Learning to Plan in High Dimensions via Neural Exploration-Exploitation Trees
This addresses path planning problems for robotics or autonomous systems in complex environments, offering a novel approach that is incremental in combining neural learning with existing planning techniques.
The paper tackles path planning in high-dimensional continuous spaces by proposing a meta algorithm, Neural Exploration-Exploitation Trees (NEXT), which learns from prior experience to improve sample efficiency and outperforms state-of-the-art methods with more compact search trees.
We propose a meta path planning algorithm named \emph{Neural Exploration-Exploitation Trees~(NEXT)} for learning from prior experience for solving new path planning problems in high dimensional continuous state and action spaces. Compared to more classical sampling-based methods like RRT, our approach achieves much better sample efficiency in high-dimensions and can benefit from prior experience of planning in similar environments. More specifically, NEXT exploits a novel neural architecture which can learn promising search directions from problem structures. The learned prior is then integrated into a UCB-type algorithm to achieve an online balance between \emph{exploration} and \emph{exploitation} when solving a new problem. We conduct thorough experiments to show that NEXT accomplishes new planning problems with more compact search trees and significantly outperforms state-of-the-art methods on several benchmarks.