[robotics-worldwide] [software] Control Toolbox (CT) - Library for MPC and Trajectory Optimization

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

[robotics-worldwide] [software] Control Toolbox (CT) - Library for MPC and Trajectory Optimization

Jonas Buchli-2
Dear Colleagues,

it is our pleasure to announce the Control Toolbox (CT), an open-source
C++ library aiming at enabling robotics researchers to solve optimal
control as well as nonlinear model predictive control problems.

The library contains a collection of tools to model dynamical systems,
design and evaluate controllers and efficiently solve trajectory
optimization problems.

The CT has been designed with efficiency and real-time robot control in
mind - it allows for fast robot modeling and optimal control problem setup:
* it includes high-performance implementations of iLQR/SLQ, Gauss-Newton
Multiple Shooting, Direct Single and Multiple Shooting, and also
provides a variety of interfaces to third-party Optimal Control and NLP
solvers such as HPIPM, IPOPT and SNOPT.
* a cost function and constraints module allows to quickly set up
optimal control problems.
efficient first- and second-order derivative code can be easily obtained
using automatic differentiation and code generation.
* it provides an interface to a state-of-the-art rigid body dynamics
modeling tool (RoCoGen), generating auto-differentiable rigid-body
dynamics and kinematics equations.
* several recent publications describe use cases and demonstrate the
toolbox’ potential, e.g. by running full body NMPC on quadruped hardware
at over 180 Hz.

The source code and a variety of application examples, including MPC on
quadrupeds, mobile manipulators, UGVs and UAVs can be found at

For questions, comments or suggestions you can reach the developers at
[hidden email]. Subscribe to
[hidden email] to receive news about the CT.

On behalf of all CT-developers,

Michael Neunert, Markus Giftthaler, Markus Stäuble and Jonas Buchli

Prof. Dr. Jonas Buchli
Agile and Dexterous Robotics Lab
Institute of Robotics and Intelligent Systems
ETH Zürich
+41 44 633 92 06
robotics-worldwide mailing list
[hidden email]