Abstract
In this work, we present several heuristic-based and data-driven active vision strategies for viewpoint optimization of an arm-mounted depth camera to aid robotic grasping. These strategies aim to efficiently collect data to boost the performance of an underlying grasp synthesis algorithm. We created an open-source benchmarking platform in simulation (https://github.com/galenbr/2021ActiveVision), and provide an extensive study for assessing the performance of the proposed methods as well as comparing them against various baseline strategies. We also provide an experimental study with a real-world two finger parallel jaw gripper setup by utilizing an existing grasp planning benchmark in the literature. With these analyses, we were able to quantitatively demonstrate the versatility of heuristic methods that prioritize certain types of exploration, and qualitatively show their robustness to both novel objects and the transition from simulation to the real world. We identified scenarios in which our methods did not perform well and objectively difficult scenarios, and present a discussion on which avenues for future research show promise.
Highlights
Robotic grasping is a vital capability for many tasks, in service robotics
An attempt is made to synthesize a grasp with the available data, and if it fails, the active vision policy is called to guide the camera to a new viewpoint after which the process repeats until a grasp is found
We presented heuristic and data-driven policies to achieve viewpoint optimization to aid robotic grasping
Summary
Robotic grasping is a vital capability for many tasks, in service robotics. We attempt to identify simple properties of the visual data that are reliable indicators of effective exploration directions These approaches use estimates of how many potentially occluded grasps lie in each direction. 4) We present an extensive simulation and experimental analysis, assessing and comparing the performance of five active vision methods against three baseline strategies, including the optimal BFS strategy. Taken together, these allow us to draw new conclusions about how well our algorithms work but how much it would be possible to improve them
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have