期刊
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
卷 13, 期 2, 页码 249-261出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCDS.2019.2928820
关键词
Feature extraction; Training; Computer architecture; Mobile robots; Robot sensing systems; Reinforcement learning; Angular velocity; Autonomous steering; deep reinforcement learning (DRL); depth image; difference image; intrinsic reward
资金
- ST Engineering-NTU Corporate Laboratory
- NRF
This paper proposes a novel end-to-end network architecture for mobile robots to achieve safe autonomous steering through deep reinforcement learning. The proposed method demonstrates significant superiority in various metrics and performs outstandingly in real-world environments containing both static and dynamic obstacles.
It is vital for mobile robots to achieve safe autonomous steering in various changing environments. In this paper, a novel end-to-end network architecture is proposed for mobile robots to learn steering autonomously through deep reinforcement learning. Specifically, two sets of feature representations are first extracted from the depth inputs through two different input streams. The acquired features are then merged together to derive both linear and angular actions simultaneously. Moreover, a new action selection strategy is also introduced to achieve motion filtering by taking the consistency in angular velocity into account. Besides, in addition to the extrinsic rewards, the intrinsic bonuses are also adopted during training to improve the exploration capability. Furthermore, it is worth noting the proposed model is readily transferable from the simple virtual training environment to much more complicated real-world scenarios so that no further fine-tuning is required for real deployment. Compared to the existing methods, the proposed method demonstrates significant superiority in terms of average reward, convergence speed, success rate, and generalization capability. In addition, it exhibits outstanding performance in various cluttered real-world environments containing both static and dynamic obstacles. A video of our experiments can be found at https://youtu.be/19jrQGG1oCU.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据