Concern and questions regarding new rules

Rules
#1

Dear RoboCup Committee,

My name is Witthawin Sae-Lee and I represent BART LAB Rescue Robotics Team. Regarding the meeting we just had, our team has the following suggestions/concerns:

  • We disagree with “4x multiplier for autonomy”. We believe that in any real situations, humans should also be able to make crucial decisions for the best outcome - in case of an emergency (e.g. further destructions within the building). Therefore, we believe that “autonomy” should be optional. We suggest that “autonomy” can be its own new task, instead of being integrated into every task for extra points, which we believe is way too much.

And questions:

  • When exactly would the rule book be released?
  • When mentioned “on-site participation considered crucial for success of such workshop”, what are the requirements for us to join? What are we going to get from participating the German Open? What is expected from us during this workshop?
  • Also, for the compensation, we briefly saw that non-Europeans will be granted up to $7000 for travel costs, will this be applicable for us, who live in Bangkok, Thailand? And where can we get more details about this compensation and the workshop as a whole?

Thank you very much. We look forward to hearing from you soon.

Best regards,
Witthawin Sae-Lee

Multipliers for Autonomy and Radio Degradataion
#2

Hi Witthawin Sae-Lee,

Thank you for your question! Here’s my attempt at an answer. Of course others are also welcome to comment or answer.

Multipliers:

Of course we are tuning the multipliers so this is a good discussion to have. I sense that this will be a discussion worth its own thread so I’ve opened a new one here.

Rulebook:

As for the rulebook, we are trying to get more engagement in the development process (and we’re very happy that you’re joining and engaging with us to do so!). So in a sense when the rules will be released will depend a bit on how long this process takes. I would hope (although cannot guarantee) that at least a draft for comment of the significant parts would be available in the next month or two. In the meantime we’re hoping that even when it hasn’t been finalized, engagement in these discussions will mean that people will have a good idea of what to expect and not be surprised.

Workshop Participation:
Regarding participation, I think there might be a slight miscommunication about the word “crucial” … it was in reference to the event needing to have an in-person component (and thus to justify the expense of holding the event and providing travel support to those interested).

It does not mean that it is crucial for all teams to attend or that teams attending will receive an advantage of some sort beyond added competition experience and involvement in discussions. This would be the case going to any regional competition and getting involved in these telecons and forum discussions.

Of course, we recognize that not all teams and organizers can make the trip out to Germany (I can’t go in person for instance) which is why I also pushed for there to be online participation in the discussions.

Really, this is a continuation of the series of gatherings, workshops, and summer schools that have been held in Rome, Alanya, Perth, Pattaya, Graz, and in conjunction with regional opens in past years, dating back to 2004. It was important that those were in-person even though of course we fully realize that not everyone could join and thus the findings were published for the involvement of everyone else.

On that note, we would love to return to Thailand for another workshop/meeting in Bangkok or Pattaya! We fondly remember Jackrit and Mahidol University hosting us and running this event for several years in a row. There’s certainly room in the schedule to host a workshop there as well, which of course would be easier for teams in Asia and Australia to participate in.

Compensation and Further Workshop Details:
This is something that I’ll have to leave Oskar to talk about.

Please let us know if you have any further questions!

Cheers!

  • Raymond
#3

Hi Witthawin Sae-Lee,

Autonomy:
No one is advocating fully autonomous rescue missions. This does not make sense as every mission is new and there is no reptition in the whole mission and the response tasks must always be fit to the specific situation based on the expertise of the first responders.

However, autonomous assistance functions robots can make the work of the first responders more effective. For example, just teleoperating a mobile robot remotely over uneven terrain and narrow environments is already very challenging and resulting in fast fatique of the operator. First responders told us that they would very much use assistance function providing, e.g., autonomous navigation from A to B (with collision avoidance) with the mobile base or the arm or functions warning them when they might oversee a potential collision while teleoperating or providing situational awareness with virtual 3D views of the scence from different perspectives or having a autonomous return to area of wireless connection or home if once the wireless communication becomes unstable.

Unlike in RoboCup, in real missions the operator can not go at and through the scene and inspect it from all angles before sending the teleoperator robot. The robot has to be send first. Time is often crucial in a mission and is also a critical factor for operator mistakes due to time pressure.

Such helpful assistance functions could and should be well tested and promoted in RoboCup to help to advance the use of robots and related research.

Roadmap Workshop at RoboCup German Open:
For the workshop we aim to bring together a not too large group of very experienced RRL committee members and team leaders to brainstorm and work out a draft for an updated roadmap for the league. For such intensive discussion and brainstorming physical presence is crucial. Nevertheless we will enable remote participation. However, overall participation is limited.

The draft outcome of this workshop will be shared and discussed among all team leaders and teams via the forum, so that we can hopefully have a final discussion in a meeting at RoboCup in Bordeaux.

This workshop will not impact the rules for 2023, presented in yesterday’s meeting, but presumably in follwing years.

The total funding available for the workshop is $7,000, mainly to be used for travel cost of participants from Asia or Americas (though only a few can be funded from this budget).

We plan to send more information to team leaders by email soon.

Best,
Oskar

#4

Hi All,

With regard to autonomy, I just want to clarify. Of course no-one is advocating that entire rescue missions should be fully autonomous, as an overall concept, for all the aforementioned reasons.

However, the proposal on the table is that a repetition in the lane must be completed fully autonomously to be awarded the 4x multiplier. Assistive functions are not eligible for the 4x multiplier if they require the operator to touch the controls during the repetition.

Assistive functions would be rewarded by the benefit they bring in terms of speed, and, if they enable the team to operate with degraded communications, the proposed 2x multiplier (see other thread for discussion).

Cheers!

  • Raymond
#5

As Oskar pointed out, autonomy is important to alleviate operators from tedious tasks so they can focus on hard tasks and decision making. Maybe I can shed some more light onto why the multiplier has now been increased to 4x.

For recent years, we already awarded autonomous operation with a 2x multiplier. This lead to the intended development of autonomous functionalities mainly for the easier tasks in the maneuvering category. Harder tasks like stair climbing, traversal of rough terrain or manipulation are still dominated by tele-operation of expert operators which also know the arena beforehand. Still, especially in these areas there is a great potential to develop autonomous capabilities to assist the operator. To make efforts of developing autonomous capabilities in these arenas feasible, we increased the autonomy multiplier.

However, if we also consider the new 2x multiplier for network degradation which can also be utilized while fully tele-operating, the difference to the new autonomy multiplier is again only 2x. I hope, with these two changes, the multipliers are fair for every team while also providing venues for innovation.