基于学习路径排序的从粗到精的Q-注意
Coarse-to-Fine Q-attention with Learned Path Ranking
摘要:
我们提出了学习路径排名(LPR),一种方法,接受一个末端执行器的目标构成,并学会排名的一组目标达到路径生成方法,包括:路径规划,贝塞尔曲线采样,和学习策略的阵列生成的路径。其核心思想是,每个路径生成模块将在不同的任务,或在任务的不同阶段是有用的。当LPR作为C2F-ARM的扩展添加时,我们的新系统C2F-ARM+LPR保留了其前身的样本效率,同时还能够完成更大的任务集;特别是需要非常特定的动作(例如打开马桶座圈)的任务,这些动作需要从演示和探索数据中推断出来。除了在16个RLBench任务中对我们的方法进行基准测试外,我们还在10-15分钟内学习了真实世界的任务,tabula rasa,只有3个演示。
英文摘要:
We propose Learned Path Ranking (LPR), a method that accepts an end-effector goal pose, and learns to rank a set of goal-reaching paths generated from an array of path generating methods, including: path planning, Bezier curve sampling, and a learned policy. The core idea being that each of the path generation modules will be useful in different tasks, or at different stages in a task. When LPR is added as an extension to C2F-ARM, our new system, C2F-ARM+LPR, retains the sample efficiency of its predecessor, while also being able to accomplish a larger set of tasks; in particular, tasks that require very specific motions (e.g. opening toilet seat) that need to be inferred from both demonstrations and exploration data. In addition to benchmarking our approach across 16 RLBench tasks, we also learn real-world tasks, tabula rasa, in 10-15 minutes, with only 3 demonstrations.
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。