Distributed Deep Reinforcement Learning based Indoor Visual Navigation

Recently, as the rise of deep reinforcement learning, it not only can help the robot to convert the complicated environment scene to motor control command directly but also can accomplish the navigation task properly. In this paper, we propose a novel structure, where the objective is to achieve navigation in large-scale indoor complex environment without pre-constructed map. Generally, it requires good understanding of such indoor environment to make complex spatial perception possible, especially when the indoor space consists of many walls and doors which might block the view of robot leading to complex navigation path. By the proposed distributed deep reinforcement learning in different local regions, our method can achieve indoor visual navigation in the aforementioned large-scale environment without extra map information and human instruction. In the experiments, we validate our proposed method by conducting highly promising navigation tasks both in simulation and real environments.

Authors

• Shih-Hsi Hsu
• Shao-Hung Chan
• Ping-Tsang Wu
• Kun Xiao
• Li-Chen Fu

Keywords

• Navigation
• Visualization
• Task analysis
• Training
• Reinforcement learning
• Robots
• Indoor environments
• Deep Reinforcement Learning
• Machine Vision
• Distributed Deep Reinforcement Learning
• Deep Learning
• Simulation Environment
• Indoor Spaces
• Navigation Task
• Large-scale Environments
• Large-scale Complex
• Rise Of Deep Learning
• Convolutional Neural Network
• Target Location
• Long Short-term Memory
• Light Detection And Ranging
• Deep Architecture
• Scene Images
• Reward Function
• Regional Results
• Embedding Vectors
• Auxiliary Task
• Global Coordinates
• Simultaneous Localization And Mapping
• Deep Reinforcement Learning Model
• Laser Ranging
• Indoor Navigation
• Image Embedding
• Learning Setup
• Navigation Problem
• Stereo Camera
• visual navigation