Aggressive Deep Driving: Model Predictive Control with a CNN Cost Model

TL;DR

This paper introduces a vision-based model predictive control framework for aggressive high-speed autonomous driving, utilizing CNNs to predict cost maps directly from video for online trajectory optimization.

Contribution

It presents a novel approach combining CNN-based cost prediction with MPC for high-speed autonomous driving using monocular camera input.

Findings

Effective in high-speed aggressive driving scenarios

Uses a single monocular camera for cost map prediction

Demonstrated on a 1:5 scale autonomous vehicle

Abstract

We present a framework for vision-based model predictive control (MPC) for the task of aggressive, high-speed autonomous driving. Our approach uses deep convolutional neural networks to predict cost functions from input video which are directly suitable for online trajectory optimization with MPC. We demonstrate the method in a high speed autonomous driving scenario, where we use a single monocular camera and a deep convolutional neural network to predict a cost map of the track in front of the vehicle. Results are demonstrated on a 1:5 scale autonomous vehicle given the task of high speed, aggressive driving.