Section 01
Problem Statement
A city project must predict traffic queues, noise cones, dust, and complaints. With only a phone: GPS + OSM + the camera count traffic; the mic samples dB with a calibrated AudioRecord reading to log site noise; GPS + camera provide a dust-visual plume proxy via an air-quality API fallback or an image-based haze classifier; complaints come in via phone voice + text — all on the phone.
Section 02
Solution Requirements
Traffic queue from GPS + camera
Hold the phone at a real street corner (or a parking-lot test site) and start a 30s road counter; TFLite vehicle detection classifies trucks vs. cars per frame, computing flow and time-to-clear, e.g. "truck flow at peak -48%" vs. baseline after a mitigation window.
Noise map from phone mic & GPS
Sample the phone microphone (calibrated dBFS→dB SPL) at 3 points (0/50/100m) from the site while walking and logging distance via GPS; draw dB contours on a map, and flag if a nearby school (OSM Overpass) exceeds 55 dBA.
Vibration proxy via accelerometer
Hold the phone on a fence or near a real moving vehicle/street to record accelerometer peak-to-peak readings as a proxy for vibration attenuation vs. distance, with gravity filtered out.
Dust model
Pull an air-quality API for the site's GPS point plus an activity schedule, generating a PM10 plume via a Gaussian-plume-lite model, or use a camera + TFLite haze classifier to flag "Air looks hazy → high dust."
Community feedback in a phone app
A "Report noise" button captures voice + location and pushes the complaint to a map dashboard with red pins and an SLA counter for last-complaint-handled, visible on a second reviewer's phone.
Unified phone dashboard
A mobile-optimised map/card view shows traffic, noise, dust, and complaints together; a "Mitigation ON" toggle (e.g. shifting delivery windows) recomputes queue and noise live.
Section 03
Expected Solution / Outcomes
- A user at a street corner/parking lot counts cars with the camera, showing a "32 veh/min" overlay.
- Walking from the site while recording mic + GPS draws a noise map.
- A calibrated dB reading near a real car/speaker/AC shows > 70 dB.
- A complaint with GPS is pinned on a map visible from a different reviewer's phone.
Section 04
Technology & Hardware Constraints
Data: OSM amenity data, PM10 history, dB calibration table.
Section 05
Scoring
Requirements Completeness
50
Each requirement (R-id) is worth 3–12 marks. A hard-coded or mocked claim earns 0 — only working code counts.
Presentation & Demo
50
3-slide pitch + video ≤ 7 min + live demo. Judges must understand the problem in ≤ 90 seconds.
Real-world Implementation & Feasibility
50
Would a PM, engineer, or worker install this with only a phone on Day 1 of the next project? Must work offline and cheaply. Field-trial videos prove real-environment execution (20+ pts of this score is the field video).
Innovation & Disruption
50
A novel fusion of phone sensors (e.g. IMU + camera + barometer = scaffold-instability detector) is rewarded over simple API-calling. Re-skinning an existing SaaS scores low.
Section 06
Bonus (+10)
Generate a temporary public-space SVG/printable PDF plan (pedestrian detour, pop-up green space) that the phone's PdfDocument prints for real on-site use.
Section 07
Deliverables
| Item | Format |
|---|---|
| Video Demo | 7 min showing the four phone-based measurements. |
| Field Trial Videos | (a) Real traffic count filmed from the phone's live view. (b) Real noise/vibration reading near a street corner or parking lot. (c) A complaint filed by a partner walking with the phone. |
| Demo Link | Android APK download link (installed on the reviewer's own phone) or a mobile-optimised web app URL. |
| GitHub | audio_dB/, traffic_cv/, dust_model/, complaints/, dashboard/. |
| Pitch Deck | 3-slide PDF (Problem → Solution → Field-Proof & Impact). |
| Technical Brief | 2-page PDF: architecture diagram, phone-sensor map, tech stack, limitations, PRIVACY.md. |
Section 08
Field-Trial Requirements
- Real traffic count filmed live from the phone.
- Real street-corner noise/vibration reading.
- Real complaint filed by a partner with GPS attached.
Section 09
Privacy Requirements
Any human face captured in a field video must be consented or blurred. A PRIVACY.md file is required in every repository.