CAAI Transactions on Intelligence Technology | |
Target-driven visual navigation in indoor scenes using reinforcement learning and imitation learning | |
article | |
Qiang Fang1  Xin Xu1  Xitong Wang1  Yujun Zeng1  | |
[1] Department of Intelligence Science and Technology, College of Intelligence Science and Technology, National University of Defense Technology | |
关键词: mobile robots; image matching; radionavigation; reinforcement learning; | |
DOI : 10.1049/cit2.12043 | |
学科分类:数学(综合) | |
来源: Wiley | |
【 摘 要 】
Here, the challenges of sample efficiency and navigation performance in deep reinforcement learning for visual navigation are focused and a deep imitation reinforcement learning approach is proposed. Our contributions are mainly three folds: first, a framework combining imitation learning with deep reinforcement learning is presented, which enables a robot to learn a stable navigation policy faster in the target-driven navigation task. Second, the surrounding images is taken as the observation instead of sequential images, which can improve the navigation performance for more information. Moreover, a simple yet efficient template matching method is adopted to determine the stop action, making the system more practical. Simulation experiments in the AI-THOR environment show that the proposed approach outperforms previous end-to-end deep reinforcement learning approaches, which demonstrate the effectiveness and efficiency of our approach.
【 授权许可】
CC BY|CC BY-ND|CC BY-NC|CC BY-NC-ND
【 预 览 】
Files | Size | Format | View |
---|---|---|---|
RO202302050004883ZK.pdf | 1058KB | download |