Validate that we can make the arena dynamic and interactive

[PeterJCLaw]

For SR2021 it's looking like we're going to have a dynamic arena. We need to work out how we'd actually do that.

• prove a mechanism for robots to communicate to the arena supervisor so they can get feedback about their actions
• prove that we can move the walls around from a supervisor

There is some overlap here with #18 though I think we can approach them separately, at least for validating the things we need to work can work.

The outcome here might just be a proof-of-concept branch, though it would obviously be great to have a full PR implementing this.

I suggest that we probably want a separate Supervisor to handle the arena interactions, if Webots supports having more than one.

[PeterJCLaw]

#197 (comment)

I'd only been looking into #18 to check how viable it is. I've also only been looking into the scoring feedback element rather than dynamic arena. Whilst I should be able to do it, i'd very much like not to, as there's enough on my plate already. (Happy to share the PoC information with someone, though).

@RealOrangeOne is that just notes? If so, could you add them to this issue? Or is it a branch? If so, perhaps you could push it and then link it from this issue?

[RealOrangeOne]

It's not a branch, it's more fundamentals.

If territories are defined as a .proto file, and have a color key, then we can dynamically change that in webots.

supervisor.getFromId(123).getField("color").setSFColor([0.5, 0.5, 0.5])

We'd need to get the ids of each territory (easily done through supervisor.getSelected()), poll the location of a robot (robot.getPosition), check if the robot is in a territory (some maths), and then change the colour and log that it happened (as above).

In theory the same can also be done for moving walls, however there are physics issues around that, so i'd recommend just having the walls vanish (.remove()).

[PeterJCLaw]

It looks like we're going to do tower claiming and knowledge via radio emitters and receivers. This allows them to also be used for location too 👍.
Each Robot will have both an emitter and a receiver, as will each tower. There will be a separate supervisor which looks after the arena stuff and logs out any changes. We have checked that multiple supervisors is supported.

We plan to put each robot's transmitter on a different frequency and the towers (together) on a third. The towers will listen for everything; the robots will only listen to the towers. Robots will be able to impersonate towers, but our rule around good sporting conduct covers not doing that.

Since location information can be extracted from the location towers, we'll drop the camera.

[WillB97]

This was done in #224

[PeterJCLaw]

Have we checked that we can interactively move objects?