Model-Structured Neural Networks to Control the Steering Dynamics of Autonomous Race Cars

TL;DR

This paper introduces MS-NN-steer, a model-structured neural network that incorporates vehicle dynamics knowledge to improve steering control in autonomous race cars, demonstrating better accuracy and robustness than traditional neural networks.

Contribution

The paper presents a novel model-structured neural network architecture that embeds vehicle dynamics into the neural network for improved steering control in autonomous racing.

Findings

MS-NN-steer outperforms general neural networks in accuracy and generalization.

MS-NN-steer is less sensitive to weight initialization.

The approach surpasses the steering control of the winning team in the Abu Dhabi Autonomous Racing League.

Abstract

Autonomous racing has gained increasing attention in recent years, as a safe environment to accelerate the development of motion planning and control methods for autonomous driving. Deep learning models, predominantly based on neural networks (NNs), have demonstrated significant potential in modeling the vehicle dynamics and in performing various tasks in autonomous driving. However, their black-box nature is critical in the context of autonomous racing, where safety and robustness demand a thorough understanding of the decision-making algorithms. To address this challenge, this paper proposes MS-NN-steer, a new Model-Structured Neural Network for vehicle steering control, integrating the prior knowledge of the nonlinear vehicle dynamics into the neural architecture. The proposed controller is validated using real-world data from the Abu Dhabi Autonomous Racing League (A2RL) competition, with full-scale autonomous race cars. In comparison with general-purpose NNs, MS-NN-steer is shown to achieve better accuracy and generalization with small training datasets, while being less sensitive to the weights' initialization. Also, MS-NN-steer outperforms the steering controller used by the A2RL winning team. Our implementation is available open-source in a GitHub repository.