suboptimalityFactor in rrt*

[yuvalgos]

This parameter is documented here http://motion.cs.illinois.edu/software/klampt/latest/pyklampt_docs/Manual-Planning.html#planner-attributes

It should be used in RRT* according to the documentation, but it's unclear what is the range and how does it work.

To my understanding, this it corresponds to the ratio between path length and optimal path length, so if I set it for example to 1.1, path length shouldn't exceed 1.1*optimal_path_len. But it seems to have no effect. I get long paths even when it's 1.01. I also tried extreme values or values that don't make sense like 0.01 and 100 and it seems to have no effect.

[krishauser]

This is a parameter used in LBT-RRT*. The parameter, when larger, is supposed to make the planner faster at the cost of allowing suboptimal asymptotic convergence: https://arxiv.org/abs/1308.0189