Xitong Wang
Papers
2
Total Citations
60
H-Index
2
About
Xitong Wang is a researcher specializing in artificial intelligence and robotics, with a primary focus on visual navigation in indoor environments. Their work addresses fundamental challenges in deep reinforcement learning, particularly the trade-off between sample efficiency and navigation performance. Wang’s most notable contribution is the development of a deep imitation reinforcement learning framework for target-driven visual navigation, published in 2021 and cited 57 times. This approach combines imitation learning with reinforcement learning to improve both learning speed and navigation accuracy, enabling agents to navigate toward specified targets using only visual input. Additionally, Wang has explored inverse reinforcement learning for the same domain, proposing methods that learn reward functions from expert demonstrations rather than requiring hand-crafted rewards. Their research has significant implications for autonomous robotics, particularly in applications such as service robots, autonomous vehicles, and assistive technologies that must operate in complex, unstructured indoor spaces. By advancing end-to-end navigation systems that bypass traditional mapping requirements, Wang’s work contributes to more practical and scalable real-world deployment of intelligent navigation agents.
Research Focus
Key Achievements
Top Papers
- 1
- 2