Call for Participation

2022 RoboCup Rescue Agent Simulation Competitions

to be held in Bangkok, Thailand — July 11-19, 2022

2022 RoboCup Rescue Simulation https://rescuesim.robocup.org/upcoming-events/2022-bangkok-rescue-simulation/

RoboCup 2022

https://2022.robocup.org

IMPORTANT DATES

Pre-Registration……………………………......: February 15, 2022 (23:55 GMT)

Team Description Paper (TDP) submission…: February 28, 2022 (23:55 GMT)

Qualified teams announcement…………..…..: March 15, 2022

Camera-ready TDP submission………….……: May 31, 2022 (23:55 GMT)

Tournament dates………………………..……..: July 11-19, 2022

WAIVER OF THE TEAM FEE FOR NEW TEAMS

The RoboCup Federation is pleased to introduce a waiver of the team fee for the 2022 international RoboCup competition for NEW teams in the major leagues. A NEW team is defined as a team all of whose team members have never participated in an annual international RoboCup competition. The waiver concerns only the team fee and does not imply any waiver of fees for team members.

NEW FEATURES IN THE SIMULATOR

Teams can now choose to implement their agents in Java or Python

Python repo: roborescue/rcrs-sample-agent-python (github.com) https://github.com/roborescue/rcrs-sample-agent-python

AGENT SIMULATION COMPETITION

This initiative focuses on promoting the use of Artificial Intelligence to support teamwork coordination and decision-making of ground rescue teams in a post-disaster urban setting. It maintains a computer simulation platform that can realistically represent natural disaster scenarios where Urban Search and Rescue (USAR) plans can be assessed. This platform is a comprehensive artificial simulation environment that can be used as a testbed to the effectiveness of coordination strategies of USAR teams, coalition formations of heterogeneous teams, rescue plans to explore large-scale environments with limited set of resources to search for survivors, disaster response public policies, among others.

The RoboCup Rescue Agent Simulation Competition involves primarily evaluating the performance of teams of agents in regards to their distributed coordination and planning algorithms for rescuing civilians in a city after an earthquake. The goal is evaluating the effectiveness of Ambulance Team, Police Force, and Fire Brigade agents on rescuing civilians in earthquake scenarios. Here are some of the challenges that can be tackled in this competition:

-

Decision Making Algorithms -

Task Allocation Methods -

Multi-Agent Coordination and Team Formation Methods -

Path Finding -

Search Algorithms

GETTING STARTED GUIDE

Here are some pointers to get started with the agents development

-

ADF Java: -

Repository: roborescue/adf-sample-agent-java: RoboCup Rescue Agent Development Framework Sample Agent Java (github.com) https://github.com/roborescue/adf-sample-agent-java -

Documentation: Java ADF manual https://github.com/roborescue/adf-sample-agent-java/blob/master/docs/manual.adoc -

[NEW] Python: -

Repository: roborescue/rcrs-sample-agent-python (github.com) https://github.com/roborescue/rcrs-sample-agent-python -

[NEW] Docker: -

Repository: roborescue/rcrs-docker (github.com) https://github.com/roborescue/rcrs-docker -

Using docker is optional for teams that wish to have an environment that is already setup for development

INFRASTRUCTURE COMPETITION

The Infrastructure competition involves the presentation of tools or simulators related to disaster management issues already developed by the participating team. The aim is to evaluate possible enhancements and expansions of the RoboCup Rescue Simulation League environment (both the agent simulator and frameworks and the virtual robot platform) based on the new ideas and concepts proposed in these tools and simulators. The evaluation will be done by a panel of experts and the winner chosen according to a set of factors related to the technical aspects of the tool or simulator and the presentation. The winner tool will be selected for further integration with the simulation platform.

The participation on Infrastructure competition requires the submission of a detailed Team Description Paper (TDP) describing the tool or simulator. In addition to the submission of the TDP, teams interested in participating on the Infrastructure competition must release their source code as an open-source project PRIOR to July 2, 2022.

Participant teams should address topics including, but not limited to

-

Improvements to the Virtual Robot Simulator -

Improvements to the Agent Development Framework -

Increase the number of participants on the RoboCup Rescue league -

Provide agent development kits that render easy the participation of newcomers -

Integrate the Agent and Virtual Robot competitions -

Improve the RoboCup Rescue sub-simulators

PRE-REGISTRATION

Teams must register via the submission system available at https://easychair.org/conferences/?conf=rcrs2022 no later than February 15, 2022 (23:55 UTC). This pre-registration requires only teams to fill the "Title" field and the name, institution, and contact information of all team members. The "Title" field should be "Agent_[TeamName]" to participate in the Agent Competition and “Infrastructure_[TeamName]” for the infrastructure competition, where [TeamName] should be replaced by the name of the team. The pre-registration does not imply that the team will be evaluated to participate in the competition, please see section QUALIFICATION below for further information.

QUALIFICATION

Teams must submit a Team Description Paper (TDP) (in English) describing the strategy and recent advancements implemented in the team. The TDP is limited to 16 pages and must be electronically submitted as PDF through the submission system available at https://easychair.org/conferences/?conf=rcrs2022 no later than February 28, 2022 (23:55 UTC).

The TDP file name has to be "Agent_[TeamName]" for the Agent Competition, where [TeamName] is the name of the team submitting the TDP. This format should be also used in the "Title" field in the submission system https://easychair.org/conferences/?conf=rcrs2022.

Teams should prepare their Agent Competition TDP based on the templates available at https://github.com/roborescue/rcrs-templates . (overleaf link)

All TDPs will be peer-reviewed by at least two PC members. The evaluation criteria of contributions will be based on originality, quality, clarity, and its relevance to the league's aims.

The suggestions of the reviewers should be incorporated into an improved version (i.e., Camera-Ready version) of the TDP. The camera-ready version should be uploaded to EasyChair no later than May 31, 2022.

-

The previous TDPs are accessible here: https://robocup-rescue.github.io/team_description_papers/ -

Qualification results will be announced on March 15, 2022. -

If you have any questions please reach out to r******2@easychair.org

RULES

Please read the Agent Simulation Competition Rules document available at 2022 RoboCup Rescue Simulation https://rescuesim.robocup.org/upcoming-events/2022-bangkok-rescue-simulation/ for further details.

ABOUT ROBOCUP RESCUE AGENT SIMULATION LEAGUE

Natural disasters are major adverse events that cause large-scale economic, human, and environmental losses. They are usually difficult to predict and even more challenging to prevent from happening. These characteristics demand disaster management strategies to be in place to mitigate damaging consequences after a natural disaster.

The mission of the RoboCup Rescue Agent Simulation League is to promote research and development in the socially significant domain of natural disaster with focus on multiagent teamwork coordination and decision support systems. The effective implementation of this mission is translated into three main objectives. First, the league aims to provide a simulator able to realistically represent natural disaster scenarios where response rescue plans can be assessed. Second, it aims to define evaluation benchmarks for response plans elaborated by policy-makers. Finally, it aims to promote research and development by organizing competitions to stimulate the exchange of ideas and experience between researchers and practitioners. These aims are designed to help in the development of more sophisticated and formalized response plans to effectively respond to natural disasters and reduce the negative impacts on society.