AeroGuard-MAS Documentation¶

AeroGuard-MAS is a university software engineering project that implements a simulated multi-agent system for airspace conflict management.

The system models aircraft, routes, flight levels, weather zones, separation constraints, conflict detection, symbolic reasoning, planning, explanation, event logging, and replay visualization.

The project combines:

• a Kotlin simulation core;
• Jason / AgentSpeak(L) BDI agents;
• tuProlog symbolic reasoning;
• a small STRIPS-inspired planning layer;
• JSON scenarios;
• JSONL event logging;
• a Python Streamlit GUI for replay-based visualization;
• tests for prevent errors and validate behavior;
• automated tests and CI/CD through GitHub Actions.

Documentation Structure¶

This documentation is organized as follows:

• Abstract: project summary and objectives.
• Domain: airspace simulation domain and main concepts.
• Design: architecture, components, and design rationale.
• Tech Stack: technologies, tools, and runtime environment.
• Code: important implementation details and intelligent behavior.
• Testing: testing strategy and validation scope.
• Deployment: CI/CD pipeline and automation.
• Conclusion: results, limitations, and future work.

Main Execution Flow¶

A typical AeroGuard-MAS run follows this flow:

flowchart TD
    Scenario["JSON Scenario"] --> Loader["Kotlin Scenario Loader"]
    Loader --> State["Initial Simulation State"]

    State --> Engine["Managed Simulation Engine"]

    Engine --> Detector["Conflict Detector"]
    Engine --> Reasoner["tuProlog Safety Reasoner"]
    Engine --> Planner["STRIPS Resolution Planner"]
    Engine --> Applier["Maneuver Applier"]
    Engine --> Replanning["Weather Replanning Service"]

    Detector --> Engine
    Reasoner --> Planner
    Planner --> Engine
    Applier --> Engine
    Replanning --> Engine

    Engine --> Events["JSONL Simulation Events"]
    Events --> GUI["Python Streamlit Replay GUI"]

Example Commands¶

Run the full Kotlin test suite:

./gradlew test

./gradlew run --args="--scenario scenarios/simple_conflict.json --events build/aeroguard/events/simple_conflict_events.jsonl --explain"

cd gui
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
streamlit run app.py

cd gui
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
streamlit run app.py

Kotlin Docs¶

Kotlin Docs are available here