Stow: Robotic Packing of Items into Fabric Pods

2025-05-07

AI Summary

To address the challenge of efficient, safe, and low-defect packing of diverse e-commerce items onto high-density overhead shelves in fulfillment centers, this paper proposes a hardware-software co-designed flexible robotic system. Methodologically, it integrates compliant manipulation, multimodal perception, real-time motion planning, and adaptive closed-loop control, introducing the first full-stack architecture supporting overhead-shelf-first operation—balancing human-robot collaboration safety and operational efficiency. Deployed in a large-scale e-commerce warehouse, the system executed over 500,000 real-world packing tasks, achieving packing density and speed comparable to human performance, with a defect rate below 0.1%. Key contributions include: (1) the first end-to-end flexible packing system tailored for densely packed fabric storage bags; (2) an overhead-first operational paradigm that jointly optimizes safety, speed, and precision; and (3) a scalable, hardware-software co-designed architecture validated at industrial scale.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Hardware-aware ML

Application Category

• Economics, Online Markets and Human Computation: The sharing economy
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

This paper presents a compliant manipulation system capable of placing items onto densely packed shelves. The wide diversity of items and strict business requirements for high producing rates and low defect generation have prohibited warehouse robotics from performing this task. Our innovations in hardware, perception, decision-making, motion planning, and control have enabled this system to perform over 500,000 stows in a large e-commerce fulfillment center. The system achieves human levels of packing density and speed while prioritizing work on overhead shelves to enhance the safety of humans working alongside the robots.

Problem

• Robotic packing of diverse items into dense fabric pods
• Meeting high-speed, low-defect warehouse packing requirements
• Achieving human-level performance while ensuring worker safety

Innovation

• Compliant manipulation system for dense packing
• Hardware and perception innovations enable high speed
• Prioritizes overhead shelves for human safety