English | 中文
We created MentorPi with a simple goal: to build an accessible, cost-effective educational robotics platform that integrates AI and ROS 2. By combining powerful large language models with a flexible hardware design, we hope to empower engineers, students, and creators to learn and experiment with AI robotics—and bring their most imaginative projects to life.
MentorPi is built around a robust STM32 + Raspberry Pi 5 control system and offers 3 chassis options: Ackermann chassis, Mecanum wheels and tank chassis, so you can pick the right setup for your application. Despite its compact size, it's packed with capable hardwares: high-speed encoder motors, LiDAR, a 3D depth camera, and an AI voice module. This lets you implement advanced AI behaviors like SLAM-based navigation, real-time object tracking, and traffic sign recognition using YOLOv11—all the way to fully autonomous driving.
As a fully open source platform, MentorPi is made for customization and expansion. This year, we've introduced multi-chassis support and integrated multimodal AI models with natural voice interaction—enabling more sophisticated embodied intelligence tasks. Whether you're building autonomous driving systems or exploring human robot interaction, we welcome you to join our community and help shape the future of MentorPi.
Get started with our MentorPi Tutorials to bring your first robot to life.
- Official Website: https://www.hiwonder.com/
- Product Page: https://www.hiwonder.com/products/mentorpi-m1
- Official Documentation: https://docs.hiwonder.com/projects/MentorPi/en/latest/
- Technical Support: support@hiwonder.com
- MentorPi Open Source Robot Car: ROS2 & Raspberry Pi 5: Watch
- Can AI Models Handle SLAM Navigation? See MentorPi Prove It!: Watch
- SLAM Mapping - Real-time simultaneous localization and mapping
- Path Planning - Intelligent route planning and navigation
- Autonomous Driving - Self-driving capabilities with obstacle avoidance
- YOLOv5 Recognition - Advanced object detection for road signs and traffic lights
- 3D Depth Perception - Stereo vision with depth camera integration
- Modular Chassis Design - Support for three chassis configurations: Mecanum wheels, Ackermann, and tank tracks
- Closed-loop Motor Control - High-precision encoder feedback control
- Servo Control - High-torque servo systems for precise movements
- Multi-sensor Fusion - Integrated sensor data processing
- ROS2 Integration - Full Robot Operating System 2 support
- Python Programming - Comprehensive Python SDK
- Multimodal AI Model - Advanced embodied AI capabilities
- Machine Learning - YOLOv5 training and inference support
- Processor: Raspberry Pi 5
- Operating System: ROS2 compatible Linux
- Motors: High-speed closed-loop encoder motors
- Vision System: 3D depth camera + Lidar sensor
- Actuators: High-torque servos
- Chassis: Supports three chassis configurations: Mecanum wheels, Ackermann, and tank tracks
mentorpi/
├── app/ # Application modules
├── bringup/ # System startup and configuration
├── calibration/ # Sensor calibration utilities
├── driver/ # Hardware drivers
├── example/ # Example applications and demos
├── interfaces/ # ROS2 message definitions
├── large_models/ # AI large model integration
├── multi/ # Multi-robot coordination
├── navigation/ # Navigation stack
├── peripherals/ # Peripheral device support
├── simulations/ # Simulation environments
├── slam/ # SLAM algorithms
└── yolov5_ros2/ # YOLOv5 ROS2 integration
- Current Version: MentorPi M1 v1.0.0
- Supported Platform: Raspberry Pi 5
- ROS2 - Robot Operating System 2
- OpenCV - Computer Vision Library
- YOLOv5 - Object Detection Framework
Note: This program is pre-installed on the MentorPi M1 robot system and can be run directly. For detailed tutorials, please refer to the Official Documentation.