Default OptimizationObjective in ProblemDefinition?

Original report by Luis G. Torres (Bitbucket: [Luis Torres](https://bitbucket.org/Luis Torres), ).

Currently, all optimizing motion planners have to implement boilerplate code like this:

https://bitbucket.org/ompl/ompl/src/01a095a5441b640f2ea7d7bdfaee5c168f505295/src/ompl/geometric/planners/rrt/src/RRTstar.cpp?at=default#cl-88

in order to default to PathLengthOptimizationObjective to account for the case when a user has not specified an optimization objective. PRMstar has an identical check, and any other optimizing planner will have to do the above as well under the current framework.

Since most optimizing planners end up defaulting to PathLengthOO, would it make sense to just make ProblemDefinition::optimizationObjective_ default to a PathLengthOptimizationObjective upon construction? This would reduce a lot of boilerplate code in any other optimizing planners.

Can we think of a reason not to do this?

Start by reading ProblemDefinition and the optimizationObjective_ initialization, then compare the fallback logic in RRTstar.cpp with the identical PRMstar behavior. Determine whether a default PathLengthOptimizationObjective preserves existing behavior for planners and callers. Done means the default policy is agreed, consistently applied, and the duplicated planner checks can be removed without changing explicit objectives.