Trajectory Tracking for Automated Driving Functions

摘要

In the future, automated vehicles with higher levels of automation will be an essential component of individual mobility [1]. To realize automation, all tasks that have previously been performed by the human driver must be taken over and executed by the vehicle. This results in a multitude of challenges to be solved, which are described in Lienke et al. [2]. In addition to the detection of the environment, the situation description, recognition, prediction and the planning of the movement of the ego vehicle, the control of the vehicle, through the actuation of the actuators, is one of the tasks to be realized. This is accomplished in a trajectory tracking control, which guides the vehicle as precisely as possible on the planned movement. The human driver has learned the connection between the steering wheel, pedal positions and the vehicle movement through experiences as a passenger and driver over a longer period of time. A trajectory tracking control system must also take these correlations into account. Due to this fact, the consideration of model knowledge for feedforward control of the steering wheel and pedal positions for the trajectory tracking is useful.