Curiosity-driven Exploration for Mapless Navigation with Deep Reinforcement Learning

Abstract

This paper investigates exploration strategies of Deep Reinforcement Learning (DRL) methods to learn navigation policies for mobile robots. In particular, we augment the normal external reward for training DRL algorithms with intrinsic reward signals measured by curiosity. We test our approach in a mapless navigation setting, where the autonomous agent is required to navigate without the occupancy map of the environment, to targets whose relative locations can be easily acquired through low-cost solutions (e.g., visible light localization, Wi-Fi signal localization). We validate that the intrinsic motivation is crucial for improving DRL performance in tasks with challenging exploration requirements. Our experimental results show that our proposed method is able to more effectively learn navigation policies, and has better generalization capabilities in previously unseen environments. A video of our experimental results can be found at https://goo.gl/pWbpcF.

arXiv comment5 pages, 4 figures; to be published in MLPC 2018

References in corpus (3)

• Learning to Navigate in Complex Environments
• Neural SLAM: Learning to Explore with External Memory
• Plugo: a VLC Systematic Perspective of Large-scale Indoor Localization

Cited by in corpus (10)

• Deep Reinforcement Learning (DRL): Another Perspective for Unsupervised Wireless Localization
• Goal-Guided Transformer-Enabled Reinforcement Learning for Efficient Autonomous Navigation
• Mapless Navigation among Dynamics with Social-safety-awareness: a reinforcement learning approach from 2D laser scans
• Learn to Navigate Maplessly with Varied LiDAR Configurations: A Support Point-Based Approach
• VR-Goggles for Robots: Real-to-sim Domain Adaptation for Visual Control
• Curriculum Learning for Safe Mapless Navigation
• Using RGB Image as Visual Input for Mapless Robot Navigation
• A review of mobile robot motion planning methods: from classical motion planning workflows to reinforcement learning-based architectures
• Learning to Navigate in a VUCA Environment: Hierarchical Multi-expert Approach
• Knowledge is reward: Learning optimal exploration by predictive reward cashing