# qiayuanl/legged_control

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

Repository: https://github.com/qiayuanl/legged_control
Canonical: https://ross.abutalabs.com/products/legged_control
Language: C++
License: BSD-3-Clause
License Family: permissive
Last push: 2025-02-13T10:09:24+00:00

## Health v2 (maintenance only)
Score: 35/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 6, release rhythm 35, longevity 100
- inputs: {"age_days": 1532, "days_push": 566, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1793, forks 361 (observed 2026-08-28T04:05:37.275103+00:00)

## What it is
legged_control is a C++ control framework for legged robots combining nonlinear model predictive control (NMPC), whole-body control (WBC), and state estimation, built on OCS2 and ros-controls. It supports simulation and real-robot deployment (e.g., Unitree A1) with a sim2real pipeline.

## Use cases
- control a quadruped robot with MPC
- deploy a walking controller on a Unitree A1
- simulate legged robot locomotion in Gazebo
- build a custom robot controller on ros-control
- study sim2real for legged locomotion
- implement whole-body control for a biped or quadruped

## When to choose
- you need a high-performance open-source model-based baseline for legged locomotion
- you want to deploy NMPC+WBC on a real quadruped quickly
- your robot integrates with the ros-control interface

## When to avoid
- you need actively maintained software - the authors have abandoned this project
- you need perception-integrated locomotion out of the box (see legged_perceptive instead)
- you want a learning-based (RL) locomotion stack

## Facets
- artifact type: framework
- maturity: abandoned
- function: robotics, simulation, machine-learning, middleware
- domain: robotics, simulation, developer-tools
- platform: cpp
- tags: legged-robots, nmpc, wbc, state-estimation, sim2real, quadruped, ocs2, ros-control, linux, ros

## Member repositories
- qiayuanl/legged_control (main) score 35

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:37.275103+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:22:56.821248+00:00, confidence not recorded.
  - readme: https://github.com/qiayuanl/legged_control (fetched 2026-08-28T04:05:37.275103+00:00, sha 88fdf3af2192)
- Data as of 2026-08-30T08:39:29.467469+00:00.