Police Support System

This project was implemented as part of my Master's Thesis. The system was designed to improve the decision-making process when managing police units in a smart city. It's achieved by using contextual data provided by agents such as a Police Patrol, a Navigation Agent, and a Smart Gun. A concept known as MAMS¹ was used to achieve independence between agents. In order to test the system, a simulation was implemented. To summarise, 6 different microservices were implemented (WebApp, Simulation, Patrol Service, Navigation Service, Gun Service, HQ Service), and also a front-end application was created using React. The microservices send messages between themselves using RabbitMQ. The simulation uses OSM maps imported to a PostgreSQL database to simulate routes on a real city map. Every microservice was dockerized and the whole infrastructure can be set up using a single docerk run (see docker-compose.yaml).

Technologies

Technologies used in this project:

• .NET 7
  • ASP.NET Core
  • Autofac
  • Npgsql
  • RabbitMQ.MessageBus
  • RabbitMQ.Client
  • SignalR
  • Serilog
  • Reinforced.Typings
  • Riok.Mapperly
  • Nito.AsyncEx
  • Newtonsoft.Json
  • DotNetZip
  • BAMCIS.GeoCoordinate
• React
  • MobX
  • Leaflet
    • React Leaflet
  • SignalR
  • Ant Design
  • Tailwind CSS
• RabbitMQ
• PostgreSQL
  • PostGIS
  • pgRouting
  • osm2pgrouting
  • osm2pgsql
• Grafana
  • Grafana Loki
• Docker

Footnotes

1. Multi-Agent MicroServices ↩