SamuRoid uses AI models to understand voice commands and respond smartly, like dancing when users ask for fun activities.
A small humanoid robot from China is pushing the boundaries of affordable embodied AI.
SamuRoid, developed by Shenzhen-based XiaoR Geek Technology, is a 22-DOF bionic platform powered by a Raspberry Pi 4 Model B, designed for researchers, educators, and developers.
Unlike earlier small robots that mainly handled simple movements and basic camera tasks, SamuRoid is much smarter. It uses the open-source Robot Operating System (ROS) and advanced AI models to see, hear, and respond—enabling it to interact more naturally through vision, voice, and intelligent behavior.
While priced higher than its predecessors, it offers tighter hardware-software integration and more advanced capabilities, positioning it as a powerful tool for next-generation robotics experimentation and human-like machine interaction.
Smart humanoid platform
SamuRoid is built around the Raspberry Pi 4 Model B, offering 4GB or 8GB RAM options for flexible performance.
The humanoid features 22 high-torque XRS-series servos, delivering strong and precise movement across its body—covering the head, shoulders, arms, hands, legs, and feet. A compact 0.96-inch OLED display provides real-time updates such as IP address, battery level, and operating status, reports Hackaday.
For perception and interaction, SamuRoid includes a 1080p wide-angle camera mounted on a 2-DOF gimbal, enabling wide-field-of-view tracking with pan-tilt motion. A built-in USB microphone supports voice commands, while connectivity is handled through dual-band Wi-Fi 5 and Bluetooth 5.0. The robot also integrates an MPU6050 6-axis IMU for balance and stability, along with an expansion board exposing all GPIO pins for adding sensors and modules.
Housed in a durable aluminum alloy chassis, the robot uses high-voltage servos with metal gears and thermal protection. Its dual-hip yaw design improves agility and natural walking, while mechanical grippers with overload protection enable safe object handling. Powered by a 12V battery, SamuRoid runs for about one hour, reports CNX Software.
Multimodal robot rise
SamuRoid runs on a software stack built around ROS, giving developers a flexible and familiar robotics environment. According to its makers, its open-source codebase supports both C++ and Python, making it accessible for a wide range of users.
For movement, the robot uses a combination of inverse kinematics and the Linear Inverted Pendulum Model, helping it stay balanced while walking or changing direction. It allows smoother, more stable motion during dynamic tasks.
The robot’s vision system is powered by OpenCV, enabling features like face recognition, color tracking, and QR code detection. These capabilities allow SamuRoid to better understand and interact with its surroundings in real time.
One of its standout features is multimodal AI integration. By connecting to advanced language models such as DeepSeek and Doubao, the robot can understand natural language commands instead of relying on fixed instructions. For example, if a user says they are tired and want something fun, the system interprets the intent and triggers a suitable action—like initiating a dance routine—through ROS, while providing voice feedback.
SamuRoid measures about 390 mm in height, weighs around 5 pounds (2.3 kilograms), and runs on a 12V 3000mAh battery. It supports dual-band Wi-Fi, Bluetooth 5.0, and even a PS2 wireless controller for manual operation.
The robot is available in multiple versions. The Professional Edition is priced at around $1,565, while higher-end Developer and Flagship editions include additional accessories but are currently out of stock. A more affordable base package is also offered directly through XiaoR GEEK’s online portal.
Get the latest in engineering, tech, space & science - delivered daily to your inbox.
Jijo is an automotive and business journalist based in India. Armed with a BA in History (Honors) from St. Stephen's College, Delhi University, and a PG diploma in Journalism from the Indian Institute of Mass Communication, Delhi, he has worked for news agencies, national newspapers, and automotive magazines. In his spare time, he likes to go off-roading, engage in political discourse, travel, and teach languages.