Discussing Radio Degradation Multiplier

Rules
1

Hi All!

I would like to open some discussion among the teams about the proposed radio degradation multiplier mentioned in our last team leader meeting.

Radio degradation is of course a fact of life in various rescue and response situations. Right now, we allow (and encourage) teams to use tethers in the arena because of the practicalities of running so many robots in the one place, but this does mean that communications are unrealistically good. This is something that we have been looking to address for many years (even back to the early years of me being a competitor around 2006 when we were developing fault and packet loss tolerant codecs).

As mentioned in the presentation by Martin, for the past 2 years, the “vSTING” software has been demonstrated and used at the RoboCupRescue German Open, including being used in the finals where it was well received. I honestly think this is our best opportunity to inject some much-needed realism into the communications part of RoboCupRescue.

vSTING is a program, running in Linux on a suitable dual-homed computer (a computer with two network ports), that simulates a degraded network channel. You plug your OCU into one of the network ports, and the robot into the other one (via an ethernet tether, or through a good wireless link), and load up a profile representing the type of degraded network link you want to simulate between the two devices.

Here is the Github for vSTING and here is a paper on it.

It is proposed that this be introduced in 2023 both in the “Technology Demonstrator”, and as a 2x multiplier in the main competition to promote partial autonomy.

I have just been informed that the instructions for deploying vSTING on more general hardware are still being written. Currently instructions are available for deploying this on an APU-4D4 embedded PC running Ubuntu 20.04.

Of course, we would like to make sure that all teams have a fair opportunity, and sufficient development time, to benefit from this multiplier. This is especially important as the teams going to the German Open have already had a couple of years of exposure to this software.

So I have a question for teams who are interested in taking advantage of this multiplier and/or participating in the “Technology Demonstrator”. When should be the deadline for the instructions and settings for vSTING in order to develop this in your entry for 2023?

Cheers!

  • Raymond
2

I would also like to follow up with some thoughts.

If it is decided that there is insufficient time to enable all teams to have a fair opportunity to develop for this multiplier, I have some thoughts as to other things we could do to ease teams in for 2024.

  • Perhaps we could use vSTING to simulate a good wireless connection. Right now teams with tethers have a gigabit (or multi-gigabit depending on equipment) link to their robot, which is clearly not realistic. How about for 2023 we use vSTING to simulate, say, an old style 802.11b connection (which teams can practice by simply dropping their router back to 802.11b mode)?
  • Perhaps we could limit this to only the “Technology Demonstrator” for 2023?
  • Perhaps we could make this a certificate of some sort instead?

What do you all think? What are some other things we could to do introduce the concept of radio degradation in a fair manner?

3

Hi All!

Following feedback, we have published the details of how vSTING will be used for the 2023 competition!

In summary, in 2023 this will be mostly a demonstration. It will be used to simulate a good 802.11b wireless connection and to encourage its use, teams running with this module enabled will have an advantage equal to 1 or 2 points on the old readiness test.

This way, teams who cannot get vSTING working at home can still test at home by flipping their router into 802.11b mode (note that 802.11b, and other 2.4 GHz protocols, are not likely to be allowed in the competition, this is just for your own testing).

Please see vSTING / Network Degradation Support Thread for support for vSTING. Team leaders should have also received an email with a presentation outlining the details.

Note that in 2024, assuming all goes well with its implementation in 2023, this will change to become a real test of semi-autonomy, and will simulate a significantly degraded communications channel and, in return, be worth a 2x multiplier.

4

Hi All!
Radio Degradation with vSTING is a nice feature for RRL teams which is also important for SAR applications and also for end users. For the competition we know WLAN communication between operator station and robot with 2.4 and 5 GHz is common, but we know there are a lot of other teams and leagues who are also working in this frequency band. The problem is if other teams are using the same channel. How can we guarantee that no other team is working in the same channel? In RoboCup 2022, it was not possible to maintain a continuous secure connection to the robot. Therefore, most teams always chose the wired variant.

5

Hi Raimund,

Sorry about the delay in getting back to you! Still trying to figure out why my forum notifications aren’t always working.

