Counterfactual Multi-Agent Policy Gradients (COMA) are often hampered by dense, irrelevant noise in the centralized critic, which stems from the full-conditioning assumption in high-dimensional action spaces and results in unstable gradient estimation. To address this fundamental limitation, this paper introduces SCAM-...
Rui-Xing Pang, Zhen-Dong Li, Xu-Yang Tao et al.· International Conference on...· 0 citations
Visual-Inertial SLAM (VI-SLAM) systems are sensitive to dynamic objects because moving foreground regions may corrupt feature tracking, optical-flow estimation, and subsequent pose optimization. To mitigate this problem, this paper presents a dynamic-aware image-domain soft suppression frontend for DBA-Fusion. The prop...
Zhen-Dong Li, Rui-Xing Pang, Xu-Yang Tao et al.· International Conference on...· 0 citations
Learned local navigation in crowded indoor environments is sensitive to how dynamic obstacle motion is represented, while collision-prone behaviour may persist after nominal policy training. We present a risk-aware reinforcement-learning framework that addresses these two issues through an uncertainty-aware Dynamic Unc...
Hao-Yun Feng, A. Rubio-Solis, Zhao-Dong Guo et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.