Yahboom Muto S2 18DOF AI Vision Hexapod Robot for Raspberry Pi 5 & Jetson NANO, 2DOF PTZ FPV

Overview

The Yahboom Muto S2 is an 18DOF hexapod robot designed for Raspberry Pi 5 or NVIDIA Jetson NANO as the main controller. It is a desktop-level bionic hexapod robot with an aluminum alloy body, 18 degrees of freedom, and built-in inverse kinematics algorithm control for simulated gaits. With Python3 programming and OpenCV image processing, it supports AI visual interaction functions such as color recognition, tracking/following, face tracking, QR code recognition, and visual line patrol. Control methods include mobile phone APP, wireless handle control, and computer web page (Jupyter Lab) control, with real-time video transmission (FPV).

Key Features

• AI Vision Hexapod Robot: Inverse Kinematics Algorithm, Bionic Gait, 18DOF Joint, AI Visual Interaction.
• 18DOF motion joints: Uses 18 high-performance servos and aluminum alloy structural parts to connect three joints on each leg.
• 35KG smart serial bus servo system: 18PCS 35KG metal servos.
• 2D camera PTZ: 2DOF camera PTZ for vision applications.
• Real-time video transmission: Connect via local area network through the mobile phone APP to view HD video footage in real time.
• Posture & movement adjustment: Supports free adjustment of walking speed and robot body height (Robot Height Adjustment / Robot Speed Adjustment; walking speed adjustment: slow/fast).
• Teaching mode: Manually control single-leg movement of the host machine; another slave machine performs the same action.
• Raspberry Pi 5 support note: “MUTO RS is equipped with a regulated power supply expansion board” adapted to Raspberry Pi 5, providing a stable 5.1V/5A power supply; the 0.6A current limit increases USB port current output to 1.6A (to help avoid freezing/restarting).

For selection help and after-sales support, contact https://rcdrone.top/ or email support@rcdrone.top.

Specifications

Main Control Comparison (as provided)

Muto S2 robot provides two main controllers, Raspberry Pi 5 and Jetson NANO 4GB SUB, and the usage methods are basically the same, both using the Ubuntu system. Different main controls only slightly affect the smoothness of the system. The course materials, product functions, and control software provided are consistent.

Function List (Courses/Examples)

Camera PTZ

• 00. Color HSV value calibration
• 01. Color recognition
• 02. Color tracking
• 03. Color following
• 04. Color recognition action group
• 05. Face Detection
• 06. Face tracking
• 07. Greet people
• 08. QR code recognition
• 09. QR code instructions
• 10. Visual line following
• 11. Motion learning
• 12. Teach pendant synchronized action

Machine deep learning

• 01. KNN
• 02. TensorFlow basic tutorial
• 03. Basic use of PyTorch
• 04. Yolov5 detects objects in real time
• 05. Jetson-inference environment construction
• 06. Object detection and action
• 07. Body movement control robot
• 08. Gesture controlled robot

Jetson NANO course

• 1. About JetsonNano system
• 2. Network configuration and Jtop
• 3. Swap space increased
• 4. API usage of GPIO library
• 5. Hardware library configuration
• 6. Pin reading function
• 7. Pin level output control
• 8. Control LED
• 9. Jetson Nano communicates with external device serial ports
• 10. Jetson nano I2C communication

Remote control course

• 1. Close APP control process
• 2. Mobile APP remote control tutorial
• 3. USB wireless handle remote control

Robot basic course

• 1. Control buzzer
• 2. Control PWM servo
• 3. Control bus servo
• 4. Robot forward and backward
• 5. Robot move left and right
• 6. Robot rotate left and right
• 7. Control height
• 8. Control head
• 9. Action group performance
• 10. Read data
• 11. Host computer control
• 12. Camera driver

Raspberry Pi course

• 1. Build Python environment
• 2. Helloworld
• 3. Pin output high and low levels
• 4. Read pin high and low levels
• 5. Output PWM
• 6. Serial communication
• 7. I2C communication
• 8. Serial communication
• 9. I2C communication

