Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

New Role Assignment/Behavior #1587

Open
MaikRe opened this issue Jan 16, 2025 · 0 comments
Open

New Role Assignment/Behavior #1587

MaikRe opened this issue Jan 16, 2025 · 0 comments

Comments

@MaikRe
Copy link
Contributor

MaikRe commented Jan 16, 2025

We need new behavior

As discusses at the team retreat our current role assignment has some major issues that should in the long run be addressed. Besides the issues with role assignment doing too much that it shouldn't (filtering team ball/other messages, path planning) it functionally also really only decides where robots should be standing on the field in order for the to wait until the ball is conveniently within reach so they can take it and make a run for it.
Preferably we need new roll assignment/behavior such that each 'role' becomes more abstract, more powerful, and less convolutedly tied in with some sort of assignment algorithm.

Ideas include:

  • A (potentially learned) behavior decision tree for role assignment
  • Better distance based planning although this is difficult as we lack perfect world state information

Besides that there are ideas to completely rework role assignment based on:

  • Potentially other sports such as volleyball or hockey?
  • Voronoi setup for optimal field coverage (although this does not guaranteee strategic positioning)
  • Some type of 'zone' assignment that can be deterministically calculated by each robot based on the available field players. Within this a focus might be on passing between zones, as to not have one striker robot run with the ball when a pass could be more efficient

Benefits include:

  • Less walking
  • Better cooperative play
  • Adaptability to opponents by priming the team with certain play formations
    Drawbacks include:
  • May be difficult with limited communication
  • Needs accurate and long distance passes to avoid robots leaving their own zones

Additionally all of this stemmed from wanting to handle/implement the ability to play with any given jersey number, so whatever new thing is designed should optimally be dynamic and not based on some fixed 1-7 assignment.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Status: Open
Development

No branches or pull requests

1 participant