Journal
NEUROCOMPUTING
Volume 503, Issue -, Pages 118-128Publisher
ELSEVIER
DOI: 10.1016/j.neucom.2022.06.102
Keywords
Deep reinforcement learning; Collision avoidance; Motion and path planning; Real-time autonomous navigation; Autonomous unmanned vehicles
Categories
Funding
- A*STAR
- National Robotics Programme (NRP) , Singapore
- National Robotics Programme (NRP), Singapore [192 2500049]
Ask authors/readers for more resources
This paper proposes a low-level motion controller based on deep reinforcement learning (DRL) for map-less autonomous navigation in dynamic scenes. It introduces three enhancements to the navigation task and outperforms conventional methods in several aspects, including not requiring prior knowledge of the environment and metric map, lower reliance on accurate sensors, learning emergent behavior in dynamic scenes, and the ability to transfer to real robots without further fine-tuning.
This paper proposes iTD3-CLN, a Deep Reinforcement Learning (DRL) based low-level motion controller, to achieve map-less autonomous navigation in dynamic scene. We consider three enhancements to the Twin Delayed DDPG (TD3) for the navigation task: N-step returns, Priority Experience Replay, and a channel-based Convolutional Laser Network (CLN) architecture. In contrast to the conventional methods such as the DWA, our approach is found superior in the following ways: no need for prior knowledge of the environment and metric map, lower reliance on an accurate sensor, learning emergent behavior in dynamic scene that is intuitive, and more remarkably, able to transfer to the real robot without further fine-tuning. Our extensive studies show that in comparison to the original TD3, the proposed approach can obtain approximately 50% reduction in training to get same performance, 50% higher accumulated reward, and 30-50% increase in generalization performance when tested in unseen environments. Videos of our experiments are available at https://youtu.be/BRN0Gk5oLOc (Simulation) and https://youtu.be/ yIxGH9TPQCc (Real experiment).(c) 2022 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available