SwiftRoutes

An Intelligent Delivery Optimization System - "Smart routes for seamless deliveries"

Description:

The system manages and optimizes delivery routes for a fleet of vehicles. Each vehicle is modeled as an autonomous agent that makes decisions based on:

1. Real-time traffic conditions.
2. Assigned delivery orders.
3. Weather conditions.
4. Interaction with other vehicles (to avoid route conflicts).

Objective:

Demonstrate how a backend system can handle the complexity of a MAS using Spring AI to enable cognitive capabilities in agents and Kotlin for business logic development.

System Architecture

1. Key Components:

   1. Agents (Vehicles): • Each vehicle is an autonomous agent. • Decides its routes using AI models integrated into the backend.
   2. Central Controller (Optional): • Coordinates communication among agents and supervises global rules.
   3. Backend (Spring AI + Kotlin): • Implements MAS logic and connects to AI models. • Exposes REST APIs for traffic data, order assignments, and delivery monitoring.
   4. External Data Sources: • Traffic APIs (Google Maps, OpenStreetMap). • Weather APIs.

2. Backend Flow Diagram: Orders -> Backend (Spring AI) -> Agents -> Decisions (Routes/Actions) -> Execution (Deliveries)

Technologies:

Spring AI: A Spring Boot application that integrates AI models for decision-making in agents.

Kotlin: A modern programming language for backend development.

OLlama: A MAS library for Kotlin.

PostgreSQL: A relational database for storing orders and traffic data.

Docker: Containerization for easy deployment.

TODO: Add InfluxDB: A time-series database for storing real-time traffic data.