In robotic domains, learning and planning are complicated by continuous state spaces, continuous action spaces, and long task horizons. In this work, we address these challenges with Neuro-Symbolic Relational Transition Models (NSRTs), a novel class of models that are data-efficient to learn, compatible with powerful robotic planning methods, and generalizable over objects. NSRTs have both symbolic and neural components, enabling a bilevel planning scheme where symbolic AI planning in an outer loop guides continuous planning with neural models in an inner loop. Experiments in four robotic planning domains show that NSRTs can be learned after only tens or hundreds of training episodes, and then used for fast planning in new tasks that require up to 60 actions and involve many more objects than were seen during training. Video: https://tinyurl.com/chitnis-nsrts
翻译:在机器人领域,学习和规划由于连续的国家空间、连续的行动空间和长期任务视野而变得复杂。在这项工作中,我们通过Neuro-Symblic 关系过渡模型(NSRTs)来应对这些挑战。 Neuro-Symblical Relational Translational Models (NSRTs)是一套新型的模型,具有数据效率,可以学习,与强大的机器人规划方法相兼容,并且可以普遍覆盖物体。NSRTs具有象征和神经两个组成部分,使双级规划计划成为双级计划,在外循环中象征性的AI规划指导与内循环中的神经模型进行连续规划。在四个机器人规划域的实验显示,NSRTs在经过数十或数百次培训后才能学习,然后用于在新任务中快速规划,需要多达60项行动,涉及比培训中看到更多的目标。视频:https://tinyurl.com/chitnis-nrts。