Yahboom DOFBOT AI Large Model Vision Robotic Arm Kit for Raspberry Pi 5, ROS2 Humble, Python3, USB Camera

Overview

The Yahboom DOFBOT AI Large Model Vision Robotic Arm is a desktop robotic arm kit designed for ROS2 development and AI vision learning on Raspberry Pi 5. It combines a 6-DOF robotic arm with a USB camera for vision-based grasping, tracking, and interactive AI applications, and supports programming with Python3, OpenCV, and ROS2 Humble (Docker + ROS2 Humble is specified).

It supports cross-platform control including mobile APP (iOS/Android), PC host computer control, controller/handle control, and JupyterLab web-based programming.

Video

Key Features

• 6 degrees of freedom with camera + robotic arm integration ("6-DOF Camera and Robotic Arm 2 in 1").
• AI large model functions (as described): multimodal fusion, natural speech dialogue, text semantic understanding, visual scene understanding, and a scalable RAG knowledge base.
• AI vision interactions: color recognition tracking, object tracking, gesture-related interactions, grasping/sorting demos, and more.
• ROS2 development stack: ROS2 Humble (noted as “ROS2 Humble” and “Docker+ROS2 Humble”), with MoveIt motion planning and RViz robot simulation listed.
• Algorithm frameworks listed: inverse kinematics algorithm, YOLOv11, OpenCV, MediaPipe; TensorRT acceleration is mentioned for millisecond-level real-time inference.
• User-friendly hardware design: OLED display for IP address and Raspberry Pi CPU information; chassis with suction cups for stability; 12V 5A adapter for continuous power.
• Extensibility: expansion board stated as compatible with Jetson NANO, Raspberry Pi, Arduino, and Micro:bit; reserved interfaces include 6 bus-servo + 6 PWM-servo, wireless controller receiver, WiFi/Bluetooth module, I2C, and ultrasonic module interfaces.

Specifications

Raspberry Pi 5 (listed in product materials)

Raspberry Pi 5 vs Raspberry Pi 4B (comparison table text)

ROS configuration differences (Standard Version vs Superior Version)

What's Included

• Assembled robotic arm
• Matching color-printed map
• 4 different color blocks
• PS2 gamepad
• TF card with image system
• Yahboom special cooling HAT
• 12V 5A power adapter

Note: A demonstration note indicates “The trash can is a display prop and is not in shipping list.”

Applications

• AI vision functions and gameplay examples listed: gesture recognition, color recognition, visual positioning, garbage sorting, catch game, face tracking, and building blocks stacking.
• AI vision interactive functions described: color recognition tracking; catch game (map recognition area); grab color blocks; color interaction; garbage sorting; color block stacking.
• Multimodal large model application examples listed: video analysis; long-command motion control; intelligent handling; 3D space sorting.
• MoveIt simulation control and trajectory planning (with collision detection and spatial gripping) are listed for virtual environment verification.
• Deep learning model training is supported; a note indicates model training needs to be done by the user.

Manuals

• Yahboom DOFBOT-Pi tutorial

For pre-sales questions, kit version selection, or technical support, contact support@rcdrone.top or visit https://rcdrone.top/.

Details

DOFBOT combines a 6‑DOF robotic arm and USB camera for Raspberry Pi 5 vision projects, grasping demos, and ROS2 development.

Multiple control options and built‑in algorithm demos help you move from basic servo control to vision interaction and AI workflows.

Designed around Raspberry Pi 5 computing power for ROS2 Humble, Python3, and real‑time vision inference workflows.

Quasi‑3D spatial demos connect camera perception with arm motion for embodied intelligence applications.

Multimodal demos cover long‑command motion control, intelligent handling, and sorting tasks driven by vision and prompts.

Use built‑in tasks like visual tracking, color sorting, and intent inference to prototype your own interactive robotics logic.

OpenCV-style vision interactions include color recognition, tracking, and grasping routines for hands‑on learning.

MediaPipe and training examples pair with inverse kinematics so vision results translate into accurate joint movement.

Develop and validate motion with MoveIt and RViz before running trajectories on the physical arm.

Cross‑platform control options include mobile app, PC host control, and a USB wired remote, alongside ROS2 Humble support.

The 6‑DOF joint layout (J1–J6) supports flexible pick‑and‑place paths and camera‑guided interaction.

Expandable hardware includes an arm control board, bus servos, optional voice module and speaker, plus Raspberry Pi 5 mounting.

The DOFBOT robotic arm expansion board uses a labeled port layout and serial bus servo support to help simplify wiring and control setup.

The USB camera mounts on the robotic arm and includes a 110° field of view with 480p resolution at 30fps, along with an AI voice module (superior version) featuring a board, speaker, and wiring.

Yahboom’s DOFBOT-Pi tutorial repository provides a tutorial link and up to 200+ structured courses for learning the DOFBOT robotic arm.

The DOFBOT course syllabus outlines step-by-step lessons covering Raspberry Pi setup, ROS2 Humble, Python programming, and camera-based vision projects.

Open-source code folders and step-by-step video tutorials help you set up and program the DOFBOT robotic arm in Python and ROS2.

DOFBOT includes a detailed dimensions layout and key specifications to help plan mounting space and system integration.

The DOFBOT kit includes the assembled robotic arm, USB camera, expansion board, OLED display, cables, power adapter, tools, and manual, with optional accessories listed separately.

The Raspberry Pi 5 accessory set includes a 64GB TF card with reader, an active cooler, a power expansion board, an I2C communication cable, and a dual Type‑C power adapter board.