ActivePusher: Active Learning and Planning with Residual Physics for Nonprehensile Manipulation

TL;DR

ActivePusher introduces an active learning framework that combines residual physics and uncertainty estimation to improve data efficiency and planning success in nonprehensile manipulation tasks.

Contribution

It presents a novel integration of residual-physics modeling with active learning and kinodynamic planning for more reliable manipulation.

Findings

Improves data efficiency in learning-based manipulation models.

Achieves higher planning success rates in simulation and real-world tests.

Abstract

Planning with learned dynamics models offers a promising approach toward versatile real-world manipulation, particularly in nonprehensile settings such as pushing or rolling, where accurate analytical models are difficult to obtain. However, collecting training data for learning-based methods can be costly and inefficient, as it often relies on randomly sampled interactions that are not necessarily the most informative. Furthermore, learned models tend to exhibit high uncertainty in underexplored regions of the skill space, undermining the reliability of long-horizon planning. To address these challenges, we propose ActivePusher, a novel framework that combines residual-physics modeling with uncertainty-based active learning, to focus data acquisition on the most informative skill parameters. Additionally, ActivePusher seamlessly integrates with model-based kinodynamic planners, leveraging uncertainty estimates to bias control sampling toward more reliable actions. We evaluate our approach in both simulation and real-world environments, and demonstrate that it consistently improves data efficiency and achieves higher planning success rates in comparison to baseline methods. The source code is available at https://github.com/elpis-lab/ActivePusher.