OSQP solver documentation¶
[Visit our GitHub Discussions page](https://github.com/orgs/osqp/discussions) for any questions related to the solver!
The OSQP (Operator Splitting Quadratic Program) solver is a numerical optimization package for solving convex quadratic programs in the form
where \(x \in \mathbf{R}^n\) is the optimization variable and \(P \in \mathbf{S}^{n}_{+}\) a positive semidefinite matrix.
Code available on GitHub.
Citing OSQP
If you are using OSQP for your work, we encourage you to
We are looking forward to hearing your success stories with OSQP! Please share them with us.
Features
- Efficient¶
It uses a custom ADMM-based first-order method requiring only a single matrix factorization in the setup phase. All the other operations are extremely cheap. It also implements custom sparse linear algebra routines exploiting structures in the problem data.
- Robust¶
The algorithm is absolutely division free after the setup and it requires no assumptions on problem data (the problem only needs to be convex). It just works!
- Detects primal / dual infeasible problems¶
When the problem is primal or dual infeasible, OSQP detects it. It is the first available QP solver based on first-order methods able to do so.
- Embeddable¶
It has an easy interface to generate customized embeddable C code with no memory manager required.
- Library-free¶
It requires no external library to run.
- Efficiently warm started¶
It can be easily warm-started and the matrix factorization can be cached to solve parametrized problems extremely efficiently.
- Interfaces¶
It provides interfaces to C, C++, Fortran, Julia, Matlab, Python, R, Ruby, and Rust.
License
OSQP is distributed under the Apache 2.0 License
Credits
Please credit the OSQP Developer Team when using OSQP.
Interfaces development
Nick Gould (Rutherford Appleton Laboratory): Fortran and CUTEst interfaces
Ed Barnard (University of Oxford): Rust interface
Bug reports and support
Please report any issues via the Github issue tracker. All types of issues are welcome including bug reports, documentation typos, feature requests and so on.
Numerical benchmarks
Numerical benchmarks against other solvers are available here.