Stalling and Final Report

This weekly report is actually contains the last 4 and a half weeks, since I was hoping that I could obtain experimental results before submitting the final report, and also had some subset of that time occupied moving back onto campus and getting started with class.

At this point in time, templates are now placed, checking against both static obstacles as well as other robots outside the template. The paste function has been into pre-paste and paste, since their code logically grew out of eachother. Multiple templates are now tried; pre-paste creates the transform and the endpoints of the template subject to the transform, an evaluation function determines a heuristic for how good the paste is based on the distances between the window endpoints and the pasted template endpoints, and the heuristic determines the order the templates are tried out. By tried out, I mean that the paste function is called, which now just adds the template to the roadmap as it does all of the types of collision checks, and returns the paste is a failure if any part of it collides. It should be working, but there is still bugs in it, and at this time it basically always fails to paste the templates. Oh, and I also started adding code for ComposeRepair, which can be used to combine repair methods in a modular way to get more advanced behaviors – but it is still very much in the early stages.

I have greatly enjoyed this DREU experience and the lab I have been working with, and will continue working with this research group into the foreseeable future.

If you are an incoming DREU in the future reading this, I wish you the best of luck, and hope you enjoy it as much as I did!