Robots are good at making identical repetitive movements such as a simple task on an assembly line. But they lack the ability to perceive objects as they move through an environment. A recent study was conducted by researchers at the University of Illinois at Urbana-Champaign, NVIDIA, the University of Washington, and Stanford University on 6D object pose estimation to develop a filter to give robots greater spatial perception, so they can manipulate objects and navigate through space more accurately.
While 3D pose provides location information on X, Y, and Z axes — relative location of the object with respect to the camera — 6D pose gives a much more complete picture. Much like describing an airplane in flight, the robot needs to know the three dimensions of the object’s orientation: yaw, pitch, and roll. In real-life environments, all six of those dimensions are constantly changing.
The filter was developed to help robots analyze spatial data. The filter looks at each particle, or piece of image information collected by cameras aimed at an object to help reduce judgement errors.
In an image-based 6D pose estimation framework, a particle filter uses samples to estimate the position and orientation. Every particle is like a hypothesis — a guess about the position and orientation that requires estimation. The particle filter uses observation to compute the value of importance of the information from the other particles and eliminates the incorrect estimations.
Previously, there was no system to estimate the full distribution of the orientation of the object. This gives important uncertainty information for robot manipulation. The new filter uses 6D object pose tracking in the Rao-Blackwel-lized particle filtering framework, where the 3D rotation and the 3D translation of an object are separated. This allows the new approach, called PoseRBPF, to efficiently estimate the 3D translation of an object along with the full distribution over the 3D rotation. As a result, Pose-RBPF can track objects with arbitrary symmetries while still maintaining adequate posterior distributions.