Robot arms

JetArm

A desktop robot arm on NVIDIA Jetson for teaching ROS, 3D vision and embodied AI.

NVIDIA Jetson
Suggested for university

JetArm is a desktop robot arm that Hiwonder designed for ROS teaching, pairing an NVIDIA Jetson controller with an STM32 robot controller. In the Standard, Advanced and Ultimate kits, a 3D depth camera on the end of the arm lets it recognise, track and pick up objects in three dimensions; the Starter Kit has a monocular camera. The Ultimate Kit adds a circular 6-microphone array, so the arm can take spoken instructions through ChatGPT and act on what it sees. It runs ROS1 or ROS2, and Hiwonder supplies manuals, video tutorials, source code and system images.

Zeta42 is an authorised Hiwonder partner in the Middle East.

Configurations

Choose how it ships.

Model: Starter Kit · Standard Kit · Advanced Kit · Ultimate Kit
Controller: Without Jetson Controller · With Jetson Nano(4GB) · With Jetson Orin Nano Super(4GB) · With Jetson Orin Nano Super(8GB) · With Jetson Orin NX Super(8GB) · With Jetson Orin NX Super(16GB)

What it does

Built for.

3D vision and grasping
The depth camera fuses colour and depth (RGB+D), so JetArm can find objects of different shapes, follow them and pick them up in a 3D scene.
Voice and large AI models
With the Ultimate Kit's microphone array and a Jetson controller, JetArm takes spoken commands through ChatGPT and pairs them with vision: sorting by colour, sorting waste, describing a scene or tracking an object it has been asked to find. Intelligent transport and the smart home assistant need a Jetson Orin Nano or Orin NX.
Train your own detection models
YOLOv11 is built in for real-time detection of several objects at once, and students can train models on their own objects, producing lightweight models optimised for TensorRT acceleration.
Metal arm on a bearing base
The arm is anodised aluminium throughout, the base uses industrial-grade bearings, and the servo wiring runs inside the body.
Simulate first
JetArm is developed on ROS and supports Gazebo, so control and algorithms can be checked in a virtual environment before they run on the real arm.
Room to grow
The base board has ports for Hiwonder's sliding rail and mecanum wheel chassis. The Advanced and Ultimate kits add a 7-inch touchscreen at the back that shows the arm's camera view and status in real time.

Where it is used

Applications.

University courses in ROS, inverse kinematics and machine vision
Embodied AI projects that connect a language model to a physical arm
Training and deploying custom object detection models on edge hardware
Sorting and pick-and-place exercises by colour, shape or object type

Videos

See it run.

Specifications

The full sheet.

Product dimensions: 339 × 165 × 532 mm
Product weight: Approx. 2.4 kg (ultimate version)
Body material: Full-metal aluminum alloy frame, anodized surface treatment
STM32 controller: ROS robot controller (STM32F407VE16), embedded with FreeRTOS real-time OS
ROS controller: Jetson Nano / Jetson Orin Nano / Jetson Orin NX
Control methods: USB serial port, CAN interface, Bluetooth app, RC transmitter, computer
USB expansion: 5A high-current USB HUB expansion board
Camera: Gemini Plus depth camera (Standard, Advanced and Ultimate Kit)
Power supply: 12V 5A DC power adapter / 19V 2.37A power adapter
Operating system: Ubuntu 22.04 LTS + ROS Humble / Ubuntu 20.04 LTS + ROS Noetic
Software: iOS / Android app
Communication: USB / Wi-Fi / Ethernet
Programming language: Python / C / C++ / JavaScript
Storage: 64GB TF card (Jetson Nano) / 128GB SSD (Jetson Orin Nano/NX)
Servo model: HTS-21H / HTD-35H / HX-12H smart bus servo motors
Optional accessories: Integrated 6-microphone array (Ultimate Kit), and 7-inch HD touchscreen (Advanced and Ultimate Kit)
Supporting materials: Full development manuals, video tutorials, ROS/STM32 source code, system images, software tools
Package weight and size: 4 kg; 385 × 295 × 175 mm