Simultaneous Localization, Mapping, and Manipulation for Unsupervised Object Discovery
This work addresses the challenge of enabling robots to autonomously discover and interact with unknown objects in unstructured environments, representing an incremental advancement by integrating existing techniques into a holistic pipeline.
The paper tackles the problem of unsupervised object discovery in robotics by developing a framework that simultaneously performs localization, mapping, and manipulation to discover and model unknown objects using RGBD cameras and a robot manipulator, achieving high-quality candidate object models with improved precision and recall through a novel spatio-temporal super-pixels approach.
We present an unsupervised framework for simultaneous appearance-based object discovery, detection, tracking and reconstruction using RGBD cameras and a robot manipulator. The system performs dense 3D simultaneous localization and mapping concurrently with unsupervised object discovery. Putative objects that are spatially and visually coherent are manipulated by the robot to gain additional motion-cues. The robot uses appearance alone, followed by structure and motion cues, to jointly discover, verify, learn and improve models of objects. Induced motion segmentation reinforces learned models which are represented implicitly as 2D and 3D level sets to capture both shape and appearance. We compare three different approaches for appearance-based object discovery and find that a novel form of spatio-temporal super-pixels gives the highest quality candidate object models in terms of precision and recall. Live experiments with a Baxter robot demonstrate a holistic pipeline capable of automatic discovery, verification, detection, tracking and reconstruction of unknown objects.