Picture a city after an earthquake: buildings collapsed, fire spreading, roads blocked and hundreds trapped. You are not the commander — instead you must program the intelligence of each responder so they decide for themselves. That is RoboCup Rescue Simulation.
The structure
In the agent-based version you write code for three kinds of responder:
- Fire brigades — extinguishing fires and stopping their spread.
- Police forces — clearing blocked roads so everyone else can move.
- Ambulance teams — rescuing and transporting casualties before it is too late.
Each agent sees only part of the city, can communicate only so much, and is racing the clock.
What scores
The final score typically combines lives saved, the remaining health of casualties, and how much of the city survived the fire. You cannot optimise one objective alone; you must balance conflicting ones.
Key skills
- Decentralised multi-agent coordination — no central commander.
- Clustering and task allocation — dividing the city between responders.
- Path planning on a dynamic graph — roads that block and reopen.
- Prioritisation and resource allocation.
- Bandwidth-limited communication — agents cannot tell each other everything.
Why it suits Iranian students
Like Soccer Simulation, this needs only a computer and programming skill. Iranian teams have a strong international record in simulation and the route in is wide open. Pedagogically the subject is also unusually motivating: students see their code translate directly into lives saved.
Frequently asked questions
What are the prerequisites?
Solid programming (usually Java, C++ or Python) and a grasp of algorithms and data structures. No special hardware.
How many rescue simulation branches are there?
Mainly two: agent-based simulation (managing city responders) and virtual robot simulation, which is closer to the physical arena.
Is it suitable for high-school students?
For strong programmers, yes — but CoSpace or 2D Soccer Simulation is usually the easier starting point.
Want to work on multi-agent AI?
Novin Zehn competition teams train from zero to the world stage — in Tehran and live online.
Register / free consultation AI & machine learning course