Collision Imminent Steering at High Speed Using Nonlinear Model Predictive Control

被引:22
作者
Wurts, John [1 ]
Stein, Jeffrey L. [1 ]
Ersal, Tulga [1 ]
机构
[1] Univ Michigan, Dept Mech Engn, Ann Arbor, MI 48109 USA
关键词
Trajectory; Safety; Road transportation; Collision avoidance; Predictive control; Optimization; Vehicle dynamics; Advanced driver assistance systems; autonomous vehicles; nonlinear control systems; trajectory optimization; vehicle dynamics; AUTONOMOUS GROUND VEHICLES; OBSTACLE AVOIDANCE; PATH TRACKING; ALGORITHM;
D O I
10.1109/TVT.2020.2999612
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Collision imminent steering is an automotive active safety feature designed to perform an aggressive lane change to avoid a forward collision in the event that the system determines there is insufficient distance to avoid a collision by braking alone.In this paper, a collision imminent steering system is developed using nonlinear model predictive control to perform a lane change at high speed in a highway environment. Two formulations are presented: one formulation to show the theoretical minimum lane change distance within which a safe lane change can be performed, and a second formulation to find the minimum aggressive maneuver for a given lane change distance. The nonlinear model predictive control formulation is posed as a constrained optimization problem, where solutions obey hard safety constraints of the highway environment. Constraint aggregation techniques are analyzed and introduced to monitor the peak tire slip throughout the maneuver. Numerical simulations of a typical luxury sedan in a highway setting show that minimum distance required to change lanes is about half the limit braking distance. Hence, there is a window of feasibility where the vehicle cannot avoid collision by braking alone, but can change lanes safely and maintain a safe trajectory throughout the maneuver. Additional simulations show that an active four wheel steering architecture improves performance over front only steering.
引用
收藏
页码:8278 / 8289
页数:12
相关论文
共 32 条
[1]   Obstacle Avoidance in Real Time With Nonlinear Model Predictive Control of Autonomous Vehicles [J].
Abbas, Muhammad Awais ;
Milman, Ruth ;
Eklund, J. Mikael .
CANADIAN JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING-REVUE CANADIENNE DE GENIE ELECTRIQUE ET INFORMATIQUE, 2017, 40 (01) :12-22
[2]   A comparison of alternative obstacle avoidance strategies for vehicle control [J].
Alleyne, A .
VEHICLE SYSTEM DYNAMICS, 1997, 27 (5-6) :371-392
[3]  
Alsterda J., 2019, P IEEE ANN AM CONTR, P718
[4]  
Altché F, 2017, IEEE INT VEH SYM, P141, DOI 10.1109/IVS.2017.7995711
[5]   An optimal-control-based framework for trajectory planning, threat assessment, and semi-autonomous control of passenger vehicles in hazard avoidance Scenarios [J].
Anderson S.J. ;
Peters S.C. ;
Pilutti T.E. ;
Iagnemma K. .
International Journal of Vehicle Autonomous Systems, 2010, 8 (2-4) :190-216
[6]  
[Anonymous], 1987, SAE TECH PAPER 87042, DOI DOI 10.4271/870421
[7]   Safe driving envelopes for path tracking in autonomous vehicles [J].
Brown, Matthew ;
Funke, Joseph ;
Erlien, Stephen ;
Gerdes, J. Christian .
CONTROL ENGINEERING PRACTICE, 2017, 61 :307-316
[8]  
Chakraborty I, 2013, P AMER CONTR CONF, P2165
[9]  
Di Cairano S, 2016, IEEE DECIS CONTR P, P709, DOI 10.1109/CDC.2016.7798351
[10]  
Erlien SM., 2013, IFAC Proceedings Volumes, V46, P831