End-to-end Multi-Modal Multi-Task Vehicle Control for Self-Driving Cars with Visual Perception

TL;DR

This paper introduces a multi-modal, multi-task end-to-end neural network for autonomous vehicle control that predicts both steering and speed from visual inputs, improving accuracy and robustness over previous single-task models.

Contribution

It presents a novel multi-task learning framework that predicts steering and speed simultaneously, incorporating previous feedback and multi-modal data for enhanced vehicle control.

Findings

Accurate prediction of steering angles and speeds on public datasets.

Improved failure data synthesis methods for real-world testing.

Enhanced robustness in vehicle control through multi-task learning.

Abstract

Convolutional Neural Networks (CNN) have been successfully applied to autonomous driving tasks, many in an end-to-end manner. Previous end-to-end steering control methods take an image or an image sequence as the input and directly predict the steering angle with CNN. Although single task learning on steering angles has reported good performances, the steering angle alone is not sufficient for vehicle control. In this work, we propose a multi-task learning framework to predict the steering angle and speed control simultaneously in an end-to-end manner. Since it is nontrivial to predict accurate speed values with only visual inputs, we first propose a network to predict discrete speed commands and steering angles with image sequences. Moreover, we propose a multi-modal multi-task network to predict speed values and steering angles by taking previous feedback speeds and visual recordings as inputs. Experiments are conducted on the public Udacity dataset and a newly collected SAIC dataset. Results show that the proposed model predicts steering angles and speed values accurately. Furthermore, we improve the failure data synthesis methods to solve the problem of error accumulation in real road tests.