Home
/
Robots
/
Surena 4

Surena 4

Surena 4 by University of Tehran is a 1.7m humanoid robot with 43 DOF, advanced AI, walking, object manipulation, and speech interaction for research and education.
Available
Iran
Humanoid
2019
Specifications
Dev Doc
Website
Made by
University of Tehran
Software Type
Closed Source
Software Package
Robot Operating System (ROS) for control and monitoring. FPGA-based real-time control system. AI modules for speech, vision, and action imitation. Whole-body motion planning and dynamic balance control.
Actuators
Custom-designed actuators enabling 43 degrees of freedom with precise torque control, optimized for lightweight and efficient operation.
Compiute
FPGA board running control loops at 200 Hz, combined with onboard processors for AI tasks, supporting real-time perception and decision-making.
Sensors
Cameras for face and object detection Microphones for speech recognition Force sensors for manipulation control Inertial measurement units (IMUs) for balance and locomotion
Max Op. time
mins

Robot Brief

Surena 4 is the latest generation in the Surena series of humanoid robots developed by the University of Tehran’s Center for Advanced Systems and Technologies (CAST). Unveiled in late 2019, Surena 4 represents a major technological advancement over its predecessors, featuring 43 degrees of freedom (DoF) and enhanced dexterity, especially in the hands, allowing it to grip various objects with precision. The robot stands 1.7 meters tall and weighs 68 kilograms, significantly lighter and more compact than Surena 3 due to improved structural design, topology optimization, custom actuators, and 3D printed covers. Surena 4 incorporates advanced AI capabilities including speech recognition and generation, face and object detection, and activity recognition. It can walk continuously at speeds up to 0.7 km/h, climb stairs, and perform complex upper and lower body motions. The robot integrates FPGA-based control with a 200 Hz control loop and uses the Robot Operating System (ROS) for real-time algorithm execution and monitoring. Surena 4 serves as a versatile research platform for bipedal locomotion, manipulation, human-robot interaction, and AI, symbolizing technological progress toward peaceful and humanitarian applications.

Use Cases

Surena 4 performs bipedal locomotion including walking on uneven terrain and climbing stairs, manipulates objects with force control (such as gripping a pen to write), and interacts with humans using speech recognition and generation. It can detect and track faces and objects, imitate actions, and engage in real-time interaction, making it suitable for research, education, and public demonstrations.

Industries

  • Research & Development: Platform for studying bipedal locomotion, manipulation, and AI integration.
  • Education: Inspires engineering students and promotes robotics careers.
  • Entertainment: Demonstrated at events to showcase humanoid robotics capabilities.
  • Human-Robot Interaction: Development of speech and gesture-based communication.

‍

Specifications

Length
mm
Width
mm
Height (ResT)
mm
Height (Stand)
-
1700
mm
Height (Min)
mm
Height (Max)
mm
Weight (With Batt.)
-
kg
Weight (NO Batt.)
-
68
kg
Max Step Height
-
mm
Max Slope
+/-
-
°
Op. Temp (min)
-
°C
Op. Temp (Max)
-
°C
Ingress Rating
-
No items found.

Intro

Surena 4 is a full-sized humanoid robot standing 1.7 meters tall and weighing 68 kg, designed with 43 degrees of freedom for enhanced mobility and dexterity. It features improved hand design for gripping diverse objects and advanced structural optimization for reduced weight. The robot’s control system runs on an FPGA board at 200 Hz, enabling precise real-time control. It uses ROS for software modularity and simultaneous execution of multiple algorithms. Surena 4 is equipped with AI modules for speech-to-text, text-to-speech, face and object detection, and action imitation. Its locomotion system supports continuous walking at 0.7 km/h and stair climbing. The robot’s design emphasizes human-robot interaction and serves as a research platform for advancing humanoid robotics.

Connectivity

  • ROS-based communication protocols
  • Likely Wi-Fi and Ethernet for network connectivity (standard in ROS systems)
  • USB and other typical robot interface ports

‍

Capabilities

  • 43 degrees of freedom enabling complex full-body motions
  • Walking on flat and uneven surfaces, including stair climbing
  • Force-controlled object manipulation with dexterous hands
  • Speech recognition and generation for interactive communication
  • Face detection and tracking, object detection, and activity recognition
  • Real-time control with 200 Hz loop frequency via FPGA
  • ROS-based software framework for modular development and monitoring

‍