AI Large-Scale Bionic Mechanical Dog
Radar Mapping and Navigation | ROS Operating System | AI Visual Recognition and Interaction
Voice Interaction Control | Inverse Kinematics Algorithm | Bionic Gait and Movement
Features
ROS Operating System: An open-source meta-operating system providing the services an operating system should offer. | OpenCV a mainstream deep learning framework, can meet the needs of most artificial intelligence projects. | Mapping and Navigation Built-in mapping and navigation algorithm packages Autonomous obstacle avoidance during navigation |
Built-in IMU Built-in IMU sensor allows for real-time adjustment of the device's posture. | Serial bus servos 13 high-performance serial bus servos with powerful motors. | Python programming A mainstream programming language Numerous developer communities |
Raspberry Pi Powerful Raspberry Pi performance supports faster and smoother operation. | TOF LiDAR Capable of mapping, navigation, path planning, and other ROS SLAM functionalities. | 3D real-scene mapping Can be used with depth cameras To achieve 3D mapping |
Package differences
| |||||
| Function | Basic Edition | Graphica Programming Edition | AI Vision Edition | ROS Advanced Edition | ROS Professional Edition |
| APP Control | ✅ | ✅ | ✅ | ✅ | ✅ |
| Wi-Fi Real-Time Video | ✅ | ✅ | ✅ | ✅ | ✅ |
| Fun Motions | ✅ | ✅ | ✅ | ✅ | ✅ |
| Custom Motions | ✅ | ✅ | ✅ | ✅ | ✅ |
| Attitude Angle Control | ✅ | ✅ | ✅ | ✅ | ✅ |
| Camera Gimbal | ✅ | ✅ | ✅ | ✅ | ✅ |
| Height Control | ✅ | ✅ | ✅ | ✅ | ✅ |
| Stair Climbing | ✅ | ✅ | ✅ | ✅ | ✅ |
| Gait Adjustment | ✅ | ✅ | ✅ | ✅ | ✅ |
Graphical Programming | ❌ | ✅ | ❌ | ❌ | ❌ |
| Infrared Sensors | ❌ | ✅ | ❌ | ❌ | ❌ |
| Ultrasonic Sensors | ❌ | ✅ | ❌ | ❌ | ❌ |
| LED Lights | ❌ | ❌ | ❌ | ❌ | ❌ |
| Visual Line Following | ❌ | ❌ | ✅ | ❌ | ❌ |
| Voice Control | ❌ | ❌ | ✅ | ❌ | ❌ |
Wireless Handheld Control | ❌ | ❌ | ✅ | ✅ | ✅ |
| Face Recognition | ❌ | ❌ | ✅ | ✅ | ✅ |
| Color Recognition | ❌ | ❌ | ✅ | ✅ | ✅ |
| QR Code Recognition | ❌ | ❌ | ✅ | ✅ | ✅ |
| Python Programming | ❌ | ❌ | ✅ | ✅ | ✅ |
| ROS System | ❌ | ❌ | ✅ | ✅ | ✅ |
SLAM Mapping and Navigation | ❌ | ❌ | ❌ | ✅ | ✅ |
| Visual SLAM | ❌ | ❌ | ❌ | ❌ | ✅ |
Modular design
The robotic corgi has a serial port and slots for additional modules.
It allows you to assemble and build various programming modules, expanding the creative ways to play.
| AI Main Control Module | ROS Main Control Module | Graphical Module |
| Voice module | RGB camera | Depth camera |
High-performance servo with metal casing
1. Durable Aluminum Alloy Shell
The Corgi robot dog features a durable aluminum alloy shell, ensuring stability and safety in various environments,
making it suitable for flexible operation in multiple scenarios.
2. 13 Degrees of Freedom (DOF) Motion Joints
Equipped with 13 high-performance servos, each leg is connected to the elbow, shoulder, and three joints via a serial bus servo.
Using inverse kinematics algorithms, complex movements can be precisely achieved, ensuring overall fluidity and closely
resembling the movement posture of a realistic quadruped.
