Control Transformer：通过 PRM 引导的回报条件序列建模实现未知环境中的机器人导航
Control Transformer: Robot Navigation in Unknown Environments through PRM-Guided Return-Conditioned Sequence Modeling
AI总结：
针对未知环境长时程机器人导航，本文提出由 PRM 规划器引导低层策略、并用 Control Transformer 建模回报条件序列的方法，实现仅依赖局部信息的迷宫穿越、未知环境迁移及 Turtlebot3 零样本 sim2real。
摘要：
学习导航等长时程任务，给强化学习在机器人领域的成功应用带来了严峻挑战。从另一个角度看，在已知环境中，基于采样的规划无需学习即可鲁棒地找到无碰撞路径。本文提出 Control Transformer，对来自低层策略的回报条件序列进行建模，而这些低层策略由基于采样的 Probabilistic Roadmap（PRM，概率路线图）规划器引导。作者证明该框架仅利用局部信息即可解决长时程导航任务。他们在使用 MuJoCo 机器人（包括 Ant、Point 和 Humanoid）的部分观测迷宫导航中评估了该方法，结果表明 Control Transformer 能够成功穿越迷宫并迁移到未知环境。此外，作者将该方法应用于差速驱动机器人 Turtlebot3，并展示了在含噪观测下的零样本 sim2real 迁移。
英文摘要：
Learning long-horizon tasks such as navigation has presented difficult challenges for successfully applying reinforcement learning to robotics. From another perspective, under known environments, sampling-based planning can robustly find collision-free paths in environments without learning. In this work, we propose Control Transformer that models return-conditioned sequences from low-level policies guided by a sampling-based Probabilistic Roadmap (PRM) planner. We demonstrate that our framework can solve long-horizon navigation tasks using only local information. We evaluate our approach on partially-observed maze navigation with MuJoCo robots, including Ant, Point, and Humanoid. We show that Control Transformer can successfully navigate through mazes and transfer to unknown environments. Additionally, we apply our method to a differential drive robot (Turtlebot3) and show zero-shot sim2real transfer under noisy observations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
