Published on Thu Sep 06 2018

Driving Experience Transfer Method for End-to-End Control of Self-Driving Cars

Dooseop Choi, Taeg-Hyun An, Kyounghwan Ahn, Jeongdan Choi

A convolutional neural network trained on a source domain can be utilized for the same task in a different target domain. A conventional CNN for the end-to-end control is designed to map a single front-facing camera image to a steering command.

0
0
0
Abstract

In this paper, we present a transfer learning method for the end-to-end control of self-driving cars, which enables a convolutional neural network (CNN) trained on a source domain to be utilized for the same task in a different target domain. A conventional CNN for the end-to-end control is designed to map a single front-facing camera image to a steering command. To enable the transfer learning, we let the CNN produce not only a steering command but also a lane departure level (LDL) by adding a new task module, which takes the output of the last convolutional layer as input. The CNN trained on the source domain, called source network, is then utilized to train another task module called target network, which also takes the output of the last convolutional layer of the source network and is trained to produce a steering command for the target domain. The steering commands from the source and target network are finally merged according to the LDL and the merged command is utilized for controlling a car in the target domain. To demonstrate the effectiveness of the proposed method, we utilized two simulators, TORCS and GTAV, for the source and the target domains, respectively. Experimental results show that the proposed method outperforms other baseline methods in terms of stable and safe control of cars.

Sat Jan 20 2018
Computer Vision
End-to-end Multi-Modal Multi-Task Vehicle Control for Self-Driving Cars with Visual Perception
Convolutional Neural Networks (CNN) have been successfully applied to autonomous driving tasks, many in an end-to-end manner. We propose a multi-task learning framework to predict the steering angle and speed control simultaneously.
0
0
0
Sat Oct 26 2019
Machine Learning
Deep Learning and Control Algorithms of Direct Perception for Autonomous Driving
CNNs (convolutional neural networks) AlexNet and GoogLeNet that map an input image to few perception indicators. We also design a controller with these indicators and the short-range sensor information of TORCS for driving simulated cars to avoid collisions.
0
0
0
Tue Nov 13 2018
Artificial Intelligence
Two-stream convolutional networks for end-to-end learning of self-driving cars
We propose a methodology to extend the concept of Two-Stream Convolutional Networks to perform end-to-end learning for self-driving cars. The system has the ability to learn spatiotemporal features by simultaneously mapping raw images and pre-calculated optical flows.
0
0
0
Sun Sep 08 2019
Artificial Intelligence
Self-driving scale car trained by Deep reinforcement learning
Self-driving based on deep reinforcement learning has become a popular topic. Most of the current self-driving methods focus on how to directly learn end-to-end control strategy from the raw sensory data. To obtain a good generalization ability in safety, a virtual simulation environment is constructed.
0
0
0
Thu May 16 2019
Artificial Intelligence
Autonomous Vehicle Control: End-to-end Learning in Simulated Urban Environments
An end-to-end approach attempts to achieve autonomous driving using a single, comprehensive software component. The results of this paper indicate that end- to-end systems can operate autonomously in simple urban environments.
0
0
0
Wed Dec 19 2018
Computer Vision
Learning On-Road Visual Control for Self-Driving Vehicles with Auxiliary Tasks
NVIDIA recently proposed an End-to-End algorithm that can directly learn steering commands from raw pixels of a front camera. In this paper, we leverage auxiliary information aside from raw images and design a novel network structure called Auxiliary Task Network.
0
0
0