AI Travel Planner: Your Intelligent Guide to Unforgettable Journeys

Abstract

The AI Travel Planner is a revolutionary platform that crafts personalized travel itineraries using the power of Google Gemini 2.5. Input your budget, travel dates, and interests, and our multi-agent system delivers tailored destination suggestions, detailed daily plans, and seamless booking options—all within a sleek, accessible interface. Built with Flask, SQLAlchemy, and Bootstrap 5, this tool redefines travel planning with AI-driven insights and a user-friendly design. This publication details the system’s architecture, agent collaboration, setup instructions, and support status, with all resources available in our public GitHub repository and a video demo.

Introduction

Planning a trip can be daunting—balancing budgets, destinations, and activities is no easy task. The AI Travel Planner simplifies it all with artificial intelligence, offering personalized recommendations, comprehensive itineraries, and booking management in one intuitive platform. Certified under AAIDC-M2 standards, this publication provides a clear, engaging overview of the system’s design, implementation, and usage.

Why It’s Awesome

• Personalized Planning: Tailored destination and itinerary suggestions based on your unique preferences.
• AI-Powered Insights: Google Gemini 2.5 delivers smart recommendations and local tips.
• Seamless Booking: Manage flights, hotels, and activities from a single platform (simulated for now).
• Accessible Design: A modern, mobile-first interface built with Bootstrap 5, meeting WCAG accessibility standards.

System Design

The AI Travel Planner is modular, scalable, and designed for ease of use, leveraging a multi-agent system and a robust tech stack.

Architecture Overview

The platform comprises four core components:

• Core Application: Flask-powered backend for handling user requests and agent coordination.
• AI & Services: Google Gemini 2.5 for recommendations, paired with mock booking services.
• Data Layer: SQLAlchemy with SQLite/PostgreSQL for storing preferences and itineraries.
• Frontend: Bootstrap 5 and Vanilla JavaScript for a responsive, dynamic UI.

Project Structure:

ai-travel-planner/
├── Core Application
│   ├── app.py              # Flask app setup
│   ├── main.py             # Application entry point
│   └── routes.py           # API routes
├── AI & Services
│   ├── gemini.py           # AI recommendation logic
│   └── travel_service.py   # Mock booking services
├── Data Layer
│   ├── models.py           # Database models
│   └── instance/           # Database files
├── Frontend
│   ├── templates/          # Jinja2 HTML templates
│   ├── static/css/         # Custom CSS
│   └── static/js/          # JavaScript for interactivity
└── Documentation
    ├── README.md           # Project guide
    └── docs/               # Architecture diagrams

Tech Stack

The Multi-Agent System

Four specialized agents collaborate seamlessly to plan your perfect trip.

Agent Roles

1. Preference Agent:
   • Collects your travel preferences (budget, dates, interests).
   • Stores data in SQLAlchemy and outputs a JSON package.
2. Recommendation Agent:
   • Suggests three destinations tailored to your preferences.
   • Uses Gemini 2.5 to provide insights and budget estimates.
3. Itinerary Agent:
   • Creates a detailed daily plan with activities and dining options.
   • Stores itineraries in the database via SQLAlchemy.
4. Booking Agent:
   • Simulates booking for flights, hotels, and activities.
   • Outputs confirmations stored and displayed to the user.

Agent Workflow

• Communication: Agents exchange data via internal API calls and the database.
• Orchestration: Flask (routes.py) manages agent interactions:
  • Submit preferences at /preferences.
  • View recommendations at /destinations.
  • Access itineraries at /itinerary.
  • Book via /book/<type>.
• Data Flow: JSON ensures modular, organized data exchange.

Agent Interaction Flow

The chart below visualizes the data flow between agents, showing the number of API calls per request cycle (sample data).

Implementation Details

Backend

• Flask App (app.py, routes.py): Handles user sessions and API routing.
• Database (models.py):
  • TravelPreference: Stores budget, dates, and interests.
  • Itinerary: Saves trip plans with activities and budgets.
• AI Integration (gemini.py): Custom prompts for Gemini 2.5 to generate recommendations.

Frontend

• Templates (templates/): Jinja2 for dynamic HTML rendering.
• Styling (static/css/style.css): Bootstrap 5 with a dark theme and WCAG compliance.
• Interactivity (static/js/): Vanilla JavaScript for form validation and dynamic features.

Key Endpoints

Installation Instructions

Follow these steps to set up the AI Travel Planner locally.

Prerequisites

• Python 3.11+
• Git
• A Google Gemini API key (sign up at Google Cloud)
• Optional: Docker for containerized deployment

Step-by-Step Setup

1. Clone the Repository:

   git clone https://github.com/AishwaryaChandel27/AI-Travel-Planner.git
   cd AI-Travel-Planner

2. Install Dependencies:

   pip install -r requirements.txt

3. Set Environment Variables:

   export GEMINI_API_KEY="your-gemini-api-key"
   export DATABASE_URL="sqlite:///travel_planner.db"  # Or PostgreSQL URL

4. Initialize the Database:

   python -c "from app import db; db.create_all()"

5. Run the Application:

   python main.py

6. Access the App: Open http://localhost:5000 in your browser.

Docker Deployment

1. Build the Docker image:

   docker build -t ai-travel-planner .

2. Run the container:

   docker run -p 5000:5000 -e GEMINI_API_KEY="your-gemini-api-key" ai-travel-planner

3. Access at http://localhost:5000.

Usage Guide

1. Enter Preferences: Visit /preferences to input your budget, dates, and interests.
2. Explore Destinations: Check /destinations for three AI-curated suggestions.
3. View Itinerary: Select a destination at /itinerary for a detailed plan.
4. Make Bookings: Use /book/<type> to reserve flights, hotels, or activities.
5. Review Bookings: See your history at /my_bookings.

Maintenance and Support

• Status: Actively maintained with regular updates to improve functionality and security.
• Support:
  • Documentation: Comprehensive guides in the repository’s Wiki.
  • Issue Tracking: Report bugs or feature requests on GitHub Issues.
  • Community: Join GitHub Discussions for help and collaboration.
• Updates: Follow the repository for release notes and changelogs.
• Contact: Reach out via GitHub for direct support.

Licensing

The AI Travel Planner is licensed under the MIT License, allowing free use, modification, and distribution. See the LICENSE file in the repository for full details. This open-source license ensures transparency and encourages community contributions while protecting user rights.

Resources

• GitHub Repository: https://github.com/AishwaryaChandel27/AI-Travel-Planner
  • Full source code, documentation, and setup guides.
  • Adheres to PEP 8 and AI/ML best practices.
• Video Demo: Available on Ready Tensor, showcasing the platform’s features.
• Architecture Diagram: Located in docs/ folder of the repository.

Contribute

We welcome contributions! To get started:

1. Fork the repository.
2. Create a feature branch (git checkout -b feature/your-feature).
3. Commit changes (git commit -m "Add your feature").
4. Push to the branch (git push origin feature/your-feature).
5. Submit a pull request.

Please follow PEP 8 guidelines and include tests for new features.

Acknowledgments

• Google Gemini: For intelligent AI recommendations.
• Flask Community: For a robust web framework.
• Bootstrap Team: For a responsive, accessible UI framework.
• Font Awesome: For intuitive icons.

Get in Touch

Have questions or ideas? Check the repository’s Wiki, report issues on GitHub, or join our Discussions community. Let’s make travel planning unforgettable together!