Mind the gap - robotic grasping under incomplete observation
Jeannette Bohg · M. Johnson-Roberson · Beatriz León · Javier Felip · Xavi Gratal · Niklas Bergstrom · D. Kragic · A. Morales
TL;DR
A robot completes the hidden side of unknown tabletop objects by fitting symmetries, then plans stable grasps on the completed shape.
BRIEF
Robots picking up unknown objects from a table usually see only part of each object, since the camera cannot view the back or the sides. This work fills in those gaps by exploiting a common property of household objects: symmetry. The method searches for the symmetry parameters that best fit the visible points, then completes the point cloud and builds a surface mesh. A simulator uses the completed shape to plan stable grasps and collision-free motions. Experiments show the predicted shapes approximate the real objects, and the approach was demonstrated on two different robots.
Abstract
We consider the problem of grasp and manipulation planning when the state of the world is only partially observable. Specifically, we address the task of picking up unknown objects from a table top. The proposed approach to object shape prediction aims at closing the knowledge gaps in the robot's understanding of the world. A completed state estimate of the environment can then be provided to a simulator in which stable grasps and collision-free movements are planned. The proposed approach is based on the observation that many objects commonly in use in a service robotic scenario possess symmetries. We search for the optimal parameters of these symmetries given visibility constraints. Once found, the point cloud is completed and a surface mesh reconstructed. Quantitative experiments show that the predictions are valid approximations of the real object shape. By demonstrating the approach on two very different robotic platforms its generality is emphasized.