Uncertainty-Aware RL-Controlled Adaptive 3D Mapping

1. 12:00
   arXiv · Machine Learning Theory

   Uncertainty-Aware RL-Controlled Adaptive 3D Mapping

   该研究提出一种基于语义熵、几何曲率和纹理丰富度的不确定性感知自适应体素映射框架，用强化学习智能体在用户指定内存预算下学习体素细分策略，替代手工调参。实验表明，相比 MAP-ADAPT 和固定分辨率基线，多分辨率 TSDF 在几何精度、语义一致性和内存-精度权衡上均有提升。代码与模型已公开。

Current heat 9·Comparable peak 10（Oct 2）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.