Autonomous Driving in Reality with Reinforcement Learning and Image Translation

Abstract

Supervised learning is widely used in training autonomous driving vehicle. However, it is trained with large amount of supervised labeled data. Reinforcement learning can be trained without abundant labeled data, but we cannot train it in reality because it would involve many unpredictable accidents. Nevertheless, training an agent with good performance in virtual environment is relatively much easier. Because of the huge difference between virtual and real, how to fill the gap between virtual and real is challenging. In this paper, we proposed a novel framework of reinforcement learning with image semantic segmentation network to make the whole model adaptable to reality. The agent is trained in TORCS, a car racing simulator.

References in corpus (3)

• Safe, Multi-Agent, Reinforcement Learning for Autonomous Driving
• Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World
• Virtual to Real Reinforcement Learning for Autonomous Driving

Cited by in corpus (4)

• Robust Multimodal Image Registration Using Deep Recurrent Reinforcement Learning
• Adversarial Deep Reinforcement Learning for Improving the Robustness of Multi-agent Autonomous Driving Policies
• Latent Space Reinforcement Learning for Steering Angle Prediction
• Learning Socially Appropriate Robot Approaching Behavior Toward Groups using Deep Reinforcement Learning