Learning to Reorient Objects With Stable Placements Afforded by Extrinsic Supports

Abstract

Reorienting objects by using supports is a practical yet challenging manipulation task. Owing to the intricate geometry of objects and the constrained feasible motions of the robot, multiple manipulation steps are required for object reorientation. In this work, we propose a pipeline for predicting various object placements from point clouds. This pipeline comprises three stages: a pose generation stage, followed by a pose refinement stage, and culminating in a placement classification stage. We also propose an algorithm to construct manipulation graphs based on point clouds. Feasible manipulation sequences are determined for the robot to transfer object placements. Both simulated and real-world experiments demonstrate that our approach is effective. The simulation results underscore our pipeline’s capacity to generalize to novel objects in random start poses. Our predicted placements exhibit a 20% enhancement in accuracy compared to the state-of-the-art baseline. Furthermore, the robot finds feasible sequential steps in the manipulation graphs constructed by our algorithm to accomplish object reorientation manipulation. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Object reorientation is a prevalent manipulation task in both domestic and industrial manufacturing scenarios. Extrinsic supporting items are often used to provide diverse object placements that allow for feasible grasp configurations for robotic manipulation. In previous methods, utilizing mesh models of objects was necessary to ascertain stable placements and construct manipulation graphs. In this work, we propose a data-driven approach to predict various object placements conditioned on point clouds. Moreover, we use predicted point cloud placements to construct manipulation graphs, which facilitate collision-free pick-and-place steps to reorient objects. Our approach demonstrates the capacity to generalize to novel objects. In future work, we will enhance the performance of our pipeline by optimizing the distance metric used for measuring pose discrepancies and improving the classifier model.