Location Tracker

Inspiration

The idea for Location Tracker was born from the desire to bridge the physical and digital worlds by:

• Addressing real-world challenges in asset tracking
• Offering a cost-effective alternative to expensive commercial tracking systems
• Showcasing the scalability and efficiency of serverless architecture
• Providing open-source infrastructure for developers building location-based applications

What it does

Location Tracker enables real-time tracking of physical assets using low-cost, GPS-enabled devices backed by a scalable cloud infrastructure. While GPS hardware integration is part of the vision, for testing purposes the system currently accepts manually submitted coordinates via an API. These coordinates are processed, demonstrating the core functionality of the system.

How we built it

• Backend: Developed entirely on AWS using a serverless architecture. Core services include AWS IoT Core, Lambda, DynamoDB, EventBridge, SQS, DynamoDB Streams, and both REST and WebSocket APIs via API Gateway.
• Deployment: Infrastructure is managed and deployed using AWS CDK for repeatable, maintainable, and scalable provisioning.

Challenges we ran into

• Optimizing cost while maintaining the performance of serverless services.
• Designing a backend capable of ingesting and processing rapid bursts of location data.
• Ensuring the solution remained both affordable and scalable without sacrificing accuracy or responsiveness.

Accomplishments that we're proud of

• Successfully processed test coordinates with minimal latency using a fully serverless architecture.
• Open-sourced the entire solution to encourage collaboration and community growth.
• Demonstrated that high-performance tracking can be achieved at a fraction of the cost of commercial systems.

What we learned

• Serverless architectures can scale seamlessly and reduce operational overhead significantly.
• Gained deep insights into secure communication between devices and cloud services using MQTT and TLS.
• Learned how open-source tooling accelerates development, fosters innovation, and broadens accessibility.

What's next for Location Tracker

• Offline Prediction: Integrating machine learning using Amazon SageMaker (Reinforcement Learning) and Amazon Bedrock (Generative AI) to predict asset behavior when offline.
• Advanced Analytics: Enhancing the dashboard to include historical route tracking, intelligent alerts, and movement analytics.

Built With

• amazon-dynamodb
• amazon-lambda
• cdk
• dynamo-streams
• event-bridge
• node.js
• rest
• serverless
• websockets