Open Source CV course

• 1. Introduction to Open Source CV
• 2. Image reading and display
• 3. Picture writing
• 4. Image Quality
• 5. Pixel operations
• 6. Image zoom
• 7. Picture cut
• 8. Picture pan
• 9. Picture mirroring
• 10. Affine transformation
• 11. Picture rotation
• 12. Perspective transformation
• 13. Grayscale processing
• 14. Binary image
• 15. Edge green detection
• 16. Line segment drawing
• 17. Rectangular circle drawing
• 18. Text and picture drawing

Video

Manuals / Tutorials

Tutorial link (official): http://www.yahboom.net/study/Muto-S2

Details

An 18DOF bionic hexapod platform combines inverse-kinematics gait control with camera-based AI interaction.

Built to run on Raspberry Pi 5 or NVIDIA Jetson Nano for robotics learning, vision projects, and gait experiments.

Key capabilities include OpenCV-based vision functions, Python programming, WiFi control, and FPV video streaming.

Choose Raspberry Pi 5 or Jetson Nano as the main controller, with a power expansion board option designed for Pi 5 stability.

A structured function list and learning path covers PTZ vision, deep learning basics, and step-by-step courses for both controllers.

Eighteen high-torque servos drive three joints per leg for stable, articulated hexapod movement.

Built-in inverse kinematics helps coordinate foot trajectories for smoother, more stable simulated gaits.

Real-time FPV lets you drive and monitor the robot from a phone app over a local network connection.

Save action groups in the app to trigger preset moves or fine-tune individual joints for custom poses.

Interactive motions support playful behaviors such as gesture-style greetings and movement routines.

Reactive movement demos highlight balance adjustments during close-range obstacles and dynamic motion.

Vision-based behaviors can support tracking and following for hands-on AI interaction experiments.

Quick commands enable common action states like curling up and moving forward.

Tune body height and walking speed to match different surfaces, demos, and indoor desktop testing.

Teaching mode allows one robot’s leg motion to be guided manually and mirrored by a second unit.

Camera-based AI features include color tracking, face tracking, and QR code recognition using OpenCV workflows.

Explore deeper AI demos such as object detection, skeleton-based pose estimation, and gesture control routines.

Program behaviors in Python and iterate quickly from a laptop for motion control, vision processing, and automation.

The MUTO S2 supports a cross-platform iOS/Android app for remote control, robot calibration, performance modes, and data monitoring.

Muto S2 supports PC control through a JupyterLab web page and 2.4G/USB wireless gamepad control for robot movement.

The Yahboom Muto S2 course catalog outlines guided modules for assembly, motion control, OpenCV vision features, and AI experiments.

Yahboom Muto S2 includes organized learning resources with AI vision and deep learning course files for step-by-step setup and coding.

The Muto S2 hexapod layout includes an OLED screen, 2DOF camera PTZ module, USB hub expansion board, and a large lithium battery mount.

The Muto S2 uses an intelligent serial bus servo with a 0–270° joint range and listed specs including 35kgf·cm torque and 6.0–8.4V operation.

The USB HD 1080P camera PTZ uses a 2DOF servo for pan/tilt control and connects via USB 2.0 with up to 80–120° field of view.

The 7.4V 9900mAh lithium battery pack uses a DC 4.0×1.7 charging interface and about 15cm leads for easy connection.

Yahboom Muto S2 comes with a dimensioned layout showing top and front views with measurements in millimeters for planning placement and clearance.

MUTO S2 is available with a Jetson Nano 4GB USB or Raspberry Pi 5 main control board, both supporting Python, 18DOF joints, and about 3.7 hours of battery life.

The Muto S2 hexapod kit includes the robot chassis, PTZ camera module, OLED components, USB hub expansion board, battery pack, charger, cables, and basic tools, with Raspberry Pi or Jetson Nano accessories listed as add‑