The idea is that during the competition teams will run wired (not with wireless LAN) to avoid the radio problems you mention. If teams wish to have the small bonus, they run through the vSTING box which will simulate the same throughput as wireless LAN for this year. This is mostly to get teams familiar with the system.

When testing in their own labs teams are encouraged to test with vSTING but if they can’t get it working, they can approximate it by using wireless 802.11b in their own labs. Of course they won’t be able to use 802.11b at the competition for the reasons you describe.

Cheers!

Raymond

6

Hi Raymond,

It is a great idea to import the radio degradation multiplier for simulating the real SAR environment while assuring the security. We are also attempting to deploy vSTING to our system. We now have the 2 questions:

  1. Is the effect of vSTING in the competition the same as that of what we are deploying? We hope the effect and the reliability are fully evaluated.

  2. We want to deploy a unique wireless image transmission equipment on our robot to improve the image definition, but the data does not go through the Ethernet and vSTING. If so, does the radio degradation multiplier still count when we use vSTING to transport our main data and use the wireless image transmission equipment?

Best regards!

Wenbang Deng

7

Hi Wenbang Deng,

Thanks for your message! I’m glad you were able to get vSTING working.

Regarding vSTING itself, I think Martin is best placed to answer that so I’ll let him do so.

Regarding your unique wireless image transmission equipment, do note that while the wireless requirements for Bordeaux have not yet been published, in general, all transmissions in RoboCupRescue need to be 802.11a or 802.11c, in the 5GHz band, on a single channel, without channel bonding.

This is because there are many robots and leagues in close proximity, each with wireless requirements, so we only have a very small number of channels (probably 2) for the whole league (and there will be more than 2 robots running at any one time). We do not know exactly which channels we will be assigned yet, those will be revealed probably at the competition.

In particular, protocols that are not 802.11a or 802.11c (e.g. DJI Ocusync, analog video senders, proprietary HDMI senders that do not operate at 802.11a or 802.11c) are not allowed.

We also highly recommend that you also be ready to run with an Ethernet cable, just in case the radio situation is bad enough that even our assigned channels are unreliable.

Cheers!

Raymond

8

Hi Raymond,

Thanks for your reply! I still have a question: If our wireless image transmission equipment meets the wireless channel requirements, is the radio degradation multiplier available when we use the image transmission equipment and the vSTING at the same time ?

Best regards!

Wenbang Deng

9

Hi Wenbang Deng,

As long as all of the communication between the robot and OCU, including your wireless image transmission equipment, passes through the vSTING module, you can have the multiplier. For example, if both your normal communication and image transmission are received at your OCU through a wireless router, you should have the vSTING module connect between the router and your control computer (and make sure we can verify that there is no connection directly to the control computer that bypasses vSTING).

If your wireless transmission system does not pass through the vSTING module (e.g. because it has a separate receiver that plugs directly into the control computer) you will not be eligible for the multiplier.

Cheers!

  • Raymond
10

Hi Raymond,

I got it. Thanks!

Best regards!

Wenbang Deng

11

Hi Wenbang Deng,

Is the effect of vSTING in the competition the same as that of what we are deploying? We hope the effect and the reliability are fully evaluated.

We will be using the same values during the competition that have been announced in the slides: Data rate limit of 10 Mbps, latency of 100 ms with +/- 80 ms jitter. This should be roughly equivalent to the old Wifi standard IEEE 802.11b. In the vSTING user interface, there is a button “RoboCup Rescue” to enable these constraints.

We want to deploy a unique wireless image transmission equipment on our robot to improve the image definition, but the data does not go through the Ethernet and vSTING. If so, does the radio degradation multiplier still count when we use vSTING to transport our main data and use the wireless image transmission equipment?

I think, this question has already been answered by Raymond. Please route all your network traffic through the vSTING module to achieve the radio degradation multiplier.

Best regards,
Martin

12

Hi Martin,

Thanks for your reply! By the way, we met a problem when we are deploying vSTING. Could you help us to solve the problem in this thread?

Best regards!

Wenbang Deng