Accurate Vision-based Manipulation through Contact Reasoning
This work addresses the problem of low accuracy and computational cost in vision-based manipulation for robotics, representing an incremental improvement over previous methods.
The paper tackles the challenge of planning contact interactions in partially observable robotic environments by disentangling contact from motion optimization and using a hybrid perception approach, resulting in improved efficiency and higher manipulation accuracy in planar pushing tasks.
Planning contact interactions is one of the core challenges of many robotic tasks. Optimizing contact locations while taking dynamics into account is computationally costly and, in environments that are only partially observable, executing contact-based tasks often suffers from low accuracy. We present an approach that addresses these two challenges for the problem of vision-based manipulation. First, we propose to disentangle contact from motion optimization. Thereby, we improve planning efficiency by focusing computation on promising contact locations. Second, we use a hybrid approach for perception and state estimation that combines neural networks with a physically meaningful state representation. In simulation and real-world experiments on the task of planar pushing, we show that our method is more efficient and achieves a higher manipulation accuracy than previous vision-based approaches.