yàn)證改進(jìn)措施的有效性.
1? ?汽車動(dòng)力學(xué)預(yù)測(cè)模型
假設(shè)車輛在水平路面上行駛,忽略空氣阻力,建立主要考慮車輛縱向、橫向以及橫擺動(dòng)力學(xué)特性的平面四輪車輛模型. 如圖1所示,oxyz為車輛坐標(biāo)系,固定于車輛質(zhì)心,OXYZ為慣性坐標(biāo)系.
2? ?路徑跟蹤目標(biāo)協(xié)調(diào)與優(yōu)化
3? ?改進(jìn)LTVMPC控制器設(shè)計(jì)
4? ?仿真實(shí)驗(yàn)結(jié)果及分析
5? ?結(jié)? ?論
針對(duì)低附著路面情況下基于LTVMPC的無(wú)人駕駛汽車路徑跟蹤控制精確性和穩(wěn)定性問(wèn)題,提出相應(yīng)的改進(jìn)措施,包括:1)建立兼顧路徑跟蹤精確性和實(shí)時(shí)性的改進(jìn)三自由度車輛模型;2)將橫擺角速度跟蹤誤差加入二次規(guī)劃性能指標(biāo),利用車輛質(zhì)心側(cè)偏角修正跟蹤目標(biāo),建立改進(jìn)的LTVMPC.
搭建Carsim-Simulink聯(lián)合仿真平臺(tái),在低附著路面情況下,車輛以30 km/h和70 km/h兩種初速度跟蹤雙移線的仿真表明:改進(jìn)后的車輛模型能提高路徑跟蹤精度且不影響實(shí)時(shí)性;將質(zhì)心側(cè)偏角和橫擺角速度加入跟蹤目標(biāo),能提高控制器的路徑跟蹤能力. 優(yōu)化后的LTVMPC路徑跟蹤控制器能更精確地跟蹤參考軌跡,且能提高車輛在低附著路面情況下的行駛穩(wěn)定性.
參考文獻(xiàn)
[1]? ? HANG P,LUO F M,F(xiàn)ANG S D,et al. Path tracking control of a four-wheel-independent-steering electric vehicle based on model predictive control[C]//第36屆中國(guó)控制會(huì)議論文集. 2017:1110—1116.
[2]? ? BAI G X,MENG Y,LIU L,et al. A new path tracking method based on multilayer model predictive control[J]. Applied Science,2019,9(13):2649—2666.
[3]? ? 龔建偉,劉凱,齊建勇. 無(wú)人駕駛車輛模型預(yù)測(cè)控制[M]. 第二版. 北京:北京理工大學(xué)出版社,2020.
GONG J W,LIU K,QI J Y. Model predictive control for self-driving vehicles[M].2nd Edition. Beijing:Beijing Institute of Technology Press,2020. (In Chinese)
[4]? ? MAYNE D Q,RAWLINGS J B,RAO C V,et al. Constrained model predictive control:Stability and optimality[J]. Automatica,2000,36(6):789—814.
[5]? ? GUO H Y,CAO D P,CHEN H,et al. Model predictive path following control for autonomous cars considering a measurable disturbance:Implementation,testing,and verification[J]. Mechanical Systems and Signal Processing,2019,118(1):41—60.
[6]? ? SWIEF A,El-ZAWAWI A,El-HABROUK M. A survey of model predictive control development in automotive industries[C]// 2019 International Conference on Applied Automation and Industrial Diagnostics(ICAAID). Elazig,Turkey:IEEE,2019:1—7.
[7]? ? VELHAL S,THOMAS S. Improved LTVMPC design for steering control of autonomous vehicle[J]. Journal of Physics:Conference Series,2017,783(1):012028.
[8]? ? JAMES B. RAWLINGS,DAVID Q. MAYNE M M. DIEHL. Model predictive control:Theory,Computation,and Design[M].2nd Edition. Santa Barbara:Nob Hill Publishing,LLC,2019.
[9]? ? CHEN S P,CHEN H Y,DAN N. Implementation of MPC-based trajectory tracking considering different fidelity vehicle models[J]. Journal of Beijing Institute of Technology,2020,2 9(03):303—316.
[10]? MING T Y,DENG W W,ZHANG S M,et al. MPC-based trajectory tracking control for intelligent vehicles[C]// SAE Technical Paper Series.400 Commonwealth Drive,Warrendale,PA,United States:SAE International,2016.
[11]? HU C,WANG R R,YAN F J,et al. Should the desired heading in path following of autonomous vehicles be the tangent direction of the desired path?[J]. Intelligent Transportation Systems,IEEE Trans- action on Intelligent Transportation Systems,2015,16(6):3084—3094.
[12]? FALCONE P,BORRELLI F,ASGARI J ,et al. Predictive active steering control for autonomous vehicle systems[J]. IEEE Transactions on Control Systems Technology,2007,15(3):566—580.
[13]? FALCONE P,TUFO M,BORRELLI F,et al.A linear time varying model predictive control approach to the integrated vehicle dynamics control problem in autonomous systems[C]//2007 46th IEEE Conference on Decision and Control. New Orleans,LA,USA:IEEE,2007:2980—2985.
[14]? RAJAMANI R.Vehicle dynamics and control[M]. Boston,MA:Springer US,2012.