"Automate Loading, Unloading, and Sorting with Robots Made Smarter by AI"

When you think of the word "robot," what is the first thing that comes to mind? While it may have felt unfamiliar at first, "AI" is a term that has become quite friendly lately due to how much we talk about robots. 🤓

What is AI? Simply put, it is Artificial Intelligence. Designed to mimic human intelligence so that robots can move smartly like people, AI serves as the "brain" for a robot. Just as human intelligence allows us to perform various tasks well, diverse AI technologies are now being integrated into robots, leading to the widespread adoption of AI robots across many fields.

 AI-related solutions available at Marosol

At Marosol, your partner for the beginning of robot automation, you can explore various applications of AI robots! 

Today, we have handpicked two favorites. We will compare and explore how AI robots are being used in Palletizing—a process many people look for—and Product Sorting.

 1. AI Palletizing using Hyundai YS140 + MechMind

 Click to view more details.

First up: A palletizing process made even more powerful with AI, handling unaligned and heterogeneous boxes with ease!

This case automates the process where the smart, deep-learning-based Mech-Mind 3D Vision acts as the robot's eyes to recognize boxes. Then, the Hyundai YS140 collaborative robot arm aligns and stacks them according to box size on a pallet before transferring them to a conveyor.

Notably, Mech-Mind 3D Vision is famous for its excellent cost-effectiveness. 🤩 Even if you apply two units to perform both palletizing and depalletizing tasks, you can secure them at a reasonable price in the late 40 million KRW range!

In addition to being reasonably priced, its ability to recognize a variety of objects is outstanding. Vision systems, which act as the robot's eyes, are typically said to have difficulty distinguishing objects that reflect light, like glass. However, the vision used in this robot boasts excellent recognition even for dark-colored products or items with heavy light reflection, such as tape.

It can even distinguish individual boxes even if they are stacked irregularly. By teaching these categorized boxes effective motion planning*, the system can avoid collisions and achieve dense stacking. (This is possible thanks to compatibility with most major robot manufacturers! 😎)

 *Motion Planning: A term used in robotics that refers to an algorithm that automatically designs the movements of a robot so that the processes for the robot to move are arranged in the correct sequence to ensure the optimal possible movement.

Furthermore, it doesn't just see "well"; it sees "wide," allowing it to perform palletizing tasks on pallets with a stacking height of over 2M.

 Photo = KIUKA AG

The task of loading and unloading heavy boxes is a process often avoided by workers because it causes musculoskeletal disorders and high fatigue over long hours. By automating this easily and comfortably with a robot, both the worker's health environment and work efficiency were improved. 

Leave the future of palletizing to the robots! 😉

2. Product Sorting Automation using UR5 and Deep Learning

 Click to view more details.

The second case is a Product Sorting Automation solution using the UR5 and deep learning. Similar to the previous case, the robot directly sorts and places products through 3D vision recognition.

The biggest advantage of this solution is that even if the position of the box where the robot moves and places products changes, the vision recognizes the Pick & Place locations to perform the task. It also doesn't matter if the position of the product itself changes instead of the box, so there is no need to worry about changing locations. (Surprisingly, many processes want to change locations after introducing robot automation. 😅)

Additionally, by utilizing motion planning, the robot learns and follows along well on its own, reducing the worker's teaching time. Through reinforcement learning, the total cycle time can also be significantly reduced. With easy operation, low maintenance, and minimized downtime, the list of benefits goes on. Let's compare them at a glance below. 😉

 Wait a minute! 👉🏻 Cycle time, tact time… what are those?

Now, it is time for the highly anticipated 'Comparison' segment that we simply cannot skip today! 👏

The biggest advantage of both solutions was their excellent recognition capabilities for various objects thanks to their deep learning-based 3D object recognition! You can choose the one that best suits your process.

As the exact price for the Classification automation solution has not yet been determined, please request a quote through Marosol's Consultation service, and we will be happy to assist you. 🙂

Looking for smart AI palletizing with a cost-effective vision system?

✅ Affordable 3D Vision: A solution priced reasonably.

✅ Superior Recognition: Deep-learning-based 3D object recognition for diverse items.

✅ High Performance: Ensures excellent recognition for dark colors or high light reflection (e.g., tape).

✅ Wide Area Recognition: Capable of performing pallet tasks for stacking heights over 2M.

A product sorting solution that shortens cycle time with easy robot teaching!

✅ Reduced Teaching Time: Automated vision calibration and teaching using deep learning.

✅ Self-Learning: Automatically learns optimal pick positions based on product shape.

✅ Easy Installation: Can be installed and operated within the existing space used by workers.

✅ Optimized Cycle Time: Shortens cycle time by optimizing work posture and paths through continuous learning.

✅ Auto-Correction: Can perform tasks and auto-calibrate even if the layout shifts or the robot ages.

✅ Stable Productivity: Secured through minimized downtime.

When I heard about an AI robot that only scored 16 points on a CSAT math exam, I dismissed it as a joke, but it’s truly fascinating to see robots now learning and mastering work methods on their own. 

Whether it's product sorting or palletizing—that is the question! The necessary solutions and options can vary depending on which process you are looking for. Please read through, and feel free to leave any questions via an inquiry anytime! :)