Stochastic predictive control based motion planning for lane change decision using a Vehicle Traffic Simulator

This paper describes the design and evaluation of a model predictive control algorithm for automated driving on a motorway using a Vehicle Traffic Simulator. For the development of a highly automated driving control algorithm, motion planning is necessary to satisfy driving condition in various road traffic situations. There are two key issues in motion planning of automated driving vehicles. One of the key issues is how to handle potentially dangerous situations that could occur in order to guarantee the safety of vehicles. The second key issue is how to guarantee the robustness of the controller under model uncertainties and external disturbances. To improve safety with respect to the future behaviors of subject vehicles, not the current states but rather the predicted behaviors of surrounding vehicles should be considered. The desired driving mode and a safe driving envelope are determined based on the probabilistic prediction of surrounding vehicles behaviors over a finite prediction horizon. To obtain the desired steering angle and longitudinal acceleration for maintaining the subject vehicle in the safe driving envelope during a finite prediction horizon, a motion planning controller is designed based on an MPC approach. The developed control algorithm has been successfully implemented on a vehicle ECU. The proposed control algorithm has been evaluated on a real-time vehicle traffic simulator. The throttle, brake, and steering control inputs and the controlled vehicle behavior have been compared to those of manual driving.

[1]  Benoit Vanholme,et al.  Highly Automated Driving on Highways Based on Legal Safety , 2013, IEEE Transactions on Intelligent Transportation Systems.

[2]  Kyongsu Yi,et al.  Human driving data-based design of a vehicle adaptive cruise control algorithm , 2008 .

[3]  Roland Siegwart,et al.  Toward automated driving in cities using close-to-market sensors: An overview of the V-Charge Project , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[4]  F. Borrelli,et al.  Stochastic Predictive Control of Autonomous Vehicles in Uncertain Environments , 2014 .

[5]  David Q. Mayne,et al.  Constrained model predictive control: Stability and optimality , 2000, Autom..

[6]  Pongsathorn Raksincharoensak,et al.  Lane Change Behavior Modeling for Autonomous Vehicles Based on Surroundings Recognition , 2011 .

[7]  Claire J. Tomlin,et al.  A probabilistic approach to planning and control in autonomous urban driving , 2013, 52nd IEEE Conference on Decision and Control.

[8]  Sterling J. Anderson,et al.  An optimal-control-based framework for trajectory planning, threat assessment, and semi-autonomous control of passenger vehicles in hazard avoidance scenarios , 2010 .

[9]  Pieter Abbeel,et al.  Finding Locally Optimal, Collision-Free Trajectories with Sequential Convex Optimization , 2013, Robotics: Science and Systems.

[10]  Basil Kouvaritakis,et al.  Explicit use of probabilistic distributions in linear predictive control , 2010, Autom..

[11]  David Q. Mayne,et al.  Robust model predictive control of constrained linear systems with bounded disturbances , 2005, Autom..

[12]  John M. Dolan,et al.  A prediction- and cost function-based algorithm for robust autonomous freeway driving , 2010, 2010 IEEE Intelligent Vehicles Symposium.

[13]  Francesco Borrelli,et al.  Robust Predictive Control for semi-autonomous vehicles with an uncertain driver model , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[14]  D. A. Schwartz.  Clothoid road geometry unsuitable for sensor fusion clothoid parameter sloshing , 2003, IEEE IV2003 Intelligent Vehicles Symposium. Proceedings (Cat. No.03TH8683).

[15]  Shekhar Verma,et al.  Prediction of Lane Change Trajectories through Neural Network , 2010, 2010 International Conference on Computational Intelligence and Communication Networks.