JetHexa Standard Kit

NVIDIA Jetson Nano B01 Controller – Runs ROS, TensorFlow, PyTorch, Keras, and TensorRT; enables YOLO model training and AI Boost NPU-level deep learning at the edge.
Monocular 2DOF HD Camera – Rotatable camera head supports real-time FPV, KCF target tracking, color recognition, line following, and April Tag detection.
Upgraded Inverse Kinematics Algorithm – Supports tripod gait and ripple gait switching; adjustable pitch angle, roll angle, direction, speed, height, and stride for ultimate terrain control.
18 Intelligent Serial Bus Servos – High-torque anodized aluminum alloy frame with 35KG servos for powerful, precise hexapod locomotion.
AI Vision Applications – Color tracking, autonomous line following, KCF target tracking, gesture recognition, MediaPipe human body recognition, and somatosensory interaction.
Multi-Platform Control – WonderAi app (iOS/Android), wireless gamepad, ROS framework with Gazebo simulation, and keyboard; 150+ dual-language tutorials included.
Open-Source Ecosystem – Full ROS source code, Linux, OpenCV, and motion-control tutorials from basics to advanced; ideal for STEAM education and research.

₹88,983 (Excl. GST)

Trusted dealerDirect manufacturer partnership
Pan-India deliveryInsured shipping, all states
Post-sale supportTraining + service included

Overview

What makes it work

JetHexa Standard Kit — ROS Hexapod Robot Powered by NVIDIA Jetson Nano | A+ Content | xBoom India

Hexapod Robot · ROS · NVIDIA Jetson Nano B01 · xBoom India

JetHexa Standard Kit
ROS Hexapod Robot

The World's Most Advanced AI Hexapod — Powered by NVIDIA Jetson Nano B01

JetHexa is an open-source hexapod robot based on the Robot Operating System (ROS), powered by NVIDIA Jetson Nano B01. It is armed with high-performance hardware including 18 intelligent serial bus servos, Lidar, and a monocular HD camera, capable of robot motion control, SLAM mapping and navigation, tracking, obstacle avoidance, and human feature recognition. With a novel inverse kinematics algorithm supporting tripod and ripple gaits and highly configurable body posture, JetHexa delivers an ultimate user experience for ROS hexapod robot development.

NVIDIA Jetson Nano B01 ROS Framework 18 DOF · 35KG Servos 2DOF Monocular HD Camera YOLO · MediaPipe · TensorRT Gazebo Simulation 150+ Bilingual Tutorials

Technical Specifications

JetHexa Standard Kit — Full Specifications

Specification	Details
Controller	NVIDIA Jetson Nano B01 — 4GB RAM, 128-core Maxwell GPU
Framework	Robot Operating System (ROS) — Ubuntu 18.04, ROS Melodic/Noetic; Gazebo simulation; RViz
Deep Learning	TensorFlow, PyTorch, Caffe/Caffe2, Keras, MXNet, TensorRT acceleration, YOLO, MediaPipe
Camera	Monocular HD Camera — 2DOF rotatable (up/down/left/right), real-time FPV streaming
Servos	18 Intelligent Serial Bus Servos — 35KG torque, over-temperature protection, anti-blocking, 240° rotation
DOF	18 Degrees of Freedom — 3 joints per leg × 6 legs
Chassis	Hard anodized aluminum alloy frame — low weight, high strength
Gaits	Tripod Gait, Ripple Gait, Moonwalk — constant speed via IK algorithm
Motion Control	Inverse Kinematics — real-time pitch angle, roll angle, direction, speed, height, stride adjustment
Self-Balancing	Built-in IMU sensor detects body posture in real time; adjusts joints to balance on complex terrain
AI Vision	KCF target tracking, color/tag recognition, line following, April Tag, gesture recognition, face detection, somatosensory control
Control Methods	WonderAi App (iOS/Android), wireless gamepad, ROS keyboard
Battery	9–12.6V LiPo battery
OS	Ubuntu 18.04 + ROS Melodic/Noetic
Tutorials	150+ bilingual tutorials: Linux, ROS, OpenCV, motion control, AI deep learning, depth camera, Lidar, SLAM
Variant Note	Standard Kit = Monocular HD Camera only; Advanced Kit adds 3D Depth Camera + EAI G4 Lidar for SLAM

Product Introduction

Built for ROS Hexapod Robot Research & AI Development

JetHexa powered by NVIDIA Jetson Nano ROS

AI at the Edge with Deep Learning & ROS

JetHexa is a hexapod robot powered by NVIDIA Jetson Nano B01 and built on the Robot Operating System (ROS). It leverages mainstream deep learning frameworks including TensorFlow, PyTorch, Caffe, Keras, MXNet, and TensorRT, incorporates MediaPipe development, and enables YOLO model training. This combination delivers a diverse range of AI applications including motion control, object recognition, KCF target tracking, line following, 3D face detection, gesture recognition, and somatosensory control.

NVIDIA Jetson Nano runs these frameworks natively, providing powerful computing power for massive AI projects. JetHexa can achieve image recognition, object detection and positioning, pose estimation, semantics segmentation, and other advanced functions — making it a true AI edge-computing platform in a hexapod form factor.

JetHexa inverse kinematics gait algorithm

Inverse Kinematics Algorithm

Tripod & Ripple Gait with Self-Balancing

JetHexa adopts a novel inverse kinematics algorithm, supporting both tripod and ripple gaits. It can perform "moonwalking" at a fixed speed and height using IK. The body posture — including pitch angle, roll angle, center of gravity, direction, speed, height, and stride — is fully adjustable in real time, giving complete control over the robot's movements across complex terrain.

