A drive-by-wire system seems to be a crucial technology for future road vehicles, including personal and commercial cars and urban road mobility platforms. The augmentation of human driving capabilities in the domain of maneuver safety and driving performance is beneficial. The most important benefit comes for autonomous driving, where the drive-by-wire represents technological enablers. The drive-by-wire architecture provides straight forward means of vehicle dynamics control system implementation. Mechatronics systems replace direct mechanical control links. Such a solution brings a possibility to change the inputs commanded by the driver according to the driving conditions.
The drive-by-wire vehicle dynamics control algorithm is proposed in this work, which helps the driver to steer, drive, and brake the car within predefined safety limits. The set of vehicle maneuvers represented by state values, where the vehicle wheels' sliding motion stays in defined limits, is defined in previous work. This set is called the Driving Envelope (inspired by aerospace application, namely the flight envelope used for modern aircraft).
The safety limits of car maneuvers are represented by the Driving Envelop in vehicle states (speed, sideslip angle, and cornering speed in the center of gravity). The control system design protecting safety limits violation is presented in this work. The control functionality based on Model Predictive Control (MPC) approach, as both states and inputs constraints, are respected during optimization. The vehicle dynamics control system is denoted Driving Envelop Protection, again inspired by the flight control system and flight envelope protection functionality. The lateral vehicle dynamics is controlled using the MPC objective designed to track reference steering angle of the front axle commanded by the driver. The longitudinal vehicle dynamics is controlled based on wheel angular velocities of the front and rear axles, commanded from pedals. The optimization process is constrained by vehicle maneuver boundaries defined by the driving envelope. Hence, the car is protected from getting into a critical spin situation, wheel blocking, and loss of wheels' traction.
The algorithm is validated using virtual ride tests on a simulator with an implemented high-fidelity twin-track model. During the experiments, an uncontrolled car was compared with the same but controlled vehicle on a couple of different maneuvers, including double-lane change tests, obstacle avoidance tests, and slalom ride tests.
Ing. Denis Efremov, Czech Technical University in Prague, Faculty of Electrical Engineering, Department of Control Engineering, CZECH REPUBLIC; Dr.-Ing. Tomas Hanis, Czech Technical University in Prague, Faculty of Electrical Engineering, Department of Control Engineering, CZECH REPUBLIC; Ing. David Vosahlik, Czech Technical University in Prague, Faculty of Electrical Engineering, Department of Control Engineering, CZECH REPUBLIC; Dr.-Ing. Martin Klauco, Institute of Information Engineering, Automation, and Mathematics, Slovak University of Technology in Bratislava, SLOVAKIA