AI indoor navigation service
Infrastructure development for indoor tracking application, the project from A to Z.
Abto Software was contracted by a mapping and accessibility organization providing indoor navigation services. Our company has covered application refactoring, software development, cloud development, cloud migration, and several other services to host the implemented inference service.
During the project’s course, our team was extended to provide DevOps practices and legacy code refactoring. Moving further over the project’s course, we provided MLOps practices to deliver a quality control solution, which allowed our team to evaluate the accuracy of the inference service.
Navigation app infrastructure development: Our goals
The indoor navigation service was presenting several challenges that impacted the overall business efficiency, so we primarily focused on optimizing:
Response times and costs
- by ensuring efficient allocation of directed computational resources
- by allowing the system to scale on demand and avoid unnecessary consumption
Software development and release, thereby facilitating:
- Competitive advantage
- Meeting evolving customer demands
- Shorter time-to-market
- Greater maintainability and adaptivity to changes
Navigation app infrastructure development: The solution in details
The indoor navigation service is designed and developed to allow any company guide visitors though spaces, thus bringing additional benefit to organizations across transportation, retail, recreation, and others.
Having built-in visual and audio guidance for accessibility, the solution can provide autonomous navigation. What’s more, the system can provide additional information about facilities, thus improving user experience and enhancing business reputation and image.
The concept is simple:
- The user first scans the location to capture required information
- The data is stitched to create a map
- The user then manages the map to control the permissions and options for interaction
- And, finally, the visitors can navigate the location using both visual and audio clues
The service is made up of:
- A custom web application for the client’s internal tech personnel
- A custom mobile application for the end user
- And the inference service that extracts specific features from images to match those features with the scanned model of the added building or point of interest
Our contribution
We covered:
- Business analysis
- Software development
- AI development
- Cloud development
- Code refactoring
- DevOps services
- Thorough testing (manual testing, unit testing)
- Technical support and maintenance
In particular:
- In-depth research of models considering performance and accuracy
- Event-driven architecture for optimized response times and costs
- DevOps practices and legacy code refactoring
- MLOps services for quality control solution
Web and mobile stack: TypeScript, Node.js, JavaScript, DynamoDB, Postgres, Python, PyTorch, Snowflake, Grafana, Plotly Dash, API Gateway Kong
DevOps and MLOps services: OpenVino, PyColMap, Fast API, EKS stack, AWS, Azure, Flask, Kubernetes, Prometheus, Bitbucket, RabbitMQ, Redis, Knative, Kserve, GraphQL, Airflow, Terraform, HELM
The challenges
During the project’s scope, we faced multiple challenges associated with:
The improvement of the SLAM system
The size and similarity of spaces within some larger buildings has complicated image capture and processing. To handle this challenge, our team has autotuned the thresholds for each pipeline microservice and minimized the false positive responses.
The hosting of the SLAM system
The application requires intensive computational resources for adequate response times at an acceptable cost. To solve this challenge, our engineers have built message brokers and divided the components into individual pipeline microservices.
The scaling of the delivered solution
To scale the capabilities and enable the scanning of different facility types, including airports, train stations, schools, museums, and more, we delivered a custom Infrastructure-as-a-code system for quick, one-click setup of clusters.
The monitoring of performance, user behavior, and scaling
To ensure high uptime and facilitate user experience, we established individual monitoring within clusters.
Summing up
Abto Software was focused around replacing an inefficient inference service to accelerate user experience, simultaneously optimize response times and costs, and streamline localization precision.
Our client can now explore wider, forward-looking opportunities and leverage:
New customers, higher demand, and expanded revenue streams
By accelerating navigation precision, thus improving user experience, the client can attract more customers and leverage future opportunities for growth. and profit
Business agility
By streamlining software development and release, optimizing time and cost, and achieving greater scalability, the client can introduce new features and updates more frequently, invest strategically, and adapt to changes without compromising business performance and reliability.