|
|
|
TOMLAB /MINOS
The TOMLAB /MINOS toolbox efficiently integrates the three well-known solvers MINOS, QPOPT, and LPOPT, developed by the Stanford Systems Optimization Laboratory (SOL), with Matlab and TOMLAB.
MINOS is a large-scale sparse general nonlinear solver, but solves also sparse linear and quadratic problems efficiently.
QPOPT is a dense convex quadratic solver that also solves non-convex quadratic programs.
LPOPT is a special version of QPOPT that solves dense linear programming problems efficiently.
Read more about the above mentioned SOL solvers in the TOMLAB /MINOS User's Guide.
Main features
- Other TOMLAB solvers are using MINOS, QPOPT and LPOPT as subproblem solvers resulting in faster and more robust code compared to only using the TOMLAB Base Module.
- The dense QPOPT code is implemented to efficiently handle Matlab sparse arrays.
- It is easy to use warm starts for the SOL solvers, and further speed up sequences of optimization solutions.
- In TOMLAB /MINOS the mixed-integer linear programming solver mipSolve runs much faster than in the TOMLAB Base Module due to the use of MINOS (or LPOPT) as linear subproblem solver with warm starts.
|
|
|