WLKATA AI Automatic Sorting Cell (Mirobot) | Touch Screen

The WLKATA AI Vision Set uses an OpenMV-based camera module to dynamically identify target objects on the conveyor. The included color blocks support color-based sorting out of the box. The identification type can be reprogrammed for other criteria: shape recognition, custom-trained AI models, barcode/QR detection, and feature-based classification. This flexibility extends the cell from introductory color sorting to research-grade vision tasks.

Two arms teach multi-robot coordination, work-zone allocation, and dual-arm scheduling: topics central to real smart-factory deployments where multiple robots share workspace. A single larger arm would teach single-arm motion planning but miss the coordination challenges that account for much of real-world industrial automation complexity. The cell pairs a Professional Kit (with Bluetooth controller for handheld programming) with an Education Kit (standard PC-based programming).

Yes. The vision processing pipeline is OpenCV-based, but the broader software architecture supports Python-driven AI integration. Students and researchers can deploy TensorFlow, PyTorch, or YOLOv5 models for advanced detection and classification tasks. A common capstone project is to extend the base cell with a deep-learning detector for a novel sort criterion beyond color.

The primary programming environment is C++ via the Arduino IDE, leveraging the Arduino Mega 2560-based master control box. Additional support is provided for Blockly (introductory graphic programming), Python (via the WLKATA Python SDK), MATLAB (for university-level coursework), and ROS2 (for advanced robotics integration). The same hardware platform supports all skill levels.

The WLKATA AI Vision Set is a calibrated integrated camera-and-stand unit using an OpenMV-based vision module designed specifically for the cell. While the same OpenCV algorithms can run on a generic webcam, the AI Vision Set provides consistent lighting, fixed-distance calibration, and a known coordinate frame relative to the conveyor. This eliminates the calibration drift and lighting variance issues that complicate generic webcam setups. The OpenMV module also offers on-board image processing capability, reducing host-PC processing load.

The AI Automatic Sorting Cell focuses on vision-guided sorting using OpenCV and a conveyor belt with two Mirobots. The Automobile Assembly Cell focuses on multi-station automotive assembly using three Mirobots, a sliding rail, a 1/18 scale car model, and a welding simulation module. Schools teaching multi-industry curriculum often deploy both cells; they share the Mirobot platform, simplifying maintenance and student knowledge transfer.

The cell supports a range. High school students using Blockly programming can complete introductory color-sorting exercises. Community college students using C++ via Arduino develop new sort criteria. University students using ROS2 and custom AI models complete research-grade integrations. The included textbook provides structured exercises across these levels.

Yes. The cell is a single integrated unit. For multi-cell production-line scenarios (e.g., chaining AI Sorting to Automobile Assembly), schools purchase multiple cells and integrate them via shared conveyor or coordination programming. WLKATA supports this configuration; schools should plan for floor space and cell-to-cell coordination protocols.

RoboticsSelect is an authorized WLKATA dealer, so the full 12-month manufacturer warranty applies to all AI Automatic Sorting Cell purchases. Authorized dealer status ensures warranty claims are processed through verified channels. For schools and institutions purchasing multiple cells or committing to multi-year curriculum partnerships, RoboticsSelect supports institutional purchase orders and provides ongoing dealer support.