What's new


YALMIP 3 is a major revision compared to earlier versions. The internal machinery of YALMIP has been rewritten, and the main result is the possibility to work with nonlinear expressions. In addition, mixed integer capabilities have been improved, and a number of new solvers have been interfaced.

YALMIP 3 [R20040306]

bullet
Determinant maximization problems can now be solved using any SDP solver.
bullet
Global bilinear programming (beta version, slow and unstable...)
bullet
Extended mixed integer code to include posynomial geometric programming (requires latest MOSEK patches, not tested seriously).
bullet
Least squares assignment in setsdpvar.
bullet
sdpvar can be called with only one argument (simplifies definition of symmetric matrices and scalars).
bullet
Fixed bug in internal mixed integer solver (might have caused YALMIP to miss the optimal solution.)
bullet
Fixed bug on overloaded >= operator.
bullet
Fixed bug on purely binary problems using MOSEK.
bullet
Fixed bug in interface to PENBMI (may have caused erroneous solutions in some rare cases).
bullet
Fixed various bugs to enable nonlinear objective functions in moment relaxations.
bullet
YALMIP does not mean Yet Another LMI Parser any more.
bullet
YALMIP now works with LMILAB in MATLAB 5.3.
bullet
Various fixes for MATLAB 5.3.

News in Version 3 compared to 2.4

bullet
Polynomial expressions.
bullet
Quadratic objective functions.
bullet
Interfaces to solvers PENSDP, PENBMI, LMILAB, NAG, CDD, OOQP, MOSEK, CPLEX, LINPROG and QUADPROG.
bullet
Added possibility to define KYP-related constraints (kyp) and related interface to the specialized code KYPD.
bullet
Multi-parametric linear and quadratic programming using MPT-Toolbox.
bullet
Integer and binary variables.
bullet
Posynomial geometric programming.
bullet
An internal (wickedly slow) branch-and bound solver for mixed integer conic programs (MILP, MIQP, MISOCP, MISDP).
bullet
Sum-of-squares decompositions.
bullet
Moment-based polynomial programming.
bullet
Simple definition of double-sided constraints (see Linear regression example).
bullet
Plot the feasible set of a constraint.
bullet
Imroved functionality in checkset (previously called checklmi).
bullet
Imroved support for subsref in set (previously called lmi) and sdpvar objects.
bullet
New feature in the option removeequalities in sdpsettings.
bullet
New feature in the option cachesolvers in sdpsettings.
bullet
New feature in the option usex0 in sdpsettings.
bullet
New feature in the option radius in sdpsettings.
bullet
New feature in the option shift in sdpsettings.
bullet
Updated interfaces to new versions of CSDP, DSDP, SDPT3 and GLPK.
bullet
Improved error detection in SeDuMi and GLPK.

Make sure to read the information on backwards compatibility.