Is the Control Toolbox still maintained?

No, as of July 2021, the library is scarcely maintained, with slow response to issues. It's best for projects that can tolerate limited support or are based on existing stable versions.

Control Toolbox vs CasADi for trajectory optimization?

Control Toolbox is C++-focused and optimized for real-time robotics with hardware integration, while CasADi is more language-agnostic and general-purpose. CT offers better performance for online MPC but has a steeper learning curve.

How to implement MPC on a quadrotor with Control Toolbox?

Use the ct_optcon module to define cost functions and constraints for the quadrotor model from ct_models, then set up the MPC wrapper with a solver like iLQR. Detailed examples are in the documentation and related publications.

Can I use Control Toolbox without ROS?

Yes, it's designed to be middleware-agnostic with minimal dependencies, so you can integrate it directly into C++ applications. For ROS, there's a separate ct_ros repository for examples.

What automatic differentiation methods does it support?

It supports first and second-order AD with code generation using Cppad for efficiency, and numerical differentiation for flexibility, as detailed in the ct_core module.

How to cite the Control Toolbox in papers?

Cite the provided SIMPAR 2018 paper reference from the README, which includes the authors and key features, ensuring proper attribution for academic use.

control-toolbox

BSD-2-ClauseC++

An efficient C++ library for robotics, optimal control, and model predictive control with a focus on online performance.

GitHub

What is control-toolbox?

The Control Toolbox (CT) is an open-source C++ library for robotics, optimal control, and model predictive control. It provides tools for modeling dynamic systems, solving trajectory optimization problems, and implementing controllers that can run online on robotic hardware. The library addresses the need for efficient, flexible software in numerical optimal control and robotics applications.

Target Audience

Researchers and practitioners in robotics and control communities who need to implement model-based controllers, estimators, or planning algorithms with a focus on efficiency and real-time performance.

Value Proposition

Developers choose the Control Toolbox for its combination of high-performance C++ implementation, support for automatic differentiation, and modular design that allows seamless integration with existing code. It simplifies the application of advanced concepts like nonlinear MPC while enabling online operation on actual hardware.

Overview

The Control Toolbox - An Open-Source C++ Library for Robotics, Optimal and Model Predictive Control

Use Cases

Best For

Implementing nonlinear model predictive control for UAVs or ground robots

Related Projects

Community-curated · Updated weekly · 100% open source

Found a gem we're missing?

Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.

Submit a project Star on GitHub

1.7k stars341 forks0 contributors

Trajectory optimization for legged robots like quadrupeds

Real-time optimal control in hardware experiments

Modeling rigid body dynamics with automatic differentiation

Solving large-scale optimal control problems with MPC

Rapid prototyping of control algorithms for robotic systems

Not Ideal For

Projects requiring active library maintenance and frequent updates
Teams preferring Python or MATLAB for rapid control system prototyping
Simple control tasks that don't need robotics-specific modeling or optimal control

Pros & Cons

Pros

High-Performance C++ Design

Optimized for real-time operation on hardware, with efficient code generation for automatic differentiation, enabling online MPC as highlighted in the README.

Comprehensive Optimal Control Suite

Includes multiple shooting methods, MPC wrappers, and interfaces to solvers like IPOPT and SNOPT, covering a broad range of algorithms from iLQR to direct multiple shooting.

Modular and Flexible Architecture

Organized into core, optcon, and rbd modules, allowing selective use without forcing a full framework, as described in the structure section.

Robotics-Focused Modeling Tools

Integrates with RobCoGen for rigid body dynamics and provides inverse kinematics solvers, specifically tailored for robotic applications like quadrupeds and UAVs.

Cons

Scarce Maintenance

The README notes it's only scarcely maintained since 2021, leading to slow bug fixes and potential compatibility issues with newer systems.

High Expertise Requirement

Requires in-depth knowledge of control theory and C++ programming, making it inaccessible for beginners or those without a robotics background.

Complex Setup and Dependencies

Relies on Eigen, Kindr, and external solvers, which can make installation and integration cumbersome, especially for users new to C++ build systems.

Frequently Asked Questions

Home

Robotic Tooling

pinocchio

A fast and flexible implementation of Rigid Body Dynamics algorithms and their analytical derivatives

Stars3,283

Forks529

Last commit4 days ago

EGO-Planner

EGO-Planner is an ESDF-free gradient-based local planner designed for quadrotor navigation. It significantly reduces computation time compared to state-of-the-art methods by avoiding the computationally expensive Euclidean Signed Distance Field (ESDF) construction, enabling real-time performance with total planning times around 1ms. ## Key Features - **ESDF-Free Planning** — Eliminates the need to compute Euclidean Signed Distance Fields, drastically reducing computational overhead. - **Lightweight Gradient-Based Optimization** — Uses a gradient-based approach for efficient local trajectory generation. - **GPU/CPU Versatility** — Offers both GPU and CPU versions of its local sensing module for depth image generation or pointcloud processing. - **Fast Computation** — Achieves planning times of approximately 1ms, suitable for real-time drone control. - **Simulation-Ready** — Includes a lightweight quadrotor simulator and supports integration with sensors like Intel RealSense for hardware testing. ## Philosophy EGO-Planner prioritizes computational efficiency and real-time performance by removing the ESDF construction bottleneck, making advanced local planning accessible for resource-constrained aerial robotics applications.

Stars2,413

Forks385

Last commit1 year ago

casADi

CasADi is a symbolic framework for numeric optimization implementing automatic differentiation in forward and reverse modes on sparse matrix-valued computational graphs. It supports self-contained C-code generation and interfaces state-of-the-art codes such as SUNDIALS, IPOPT etc. It can be used from C++, Python or Matlab/Octave.

The Open Motion Planning Library (OMPL)

Stars2,019

Forks685

Last commit15 days ago

#robotics

#motion-planning

#cplusplus-library

#automatic-differentiation