Skip to main content

MuJoCo Humanoid

by Google DeepMind

MuJoCo Humanoid

A 21 degree-of-freedom bipedal humanoid model for advanced locomotion research. Features full-body articulation including torso, hips, knees, ankles, shoulders, and elbows. Ideal for studying walking, running, balance control, and complex whole-body coordination tasks. This high-fidelity physics model enables realistic biomechanics simulation and is widely used in reinforcement learning research for bipedal locomotion.

Specifications

Manufacturer
Google DeepMind
Type
Humanoid Robot
Environment
Simulated
Category
MuJoCo Physics Simulator
Controller
MuJoCo
Sensors
Joint position
Joint velocity
Torso orientation
Center of mass
Ground contact
Actuators
Position servo
Velocity servo
Torque control
Connectivity
ZMQ
Local

Getting started

MuJoCo Humanoid Integration Guide

Overview

The MuJoCo Humanoid is a 21-DOF bipedal robot model designed for advanced locomotion research. This integration allows FEAGI's neural circuits to control the humanoid's movements, learning to walk, balance, and perform complex whole-body coordination tasks.

Model Specifications

  • Degrees of Freedom: 21
  • Actuators: 21 (torso, hips, knees, ankles, shoulders, elbows)
  • Body Segments: Torso, pelvis, two legs (thighs, shins, feet), two arms (upper arms, forearms, hands)
  • Control Frequency: 50 Hz
  • Simulation Timestep: 0.002s

Sensory Capabilities

Joint Sensors

  • Joint Position: Current angle of each actuated joint (21 values)
  • Joint Velocity: Angular velocity of each joint (21 values)

Body State Sensors

  • Torso Orientation: Quaternion representing torso rotation in 3D space
  • Center of Mass: XYZ position of the robot's center of mass
  • Ground Contact: Binary sensors for each foot indicating ground contact

Motor Capabilities

Actuator Types

  1. Position Control: Direct joint angle targets
  2. Velocity Control: Desired angular velocity for each joint
  3. Torque Control: Direct torque commands (advanced)

Joint Ranges

  • Hip joints: ±90°
  • Knee joints: 0° to 150°
  • Ankle joints: ±45°
  • Shoulder joints: ±90°
  • Elbow joints: 0° to 150°

FEAGI Integration

Cortical Mapping

Input (Sensory):
  • i_joint_position: 21 neurons (one per joint angle)
  • i_joint_velocity: 21 neurons (angular velocities)
  • i_orientation: 4 neurons (quaternion)
  • i_com_position: 3 neurons (XYZ)
  • i_ground_contact: 2 neurons (left foot, right foot)
Output (Motor):
  • o_joint_control: 21 neurons (target positions or velocities)

Neural Circuit Design Tips

  1. Balance Control: Use torso orientation and COM sensors to maintain upright posture
  2. Gait Generation: Create CPG (Central Pattern Generator) circuits for rhythmic leg movements
  3. Coordination: Hip-knee-ankle coordination is critical for stable walking
  4. Ground Contact Feedback: Use foot contact sensors to trigger stance/swing phase transitions

Controller Setup

Prerequisites

  • MuJoCo physics engine (included with controller)
  • Python 3.8+
  • FEAGI Python SDK

Installation

The generic MuJoCo controller handles all models. When launching the humanoid:

python controller.py --model_xml humanoid.xml --agent_id mujoco_humanoid_01

Controller Parameters

  • --ip: FEAGI IP address (default: 127.0.0.1)
  • --model_xml: Model file (humanoid.xml)
  • --agent_id: Unique agent identifier
  • --cortical_input: Input cortical area name (default: i_vision)
  • --cortical_output: Output cortical area name (default: o_motor)

Beginner

  1. Stand Upright: Learn to maintain balance without falling
  2. Shift Weight: Transfer weight between feet

Intermediate

  1. Step in Place: Lift one foot while maintaining balance
  2. Forward Steps: Take coordinated steps forward
  3. Turn in Place: Rotate the body while standing

Advanced

  1. Continuous Walking: Maintain stable forward locomotion
  2. Obstacle Navigation: Walk around or over obstacles
  3. Terrain Adaptation: Adjust gait for uneven surfaces

Troubleshooting

Model Falls Immediately

  • Check that torso orientation sensors are connected
  • Verify ground contact sensors are providing feedback
  • Ensure initial joint angles are within safe ranges

Jerky or Unstable Movement

  • Reduce motor command magnitudes
  • Implement smoothing in motor output layer
  • Check simulation timestep and control frequency match

