Key Advantages
128-Core NVIDIA GPU for Edge AI
Jetson Nano variant runs real-time object detection and image recognition entirely on-device. Train and deploy vision models without external hardware — GPU-accelerated inference at the edge.
Full Arduino Ecosystem Compatibility
Supports UNO, MEGA2560, and MKR boards out of the box. Leverage thousands of existing Arduino libraries, shields, and community resources — the world's largest open-source hardware ecosystem.
780 g Ultra-Lightweight Body
The Arduino variant weighs just 780 grams — the lightest in the myCobot lineup. Move it between workstations, pack it for field demos, or mount it anywhere with minimal support.
Native ROS 1 + ROS 2 on Ubuntu
Jetson Nano runs full Ubuntu Mate 20.04 with ROS 1 and ROS 2 pre-configured. Drop into any existing ROS pipeline, access 90+ control interfaces, and integrate with Gazebo simulation.
Multi-Language Visual Programming
myBlockly drag-and-drop interface supports Python, JavaScript, C++, C# generation. Arduino IDE for embedded developers; OpenCV-ready for computer vision engineers — one platform, no coding silos.
±0.5 mm Precision at 280 mm Reach
Same 6-axis kinematics across both variants deliver repeatable sub-millimeter positioning. From PCB assembly to lab automation, the mechanical core is proven across thousands of units worldwide.
Proven Across Industries
From GPU-accelerated AI research to Arduino-powered embedded learning, two controllers open two worlds — both built on the same precision 6-axis platform with 280 mm reach and ±0.5 mm accuracy.
AI & Computer Vision Research
Der 128-Kern-Maxwell-GPU des Jetson Nano und die OpenCV-Integration verwandeln dieses Arm in einen Edge-AI-Arbeitsplatz. Führen Sie Objekterkennung durch, trainieren Sie benutzerdefinierte Vision-Modelle und setzen Sie Echtzeit-Inferenz um – alles innerhalb des auf dem Roboter integrierten Prozessors, ohne externe GPU-Cluster erforderlich.
Der 128-Kern-Maxwell-GPU des Jetson Nano und die OpenCV-Integration verwandeln dieses Arm in einen Edge-AI-Arbeitsplatz. Führen Sie Objekterkennung durch, trainieren Sie benutzerdefinierte Vision-Modelle und setzen Sie Echtzeit-Inferenz um – alles innerhalb des auf dem Roboter integrierten Prozessors, ohne externe GPU-Cluster erforderlich.