Mobile and ROS robots

JetAuto

A mecanum wheel ROS car on NVIDIA Jetson or Raspberry Pi 5, with lidar SLAM and 3D vision.

Raspberry Pi
NVIDIA Jetson
Suggested for university

JetAuto is a ROS education robot car on a mecanum wheel chassis with pendulum suspension, sold with an NVIDIA Jetson or a Raspberry Pi 5 controller. Every kit has a lidar for SLAM mapping, path planning and obstacle avoidance, and from the Standard Kit up a 3D depth camera adds 3D mapping, point clouds and AI vision. The Advanced and Ultimate kits include a 7-inch touchscreen and a circular 6-microphone array; paired with a Jetson Orin Nano or Orin NX controller, they add multimodal large AI models that turn spoken instructions into navigation and describe what the robot finds. With those models, JetAuto also works with the OpenClaw agent. Hiwonder supplies tutorials, schematics, source code and videos.

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 Controller · With Raspberry Pi 5(8GB) · 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.

Mecanum chassis with suspension
Four mecanum wheels move JetAuto forwards, sideways, diagonally and on the spot, and a pendulum suspension keeps all four on uneven ground.
Lidar SLAM
The lidar maps a space with Gmapping, Hector, Karto or Cartographer, then plans paths, navigates to fixed points and avoids moving obstacles. The Ultimate Kit uses an EAI G4 lidar, the other kits a SLAMTEC A1.
3D depth camera
From the Standard Kit up, the depth camera on a 240° pan-tilt captures point clouds for RTAB-VSLAM 3D mapping, and supports KCF tracking, AprilTag and colour recognition, YOLO object recognition and MediaPipe body, fingertip and face detection.
Large AI models
In the Advanced and Ultimate kits with a Jetson Orin Nano or Orin NX controller, language, speech and vision models run online or locally. A language model turns spoken commands into multi-point navigation, and a vision language model describes the objects and events JetAuto finds. Hiwonder's kit comparison states that these features need that pairing.
OpenClaw agent
With the large AI models, the OpenClaw agent takes text or voice instructions from a PC or app, plans the steps, carries them out and reports on what it finds.
Simulation
JetAuto supports Gazebo simulation, and its URDF model can be viewed in RViz to check mapping and navigation while debugging.

Where it is used

Applications.

University ROS courses covering SLAM, navigation and path planning
Embodied AI projects with language and vision models (Advanced and Ultimate kits with Jetson Orin)
3D vision and deep learning with the depth camera (Standard Kit and up)
Multi-robot navigation and formation

Videos

See it run.

Specifications

The full sheet.

Weight: 3500g
Material: Full-metal hard aluminum alloy bracket
Battery: 11.1V 6000mAh lithium battery
Continuous working time: 60 min
Drive motors: 520 Hall encoder geared motors, 1:90 ratio, 15kg.cm torque
Pan-tilt: 240° high-performance pan-tilt
Operating system: ROS1: Ubuntu 18.04 (Jetson Nano); ROS2: Ubuntu 22.04 (all three Jetson boards)
Lidar: SLAMTEC A1 (Starter, Standard and Advanced Kit); EAI G4 (Ultimate Kit)
3D depth camera: Standard, Advanced and Ultimate Kit
Screen: 7-inch HD LCD touchscreen, 1024 × 600 (Advanced and Ultimate Kit)
Microphone: Circular 6-microphone array and speaker (Advanced and Ultimate Kit)
Software: iOS / Android app
Communication: USB / WiFi / Ethernet
Programming language: Python / C / C++ / JavaScript
Control method: Phone / wireless handle
Package size (Advanced Kit): 34 × 32 × 23 cm
Package weight (Advanced Kit): About 4.5kg