Soft Tracking Using Contacts for Cluttered Objects to Perform Blind Object Retrieval

Retrieving an object from cluttered spaces such as cupboards, refrigerators, or bins requires tracking objects with limited no visual sensing. In these scenarios, contact feedback is necessary to estimate the pose of objects, yet are movable while their shapes and number may be unknown, making association contacts extremely difficult. While previous work has focused on multi-target tracking, assumptions therein prohibit using prior methods given only contact-sensing modality. Instead, this paper proposes method Soft Tracking Using Contacts for Cluttered Objects (STUCCO) that tracks belief over point locations implicit associations a particle filter. This allows ambiguous past revised new information becomes available. We apply STUCCO Blind Object Retrieval problem, where target known shape but unknown must retrieved clutter. Our results suggest our outperforms baselines in four simulation environments, real robot, sensing noisy. simulation, we achieve grasp success at least 65% all environments 5%.