)
Xweather RoadAI
Road-tested AI for road surveys
- Same-day results
- No specialist equipment
- Independently validated
Automated AI detection of road defects, markings, and assets, independently validated by national transport authorities in the UK, Denmark, and France.
)
Proven technology used by agencies worldwide
Xweather RoadAI has been in continuous use since 2018, surveying millions of kilometers annually for road authorities, municipalities, and infrastructure managers across multiple countries.
0 ×
more data at half the cost
0 million km
surveyed annually
Since 0
in continuous operational use
0 ×
more consistent than human surveyors
Detect every defect, marking, sign, and asset in a single pass
RoadAI's detection catalog covers the full range of road condition and asset data that agencies need for planned maintenance, safety inspections, and reporting.
- Surface conditions
- Road furniture
- Same-day processing
)
Continuous surveying whenever vehicles are on the network
Traditional survey methods restrict agencies to one expensive survey per year. Minor defects go unrecorded until they deteriorate into costly repairs. RoadAI enables planned preventative maintenance.
- Data-driven maintenance planning
- Find and fix minor defects early
- From reactive repair to early interventions
)
Road asset management on your schedule
RoadAI replaces expensive annual contractor surveys with continuous data collection that you control, enabling earlier intervention and better allocation of maintenance budgets.
Planned maintenance
Comprehensive network condition data
RoadAI captures surface defects, IRI values, and a complete video record in a single survey pass. Agencies build a continuously updated picture of network condition without commissioning separate surveys for each data type. Results are available the same day as the drive, replacing the weeks-long turnaround of traditional contractor assessments.
Safety inspections
Safety inspections
RoadAI captures a complete video record of every surveyed route alongside classified defect data. Inspectors can record, annotate, and verify conditions remotely without return site visits.
The video archive provides a time-stamped visual record for each road segment, supporting safety audits and reducing the need for repeated physical inspections. Data is available same-day, compared to weeks or months with traditional approaches.
)
Asset management
Road asset inventories
RoadAI automatically detects and catalogs road furniture including signs, guardrails, barriers, drainage structures, lighting, and utility poles during every survey pass. Agencies build and maintain a comprehensive asset inventory without dedicated inventory surveys. As vehicles drive the network for any purpose, the inventory updates continuously, keeping records current without additional cost or effort.
Liability protection
Liability protection
Complete video records and classified condition data provide agencies with the evidence base needed for budget requests, funding applications, and liability defense. When a claim is made about a road defect, agencies can reference the video archive and survey data to demonstrate the condition of the road at a specific date, the maintenance actions taken, and the decision-making process. This documentation supports both proactive budget cases and reactive liability responses.
4× more data, half the cost
Affordable, repeatable surveys replace expensive annual contractor assessments.
Actionable from day one
Classified defect data is available the same day as the survey drive. RoadAI's AI models detect defects, markings, and assets accurately from the first pass, with no tuning or calibration required for each network.
Anyone can collect data
Install the app, mount a smartphone in any vehicle, and drive. No specialist surveyors, training, or equipment.
Complete network documentation
A time-stamped video history and classified data record for every surveyed route supports reporting and liability defense.
RoadAI turns any vehicle into a survey platform. No specialist equipment, training, or personnel required.
01
Install
Download and install the RoadAI app on a standard smartphone. No specialist hardware, calibration, or vehicle modifications are needed to begin collecting survey data.
02
Mount
Mount the smartphone on the windshield of any vehicle. RoadAI works in cars, vans, trucks, or any vehicle that drives the road network as part of normal operations.
03
Drive your network
Drive the routes you need to survey. Data collection happens automatically during normal driving. Surveys can run during routine operations, inspections, or dedicated passes.
04
AI processes and classifies
Automated AI processes the captured footage, detecting and classifying defects, markings, signs, and assets with geo-located precision. Results are available same-day.
Computer vision built for road networks
RoadAI's computer vision models detect and classify defects, markings, and assets from standard smartphone video. The same models tested and validated by national transport authorities in the UK, Denmark, and France process every survey. When agencies use RoadAI, they use the proven AI, not a model that needs tuning to their network.
