Elephant Robotics Technology Mycobot 320 AI Kit 2023 (Adaptive Gripper Version) (4010800016)

Teaching a robot arm to move is one problem; teaching it to see what it is moving toward is another. The Elephant Robotics Technology Mycobot 320 AI Kit 2023 (Adaptive Gripper Version) (4010800016) is built for the second problem. It surrounds a desktop arm with a camera, ready-made vision software and a gripper that adapts to the object in front of it, so a classroom or lab can move from theory to a block actually being picked up and sorted.

This product is available through this page, by quotation.

Background

The company and its lineage

Elephant Robotics describes itself as a specialized and nationally recognized high-tech enterprise that provides robot solutions for scientific research, education and maker development. That mix of audiences explains the character of the company's products: approachable enough for a first-time student, yet programmable enough for a researcher who wants to test an idea. The myCobot family of compact collaborative arms is the clearest example, and the 320 sits within it alongside smaller and larger siblings. You can browse related arms in the Cobots category or see more of the maker's range under Elephant.

The myCobot 320 has appeared in M5 and PI editions, named for the controller platform that each version pairs with. This 2023 AI Kit is designed to work with both adapters, so an arm already in a lab does not have to be replaced to gain vision capabilities.

Positioning and key technology

The kit is an entry-level AI package that integrates positioning, automatic sorting and object grasping modules. In practical terms, that means the three things a beginner struggles to assemble separately, namely finding an object, deciding what it is and picking it up, arrive as one coordinated set. It features 5 major visual recognition algorithms, 7 application scenarios and 2 gripping methods, which gives a teacher or developer a structured path through machine vision rather than a blank page.

The software side is built on the Python platform, which allows software development to control the robotic arm. Python is the language most students and researchers already know, so the barrier between an idea and a working test stays low. The set also includes the YOLOv5 algorithm, which helps learners understand neural networks, and four ID QR codes that illustrate the projection relationship between the 2D and 3D worlds. For hands-on practice, four colored wooden blocks and cards of different shapes can be freely chosen.

Two types of end effector exist for the kit: the myCobot Pro Adaptive Gripper and the myCobot Pro Single Head Suction Pump. This page covers the Adaptive Gripper Version.

Technical Specifications

Specifications as published by the manufacturer; confirmed against the datasheet at quotation.

The manufacturer has not published the kit weight, the power supply of the kit, the camera mounting details, the full box contents list or the IP rating; they are confirmed at quotation.

What makes the Adaptive Gripper Version different from the suction pump option?

The kit comes with a choice of two end effectors, and the choice shapes what the arm can pick up. The adaptive gripper closes around an object, with a clamping range of 0-90mm, so it suits blocks and other items with solid sides that a pair of fingers can hold. The single head suction pump works differently, holding an object by suction rather than by clamping, and it is the better fit for objects with a flat surface to seal against.

For teaching, the gripper version has a clear advantage: students see how grip width, object position and recognition accuracy interact. When the camera reports a position within the 3mm recognition accuracy, the gripper still has to close at the right width and the right moment. That makes mistakes visible and instructive. The gripper also weighs 350 grams, a figure worth including when you plan the payload budget of the arm, which is rated for a maximum of 1kg.

How does the AI Kit see and sort objects?

Vision begins with a USB distortion free camera delivering 2M 1080p images through a lens with a 2.8mm focal length and a 110° field of view with no distortion. A wide, distortion-free view matters because positioning calculations depend on straight lines staying straight; a warped image would put the arm's target off before any algorithm had a chance to run.

From there, the software applies its recognition algorithms and works out where an object sits. The four ID QR codes help explain how a flat camera image maps to real positions in three-dimensional space, and the YOLOv5 algorithm introduces neural-network detection. The arm, with 6 degrees of freedom and a working radius of 350mm, then reaches the computed position and the end effector grasps. Because everything is controlled through Python, a developer can read, change and extend each stage. The kit runs on Windows 10, Windows 11, Ubuntu 18.04 and Ubuntu 20.04.

Who is the Mycobot 320 AI Kit 2023 (Adaptive Gripper Version) for?

Educators are the first group. A teacher building a robotics or computer vision course gets a defined set of algorithms and scenarios, plus physical materials such as the colored blocks and shape cards, so lessons do not depend on improvised parts.

University labs and researchers are the second. They can use the kit as a quick prototyping bench for vision-guided grasping, writing their own Python code on top of the supplied modules and testing it on an arm that fits on a desk.

Makers and developers form the third group. If you already own a myCobot 320 in its M5 or PI edition, the compatibility with both adapters means the kit extends a machine you have rather than replacing it. Buyers who want a larger arm for heavier work may also look at the Elephant Robotics myCobot Pro 630 Robot Arm, while those exploring smaller platforms can compare the Mycobot-280-Pi.

What is the Mycobot 320 AI Kit 2023 used for?

1. Teaching machine vision. Students learn how a camera image becomes a position the arm can act on.
2. Automatic sorting. The sorting module separates objects such as colored blocks by what the software recognizes.
3. Object grasping practice. The gripper handles pick-up tasks, with objects up to 0-90mm in clamping range.
4. Neural network study. The included YOLOv5 algorithm gives learners a working example of neural-network detection.
5. Python development. Developers write and test their own control code on the Python platform.
6. Research prototyping. Labs use the kit to test vision-guided manipulation ideas before scaling to larger equipment.

Pricing and Availability

We do not publish prices on this page. Quotation is the way to proceed: tell us your quantity and your location, and we prepare an offer for the Adaptive Gripper Version. Delivery, customs and support are arranged with the quotation, so you know the full picture before you commit. The Elephant Robotics Technology Mycobot 320 AI Kit 2023 (Adaptive Gripper Version) (4010800016) is available on this page.

Frequently Asked Questions

What is included in the AI Kit?

The set includes four colored wooden blocks, cards of different shapes, four ID QR codes and the YOLOv5 algorithm, together with a USB distortion free camera and the adaptive gripper end effector.

Which robotic arms is it compatible with?

It is compatible with M5 and PI adapters, which means it works with the myCobot 320 arm in both editions.

What programming language does it use?

The kit is built on the Python platform, which allows software development to control the robotic arm.

Which operating systems are supported?

Windows 10, Windows 11, Ubuntu 18.04 and Ubuntu 20.04 are supported.

How much can the arm carry?

The maximum payload is 1kg. The adaptive gripper itself weighs 350 grams, and its clamping range is 0-90mm.

How accurate is the recognition?

Recognition accuracy is 3mm, using a USB distortion free camera with 2M 1080p images and a 110° field of view.

Where can I buy the Mycobot 320 AI Kit 2023 (Adaptive Gripper Version) and what does it cost?

You can request it through this page. The cost is provided by quotation, together with delivery, customs and support arrangements.

Summary

Seeing is the hard part of robotics, and this kit treats it that way. A 2M 1080p camera with a 110° undistorted view, 5 recognition algorithms, 7 application scenarios, YOLOv5 and a Python platform give the myCobot 320 a complete route from image to grasp, while the adaptive gripper turns the result into a physical pick-up. For schools, labs and makers across the Caribbean who want to learn vision-guided handling on a real desktop arm, this page is the starting point.