Vision-controlled obstacle crossing and agile companion
Video Pan/Tilt The robot dog has multiple built-in gait modes: Pace, Walk, Trot, and Hop, allowing it to adapt to different environments and task requirements. | Adjustable Height The robot dog supports height adjustment, allowing it to easily navigate over suspended obstacles |
| Composite posture and motion control Supports simultaneous combined control of speed and angle in the X, Y, and Z directions, enabling a wide range of running movements | Fun High-Leg Climbing Function The robot dog has a built-in stair-climbing function; while walking, it can step onto stairs from flat ground by raising its legs high. |
The robot dog has multiple built-in gait modes
Pace, Walk, Trot, and Hop. These allow the robot dog to adapt to different environments and task requirements.
1) Walk gait: Stepping forward with overlapping steps. | 2) Trot gait: Rapid alternating left and right movemen | 3) Hop gait: Leaning forward with a bouncing motion. | 4) Pece gait: Slow alternating left and right movement |
|
|
|
|
Expandable Visual AI Features
Optional RXR RPI Extend BOX AI Vision Edition expansion box with built-in Raspberry Pi development board
Integrates a wealth of machine vision AI features, showcasing the charm of artificial intelligence.
1. Voice Control Control the robot dog to perform actions via voice. | 2. Face Recognition Face recognition via APP/voice control |
3. Color Recognition Trigger the robot dog to recognize colors via the app/voice. | 4. QR Code Recognition QR code recognition via APP/voice trigger |
5. Visual Line Following The app triggers the robot dog to automatically follow lines visually. | 6. Handheld Control Control the robot dog to perform actions via handheld controller |
Forward/Inverse Kinematics Algorithm Development:
The Corgi robot dog's legs utilize 12 serial bus servos to control the rotation of each joint. Based on the biomechanical characteristics of the foreleg bones of real-life quadrupeds, these servos are linked to the shoulders and legs, achieving a very close resemblance to realistic biological movements. Using forward/inverse kinematics algorithms, the angles of each joint are calculated, and then, through a hardware interface, the servos of each leg joint are driven to rotate to the corresponding angles.
Built-in ROS and Machine Vision Basic Tutorial
11 lessons (PDF lecture notes + videos + source code) Getting started with ROS is no longer difficult
(Exclusive to AI Vision and ROS versions)
Expandable Graphical Programming
An optional RBlockExtendBox graphical programming expansion box is available, allowing programming of the robot dog using XRBlock graphical programming software based on Scratch 3.0. Suitable for beginner programmers in elementary and middle school.
LiDAR Mapping and Navigation
It not only supports Hector algorithm mapping, but also allows for secondary development and installation of other algorithms such as Gmaping and Cartograph for mapping, enabling path planning, point-to-point navigation, and multi-point navigation.
Multi-point navigation and dynamic obstacle avoidance
The TOF lidar can detect the surrounding environment in real time and dynamically avoid obstacles during navigation.
Upon detecting an obstacle, it replans the path using a local path planner.
Built-in IMU Attitude Sensor
The ROS expansion box is equipped with an IMU sensor, which can obtain attitude data through the IMU; it employs a closed-loop control algorithm to monitor the robot's attitude in real time and automatically keep the robot dog's body in a balanced state.
Equipped with a 3D depth camera,
the robot dog's head can be optionally equipped with an Orbbec series depth camera, which not only enables all the AI vision functions of an RGB camera, but also allows for the development of more depth vision functions such as depth image data processing, RTAB 3D visual mapping and navigation.
RTABSLAM 3D Vision Mapping and Navigation
The robot dog utilizes the RTABSLAM algorithm, fusing visual and radar data to construct a 3D color map. The robot can autonomously navigate and avoid obstacles within this map, supporting global relocalization and autonomous localization functions.
Depth Image Data Point Cloud Image
Through the corresponding depth camera, depth images, color images, point cloud images, and other data can be acquired.
