Description
Tutorial Link
ROSMASTER M3 Pro is a highly integrated embodied intelligent robot platform developed by Yahboom specifically for ROS education, scientific research experiments, and AI application teaching. It utilizes Mecanum wheels and pendulum suspension chassis for omnidirectional movement. Developed based on the ROS2 Humble system, equipped with a 6DOF robotic arm and a binocular structured light depth camera to perform tasks such as visual recognition, 3D grasping, and precise handling. With dual TOF LiDAR, it enables stable and reliable SLAM mapping and autonomous navigation, as well as LiDAR obstacle avoidance and path planning. Unlike traditional ROS robots, the ROSMASTER M3 Pro deeply integrates cutting-edge AI large-scale model technology. Built-in speech recognition and natural language understanding modules, can realize voice command control, multimodal interaction with text/image/voice, task planning and execution, and dynamic environment perception. Whether used in AI courses, robotics algorithm teaching, or university research projects, the ROSMASTER M3 Pro provides a stable, powerful, and easily scalable experimental platform, making it an ideal choice for AI and robotics education.
【Top-level Hardware Configuration】
- Raspberry Pi 5, Jetson NANO B01, Jetson ORIN NANO SUPER, Jetson ORIN NX SUPER development boards, four main control boards for choice.
- An aluminum alloy chassis, 80mm Mecanum wheels, and a rear-wheel pendulum suspension structure allow for easy navigation across diverse terrains.
- Combining a depth camera and a 6DOF robotic arm, it enables 3D grasping, precise handling, and MoveIt simulation.
- Two TOF ranging LiDAR on the front left and rear right provide 360° scanning, enhancing mapping and navigation accuracy.
【Fully Integrates Intelligent Perception and Semantic Understanding】
- Built-in speech recognition and natural language processing, combined with speakers, enable easy voice commands and question-and-answer interactions.
- Integrating multimodal interaction capabilities, including text, image, and voice, it can adjust actions in real time based on environmental changes, supporting free conversation interruptions and dynamic feedback reasoning.
- Integrating a large model and an extensible RAG knowledge system enhances task awareness and complex problem-solving capabilities.
【A full-stack robotics platform for teaching and research】
- Based on the ROS2 Humble system, compatible with a variety of robotics algorithms and AI course content.
- Integrated hardware and software, along with comprehensive documentation and teaching resources, facilitate efficient progress from introductory learning to scientific research experiments.
- From AI and SLAM to visual recognition and robot control, it's widely applicable to university teaching, scientific research experiments, and robotics competition platform development.
Shipping List:
Shipping & Delivery
Payment & Security
Payment
PayPal
The main Payment method is Paypal. If you have paypal, that is great. You can simply place order and check out with few steps. If you don't have Paypal account, you can register it and pay by Credit or Debit Card.
Pay with Paypal
Pay with Credit or Debit Card
Your payment information is processed securely. We do not store credit card details nor have access to your credit card information.
Customer Reviews
Based on 3 reviews
67%
(2)
0%
(0)
0%
(0)
0%
(0)
33%
(1)
C
I am an expert in ROS, and I initially considered this robot a good choice for my research. However, all sensor feedback—such as wheel speed data, arm joint positions, LiDAR data, and IMU measurements—is processed internally through their STM32 bridge. The user has no access to these variables, despite the official specifications clearly stating that the robot provides this sensor data.
The STM32 firmware is not open-sourced, which makes it impossible to access even basic robot telemetry. The robot only exposes the limited data that the manufacturer chooses to provide, which is essentially useless for serious research purposes. In its current state, the robot is basically an expensive toy.
Technical support has been extremely poor. They have not provided any viable solution to the sensor access issues, and there is no clear justification for keeping the STM32 firmware closed. I have identified a competing robot platform that provides full access to all required sensor data for research use. As a result, I will be purchasing our team’s nine units from another supplier.
Poor service.
D
ROSMASTER M3 Pro ROS2 Robot with Multimodal AI Large Model, AI Voice Interaction, Vision Recognition, Tracking & Sorting
D
Powered by ROS2 and unlike conventional systems,it interprets my commands by inferring intent and striving to comprehend and execute them—a significant leap forward in robotics. it occasionally interprets my intentions differently, but that's part of the fun during debugging and research. Equipped with omnidirectional Mecanum wheels and dual radar sensors, its responsiveness is enhanced, while the dual-radar design efficiently conserves port resources.
QUESTIONS & ANSWERS
-
If I start with RPi 5-16GB, can I upgrade to other board such as ORIN without much reconfiguration?
Hello, yes you can do it so and you need to purchase a corresponding control board accessory kit. Please contact market@yahboom.com
-
Can I use ChatGPT or Gemini API call with ROSMASTER M3 Pro?
Hello,
This is completely different from the method we use.
If you want to use it this way, you'll have to develop it yourself. We provide open-source documentation.
if you need more info please email to support@yahboom.com
thanks for your understanding.
have a nice day!
-
I understand that this product uses online large model APIs, not an offline LLM on device. Can you confirm if ROSMASTER can uses hosted APIs of LLM-MLLMs hosted by me on my GPU servers?
Sorry it can not use hosted APIs of LLM-MLLMs hosted by your GPU servers
-
does it has ondevice LLM , MMLLMs?
It uses an online large model API, not an offline LLM. For more reply please email to support@yahboom.com.
-
Just curious is it battery powered? I am not seeing any information about this at all.
hi, yes it is battery powered by 9600mAh battery pack.
-
Could I communicate with this robot remotely from across the internet without additional development?
hi, yes you can do so.
-
Hello, How can I change the language of the robot to english when using voice control? It returns chinease voice and english text.
hi,
thanks for your waiting.
for the tech support please email to support@yahboom.com
and we will help u to change the language of the robot step by step.
please do not worry.
have a nice time!
Yahboom Team
-
What is the biggest hard drive i can purchase for the Orin Nx super 16 GB version?
Hello,
We're sorry we can't make this recommendation. We haven't tested SSDs with this large capacity.Best wishesYahboom Team