Using Deep Learning to Bootstrap Abstractions for Hierarchical Robot Planning
Using Deep Learning to Bootstrap Abstractions for Hierarchical Robot Planning
复制标题
DOI：
10.5555/3535850.3535982
复制
发表时间：
2022-02
期刊：
影响因子：
--
通讯作者：
Naman Shah;Siddharth Srivastava
中科院分区：
文献类型：
--
作者：
Naman Shah;Siddharth Srivastava
本文解决了学习抽象的问题，提高了机器人的规划性能，同时为可靠性提供了强有力的保证。虽然最先进的分层机器人规划算法允许机器人有效地计算长期运动计划，以实现用户期望的任务，但这些方法通常依赖于环境相关的状态和动作抽象，需要由专家手工设计。我们提出了一种新的方法来引导整个分层规划过程。这允许我们使用深度神经网络预测的关键区域自动计算新环境的抽象状态和动作，该网络具有自动生成的机器人特定架构。我们证明了学习到的抽象可以与一种新的多源双向分层机器人规划算法相结合，该算法是可靠的和概率完备的。对使用完整和非完整机器人的20种不同设置的广泛经验评估表明：(a)我们的学习抽象为有效的多源分层规划提供了必要的信息；并且(b)这种学习、抽象和计划的方法，在训练期间没有看到的测试环境的计划时间方面，几乎比最先进的基线高出十倍。
This paper addresses the problem of learning abstractions that boost robot planning performance while providing strong guarantees of reliability. Although state-of-the-art hierarchical robot planning algorithms allow robots to efficiently compute long-horizon motion plans for achieving user desired tasks, these methods typically rely upon environment-dependent state and action abstractions that need to be hand-designed by experts. We present a new approach for bootstrapping the entire hierarchical planning process. This allows us to compute abstract states and actions for new environments automatically using the critical regions predicted by a deep neural network with an auto-generated robot-specific architecture. We show that the learned abstractions can be used with a novel multi-source bi-directional hierarchical robot planning algorithm that is sound and probabilistically complete. An extensive empirical evaluation on twenty different settings using holonomic and non-holonomic robots shows that (a) our learned abstractions provide the information necessary for efficient multi-source hierarchical planning; and that (b) this approach of learning, abstractions, and planning outperforms state-of-the-art baselines by nearly a factor of ten in terms of planning time on test environments not seen during training.