AI Palletizing using Hyundai YS140 and Mech-Mind

Industry > LogisticsIndustry > Food·DrinkSector > ManufacturingSector > Logistics

This automation solution utilizes Mech-Mind 3D Vision and the Hyundai YS140 industrial robot to automatically sort and stack boxes onto pallets by size, as well as transfer them to conveyors.

By employing two sets of highly cost-effective 3D vision sensors, the system can stably perform both palletizing and depalletizing tasks even when various product boxes are mixed together (Mixed-Case Palletizing). The project was implemented to improve worker health and safety by automating tasks known to cause musculoskeletal disorders.

Components

Robot	Robot: Hyundai YS140 6-axis high-payload articulated robot. Payload: 140kg (Capable of handling multiple or heavy boxes) Reach: 2,339mm Weight: 670kg Repeatability: ±0.17mm Vision System: 2 Sets of Mech-Mind 3D Vision (High-speed AI recognition).

Workflow

STEP 1.	Palletizing Recognition: Two 3D cameras simultaneously identify the shape and location of boxes on the conveyor while searching for available stacking space on the pallet.
STEP 2.	Stacking: The robot picks the recognized box using a vacuum gripper and unloads it into the optimal remaining space on the pallet based on box dimensions.
STEP 3.	Depalletizing Recognition: The overhead 3D vision identifies the shapes and coordinates of boxes currently stacked on the pallet.
STEP 4.	Unloading: The robot picks the identified boxes and unloads them onto the discharge conveyor.

※ The content provided by Bigwave Robotics is protected under U.S. and Canadian copyright and intellectual property laws. Unauthorized reproduction, distribution, or use of this content is strictly prohibited.

Features

Unbeatable Vision Value & Performance

Cost-Efficient Dual-Vision: Even with a two-camera setup for simultaneous palletizing and depalletizing, the 3D vision solution remains highly affordable, priced in the late 40 million KRW range.

Deep Learning Recognition: Exceptional accuracy in identifying diverse objects using AI-based 3D vision.

Robust in Challenging Environments: Maintains high recognition rates even for dark-colored products or surfaces with high light reflection, such as packaging tape.

High-Density Detection: Successfully distinguishes individual boxes even when they are tightly packed with minimal gaps.

Intelligent Motion Planning: Easily performs collision avoidance and high-density stacking (compatible with most major robot brands).

Wide Field of View: Capable of recognizing a wide area, enabling tasks on pallets with stacking heights exceeding 2 meters.

Results

ROI Period	16.8 months
Key Benefits	Significant reduction in personnel required for manual box loading and unloading. Improved workplace safety by automating high-risk ergonomic tasks.
Client Feedback	Automating robotic depalletizing was previously difficult to implement in mass production due to the limitations of 2D vision and varying product specs. However, through ROAS's affordable and accurate solution, we achieved a successful deployment. For palletizing, where workers often had to restack boxes multiple times to achieve high density, the robot now performs the task to a very satisfying standard.