Goal-constrained Sparse Reinforcement Learning for End-to-End Driving
arXiv:2103.09189
Abstract
Deep reinforcement Learning for end-to-end driving is limited by the need of complex reward engineering. Sparse rewards can circumvent this challenge but suffers from long training time and leads to sub-optimal policy. In this work, we explore full-control driving with only goal-constrained sparse reward and propose a curriculum learning approach for end-to-end driving using only navigation view maps that benefit from small virtual-to-real domain gap. To address the complexity of multiple driving policies, we learn concurrent individual policies selected at inference by a navigation system. We demonstrate the ability of our proposal to generalize on unseen road layout, and to drive significantly longer than in the training.
arXiv commentConference submission 6 pages, 8 figures
References in corpus (18)
- Continuous control with deep reinforcement learning
- Playing Atari with Deep Reinforcement Learning
- Dota 2 with Large Scale Deep Reinforcement Learning
- QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
- Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewards
- Combining Planning and Deep Reinforcement Learning in Tactical Decision Making for Autonomous Driving
- A Survey of End-to-End Driving: Architectures and Training Methods
- Learning by Playing - Solving Sparse Reward Tasks from Scratch
- Curriculum Learning by Transfer Learning: Theory and Experiments with Deep Networks
- Learning to Utilize Shaping Rewards: A New Approach of Reward Shaping
- Learning to Generalize from Sparse and Underspecified Rewards
- Dealing with Sparse Rewards in Reinforcement Learning
- Automated curricula through setter-solver interactions
- Interpretable End-to-end Urban Autonomous Driving with Latent Deep Reinforcement Learning
- Ctrl-Z: Recovering from Instability in Reinforcement Learning
- Robot Navigation with Map-Based Deep Reinforcement Learning
- Balance Between Efficient and Effective Learning: Dense2Sparse Reward Shaping for Robot Manipulation with Environment Uncertainty
- Learning sparse representations in reinforcement learning