PDSketch: Integrated Planning Domain Programming and Learning
This work addresses the challenge of model learning and online planning for robotics, offering a novel approach to improve efficiency and generalization, though it appears incremental as it builds on existing methods by integrating programming and learning.
The paper tackles the problem of building flexible and general robots by proposing PDSketch, a domain definition language that integrates planning domain programming and learning to exploit locality and sparsity in environmental transition models, resulting in improved model generalization, data-efficiency, and runtime-efficiency with automatically generated domain-independent planning heuristics that accelerate performance-time planning for novel goals.
This paper studies a model learning and online planning approach towards building flexible and general robots. Specifically, we investigate how to exploit the locality and sparsity structures in the underlying environmental transition model to improve model generalization, data-efficiency, and runtime-efficiency. We present a new domain definition language, named PDSketch. It allows users to flexibly define high-level structures in the transition models, such as object and feature dependencies, in a way similar to how programmers use TensorFlow or PyTorch to specify kernel sizes and hidden dimensions of a convolutional neural network. The details of the transition model will be filled in by trainable neural networks. Based on the defined structures and learned parameters, PDSketch automatically generates domain-independent planning heuristics without additional training. The derived heuristics accelerate the performance-time planning for novel goals.