build details

Show: section status errors & todos local changes recent changes last change in-page changes feedback controls

Lane Following Quickstart

Modified 2018-10-30 by breandan

Maintainer: Liam and Andrea C.

You have done the Unit A-1 - Preliminaries

You have made a submission to the Lane Following AI-DO challenge and you know how to try to make it better

1) Do a submission by following one of the templates: Random template Tensorflow template PyTorch template ROS template

Congrats! you have now made a submission, but it probably wasn’t very good (unless you got very lucky).

2) Try to make your score go up. Now is when you might want to take a look at Part C - Rules which describe in detail how your score is generated for the specific challenges. For the lane following challenge, we are currently offering 4 suggested methods to do this (our baseline templates for these options are at various stages of readiness but will be getting updated very soon):

Of course you may also choose to use these methods in combination.

If you choose on the “Learning-based” methods (i.e. 2-4) then you may want to look at the Unit D-10 - Using Machine Learning in Duckietown

Because of mathjax bug

No questions found. You can ask a question on the website.