📄 中文摘要
随着并行计算和GPU加速的进步，计算密集型学习问题如主动SLAM迎来了新的机遇。主动SLAM通过选择行动来减少不确定性并改善联合地图构建和定位。然而，现有的基于深度强化学习（DRL）的方法受到可扩展并行训练能力不足的限制。为了解决这一挑战，提出了一种可扩展的端到端DRL框架，用于主动SLAM，支持大规模并行训练。与现有最先进的方法相比，该方法显著减少了训练时间，支持连续动作空间，并促进了更现实场景的探索。此外，该框架作为开源项目发布，以促进可重复性和社区的进一步研究。
大规模并行深度强化学习用于主动SLAM
出处: Massive Parallel Deep Reinforcement Learning for Active SLAM
发布: 2026年3月30日
📄 English Summary
Massive Parallel Deep Reinforcement Learning for Active SLAM
Recent advancements in parallel computing and GPU acceleration have opened new avenues for computation-intensive learning problems such as Active SLAM, where actions are selected to minimize uncertainty and enhance joint mapping and localization. However, existing deep reinforcement learning (DRL) approaches are limited by their inability to scale for parallel training. This research proposes a scalable end-to-end DRL framework for Active SLAM that enables massively parallel training. Compared to state-of-the-art methods, this framework significantly reduces training time, supports continuous action spaces, and facilitates the exploration of more realistic scenarios. It is released as an open-source framework to promote reproducibility and further research in the community.