Author:
Lu Chen-Lung,Liu Zi-Yan,Huang Jui-Te,Huang Ching-I,Wang Bo-Hui,Chen Yi,Wu Nien-Hsin,Wang Hsueh-Cheng,Giarré Laura,Kuo Pei-Yi
Abstract
Facilitating navigation in pedestrian environments is critical for enabling people who are blind and visually impaired (BVI) to achieve independent mobility. A deep reinforcement learning (DRL)–based assistive guiding robot with ultrawide-bandwidth (UWB) beacons that can navigate through routes with designated waypoints was designed in this study. Typically, a simultaneous localization and mapping (SLAM) framework is used to estimate the robot pose and navigational goal; however, SLAM frameworks are vulnerable in certain dynamic environments. The proposed navigation method is a learning approach based on state-of-the-art DRL and can effectively avoid obstacles. When used with UWB beacons, the proposed strategy is suitable for environments with dynamic pedestrians. We also designed a handle device with an audio interface that enables BVI users to interact with the guiding robot through intuitive feedback. The UWB beacons were installed with an audio interface to obtain environmental information. The on-handle and on-beacon verbal feedback provides points of interests and turn-by-turn information to BVI users. BVI users were recruited in this study to conduct navigation tasks in different scenarios. A route was designed in a simulated ward to represent daily activities. In real-world situations, SLAM-based state estimation might be affected by dynamic obstacles, and the visual-based trail may suffer from occlusions from pedestrians or other obstacles. The proposed system successfully navigated through environments with dynamic pedestrians, in which systems based on existing SLAM algorithms have failed.
Funder
Ministry of Science and Technology, Taiwan
Qualcomm
Subject
Artificial Intelligence,Computer Science Applications
Reference40 articles.
1. Distributional Policy Gradients;Barth-Maron,2018
2. Accurate Position Tracking with a Single Uwb Anchor;Cao,2020
3. Socially Aware Motion Planning with Deep Reinforcement Learning;Chen,2017
4. Deep Trail-Following Robotic Guide Dog in Pedestrian Environments for People Who Are Blind and Visually Impaired - Learning from Virtual and Real Worlds;Chuang,2018
5. Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learning;Everett,2018
Cited by
15 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献