This has
- Separate Scatter class
- change from updateATA to updateHessian as a virtual method
- Fixed-size BinaryJacobianFactor that overloads updateHessian
- SFM Timing Script that takes BAL files
Instead of turning Hessian factors into Jacobian factors -- so that they can be eliminated with constrained Jacobian factors using the special QR in Constrained's noise model -- we combine all Hessian factors, eliminate the variable first to have a conditional and a new factor 1, then combine the constrained Jacobians with this conditional (also a Jacobian) to eliminate again, producing the final conditional, and a new factor 2. The two new factors are then combined into a new Hessian factor to be returned.
Currently, when eliminating a constrained variable, EliminatePreferCholesky converts every other factors to JacobianFactor before doing the special QR factorization for constrained variables. Unfortunately, after a constrained nonlinear graph is linearized, new hessian factors from constraints, multiplied with the dual variable (-lambda*\hessian{c} terms in the Lagrangian objective function), might become negative definite, thus cannot be converted to JacobianFactors.
Following EliminateCholesky, this version of EliminatePreferCholesky for constrained var gathers all unconstrained factors into a big joint HessianFactor before converting it into a JacobianFactor to be eliminiated by QR together with the other constrained factors.
Of course, this might not solve the non-positive-definite problem entirely, because (1) the original hessian factors might be non-positive definite and (2) large strange value of lambdas might cause the joint factor non-positive definite [is this true?]. But at least, this will help in typical cases.
Use mixed constrained noise with sigma < 0 to denote inequalities.
Working set implements the active set method, turning inactive inequalities
to active one as equality constraints by setting their corresponding sigmas to 0
and vice versa. Dual graph now has to deal with mixed sigmas.
Currently, when eliminating a constrained variable, EliminatePreferCholesky converts every other factors to JacobianFactor before doing the special QR factorization for constrained variables. Unfortunately, after a constrained nonlinear graph is linearized, new hessian factors from constraints, multiplied with the dual variable (-lambda*\hessian{c} terms in the Lagrangian objective function), might become negative definite, thus cannot be converted to JacobianFactors.
Following EliminateCholesky, this version of EliminatePreferCholesky for constrained var gathers all unconstrained factors into a big joint HessianFactor before converting it into a JacobianFactor to be eliminiated by QR together with the other constrained factors.
Of course, this might not solve the non-positive-definite problem entirely, because (1) the original hessian factors might be non-positive definite and (2) large strange value of lambdas might cause the joint factor non-positive definite [is this true?]. But at least, this will help in typical cases.
Use mixed constrained noise with sigma < 0 to denote inequalities.
Working set implements the active set method, turning inactive inequalities
to active one as equality constraints by setting their corresponding sigmas to 0
and vice versa. Dual graph now has to deal with mixed sigmas.
Abe
yao
Alex Hagiopol
Frank
Chris Beall
lvzhaoyang
dellaert
Sungtae An
thduynguyen
krunalchande
Luca
hchiu