arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1907.09591cs.RO
收藏

可微高斯过程运动规划

Differentiable Gaussian Process Motion Planning

Georgia Institute of Technology(佐治亚理工学院);University of Washington(华盛顿大学);Facebook AI Research(脸书人工智能研究院)

Mohak Bhardwaj, Byron Boots, Mustafa Mukadam

摘要:

基于现代轨迹优化的运动规划方法快速、易于实现,并且对广泛的机器人任务有效。然而,轨迹优化算法具有通常预先设置的参数(并且很少详细讨论)。正确地设置这些参数可以对算法的实际性能产生重大影响,有时会在找到可行的计划或完全失败之间产生差异。我们提出了一种方法,利用过去的经验,学习如何自动适应高斯过程运动规划(GPMP)算法的参数。具体来说,我们对GPMP2算法提出了一个可微分扩展,这样它就可以从数据中进行端到端的训练。我们进行了几个实验,验证我们的算法,并说明我们提出的基于学习的方法运动规划的好处。

英文摘要:

Modern trajectory optimization based approaches to motion planning are fast, easy to implement, and effective on a wide range of robotics tasks. However, trajectory optimization algorithms have parameters that are typically set in advance (and rarely discussed in detail). Setting these parameters properly can have a significant impact on the practical performance of the algorithm, sometimes making the difference between finding a feasible plan or failing at the task entirely. We propose a method for leveraging past experience to learn how to automatically adapt the parameters of Gaussian Process Motion Planning (GPMP) algorithms. Specifically, we propose a differentiable extension to the GPMP2 algorithm, so that it can be trained end-to-end from data. We perform several experiments that validate our algorithm and illustrate the benefits of our proposed learning-based approach to motion planning.

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。

↑