Situational awareness begins long before an aircraft leaves the gate. For pilots training in high-fidelity flight simulators, the digital environment must mirror the real world down to the precise latitude and longitude of runway thresholds, the exact height of approach lighting, and the correct geometry of taxiway intersections. When Aerosimulations, a developer of advanced virtual airfields, identified systematic geographic inaccuracies across its training environments, the company recognized an existential threat to its product's credibility. The solution required not just a one-time data scrubbing, but a fundamental restructuring of how geographic information was ingested, validated, stored, and distributed. By adopting Directus as a headless CMS and centralized data backend, Aerosimulations transformed its fragmented data landscape into a single, authoritative source of truth, dramatically improving pilot navigation accuracy and training outcomes.

The Anatomy of Geographic Error in Virtual Airfields

Geographic errors in simulation environments stem from a common origin: data drift. This occurs when the underlying data used to build a virtual environment becomes outdated, is transcribed incorrectly, or originates from conflicting sources. For Aerosimulations, the problems were diverse and pervasive.

Misplaced Runway Coordinates and Orientation

One of the most critical errors involved runway coordinate offsets. A runway threshold placed even 20 feet away from its true position can cause a pilot to misjudge the final approach fix. In several Aerosimulations airfields, the magnetic bearing published in the simulation differed from the real-world Aeronautical Information Publication (AIP) by several degrees. This discrepancy introduced negative transfer of training, where pilots would learn an incorrect visual scan pattern in the simulator that had to be unlearned in the aircraft.

Incorrect Obstacle and Terrain Elevation Data

An airfield's digital elevation model (DEM) must accurately represent surrounding terrain. Aerosimulations discovered that obstacle heights—such as control towers, hangars, and antenna masts—were frequently stored as static values disconnected from any authoritative survey. In one instance, a virtual airfield depicted a 200-foot obstacle where the real-world object had been demolished five years prior. Such errors create false hazard awareness, either desensitizing pilots to real dangers or causing unnecessary caution.

Outdated Taxiway and Ramp Configurations

Airports are living infrastructures. Taxiways are realigned, holding points are moved, and ramp areas are expanded to accommodate larger aircraft. Aerosimulations found that several of its virtual airfields retained taxiway layouts that were over a decade old. This led to confusion during ground operations training, where pilots relied on the simulator to learn airport surface markings and signage.

The Data Bottleneck: Spreadsheets, Silos, and Static Files

Prior to the correction project, Aerosimulations managed geographic data through a combination of spreadsheets, static JSON configuration files, and direct hard-coding within its Unreal Engine project. This workflow was brittle. When an FAA Notice to Air Missions (NOTAM) announced a change to a runway declared distance, updating the simulation required a developer to manually locate the relevant spreadsheet, update the value, re-export the data, and rebuild the simulation package. There was no audit trail, no version control, and no way for subject matter experts (SMEs) to verify changes without accessing the development environment directly. This data bottleneck stifled the company's ability to maintain its fleet of over 50 virtual airfields.

Architecting a Single Source of Truth with Directus

The cornerstone of the correction project was the adoption of Directus as the central data management platform. Aerosimulations needed a system that could model complex geographic relationships, support role-based editing, and integrate seamlessly with its existing 3D development pipeline. Directus provided an open, API-driven foundation that met all these requirements.

Relational Data Modeling for Airfield Infrastructure

Directus's flexible schema designer allowed the team to construct a precise relational model of an airfield. Instead of a monolithic data file, each airfield became a collection of interconnected tables:

  • Airports: Containing ICAO code, IATA code, city, country, magnetic variation, and master elevation.
  • Runways: Linked to the parent airport, with fields for length, width, true bearing, surface type, and a GeoJSON field capturing the centerline polygon.
  • Navaids: Linked to specific runways, storing ILS frequencies, DME offsets, and VOR coordinates.
  • Taxiways: Storing centerline paths as LineStrings, along with width and lighting status.
  • Obstacles: Point geometries with elevation, height above ground, and obstacle type (e.g., tower, building, tree).

