Learning When to Trust a Dynamics Model for Planning in Reduced State Spaces
This addresses the challenge of efficient and reliable planning for systems with complex dynamics, such as deformable objects, though it is incremental by building on existing reduction and planning methods.
The paper tackles the problem of motion planning in reduced state spaces where simplified dynamics models can be unreliable, by using a classifier to avoid state-action pairs that are not feasible under the true dynamics. In rope manipulation experiments, this approach significantly improved the success rate of an RRT-based planner in difficult scenarios.
When the dynamics of a system are difficult to model and/or time-consuming to evaluate, such as in deformable object manipulation tasks, motion planning algorithms struggle to find feasible plans efficiently. Such problems are often reduced to state spaces where the dynamics are straightforward to model and evaluate. However, such reductions usually discard information about the system for the benefit of computational efficiency, leading to cases where the true and reduced dynamics disagree on the result of an action. This paper presents a formulation for planning in reduced state spaces that uses a classifier to bias the planner away from state-action pairs that are not reliably feasible under the true dynamics. We present a method to generate and label data to train such a classifier, as well as an application of our framework to rope manipulation, where we use a Virtual Elastic Band (VEB) approximation to the true dynamics. Our experiments with rope manipulation demonstrate that the classifier significantly improves the success rate of our RRT-based planner in several difficult scenarios which are designed to cause the VEB to produce incorrect predictions in key parts of the environment.