Pickalo：用 6D 位姿估计实现低成本工业料箱抓取

View PDF HTML (experimental)

Abstract:Bin picking in real industrial environments remains challenging due to severe clutter, occlusions, and the high cost of traditional 3D sensing setups. We present Pickalo, a modular 6D pose-based bin-picking pipeline built entirely on low-cost hardware. A wrist-mounted RGB-D camera actively explores the scene from multiple viewpoints, while raw stereo streams are processed with BridgeDepth to obtain refined depth maps suitable for accurate collision reasoning. Object instances are segmented with a Mask-RCNN model trained purely on photorealistic synthetic data and localized using the zero-shot SAM-6D pose estimator. A pose buffer module fuses multi-view observations over time, handling object symmetries and significantly reducing pose noise. Offline, we generate and curate large sets of antipodal grasp candidates per object; online, a utility-based ranking and fast collision checking are queried for the grasp planning. Deployed on a UR5e with a parallel-jaw gripper and an Intel RealSense D435i, Pickalo achieves up to 600 mean picks per hour with 96-99% grasp success and robust performance over 30-minute runs on densely filled euroboxes. Ablation studies demonstrate the benefits of enhanced depth estimation and of the pose buffer for long-term stability and throughput in realistic industrial conditions. Videos are available at this https URL

Subjects:	Robotics (cs.RO); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2604.04690 [cs.RO]
	(or arXiv:2604.04690v3 [cs.RO] for this version)
	https://doi.org/10.48550/arXiv.2604.04690 arXiv-issued DOI via DataCite
Journal reference:	Robotics 2026, 15, 197
Related DOI:	https://doi.org/10.3390/robotics15100197 DOI(s) linking to related resources

Submission history

From: Alessandro Tarsi [view email]
[v1] Mon, 6 Apr 2026 13:56:57 UTC (17,256 KB)
[v2] Sun, 12 Jul 2026 09:59:06 UTC (17,255 KB)
[v3] Tue, 13 Oct 2026 08:19:51 UTC (17,254 KB)