The built-in IMU sensor detects body posture in real time and arranges for the robot to adjust its joints to balance the body. Through IK, JetHexa can maintain stability during SLAM mapping and execute highly configurable locomotion patterns. The self-balancing function allows JetHexa to conquer complex terrains with ease, making it ideal for outdoor field robotics research.

Tripod Gait — fast, stable locomotion on level ground
Ripple Gait — enhanced balance on irregular terrain
Stepless adjustment of linear velocity, angular velocity, stance, height and stride
IMU-based real-time posture correction for self-balancing

Robot Control Across Platforms

Multi-Method Control & Open-Source Python

JetHexa provides multiple control methods including the WonderAi app (compatible with iOS and Android), a wireless gamepad, the Robot Operating System (ROS), and keyboard control. Through the WonderAi app, you can control its movement in real time and access the live camera FPV feed. You can set the coordinate of each leg's endpoint and JetHexa automatically calculates the servo angles for instant action debugging.

JetHexa employs the ROS framework and supports Gazebo simulation, which lets you validate algorithms in a simulated environment without physical hardware — reducing experimental requirements and improving efficiency. All source code is open-source Python with detailed annotations for easy self-study. 150+ bilingual tutorials cover the full spectrum from Linux and ROS to deep learning and SLAM.

JetHexa MediaPipe gesture control AI vision

AI Vision & MediaPipe

Color Tracking, Gesture Control & Human Recognition

By incorporating AI, JetHexa can implement KCF target tracking, vision line following, color/tag recognition and tracking, and April Tag detection. It is skilled in color recognition and tracking — the robot can be set to execute different actions according to the colors it detects. Relying on the KCF filtering algorithm, JetHexa can track a selected target accurately while walking.

Using the MediaPipe development framework, JetHexa accomplishes a wide range of human-machine interaction functions: human body recognition, fingertip recognition, face detection, 3D detection, and gesture-based control. Adopting GoogLeNet, YOLO, mtcnn, and other neural networks, JetHexa masters deep learning to train models and can recognize targets quickly to implement complex AI projects including waste sorting, mask identification, and emotion recognition.

JetHexa Standard Kit

Spec Infographic

JetHexa Standard Kit — Key Specifications at a Glance

JetHexa Standard Kit — Powered by NVIDIA Jetson Nano B01

Open-Source ROS Hexapod · 18 DOF · AI Vision · SLAM · 150+ Tutorials

Total DOF

35KG Servos

Legs

3 joints each

ROS

Framework

Gazebo + RViz

2DOF

HD Camera

360° FPV

150+

Tutorials

Bilingual

YOLO

AI Model

TensorRT

IMU

Self-Balance

Real-time

4GB

Jetson RAM

128-core GPU

Product Videos

JetHexa Standard Kit — Demo & Tutorial Videos

JetHexa — Official Demo: ROS, SLAM, AI Vision & Gait Control

Frequently Asked Questions

JetHexa Standard Kit — FAQ

What is the difference between JetHexa Standard and Advanced Kit?

The Standard Kit includes a 2DOF monocular HD camera for AI vision tasks such as color tracking, line following, April Tag recognition, and KCF target tracking. The Advanced Kit additionally includes a 3D depth camera and EAI G4 Lidar, which unlock full SLAM mapping, RTAB 3D navigation, TEB path planning, point cloud imaging, depth obstacle avoidance, and supporting diverse algorithms including Cartographer, Hector, Karto, and Gmapping. Choose Standard for learning AI vision fundamentals; choose Advanced for full autonomous navigation research.

Can JetHexa walk on uneven terrain and self-balance?

Yes. JetHexa uses a proprietary inverse kinematics algorithm with the built-in IMU sensor to detect body posture in real time and adjust its joints accordingly. Body pitch angle, roll angle, center of gravity, height, and stride are all adjustable. This enables stable navigation on uneven surfaces and complex terrain using both tripod gait (level surfaces) and ripple gait (irregular terrain). JetHexa can also perform "moonwalking" with fixed speed and height via IK.

Does JetHexa support Gazebo simulation?

Yes. JetHexa fully supports Gazebo simulation within the ROS framework. You can verify kinematic algorithms in a simulated environment, observe the robot's end-effector and center-of-gravity trajectory in RViz, and validate path planning without physical hardware — reducing experimental risk and significantly improving development efficiency.

What deep learning frameworks and AI functions does JetHexa support?

JetHexa supports TensorFlow, PyTorch, Caffe/Caffe2, Keras, MXNet, and TensorRT acceleration on the NVIDIA Jetson Nano. AI functions include YOLO model training, KCF target tracking, color/tag recognition, line following, MediaPipe-based human body tracking, hand detection, posture detection, 3D face detection, gesture recognition, and somatosensory control. Advanced deep learning applications include waste sorting, mask identification, and emotion recognition.

What control methods are available?

JetHexa can be controlled via the WonderAi mobile app (iOS and Android) for live FPV and motion control, a wireless gamepad for manual joystick control, ROS keyboard commands for programmatic control, and via custom Python scripts through the open API. Leg endpoint coordinates can be set through the app — JetHexa automatically calculates servo angles for instant action debugging. Gazebo simulation can also be used for virtual control and algorithm testing.

Packing List

JetHexa Standard Kit — What's in the Box