Course List
1. Basic Courses
| Understanding the Robot Dog | Identify parts of the robot dog , Robot parameters |
| Remote Control Setup | Install mobile app , Connect mobile app with robot , Parameter adjustment |
| Hardware Setup | Install Raspberry Pi ROS expansion board and camera |
| Power | Charging |
2. Advanced Courses
| Basic Operations | Video streaming , Move forward , Move backward , Turn left , Turn right , Move left , Move right , Raise head , Tilt left , Tilt right , Combined movements , Forward kinematics control , Pan-tilt (gimbal) control , Wireless controller control , Keyboard control |
| Fun Operations | Function , Punch , Wave hand , Nod , Defense , Beckon , Throw , Pick up , Stretch , Urinate action , Jump , Play dead |
3. Quadruped Robot Principles and Architecture Learning
| System Architecture | Hardware architecture , Software architecture |
| Motion Theory | Quadruped robot gait , Center of mass dynamics , Forward / inverse kinematics algorithm introduction |
4. ROS SLAM Quadruped Robot Mapping and Navigation
| ROS Mapping & Navigation | Install virtual machine , Connect robot dog , Check network , Synchronize system time between robot and VM , LiDAR mapping , LiDAR navigation |
5. ROS SLAM Quadruped Robot Secondary Development
| Development Tool Preparation | Software preparation , Hardware preparation , Remote connection |
| Basic Secondary Development Operations | Code framework , End controller process , Modify code , Run code , Principle block diagram , TF tree relationship |
| Advanced SLAM Development | Depth camera usage , Robot self-balancing , RTABMAP visual SLAM mapping , KCF target detection and tracking , Face recognition , Edge detection , ArUco augmented reality , ROS and lower-level communication interface , Gesture recognition |
Cross-platform control
1. Voice interaction
Integrated with the voice module, it enables voice control and voice interaction functions.
Different actions and gameplay can be completed via voice commands. (Exclusive to AI Vision and Voice Control versions)
2. App Control
The Corgi robot dog supports control via a mobile phone and tablet app,
providing detailed instructions to help you quickly use all the robot dog's functions.
3. Gamepad Control
With a gamepad, similar control effects to those achieved with a mobile app can be achieved.
Built-in multiple fun actions
| Goodbye | Say hello | Jumping | Lucky Cat |
| Surrender | Play dead | Stretching | Peeing |
Secondary Development Stage
The robot dog has a dedicated serial port for further development and expansion. It provides APIs for almost all of the robot dog's functions, including basic movements, fun movements, forward/backward movement control, and head control. It can connect to external processors such as Arduino, STM32, Raspberry Pi, and Jetson Nano via the serial port to receive and process control commands sent from those devices. You can control the robot dog using familiar programming languages such as C and Python.
Application scenarios
Practical Training Suitable for practical training courses in fields such as the Internet of Things and Electronic Information. | High-End Toys As high-end intelligent toys, they are used for entertainment and as gifts. | Science Popularization Exhibitions Suitable for science museums, cience festival performances and experiences | Competitive Competitions Suitable for whitelisted events, provincial competitions, national competitions, and other science and technology competitions |
Product Parameters
| Product Name | Corgi Quadruped Dog – Keji Intelligent Quadruped Robot Dog | Operating Time | About 1 hour |
| Main Processor | Dual-core Xtensa LX6 CPU | SRAM | 520 KB |
| Main Frequency | 240 MHz | PSRAM | 4 MB |
| Core Controller | ESP32 | Interfaces | UART / SPI / I2C / PWM |
| Control Methods | iOS / Android App / Controller (AI Vision Version) / Voice | Servo Model | XR-S300 |
| Control Distance | 10 m | Camera | Wide-angle camera, 480P, 2MP |
| Degrees of Freedom | 13 DOF | Programming Languages | Python / Arduino C |
| Battery Capacity | 2200 mAh | Graphical Programming | XR Block Scratch |
Shipping list
1. Corgi intelligent quadruped robot dog controller / 2. User manual and charger / 3. Packaging box and warranty card
4. AI Vision Package (Optional): AI Vision Module + RGB Camera
5. Graphical Programming Package (Optional): Graphical Programming Module
6. ROS Advanced Package (G Optional): ROS Controller Module + RGB Camera
7. ROS Professional Package (Optional): ROS Main Module + Depth Camera
