Waveshare UGV Beast

The UGV Beast is an AI robot designed for exploration and creation, with great expansion potential. It seamlessly connects your imagination to reality, suitable for tech enthusiasts, makers, or beginners in programming, it is your ideal choice for exploring the world of intelligent technology.

Powered by the high-performance Raspberry Pi computer (not included) to handle complex strategies and functions while inspiring your creativity. Adopts dual-controller design, combines the high-level AI functions of the host controller with the high-frequency basic operations of the sub controller, ensuring that every operation is accurate and smooth.

You can easily control it remotely via the UGV Beast Web Application without downloading any software, just open your browser and start your journey. Supports high-frame rate real-time video transmission and multiple AI Computer Vision functions, the UGV Beast is an ideal platform to realize your ideas and creativity!

Dual-controller design, provides efficient collaboration and upgraded performance

The host controller adopts Raspberry Pi for AI vision and strategy planning, and the sub controller uses ESP32 for motion control and sensor data processing.

Continuing the adventure as night falls

High-brightness LED light for ensuring clear images in low-light conditions.

Standard aluminum rail

Comes with 2×1020 European standard profile rails, and supports installing additional peripherals via the boat nuts to meet different needs, easily expanding the special operation scenarios.

Supports driving in complex terrain

Adopts tracked mobile robot chassis with independent suspension systems for more stable off-road crossing ability.

Comes with a multi-functional mounting plate

Can be used to install a variety of peripherals, including D500 Lidar, STL-27L Lidar, camera, or other modules. Open source for DXF drawing and 3D model, which is more convenient for secondary development.

Easy to control via cross-platform web application

No App installation required, allows users to connect and control the robot via mobile phones, tablets and computers via browser Web App. Supports shortcut key control such as WASD and the mouse via a PC with keyboard.

Real-time wireless transmission of high frame rate video

Adopts Flask lightweight web application, based on WebRTC ultra-low latency real-time transmission, using Python language and easy to extend, working seamlessly with OpenCV.

Vision line tracking for autonomous driving

Integrated with vision line tracking function on the Computer Vision demo, comes with yellow tape for easier planning a path for the robot. Users can understand the basic algorithms of autonomous driving with this simple demo.

Reserved space for expanding peripherals

Reserved space on the mounting plate with M2.5 mounting holes, for installing D500/STL27L Lidar or other peripherals. Comes with screws and cables for installing the D500 Lidar, please refer to wiki page for more details. The D500 Lidar and battery set are note included.

40PIN GPIO extended header

The robot only occupies the URAT interface of the Raspberry Pi GPIO for communication, adapting outer side 40-pin header on the driver board for expanding more peripherals and functions.

Multiple audio interaction methods

Equipped with audio driver board, dual-track speaker, and onboard microphone, built in 3.5mm audio jack for connecting peripherals to achieve text-to-speech and audio playing.

Obtains real-time information feedback

Real-time monitoring the operating status of the robot.

Web page command line tool, multiple functions for easier expansion

Quick to set up, easy to expand. Easily customize and add new functions without modifying front-end code.

ESP-NOW wireless communication between robots

Based on ESP-NOW communication protocol, multiple robots can communicate with each other without IP or MAC address, achieving multi-device collaboration with 100-microsecond low-latency communication.

Gamepad control for better operation experience

Comes with a wireless gamepad, making robot control more flexible. You can connect the USB receiver to your PC and control the robot remotely via the internet. Provides open source demo for customizing your own interaction method.

Multi-functional driver board

Used as the sub controller of the robot, onboard ESP32 module and can be controlled via serial port, http request, ESP-NOW and other communication protocols, provides a web application for some basic control of the robot. ESP32 can handle low-level computations for the robot, such as high-frequency PID controller for controlling wheel speed, high-frequency inverse kinematics calculations for robotic arm linkages, position interpolation calculation, Pan-Tilt angle control and OLED screen control, etc. ESP32 is also responsible for data reading of various sensors, such as IMU (9-axis attitude sensor) and INA219 (battery voltage detection), and can independently achieve the camera vertical stabilization function, effectively saving computing and IO resources for the host controller.