Controller Not Connecting

  • Verify FEAGI is running and accessible
  • Check firewall settings for ZMQ ports (5555, 30005, 30001)
  • Review controller logs for connection errors

Resources

Support

For integration questions or issues:

Connecting to FEAGI

MuJoCo Humanoid - Connectivity

Connection Type

Local WebSocket + ZMQ

The MuJoCo Humanoid runs on your local machine and connects to FEAGI via:

  • ZMQ for sensory data (joint positions, velocities, orientation)
  • ZMQ for motor commands (joint control signals)
  • HTTP for registration and metadata

Network Requirements

Ports Used

  • 5555: Sensory data (ZMQ PUB socket)
  • 30005: Motor commands (ZMQ SUB socket)
  • 30001: Agent registration (TCP)

Firewall Configuration

Ensure these ports are open for localhost communication:

# macOS
sudo pfctl -d  # Disable firewall temporarily for testing

# Linux
sudo ufw allow 5555
sudo ufw allow 30005
sudo ufw allow 30001

# Windows
# Open Windows Firewall settings and allow these ports

Controller Setup

Installation

The generic MuJoCo controller is installed once and can run any MuJoCo model.

Via FEAGI Desktop:
  1. Navigate to Embodiment Explorer
  2. Select "MuJoCo Physics Simulator" category
  3. Click on "MuJoCo Humanoid"
  4. Click "Launch" (controller installs automatically if needed)
Manual Installation:
# Clone controller repository
git clone https://github.com/feagi/embodiment-controllers.git
cd embodiment-controllers/simulators/mujoco

# Create Python virtual environment
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Install FEAGI Python SDK
pip install feagi

Launch Command

# From controller directory with venv activated
python generic_controller.py \
  --model_xml humanoid.xml \
  --agent_id mujoco_humanoid_01 \
  --ip 127.0.0.1 \
  --cortical_input i_vision \
  --cortical_output o_motor

Configuration Options

ParameterDefaultDescription
--ip127.0.0.1FEAGI host IP address
--model_xml(required)MuJoCo model XML file
--agent_id(required)Unique agent identifier
--cortical_inputi_visionInput cortical area name
--cortical_outputo_motorOutput cortical area name

Connection Process

  1. Startup: Controller loads MuJoCo model and initializes physics engine
  2. Check FEAGI: Verifies FEAGI is reachable on specified IP
  3. Register: Sends agent registration to FEAGI (port 30001)
  4. Establish ZMQ: Creates publisher (port 5555) and subscriber (port 30005)
  5. Simulation Loop:
    • Reads joint states from MuJoCo
    • Publishes sensory data to FEAGI
    • Receives motor commands from FEAGI
    • Applies commands to MuJoCo actuators
    • Steps physics simulation at 120 Hz

Standalone Mode

If FEAGI is not running, the controller automatically enters standalone mode:
  • Physics simulation continues normally
  • MuJoCo viewer allows manual joint manipulation
  • No neural control (useful for testing physics)

To exit standalone and connect to FEAGI:

  1. Start FEAGI
  2. Restart the controller

Troubleshooting

"Connection refused" Error

  • Verify FEAGI is running: docker ps or check FEAGI logs
  • Check IP address matches FEAGI host
  • Ensure ports 5555, 30005, 30001 are not blocked

Model Freezes or Stops Responding

  • Check FEAGI logs for errors
  • Verify ZMQ sockets are not blocked by firewall
  • Restart controller and FEAGI if connection is lost

Model Loads but No Neural Control

  • Verify cortical area names match FEAGI genome
  • Check motor commands are being published by FEAGI
  • Review FEAGI logs for mapping errors

High Latency or Lag

  • Reduce control frequency if CPU is overloaded
  • Check network latency if FEAGI is remote
  • Close unnecessary applications

Remote FEAGI Connection

To connect to FEAGI running on a different machine:

# Replace 127.0.0.1 with FEAGI host IP
python generic_controller.py \
  --model_xml humanoid.xml \
  --agent_id mujoco_humanoid_01 \
  --ip 192.168.1.100
Security Note: Ensure firewall rules allow ZMQ traffic between machines.

Performance Optimization

For Best Performance:

  • Run controller and FEAGI on the same machine (localhost)
  • Close MuJoCo viewer if visual feedback not needed (headless mode)
  • Reduce control frequency for complex genomes
  • Use dedicated GPU for FEAGI neural computations

Support

For connectivity issues:

View marketplace for this embodiment · Back to embodiment explorer