- Proven AI from day one
The AI models are pre-trained and validated. No network-specific configuration, calibration, or training period required.
- Same-day results
AI processes footage and delivers geo-located, classified data within hours of the survey drive.
- Complete video archive
A time-stamped video record of every surveyed route. Navigate the network remotely and reference conditions at any point in time.
Reliable detection of defects and assets
Powered by computer vision, RoadAI detects multiple distresses, fixes, and road assets. Talk to our team for the complete catalog.
Distress detection
Identify distress by type and severity. Work with asset-specific data and overall road surface condition ratings.
Potholes
Alligator cracking
Longitudinal cracking
Transverse cracking
Wheel track cracking
Fretting / raveling
Subsidence
Edge deterioration
Binder bleeding
PotholesCrackingFretting
Fix detection
Detect and categorize repairs to see where fixes were made and how they’re performing.
Crack sealing
Area patching
Spot patching
PatchingSealing
Traffic signs
Build and maintain accurate asset inventories with automatic categorization and condition assessment of traffic signs.
Prohibitory
Indication
Mandatory
Regulatory
Temporary
NameCodeCategory
Line and surface markings
Automatic detection and condition assessment of line markings. Detect and categorize other surface markings automatically.
Edge and center lines
Yield lines
Directional and lane use arrows
Speed limit markings
Crosswalks
Text and symbols
Regulatory markings
LongitudinalTransverseSymbols
Street furniture and drainage
Recognize and catalog roadside assets, including barriers, lighting, and drainage features.
Guard rails
Bollards
Traffic cones
Street lights
Maintenance holes
Storm drains
RailsLightsDrains
Special objects
Record annotations while driving to capture other assets and hazards. Add tags to categorize annotations for filtering and reporting.
Organization-specific assets
Debris and obstructions
Overhanging vegetation
Flooded areas
Safety hazards
AnnotationsTagsComments
Independently validated
RoadAI has been validated by transport authorities in three countries, meeting the highest standards required for official road survey and reporting use.
PAS 2161
RoadAI is approved by the UK Department for Transport for surveying road conditions under the PAS 2161 standard, delivering accurate, consistent, and repeatable network‑level data for compliant reporting and asset management.
Vejdirektoratet
The Danish Road Directorate selected RoadAI after formal tendering and acceptance tests against LCMS reference data, validating its repeatability, reproducibility, and ability to classify network‑level pavement condition on the Danish road network.
Cerema
Winner of the French national CIRR 2019 innovation call, with a Cerema Certificate of Achievement confirming the quality of RoadAI’s measurements relative to the LCPC ME 38‑2 M3 reference test method.
Privacy compliant
Every video collected with RoadAI is anonymized before it is available in the web app. The process masks vehicles and people in the video, ensuring no personally identifiable information is visible.
Security and quality built in
Certified to industry standards, ensuring trusted data, safe operations, and reliable performance.
ISO
9001
Standard for quality management systems
ISO
27001
Standard for security governance
See RoadAI in action
Watch how RoadAI collects, processes, and delivers road condition data from a standard smartphone mounted in any vehicle.
)
See what RoadAI finds on your network
Talk to our team about running a pilot survey on your roads. Results are typically available the same day as the survey drive.
- Works with any vehicle
- Results in hours, not months
- Unlimited seats included
)
Frequently asked questions
RoadAI delivers the same accuracy as laser scanning at a fraction of the cost and complexity.
In independent testing by the Danish Road Directorate (2025), RoadAI matched the accuracy of LCMS reference data with only 1% variation in measurements.
Repeatability: 99% consistency of results across 5 runs
Classification accuracy: 100% match with laser scanning rankings across 5 test sections
Reproducibility: RoadAI indices tracked LCMS reference curves throughout the test sections
In the UK, the Department for Transport has approved RoadAI for surveying road conditions under the PAS 2161 standard. This independent certification confirms that RoadAI delivers accurate road-condition data and can serve as a reliable substitute for laser-scanning surveys. Developed with the British Standards Institution and the UK’s Transport Research Laboratory, PAS 2161 is the most comprehensive technical evaluation of road condition monitoring technology ever carried out.