1. ESP32 module
2. ESP32 module antenna connector
3. Motor control interfaces
4. Battery voltage detection IC
   INA219
5. Power supply
6. USB communication / downloading
7. Lidar USB connector
8. ESP32-IO0 button
9. ESP32-EN button
10. I2C device interface
11. Lidar UART interface
12. Bus servo control interfaces
13. Motor driver IC
    TB6612FNG
14. Host controller 40PIN extended header
15. Host controller connection header
16. I2C device interface
17. 9-axis IMU attitude sensor
    ICM20948
18. 12V switch controlled by ESP32-IO4
19. 12V switch controlled by ESP32-IO5

Technical data

Supports 4G/5G expansion

Supports installing 4G/5G module for the application scenarios without WiFi. You may need to use Tunneling Service such as Ngrok, Cpolar, or LocalTunnel to expose the local network service of the robot (Flask application) to the internet so that you can control the robot from anywhere.

Realizes remote control across the internet

Our web application demos are based on WebRTC for real-time video transmission. WebRTC (Web Real-Time Communications) is a technology that enables web applications and sites to establish peer-to-peer connection and capture optionally stream audio and/or video media, as well as to exchange arbitrary data between browsers without requiring an intermediary. We provide comprehensive Ngrok tutorials to help you get started quickly and realize robot control across the internet.

Supports installing smartphone holder

If you have a spare phone, you can install it on the robot via holder as below, using the phone to create a hotspot for the robot and achieving remote control across the internet at a lower cost. It comes with a smartphone holder with 1/4″ screw in the package.

Cross-platform interactive tutorial, develop while you learn

Supports accessing Jupyter Lab via devices such as mobile phones and tablets to read the tutorials and edit the code on the web page, making development easier.

Rich tutorial resources

We provide complete tutorials and demos to help users get started quickly for learning and secondary development.

18650 lithium battery support

Onboard high-efficiency UPS module, supports charging and discharging at the same time to ensure continuous standby time. The 18650 batteries are not included.

Large internal space for extension

Supports installing larger lithium battery set with XH2.54 interface to replace the UPS module. Please refer to the open-source drawing for the detailed internal space dimensions.

High strength aluminum body

2mm thickness aluminum alloy body, durable and reliable. Tracked robot chassis, excellent shock absorbing performance.

High-speed motor & high-frequency PID closed-loop speed control

Supports quick response without occupying the resources for the host controller

Color Recognition

Based on OpenCV to achieve Color Recognition. Supports automatic LED lighting, allows expansion for more functions.

Face Detection: Automatic picture or video capturing

Based on OpenCV to achieve Face Recognition, supports automatic photo taking or video recording once a face is recognized.

Intelligent Object Recognition

Supports recognizing for many common objects with the default model.

Gesture Recognition: AI interaction with body language

Combines OpenCV and MediaPipe to realize gesture control.

More MediaPipe demos for easily creating complex video processing tasks

MediaPipe is an open-source framework developed by Google for building cross-platform multimedia processing pipelines, provides a set of pre-built components and tools, its high-performance processing capability enables the robot to respond to and process complex multimedia inputs such as real-time video analytics.

Audio driver board

Audio driver board is designed for robot host controller, onboard USB interface, suitable for Raspberry Pi 4B / Raspberry Pi 5 and other motherboards, multi-system compatible. Adopts SSS1629A5 audio control chip, provides reliable audio input and output functions, supports USB interface communication, driver-free, plug and play. Onboard APA2068 audio amplifier chip, provides high-quality audio amplification to ensure clear and stable audio signal output. Integrated FE1.1S USB 2.0 HUB and CH340 chips for sending data from serial peripherals such as Lidar to the host controller via USB cable without occupying too many resources of the host.

1. Lidar UART to USB IC
2. USB HUB IC
3. Type-C port
4. Lidar UART Interface
5. 3.5mm AUX audio jack
6. Audio amplifier IC
   APA2068KAI-TRG
7. Audio control IC
   SSS1629A5
8. Cooling fan power supply header
9. Dual-track microphone
10. Dual-track speaker header