ROGIBOT: Development of a Dual-Arm Autonomous Container Unloading Robot System for Various Logistics Packages
Eugene Auh, Ilho Oh, JangHoon Lee, Nabih Andres Pico Rosas, Yeong-Jae Park, Jae Hyuck Jang, Haneol Lee, Altair Coutinho, jongsam park, Byungtaek Jung, Seongyong Koo, Kyung-Hoon Kim, Hugo Rodrigue, Hyungpil Moon
AI summary
Problem
Existing autonomous unloading robots are limited to specific package types and low throughput, failing to match the speed and versatility required for general logistics environments.
Approach
The system uses a dual-arm mobile robot with multi-functional end effectors, including a novel on-hand sliding surface with rotating elastic belts, combined with deep learning-based package recognition and an FSM task planner to sweep and unload mixed packages rapidly.
Key results
- Achieved 2,030 packages per hour in mock container tests
- Surpassed state-of-the-art robots by over 35% and human worker benchmarks
- Integrated a novel on-hand sliding surface for high-speed multi-package sweeping
- Developed a custom vision pipeline and FSM planner for diverse package recognition and automation
Why it matters
It offers a viable, high-throughput automation solution for e-commerce logistics, directly addressing labor shortages and reducing workplace injuries in warehouse operations.
Abstract
This article presents ROGIBOT, an autonomous robot designed for high-speed unloading inside standard con- tainers. Previous unloading robots have typically been limited to handling only box-shaped packages and a small number of items at a time. These limitations hinder their applicability in general logistics environments, which involve a wide variety of package types and require high throughput. To address these challenges, we propose a novel dual-arm robot equipped with multi-functional end effectors, including a vacuum array, a two- finger gripper, and an on-hand sliding surface. The components are designed to handle the most common types of logistics packages, such as boxes, plastic pouches, and bundle sacks. In particular, the on-hand sliding surface can draw packages efficiently using the frictional force generated by a rotating elastic conveyor belt. This enables the robot to unload multiple packages consecutively at high speed by sweeping the end effectors across the pile. In addition to the hardware, we introduce a package recognition method and a finite-state- machine-based task planner that detects logistics packages and selects actions to maximize operational efficiency. In our evaluation, ROGIBOT was tested in a mock container and achieved a throughput of 2,030 pieces per hour, surpassing both state-of-the-art robots and typical human performance.