ROADRELAYOperations console
Hosted demoRC
RoadRelay / Demo workspace

ROAD
OPERATIONS.

Review the signal. Coordinate the next inspection.

iHosted simulation — 20 synthetic vehicles. Changes stay in this browser. Real sensor input uses the local application.
Scenario vehicles
—synthetic fleet
Recorded passes
—not sample count
Awaiting review
—inspection candidates
Checkpoint coverage
—/ 3demonstration route

Los Angeles · USC area

Illustrative road network

Simulated location
W Jefferson BlvdExposition BlvdS Figueroa StS Vermont AveW 30th St University ofSouthern CaliforniaExposition Park 110
N
High impactModerateUnobserved
Schematic · not for navigation · no measured GPS
Illustrative fleet route

Inspection queue 0

Evidence first. Human decisions.

Highest impact first
iRepeated signals support inspection, not a confirmed pothole.

Impact evidence Synthetic recording

Select an anomaly to inspect its strongest recorded pass.

Peak impact—gNo observation

Measured acceleration is not pothole depth.

ImpactModerate > 0.20 gHigh > 0.60 gSynthetic waveform · recorded scenario
Observation

Connect your prototype

Keep the original sketch. The separate RoadRelayWiFi sketch adds telemetry without changing your thresholds or LEDs.

1

Join the same network

Connect this computer and the UNO R4 WiFi to a trusted 2.4 GHz network or hotspot with device-to-device access.

2

Configure the Wi-Fi sketch

Open firmware/RoadRelayWiFi/RoadRelayWiFi.ino. In arduino_secrets.h, set the network, this computer’s LAN IP, and the receiver token below.

Loading…

UDP port 4210

3

Upload, then record

Select Arduino UNO R4 WiFi in the IDE and upload. Choose Live prototype here, select checkpoint A, B or C, then press Start recording before rolling the toy car.

USB fallback · original sketch supported

Close Serial Plotter and Serial Monitor, then run:

python3 serial_bridge.py --port /dev/cu.YOUR_ARDUINO

The bridge reads the same Impact / Small / Big data and forwards it to this console.

No sensor connected yet.

The receiver runs locally. No data is sent to a government agency. AI requests only run when you explicitly generate an AI brief.

A 30-second RoadRelay demo

  1. Show the real signal. Connect the toy car, switch to Live prototype and start recording at checkpoint B. Roll it over a bump.
  2. Follow the evidence. The waveform spikes and an inspection candidate appears. Finish recording. Repeat to add evidence to the same candidate.
  3. Make a decision. Open the candidate, review its signal, choose a team and assign an inspection. Generate a rule-based or AI brief.
  4. Show the larger scenario. Switch to Fleet simulation to demonstrate 20 synthetic vehicles and separate road authorities. Say explicitly that the fleet and locations are simulated.
Presentation wording

“We demonstrate live impact sensing with simulated location tagging. A future field deployment would add GNSS or authorized fleet-location data.”

Green: this observation did not exceed the threshold. Yellow: moderate impact. Red: high impact, suspected road anomaly, inspection needed.

Scenario fleet

20 synthetic vehicle IDs. Three illustrative passes per vehicle. These are not connected vehicles, customers, or field results.

Research & prototype limits

Secondary research informs the workflow. No agency affiliation, interviews, or field deployment is claimed.

What this prototype can establish

A live sensor can register an impact and attach it to a manually selected, simulated location. An impact is not a confirmed pothole or a measurement of its depth. No actual GPS, road accuracy, repair savings, automated dispatch, or engineering pavement rating is demonstrated.