robocup-rescue-v

robocup-rescue-v@lists.robocup.org

February 2025

1 participants
1 discussions

[Reminder] RoboCup 2025 Rescue Agent Simulation Competition - Call for Participation
by Farshid Faraji 08 Feb '25

08 Feb '25

Call for Participation 2025 RoboCup Rescue Agent Simulation Competitions to be held in Salvador, Brazil – July 15-21, 2025 Co-located with the RoboCup 2025 https://2025.robocup.org IMPORTANT DATES Registration……………………............……..: February 20, 2025 (23:55 UTC) Acceptance Notification……………….……..: February 25, 2025 TDP submission……………………………….: April 15, 2025 (23:55 UTC) Reviewed TDP submission…………….....…: May 04, 2025 Camera-ready TDP submission……............: June 04, 2025 Tournament dates………..............………….: July 15-21, 2025 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 ADF Java: - Documentation: Java ADF manual: https://roborescue.github.io/adf-sample-agent-java/ - Quick Start Example in Java: https://github.com/roborescue/adf-sample-agent-java Python (New): - Adf-core-python documentation: https://roborescue.github.io/adf-core-python/ - Quick Start Example in Python: https://github.com/adf-python/adf-sample-agent-python - You can reuse JAVA ADF modules in ADF python. We encourage all the teams to develop their new modules using Python ADF Docker: - Repository: roborescue/rcrs-docker ( 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 in the 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 in the Infrastructure competition must release their source code as an open-source project PRIOR to July 1, 2025. 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 REGISTRATION Teams must register via the submission system available at https://easychair.org/conferences/?conf=rcrs2025 no later than February 15, 2025 (23:55 UTC). This 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. Teams are not requested to submit a TDP for registering to participate in the 2025 RoboCup Agent Competition. Registered teams will be accepted to participate in the competition based on a first-come first-serve basis constrained by the allowed number of teams per country. We encourage teams to develop their code using the new RCRS Core Python in order to take advantage of state-of-the-art Machine Learning tools and algorithms, such as SciKit, TensorFlow, PyTorch, among others. Qualified teams have to submit their TDP by April 15, 2025 (23:55 UTC), otherwise the team will be disqualified. Qualified teams will be allowed to participate remotely. GETTING STARTED GUIDE Here are some pointers to get started with the agents development TDP SUBMISSION Accepted 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=rcrs2025 no later than April 15, 2025 (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=rcrs2025. Teams should prepare their Agent Competition TDP based on the templates available at https://github.com/roborescue/rcrs-templates. All TDPs will be peer-reviewed by at least two PC members. The suggestions of the reviewers will be released on May 04, 2025 and they should be incorporated into the camera-ready version of the TDP. The camera-ready version should be uploaded to EasyChair no later than June 04, 2025. * Previous TDPs can be found in the Participating Teams links https://rescuesim.robocup.org/past-events/past-competitions/ * If you have any questions please reach out to rcrs2025(a)easychair.org RULES Please read the Agent Simulation Competition Rules document available at 2025 RoboCup 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. Best Regards Farshid Faraji on behalf of RoboCup Rescue Simulation League

1 0