The myCobot 280 RDK X5 is built with the RDK OS operating system and equipped with the D-Robotics RDK X5 Robot Developer Kit. It offers up to 10 TOPS of computing power and supports various complex models, including Transformer, RWKV, Occupancy, Stereo Perception, and the latest algorithms. This enables rapid deployment of intelligent applications. With a focus on intelligent computing and robotics applications, it provides rich interfaces and exceptional ease of use.

Support for Multiple Open-Source Vision and Large Language Models

Large Language Models (LLMs)

It supports the deployment and efficient operation of large language models (LLMs) based on Transformer and RWKV, such as LLaMA, RWKV V4 0.5B, Qwen2-0.5B, CLIP, and other vision-language models. It enables features like image-text search and language interaction on the RDK X5.

YOLO-World Vision

It supports edge-side open vocabulary detection models, combining language and visual features, offering strong generalization capabilities. It runs efficiently on the RDK X5 and can address the long-tail problem in robotics.

Mobile SAM

It supports lightweight, general-purpose segment-ation networks on the edge side, utilizing the combination of prompts and images to achieve universal object segmentation. It runs efficiently on the RDK X5 and, combined with YOLO-World, enables pixel-level localization of obstacles in images.

PRODUCT VIDEO

Compatible with RDK AI Kit

It integrates seven visual algorithms and supports the local deployment of the Deepseek large model. It covers applications such as object recognition with YOLOv8, waste classification detection, ArUco code recognition for palletizing, and OCR text recognition and extraction. From image feature extraction to neural network deployment, it builds a complete algorithm application-engineering practice workflow. With Python code implementation for neural network deployment, it enables quick mastery of AI vision algorithm development and robot collaboration capabilities. This makes it suitable for university training, robot labs, and personal technical exploration.

The RDK AI Kit includes: 1× suction pump, 1× G-shaped base, 1× camera flange, 1× mat, and a set of blocks.

Kindly note that the gripper is not included and must be purchased separately.

Empowering Collaboration Between RDK X5 and myCobot 280 Ecosystem

The RDK X5 boasts an outstanding community ecosystem, bringing together nearly 100,000 developers. It offers hundreds of accessories and kits, along with a wealth of content and events, helping developers solve problems and enhance their technical abilities. Serving as the control core, the myCobot 280 RDK X5 is specifically developed for scientific research, education, and innovative technology applications. It supports various control methods, including puzzle-style programming, code-based programming, and manual dragging. It comes with built-in motion control algorithms and offers over 90 open-source drive interfaces to aid in the development of complex applications. The integration of both ecosystems provides strong support for robot development, driving technological innovation and expanding applications.

PARAMETERS

High-Performance Configuration

With its powerful product configuration, it can meet your embedded development needs. Through button interactions, screen displays, buzzers, and IO interfaces, whatever you think of can be developed.

Based on the RDK X5 from D-Robotics

Robot Operating System – RDK OS Simulation Support

RDK OS — A robot operating system specifically developed for the RDK series robot developer kits, built on Linux and Ubuntu as the system kernel. It is deeply optimized based on ROS 2, custom-designed to provide high performance, high reliability, and high real-time capabilities through the robot development middleware TogetheROS.Bot. RDK OS offers a wealth of core algorithms and model inference interfaces for robot development. It also integrates a variety of application functions, including VSLAM, Occupancy, Stereo Perception, Yolo World, and large models, helping developers quickly build and deploy robot applications.

Robot Programming Made Easy

By offering the image-based drag & drop programming software myBlockly specifically designed for entry-level education, along with puzzle-based programming and graphical debugging tools, even beginners can easily master the robotic arm. The Python 2/3 development environment supports a variety of development libraries, including OpenCV, QT, pymycobot, and others, fulfilling users' advanced development needs in robot control, image recognition, and front-end development.