Legged robots

PuppyPi

A Raspberry Pi 5 robot dog with eight coreless servos, AI vision and ROS1 and ROS2 support.

Raspberry Pi
Suggested for secondary
Suggested for university

PuppyPi is a quadruped robot dog built on the Raspberry Pi 5. Eight coreless servos drive legs with a linkage design, and it runs both ROS1 and ROS2 with Python programming. A 2DOF HD wide-angle camera and OpenCV give it target tracking, face detection, line following, and the vision to climb stairs and clear obstacles. The Advanced Kit adds an AI voice interaction module for voice control through ChatGPT, the Pro Kit adds a lidar for mapping and navigation, and the Pro Ultimate Kit adds a robotic arm for picking up and carrying objects.

Zeta42 is an authorised Hiwonder partner in the Middle East.

Configurations

Choose how it ships.

Model: PuppyPi Standard Kit · PuppyPi Advanced Kit · PuppyPi Pro Kit · PuppyPi Pro Ultimate Kit
Controller: With Raspberry Pi 4B (4GB) · With Raspberry Pi 5 (2GB) · With Raspberry Pi 5 (4GB) · With Raspberry Pi 5 (8GB) · With Raspberry Pi 5 (16GB)

What it does

Built for.

Linkage legs and coreless servos
Eight high-performance coreless servos give precise, flexible motion. PuppyPi can switch between walk, amble and trot gaits, and the PC software adjusts its height and lean as it moves.
AI vision with OpenCV
The head camera finds objects by colour and tracks them, reads tags, detects faces with MediaPipe and follows coloured lines. It also judges stairs and obstacles in its path and adjusts its posture to get over them.
Voice and large AI models
From the Advanced Kit up, ChatGPT turns spoken commands into actions: tracking and kicking a ball, patrolling a line, describing the scene or adjusting its posture on command.
Lidar and robotic arm
The Pro Kit adds a lidar for SLAM mapping, path planning and obstacle avoidance, plus a sensor pack with temperature and humidity, ultrasonic and touch sensors and a dot matrix display. The Pro Ultimate Kit adds a robotic arm for grasping and carrying.
Several ways to control it
Program it in Python with open, annotated code, use the PC software, or drive it from the iOS and Android app with a live view from its camera. The Pro kits include a Bluetooth wireless controller.

Where it is used

Applications.

Teaching legged locomotion and gait control
Computer vision projects in Python and OpenCV
Embodied AI with voice commands (Advanced Kit and up)
SLAM mapping and navigation on a legged robot (Pro kits)

Videos

See it run.

Specifications

The full sheet.

Size: 226 × 149 × 190 mm
Weight: 720g
Material: Aluminium alloy
Camera resolution: 480P
DOF: 8DOF
Power: 7.4V 2200mAh LiPo battery
Hardware: Raspberry Pi 4B/5 and Raspberry Pi expansion board
Software: PC software, iOS / Android app
Operating system: Ubuntu 22.04 LTS + ROS Humble & Ubuntu 20.04 LTS + ROS Noetic
ROS version: Raspberry Pi 5: ROS1 & ROS2; Raspberry Pi 4B: ROS1
Servo: HPS-0618SG coreless servo
Control method: PC, app and wireless handle control (wireless handle: Pro and Pro Ultimate Kit)
Package size: 32 × 32 × 17 cm
Package weight: About 1.5kg