Re: Cable in the RMRC tournament

[Raymond Sheh]

Hi All,

Odin - Thanks for your question!

I'm replying to the mailing list just so everyone has the same
information. This is also a good opportunity to talk more about cables
and tethers.


We highly encourage teams to use cables in the competition. RoboCup
involves many robots all running at once and we run out of spectrum very
quickly. In general all of RoboCupRescue (both Major and RMRC) are only
allocated two bands at 802.11ac. With up to 10 robots running at once
during preliminaries, all in close proximity, things quickly become
impossible, especially given that RMRC robots usually don't have
antennas as big as the ones on the Major RMRC robots and won't win a
radio frequency shouting match.


We have two main rules around cables.

- The cables should not present a trip or other safety hazard (to the
team members, other competitors, or competition staff). In general this
means it must lay flat along the ground (bring tape and/or some kind of
cable conduit if necessary - you can assume the operating table will be
no more than 6 m (20 ft) from the lane where you will operate).

- The team is allowed to have someone (the "Cable Wrangler") handle the
cable so that it doesn't get caught on objects in the lane but they
cannot use the cable to mechanically assist the robot. The robot should
run as if the cable wasn't there and it was wireless. In general, this
means that the cable should remain visibly slack during the run. Usually
the "Cable Wrangler" stands around mid-way on the lane and holds the
cable up, adjusting position to avoid having the cable touch the terrain.


That second rule has some nuance. If you have a single thin cable (e.g.
an Ethernet cable) it's easy to tell when the cable is slack.

On the other hand, if you have a bunch of cables that form a thick
bundle (e.g. in the limiting factor, you have no computer at all on the
robot and instead, the wires from the motors, cameras, and sensors all
come back to the operator station), you're probably going to have some
trouble not influencing which way the robot moves when you handle the
end of the cable. In such a situation, it will be quite hard to satisfy
that second rule unless you just let the cable drag, which will impose
quite a significant penalty (especially if the robot is very light
weight). For this reason, we highly discourage cables that are more than
just a lightweight Ethernet cable.


There is another angle to the question of cables. Real response robots,
be it ground, underwater, or aerial, often do run with a cable (often
called a 'tether') and this provides several advantages. For instance,
ground robots that go into complex highly shielded environments, such as
industrial buildings and nuclear facilities, cannot use radio as those
buildings are designed to block radiation.

Such robots are designed with the tether in mind. Instead of dragging
the tether through the environment, they have the tether on an
intelligent, motorized spool on the robot itself. It 'lays' the cable on
the ground as it drives forward and picks it up again as it drives
backwards. This way if the cable gets caught, it doesn't really affect
the ability of the robot to continue forward.

Some smaller robots can even use the tether to help with mobility. The
robot can use the motorized spool to pull on the tether at its end to
help 'winch' it backwards if it gets into trouble, for instance.

Teams are, of course, welcome to implement such capabilities. In such a
case a "Cable Wrangler" would not be required. I would suggest that we
could even allow teams to affix the cable to the start of the terrain
bay so that the robot's tether spool has something solid to pull on. Of
course you will need a long cable to run in the finals, where the robot
must run a significant distance. The robot is expected to carry its full
spool of cable in the preliminaries - teams are not allowed to run with
a short spool in the preliminaries and a long spool in the finals. (If
nothing else, when the team reaches the finals, the first few terrains
will need to be done with a full spool anyway.)


If anyone has any further questions, please do let us know!


Cheers!

- Raymond



On 2023-03-20 3:54 PM, Odin Haro wrote:
> Hi!, I’m hoping you’re having a nice day, I have a question about the
> use of cables in the tournament, I have seen in videos that in the
> past years, there’s people that uses a cable to send signals from the
> robot to the control station, if we can do it, how many cables can we
> have from the robot to the control station to send signals and
> communicate?