Physics-Based Object 6D-Pose Estimation during Non-Prehensile Manipulation
Zisong Xu, Rafael Papallas, and Mehmet Dogar
International Symposium on Experimental Robotics (ISER)
2023 | Chiang Mai, Thailand 🇹🇭
The paper is accepted for publication to ISER and the lead author, Zisong Xu, will present the work in Thailand in November 2023. The current version of the arxiv is slightly out-dated and we are planning to upload the most recent pre-print very soon.
We propose a method to track the 6D pose of an object over time, while the object is under non-prehensile manipulation by a robot. At any given time during the manipulation of the object, we assume access to the robot joint controls and an image from a camera. We use the robot joint controls to perform a physics-based prediction of how the object might be moving. We then combine this prediction with the observation coming from the camera, to estimate the object pose as accurately as possible. We use a particle filtering approach to combine the control information with the visual information. We compare the proposed method with two baselines: (i) using only an image-based pose estimation system at each time-step, and (ii) a particle filter which does not perform the computationally expensive physics predictions, but assumes the object moves with constant velocity. Our results show that making physics-based predictions is worth the computational cost, resulting in more accurate tracking, and estimating object pose even when the object is not clearly visible to the camera.