By establishing foreign key relationships, the team ensured referential integrity. Updating the elevation of an airport in the Airports table could automatically flag all dependency-checking logic in the simulation backend.

Leveraging GeoJSON for Precision Geometry

Geospatial accuracy demands a robust data format. GeoJSON became the standard for storing all location-specific data within Directus. The use of the Point geometry type for navaids and the Polygon type for runway surfaces allowed the simulation engine to parse precise coordinates directly from the API without intermediate conversion. This eliminated the rounding errors that had previously occurred when coordinates were transferred between spreadsheets and configuration files. The Directus Maps interface was configured to provide a visual preview of these geometries, enabling SMEs to visually verify that a runway polygon overlay matched the satellite imagery reference.

Role-Based Access Control for Data Governance

Data integrity required strict governance. Aerosimulations used Directus's built-in Role-Based Access Control (RBAC) to segment responsibilities:

  • Viewers: Instructors and quality assurance personnel could review data and flag discrepancies but could not make changes.
  • Editors: Geographic SMEs could update coordinate data, adjust elevations, and modify geometry. All edits were subject to review.
  • Admins: Senior engineers and the project lead managed the schema, user permissions, and final publication rights.

This tiered approach prevented unauthorized changes while empowering the domain experts who understood the data best. No longer did a developer need to interpret a pilot's verbal request for a runway shift; the pilot (as an Editor) could make the change directly, and the developer (as an Admin) could approve the data flow.

Automating the Correction Workflow with Directus Flows

The true power of the system came from automating the correction pipeline. Aerosimulations leveraged Directus Flows to create an event-driven workflow that connected data validation, simulation export, and team communication.

Trigger-Based Data Validation

When an SME updated the displaced_threshold_latitude field on a runway, a Directus Flow was triggered. This Flow ran a series of automated validation scripts:

  1. Range Check: Did the new latitude fall within the expected geographic bounds of the airport (e.g., between 32 and 33 degrees North for KLAX)?
  2. Cross-Reference: Did the new coordinate match the value stored in the linked source_document table, which contained the official FAA 5010 data? If a discrepancy existed, the flow flagged the entry for manual review.
  3. Impact Analysis: The system queried the Approach Procedures table to determine if any instrument procedures were affected by the change.

If validation passed, the Flow automatically changed the runway status from Draft to Validated. If it failed, a detailed error log was sent to the editing team.

Bidirectional Data Synchronization

Aerosimulations designed the system to avoid a fragile, one-way export. Using Directus's REST API and WebSocket triggers, the simulation engine could query data directly at runtime. More importantly, the simulation could write data back to Directus. When a pilot completed a training session, the simulator logged any perceived geographic discrepancies. This feedback was automatically ingested into a Pending_Issues table within Directus, creating a closed-loop system where operational feedback directly informed data corrections.

Impact on Pilot Navigation and Training Fidelity

The results of the Directus-led correction project were quantified and significant. Aerosimulations rolled out the updates across its fleet of virtual airfields over a six-month period. The impact was measured across three dimensions: data accuracy, training efficiency, and regulatory compliance.

Quantifiable Data Accuracy Improvements

The correction project audited and updated over 1,500 geographic data points across 54 airfields. The results were stark:

  • 99.7% of runway coordinates were brought within 3 meters of surveyed positions.
  • 1,200+ obstacle records were updated or added, using survey-grade data from the FAA Airport Data 5010 database.
  • 67 taxiway routing errors were corrected, eliminating a common source of student confusion during ground operations training.

Reducing Negative Transfer of Training

The most critical KPI for Aerosimulations was the reduction of "negative transfer"—the phenomenon where simulator training actively harms real-world performance. By aligning the virtual environment with the most current ICAO charts and NOTAMs, the company saw a measurable drop in pilot errors during check-rides. Instructors reported that students transitioning from the simulator to the aircraft required fewer corrections on airport surface navigation and instrument approach procedures. One senior instructor noted that the updated airfields allowed students to focus on decision-making skills rather than compensating for simulator inaccuracies.

