NMPC, WBC, state estimation, and sim2real framework for legged robots based on OCS2 and ros-controls

Official implementation for the paper "Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing". DIAL-MPC is a novel sampling-based MPC framework for legged robot full-order torque-level control with both precision and agility in a training-free manner.

An Architecture for the Versatile Control of Legged Robots

CoRL2025 UniFP: Learning a Unified Policy for Position and Force Control in Legged Loco-Manipulation

Source code for the paper DiffuseLoco: Real-Time Legged Locomotion Control with Diffusion from Offline Datasets

MultiModalWBC is a fully open-source, IsaacLab-based framework for multi-modal whole-body control, designed for motion imitation, motion tracking, and task-conditioned control in legged robots. The framework unifies robot proprioceptive states and multi-modal human motion conditions into a consistent interface

Whole-Body Model-Predictive Control of Legged Robots with MuJoCo

Welcome to the HexaPod Robot Tutorial! This repo offers a step-by-step guide to build a hexapod spider robot. From hardware specs and Raspberry Pi-driven servo control, to software algorithms and sensor integrations, we've got you covered. Ideal for both hobbyists and engineers eager to explore multi-legged robotics.

Collection of centroidal control for legged robots

Quaternion Model Predictive Control for Legged Robots

A legged robot controller for Unitree A1 and Go1 robot using different MPC algorithms

TrueWalk is a motion planning and gait generation system for controlling virtual and real-world multi-legged robots. This repository is for the Unity simulator version, that can also control tethered robots over serial.

Code and benchmark release for paper: GenTe: Generative Real-world Terrains for General Legged Robot Locomotion Control

Terrain mapping algorithm for motion planning and control in legged locomotion

A flexible framework for connecting legged robots, input devices, and control algorithms.

No description available

Use legged control to control GO2 on both simulation and real

This repository provides a framework for low-level control of a legged robot (Unitree Go2), using ROS 2 as the communication middleware. The MuJoCo simulator is used to validate the control policy in simulation. Once the policy performs well in MuJoCo, you can deploy it on the real robot by simply setting the ROS parameter is_simulation to false.

RL-Augmented Model Predictive Control for Non-Gaited Legged and Hybrid Locomotion

Modeling and Control of Two-Legged Wheeled Robot: In this project, we will be building a two wheeled robot (with a hip and knee joint) which can balance at variable heights including a provision for jumping.

Mobile Robotics has been evolving as one of the most promising domains in the field of Robotics. The ability of these robots to explore and maneuver in complex environments without human intervention attracts the attention of researchers across the globe. The mobile robots are classified into three different areas viz. wheeled robots, tracked robots, and legged robots. Robot locomotion system is an essential characteristic of mobile design, which depends not only on working space but also on technical measures like maneuverability, controllability, terrain condition, efficiency, and stability. Applications involving locomotion over rough terrains or disaster management where the robot is needed to access the remote areas within the debris demand the use of legged robots. Legged robots are further classified depending on the number of legs the robot has. Hence the types of legged robots are pogo-stick robots or one-legged robots, bi-pedal or two-legged robots, quadrupedal or four-legged robots, six-legged or hexapod robots, and eight-legged robots. Each of the types has unique applications and special locomotion mechanisms. The GAIT behavior of the quadrupedal robots is inspired by the quadruped animals like horses, dogs, etc. This project is focused on the Simulation & Control of a Quadrupedal Robot, using trajectory generation for the locomotion and describing three types of GAIT behaviors for the robot, viz. Walking, Trotting and Galloping. These are based on the speed and leg-movement patterns of the robots. These behaviors can be transitioned depending on the application and the terrain pattern. All the mechanisms are designed and simulated in MATLAB, and Simulink.

Gait planning and control software for legged robots

Some tutorials of trajectory optimization (TO) and model predictive control (MPC) for legged robots.

legged control of g1_12dofs with ros2 and ocs2

University Project: Design of a six-legged Hexapod with 3 DoF at each leg. Communication and control implementation on an Arduino 2560.

Personal legged_gym Unitree A1 implementation for paper 'Reinforcement Learning for Versatile, Dynamic, and Robust Bipedal Locomotion Control'. Homework repo for SJTU ACM class RL courses

No description available

SoccerLab is a specialized extension built on the NVIDIA IsaacLab framework, designed for developing, training, and deploying advanced control strategies for legged robots in a dynamic, multi-agent soccer environment.

Unitree Go1 legged Robot control by voice

多足机器人MPC控制器+仿真环境