Coupled Longitudinal and Lateral Control of a Vehicle using Deep Learning
This work addresses vehicle automation by improving integrated control, but it is incremental as it applies existing deep learning methods to a known control challenge.
This paper tackled the problem of coupled longitudinal and lateral vehicle control by training deep neural networks (MLP and CNN) on simulation data to compute steering and torque inputs, achieving performance evaluated on a challenging test track with comparisons to conventional decoupled controllers.
This paper explores the capability of deep neural networks to capture key characteristics of vehicle dynamics, and their ability to perform coupled longitudinal and lateral control of a vehicle. To this extent, two different artificial neural networks are trained to compute vehicle controls corresponding to a reference trajectory, using a dataset based on high-fidelity simulations of vehicle dynamics. In this study, control inputs are chosen as the steering angle of the front wheels, and the applied torque on each wheel. The performance of both models, namely a Multi-Layer Perceptron (MLP) and a Convolutional Neural Network (CNN), is evaluated based on their ability to drive the vehicle on a challenging test track, shifting between long straight lines and tight curves. A comparison to conventional decoupled controllers on the same track is also provided.