Search papers, labs, and topics across Lattice.
The paper introduces ActivePusher, a framework for nonprehensile manipulation that combines residual-physics modeling with uncertainty-based active learning to improve data efficiency and planning success. ActivePusher uses uncertainty estimates from the learned dynamics model to guide data acquisition towards informative skill parameters and biases control sampling in a kinodynamic planner towards more reliable actions. Experiments in simulation and the real world demonstrate that ActivePusher achieves higher planning success rates compared to baseline methods by actively exploring the skill space.
Stop wasting robot learning data: ActivePusher intelligently explores the action space, slashing data requirements for nonprehensile manipulation by focusing on the most informative skills.
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.
