Ross ROSS = Recommend OSS · open-source software intelligence for agents

acados/acados

Fast and embedded solvers for nonlinear optimal control and nonlinear model predictive control observed · 2026-08-28

github.com/acados/acados · homepage · C · NOASSERTION (other) observed · 2026-08-28

Health v2 · maintenance only

93/100

  • Activity 97
  • Release rhythm 84
  • Longevity 100

Flags: no_license

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 36.0
  • age_days: 3802
  • days_rel: 27
  • days_push: 21
  • n_releases_24m: 13

Full methodology

Adoption not part of the score

1450 stars · 359 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

acados is a modular C library providing fast embedded solvers for nonlinear optimal control problems, designed for real-time applications like model predictive control (MPC) and moving horizon estimation (MHE). It offers high-level interfaces for Python, MATLAB, Octave, and Simulink, and can generate self-contained C code for deployment on embedded systems.

Use cases

  • solve nonlinear model predictive control problems in real time
  • implement moving horizon estimation for state and parameter estimation
  • deploy an MPC solver as generated C code on an embedded system
  • simulate ODE and DAE systems with sensitivity propagation
  • embed a differentiable NLP solver as a layer in a machine learning pipeline
  • formulate multi-phase optimal control problems
  • prototype MPC controllers from Python or MATLAB/Simulink

When to choose

  • you need fast, real-time nonlinear MPC or MHE with millisecond solve times
  • you target embedded hardware or high-frequency control loops
  • you want code-generated, self-contained solvers deployable without heavy dependencies
  • you need solution sensitivities for reinforcement learning or gradient-based tuning
  • you work with ODEs and DAEs and need efficient integrators with first- and second-order sensitivities

When to avoid

  • you only need to solve general-purpose NLPs without optimal control structure (use IPOPT or similar)
  • you need a pure high-level Python solver without compiling C code
  • your problem is linear or small enough that simpler tools like scipy suffice
  • you require a permissively licensed library without checking its custom license terms

Facets

library · maturity active

simulation math machine-learning embedded robotics autonomous-vehicles simulation embedded-systems machine-learning windows python cpp c embedded cross-platform optimal-control model-predictive-control moving-horizon-estimation nonlinear-programming real-time differentiable-optimization casadi simulink matlab octave algorithms linux macos

7 sources

Member repositories

RepositoryRoleHealth v2
acados/acadosmain93

For agents

markdown · JSON · MCP: product_card(name="acados/acados")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem