arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1903.02308cs.RO
收藏

高维状态空间中运动规划抽象表示的学习

Towards Learning Abstract Representations for Locomotion Planning in High-dimensional State Spaces

University of Bonn(波恩大学)

Tobias Klamt, Sven Behnke

摘要:

在许多领域中需要能够在各种地形中导航的地面机器人。其中一个关键方面是适应地面结构的能力,这可以通过带有附加自由度(DoF)的可移动车身部件来实现。然而,规划各自的运动是具有挑战性的,因为合适的表示导致大的状态空间。采用额外的抽象表示-这是粗糙的,低维的,语义丰富-可以支持规划。虽然可以容易地定义期望的机器人表示和这种抽象表示的动作集,但是成本函数需要大量的调整工作。我们提出了一种将成本函数表示为CNN的方法。网络的训练是在生成的人工数据上完成的,同时它很好地概括了现实世界场景的抽象。我们进一步将我们的方法应用到基于搜索的混合驱动步进运动规划问题。抽象表示被用作一个强大的知情启发式加速规划的多个数量级。

英文摘要:

Ground robots which are able to navigate a variety of terrains are needed in many domains. One of the key aspects is the capability to adapt to the ground structure, which can be realized through movable body parts coming along with additional degrees of freedom (DoF). However, planning respective locomotion is challenging since suitable representations result in large state spaces. Employing an additional abstract representation---which is coarser, lower-dimensional, and semantically enriched---can support the planning. While a desired robot representation and action set of such an abstract representation can be easily defined, the cost function requires large tuning efforts. We propose a method to represent the cost function as a CNN. Training of the network is done on generated artificial data, while it generalizes well to the abstraction of real world scenes. We further apply our method to the problem of search-based planning of hybrid driving-stepping locomotion. The abstract representation is used as a powerful informed heuristic which accelerates planning by multiple orders of magnitude.

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。

↑