Overview
TectoTrack is a cutting‑edge social digital twin platform by Morphotect Design Group Inc. It creates a live virtual replica of real-world environments, enabling real-time simulation and prediction of pedestrian movements and social interactions. This empowers architects and planners to harmonize spaces and human experiences.
The platform integrates advanced agent-based modeling with real-world data streams to create dynamic simulations that evolve with actual usage patterns. By bridging the physical and digital worlds, TectoTrack provides unprecedented insights into how people interact with built environments.
Developed as a research-driven tool for urban designers and architects, TectoTrack combines academic rigor with practical applications, helping professionals create more human-centered, responsive spaces.
Key Features
Core Technology
Real-time Simulation
- Continuously updates the digital twin using live input data to reflect dynamic changes
- Synchronizes with IoT sensors and environmental data streams
- Supports both real-time and time-lapsed simulation modes
Agent-based Modeling
- Simulates diverse human behaviors for wayfinding, social clustering, and safety scenarios
- Customizable agent profiles with different movement patterns and preferences
- Adaptive learning algorithms that evolve based on observed behaviors
Analytics & Visualization
Spatial Analytics
- Visualizes flow metrics, density maps, and interaction hotspots
- Heatmaps showing concentration and movement patterns
- Conflict detection and bottleneck analysis
Scenario Testing
- What-if analysis for different design configurations
- Emergency egress simulation and safety evaluation
- Capacity planning and utilization forecasting
My Contributions
Data Pipeline
- Extract geometry and metadata from BIM files in Revit
- Convert to standardized format for Unity imports
- Implement automated update triggers
Structured raw BIM data flows into simulation-ready formats, ensuring seamless integration between design and analysis tools.
Simulation & Data Analysis
- Process simulation outputs to derive performance metrics
- Define evaluation indices and benchmarks
- Generate comparative reports
- Conduct exploratory and statistical analyses on sensor and simulation datasets
- Transform raw output into actionable visualizations and dashboards
- Validate model accuracy through comparative trend analysis
Processed and analyzed simulation and sensor data to extract key performance indicators and actionable insights, guiding model validation and iterative improvements.
Perception Algorithms
- Develop vision tracking and gaze direction models
- Implement isovist computation for spatial analysis
- Integrate cognitive wayfinding models
Built agent perception frameworks that simulate human vision and decision paths within virtual environments.
Computational Geometry & Linear Algebra
- Compute isovists and visibility graphs for spatial representation
- Apply vector and matrix operations for agent movement and rotations
- Optimize geometric algorithms to enhance performance
Leveraged mathematical frameworks to support precise spatial computations fundamental for agent-based simulations.
Playground Digital Twin Implementation
Case Study: Urban Play Space Analysis
Implemented a digital twin for a public playground to study children's movement patterns and equipment usage. The system integrated:
- Real-time position tracking from wearable devices
- Equipment interaction monitoring
- Social grouping detection algorithms
- Safety zone violation alerts
Figure: Simulation output showing heatmap of playground usage patterns during peak hours
Reference
Founder and CEO of Morphotect
-
Ali Jabbari
Principal Investigator
Senior Construction Manager