Self-Calibrating Sensing
Continuous Median Absolute Deviation (MAD) envelope tracking automatically adapts to heavy bus leaf springs, passenger cars, or scooters.
NO MANUAL CONFIGURATIONGLOBAL ROAD SENSING
Hodos turns everyday phone observations into high-precision road condition maps. No manual logging, no phone-buzzing distraction, and zero privacy compromise. Just drive, ride, or cycle—the road maps itself.
Production release 1.2.0. Clean, autonomous road detection on any paved or unpaved route worldwide.
Inspect crowdsourced potholes, speed humps, and pavement roughness corridors across hundreds of kilometers. Filter defects and export GIS layers.
Open Full-Screen Map →Road conditions evolve every day after rains, frost, and heavy vehicle traffic. Hodos extracts pavement intelligence quietly in the background without needing you to touch your phone.
Continuous Median Absolute Deviation (MAD) envelope tracking automatically adapts to heavy bus leaf springs, passenger cars, or scooters.
NO MANUAL CONFIGURATIONA single bump never publishes a defect. Hodos requires multiple independent passes and distinct devices before a hazard is confirmed.
BAYESIAN VALIDATIONRaw impact clusters are converted into Uber H3 hexagonal spatial bins and International Roughness Index (IRI) data for city planners.
OPEN GEOSPATIALTraditional inertial profiler surveys are slow and expensive. Hodos provides open, crowdsourced road condition maps directly to city road departments, municipal engineers, and citizens worldwide—completely free and public.
Quarter-car dynamic simulation calculates true International Roughness Index (m/km) per street segment.
Discrete global spatial indexing provides uniform, boundary-distortion-free cell aggregations ready for BigQuery and PostGIS.
Export GeoJSON and GPX files directly into QGIS, ArcGIS, or civil maintenance dispatch work orders.
8a3c029910affff (Res 10)Download open GeoJSON datasets to plan resurfacing or verify repairs in your neighborhood.
Hodos runs quietly in pockets, backpacks, and dashboard mounts of everyday commuters. Together, our daily rides build a living public record of road conditions without phone distraction.
Real road observations actively mapped and recorded in the database.
Impulse shocks and deep surface craters detected by accelerometer motion sensors.
Continuous surface rutting, aggregate wear, and sunken utility trenches.
Elevated speed tables and traffic calming humps identified along mapped routes.
The first and last 250 meters of every mapped trip are trimmed on-device so your home, school, and workplace are never exposed.
Only compact, aggregated hazard events are uploaded. Continuous accelerometer sensor streams never leave your device.
The public hazard map and GIS exports are free to browse and audit without handing over an email or phone number.
Client software is MIT for public auditability. The crowdsourced road dataset is protected under CC BY-NC-SA 4.0: free for public and civic use, commercial licensing required for corporate fleets.
Hodos runs a lightweight background service that evaluates continuous accelerometer and gyroscope data with Median Absolute Deviation (MAD) envelope tracking. When you pass over a pothole or severe bump, the sudden impulse is recorded and verified without manual input.
Continuous noise-floor tracking dynamically adjusts sensitivity. Continuous chassis vibration from buses, trucks, or motorcycles automatically raises the baseline filter so only true road surface shocks trigger an observation.
The client app and core algorithms are open-source under the MIT License. However, the crowdsourced road defect dataset is protected under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 (CC BY-NC-SA 4.0) license. Citizens, universities, and public communities can explore and download it freely for non-commercial use. Commercial operators, logistics fleets, and ride-hailing platforms cannot exploit or resell the data without an enterprise commercial license.
No. Haptic vibration is disabled so your phone stays quiet in your pocket or vehicle mount without distracting you.
Available on Android & Web worldwide. Software MIT · Dataset CC BY-NC-SA 4.0.