Incorporating Orientations into End-to-end Driving Model for Steering Control

TL;DR

This paper introduces a novel end-to-end neural network for autonomous steering control that incorporates pixel-wise orientations and a cost-sensitive loss to improve accuracy across diverse driving scenarios.

Contribution

It presents a new model integrating orientation-aware features, a cost-sensitive loss function, and a comprehensive dataset for end-to-end driving tasks.

Findings

Model predicts steering angles accurately.

Incorporating orientations improves direction-awareness.

Proposed loss function handles data imbalance effectively.

Abstract

In this paper, we present a novel end-to-end deep neural network model for autonomous driving that takes monocular image sequence as input, and directly generates the steering control angle. Firstly, we model the end-to-end driving problem as a local path planning process. Inspired by the environmental representation in the classical planning algorithms(i.e. the beam curvature method), pixel-wise orientations are fed into the network to learn direction-aware features. Next, to handle the imbalanced distribution of steering values in training datasets, we propose an improvement on a cost-sensitive loss function named SteeringLoss2. Besides, we also present a new end-to-end driving dataset, which provides corresponding LiDAR and image sequences, as well as standard driving behaviors. Our dataset includes multiple driving scenarios, such as urban, country, and off-road. Numerous experiments are conducted on both public available LiVi-Set and our own dataset, and the results show that the model using our proposed methods can predict steering angle accurately.