ASAS - Active Situational Awareness System

Inspiration

The ASAS concept was inspired by a real-world incident during a major public event—the Frankfurt Marathon. On the race day, temperatures dropped suddenly. Under these cold and windy conditions, a professional athlete withdrew and remained stationary at an aid station without adequate thermal protection. Coordinated road closures and fragmented operational domains prevented both event logistics and emergency services from reaching the athlete in time. While relevant data existed across weather services, event management, and transportation systems, no unified situational awareness emerged to highlight the escalating risk.

This incident revealed a broader pattern of urban systemic failure: critical risks often arise not from a single anomaly, but from the interaction of multiple subsystems that lack a shared operational picture.

What it does

ASAS doesn’t make decisions for the city — it makes the city legible to decision-makers.

ASAS (Active Situational Awareness System) is a cognitive situational awareness layer for cities. It actively reduces information entropy by denoising and integrating fragmented signals across sensors, events, weather, mobility, and infrastructure.

Rather than making decisions, ASAS transforms weak, distributed signals into decision-grade intelligence, enabling human operators to clearly understand what is happening, what is likely to happen next, and where attention and resources should be prioritized.

How we built it

ASAS is powered by Gemini 3 as its core cognitive engine. The system grounds sensor anomalies against live external evidence, reasons across interdependent urban sectors under pressure, and integrates multimodal inputs such as drone imagery and telemetry.

All intelligence outputs follow structured schemas, allowing insights to remain traceable, auditable, and directly usable by existing command and monitoring systems.

Challenges we ran into

The primary challenge was avoiding information overload while preserving critical context. Urban systems generate massive volumes of data, but clarity — not quantity — is what decision-makers lack under pressure. Designing ASAS to reduce noise without oversimplifying complex cross-domain interactions required careful abstraction and constraint.

Accomplishments that we're proud of

• Abstracting a real-world failure into a generalized, city-scale awareness problem.
• Designing a system that prioritizes cognitive clarity over automation.
• Demonstrating how multimodal AI can elevate weak signals into actionable insight without replacing human judgment.

What we learned

We learned that many urban risks remain invisible not because data is missing, but because no system is responsible for integrating meaning across domains. Situational awareness is not a dashboard problem — it is a cognition problem.

What's next for ASAS - Active Situational Awareness System

Next, we plan to expand ASAS across additional urban domains, refine entropy metrics for different risk classes, and explore deeper integration with city command centers. Our goal is to make ASAS a foundational awareness layer that cities can rely on before crises fully emerge.

Built With

• entropy
• gemini-3-pro
• google-search-grounding
• lucide
• multimodal
• react
• reasoning-engine
• recharts
• svg-mapping
• tailwind
• thinking
• typescript