Streamlined Regulatory Compliance

Flight simulation training devices (FSTDs) are subject to strict qualification standards under regulations such as 14 CFR Part 60 and CS-FSTD. These regulations require that the simulator's aerodynamic and visual databases accurately represent the real world. Aerosimulations used Directus to generate comprehensive audit reports for each airfield, demonstrating the source, validation status, and revision history of every geographic data point. This drastically reduced the administrative burden of regulatory audits.

Technical Deep Dive: The Digital Backbone

The correction project required a stack of specialized technologies, each playing a specific role in the data pipeline. Directus sat at the center, orchestrating the flow of geographic intelligence.

Directus as the Data Orchestrator

Directus served as the master system of record. Its primary functions included:

  • Content Repository: Storing all structured geographic data with relational integrity.
  • User Interface: Providing a no-code interface for SMEs to manage complex geospatial data without writing SQL or JSON.
  • API Gateway: Exposing data via a standardized REST and GraphQL API for consumption by the simulation engine and web-based dashboards.
  • Workflow Engine: Using Directus Flows to automate validation, transformation, and notification tasks.

GIS and 3D Mapping Integration

To visualize and validate geographic corrections, Aerosimulations integrated Cesium for Unreal into its development pipeline. Corrected GeoJSON data from Directus was streamed into the Unreal Engine environment using the Cesium ion platform. This allowed the team to overlay the corrected runway polygons and obstacle points directly onto high-resolution satellite imagery and terrain elevation. Any remaining misalignment between the simulated surface and the real-world imagery was immediately visible, validating the correction before it was deployed to training devices.

Continuous Integration and Deployment

Aerosimulations established a CI/CD pipeline that automatically packaged updated airfield data into simulation builds. When an SME published a change in Directus, a webhook was sent to the team's build server. The server queried the Directus API for the updated dataset, compiled the assets into the Unreal Engine package, and deployed it to a staging environment for final validation. This reduced the deployment cycle from a manual, multi-day process to an automated, hour-long pipeline.

Future Innovations: A Living Data Platform

The success of the correction project has positioned Aerosimulations to build a truly dynamic training ecosystem. The lessons learned from the Directus implementation are paving the way for several ambitious enhancements.

Real-Time NOTAM and Data Integration

The team is developing a system to ingest real-time aeronautical data feeds directly into Directus. When a NOTAM is published for an airport in the Aerosimulations fleet, it will be automatically parsed and correlated with the affected runways or taxiways. A Directus Flow will evaluate the impact of the NOTAM and automatically create a task for an SME to review and update the virtual airfield. This will allow the simulation environment to reflect the real-world operational status within hours of an official notice.

AI-Assisted Discrepancy Detection

Manual audits are thorough but slow. Aerosimulations is exploring the use of machine learning models to automate discrepancy detection. By feeding the Directus-stored geographic data into an AI model trained on satellite imagery, the system can proactively identify changes in the real world—such as a new construction project near an airport—and flag the corresponding data entry for review. This shifts the maintenance model from reactive correction to proactive intelligence.

Community-Driven Contribution Loops

Pilots and instructors using Aerosimulations' products will soon be able to submit geographic feedback directly through a web portal powered by Directus's frontend API. This feedback will be structured using a custom feedback form (built with Directus's intuitive UI), ensuring that bug reports are detailed and actionable. These submissions will populate a Community_Reports table, giving the development team a prioritized list of data corrections from the people who fly the simulators every day.

Setting a New Standard for Geographic Fidelity

The Aerosimulations case study demonstrates a fundamental truth about modern simulation: the fidelity of the virtual environment is ultimately limited by the quality of the underlying data. By confronting the geographic errors in its virtual airfields and adopting Directus as a centralized, open, and automated data platform, the company did more than fix a few coordinate offsets. It built a sustainable system for maintaining data accuracy over time.

For the pilots training in those corrected environments, the benefit is tangible. The runway is where it should be. The approach lights align with the procedures. The obstacles exist only where real obstacles stand. This fidelity reduces cognitive load, builds accurate mental models, and ultimately contributes to safer flight operations. Aerosimulations has proven that with the right data architecture, virtual airfields can be trusted as authoritative representations of the real world—a standard that every training organization should aspire to achieve.