Active Learning of Abstract Plan Feasibility

Abstract

Long horizon sequential manipulation tasks are effectively addressed hierarchically: at a high level of abstraction the planner searches over abstract action sequences, and when a plan is found, lower level motion plans are generated. Such a strategy hinges on the ability to reliably predict that a feasible low level plan will be found which satisfies the abstract plan. However, computing Abstract Plan Feasibility (APF) is difficult because the outcome of a plan depends on real-world phenomena that are difficult to model, such as noise in estimation and execution. In this work, we present an active learning approach to efficiently acquire an APF predictor through task-independent, curious exploration on a robot. The robot identifies plans whose outcomes would be informative about APF, executes those plans, and learns from their successes or failures. Critically, we leverage an infeasible subsequence property to prune candidate plans in the active learning strategy, allowing our system to learn from less data. We evaluate our strategy in simulation and on a real Franka Emika Panda robot with integrated perception, experimentation, planning, and execution. In a stacking domain where objects have non-uniform mass distributions, we show that our system permits real robot learning of an APF model in four hundred self-supervised interactions, and that our learned model can be used effectively in multiple downstream tasks.

arXiv commentTo appear in Robotics: Science and Systems 2021

References in corpus (10)

• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• Graph Neural Networks: A Review of Methods and Applications
• Deep Bayesian Active Learning with Image Data
• Interaction Networks for Learning about Objects, Relations and Physics
• Bayesian Active Learning for Classification and Preference Learning
• Graph networks as learnable physics engines for inference and control
• A Compositional Object-Based Approach to Learning Physical Dynamics
• Active Learning with Statistical Models
• Learning Physical Intuition of Block Towers by Example
• Active model learning and diverse action sampling for task and motion planning

Cited by in corpus (1)

• A Survey of Optimization-based Task and Motion Planning: From Classical To Learning Approaches