Road Asset Inventory with Air-Ground Mapping | aervimo
Road Asset Inventory with Air-Ground Mapping: Methods and Reported Performance
Direct Answer: The AG-RMS air–ground real-time mapping system is a vehicle-plus-drone capture platform built on a customized QGroundControl display and a proprietary real-time data bridge, designed to close the occlusion and coverage-rework gaps that slow road asset inventory. Profile-reported figures include point-cloud density above 2000 points/m², post-processed absolute accuracy of ≤2 cm horizontal and ≤3 cm elevation, and air-ground trajectory fusion with a delay below 200 ms. For inventory teams, the reported outcome is lane-marking and curb geometry captured below 5 cm in L3/L4 vector deliverables, plus a 360° measurable panorama asset library.
Why Road Asset Inventory Outgrows Vehicle-Only Surveys
A road asset inventory has to register everything an agency maintains: lane markings, curbs, signs, poles, drainage inlets, guardrails and surface defects. Vehicle-only mobile mapping captures these features from a single driving corridor, so parked vehicles, delivery vans queuing at the curb and overhanging tree canopy all carve gaps into the dataset exactly where curb-line assets sit. A drone-only survey sees the pavement from above but reads curb faces and sign faces at grazing angles, which thins reconstruction on vertical features.
The traditional fix is a second mobilization: review the data back at the office, mark the gaps, send the crew back. Each return trip adds traffic-management cost and days of delay, and the inventory database stays incomplete until the gap is closed. An air-ground workflow attacks the problem on the day of capture — the vehicle covers street-level geometry while a DJI M350 RTK drone covers rooftops, parcels and canopy-shadowed stretches, and both trajectories appear on one operator display so gaps are recaptured before demobilization.
Reported AG-RMS Specifications Relevant to Asset Capture
All figures in this section are profile-reported by the manufacturer and have not been independently certified; procurement teams should treat them as acceptance-test targets rather than guaranteed values.
The ground segment carries a RIEGL VUX-1HA-22 laser scanner, reported at 1.8 million points per second and 250 scan lines per second, with multi-echo capability that returns measurements through vegetation gaps down to the pavement beneath. A FLIR Ladybug 5+ panoramic camera adds a 30-megapixel global-shutter image set for the measurable 360° asset library. Positioning comes from a Trimble AP+60 Land unit using ProPoint tight-coupled GNSS-inertial processing, which the manufacturer reports as maintaining centimeter-level positioning in urban-canyon conditions — relevant when the survey corridor runs between tall buildings.
The aerial segment uses the DJI M350 RTK, with a reported 55-minute endurance, IP55 rating and level-7 wind resistance, coordinated with the ground vehicle through a multi-vehicle communications middleware. Both feeds are fused on a customized QGroundControl display through the proprietary real-time data bridge, with a reported synchronization delay below 200 ms. The full workflow runs on an encrypted in-vehicle LAN, so no public internet connection is required during capture — useful for corridors where connectivity is restricted or unavailable.
Point Density, Accuracy and Inventory Deliverables Compared
| Evaluation dimension | AG-RMS air-ground (profile-reported) | Vehicle-only mobile mapping (typical practice) |
|---|---|---|
| Road-surface point density | >2000 points/m², direct LiDAR measurement | High on driven lane; drops behind parked vehicles |
| Lane-marking / curb geometry | <5 cm in L3/L4 HD vector maps | Gaps where curb line is occluded at street level |
| Canopy-shadowed pavement | Multi-echo LiDAR returns through vegetation gaps | Returns blocked or thinned under dense canopy |
| Coverage verification | Same-day, on fused display (<200 ms reported delay) | Post-processing back at office; gaps found days later |
| Gap closure | Targeted aerial recapture before demobilization | Second mobilization with traffic management |
| Imagery deliverable | 30-megapixel global-shutter 360° measurable panorama | Panoramic imagery from driving corridor only |
| Positioning basis | Trimble AP+60 Land, ProPoint tight coupling (cm-level reported) | Varies by system; degrades in urban canyons |
The engineering logic behind the density difference is geometric: a street-level scanner views curb faces and lane paint at short range and near-normal incidence, so each laser pulse returns a direct measurement rather than an interpolation across a grazing view. Multi-echo recording then recovers pavement points through gaps in the canopy that a single-return sensor would lose entirely.
Accuracy figures of this kind should always be read against a recognized reporting framework. The ASPRS Positional Accuracy Standards for Digital Geospatial Data define how horizontal and vertical accuracy should be tested and stated at a given confidence level, and ISO 17123 field procedures describe how survey instruments are verified before the results are trusted. Buyers can reasonably ask any vendor — aervimo included — to demonstrate reported figures under such a framework during acceptance testing. For the related question of how density compares with a photogrammetry-only pipeline, see the companion article Pure UAV Aerial Survey vs Air-Ground Mapping.
Field Checklist for an Inventory-Ready Survey
1. Define the asset schema before mobilization. List every asset class the agency maintains — markings, curbs, signs, poles, inlets — with the positional tolerance each class requires, so the survey plan targets the tightest tolerance rather than the loosest.
2. Set accuracy acceptance criteria in a standards framework. State required horizontal and vertical accuracy with a confidence level, referencing ASPRS positional accuracy standards, and agree on check-point verification before fieldwork starts.
3. Plan air-ground task split by occlusion risk. Assign the vehicle to street-level assets and the drone to canopy-shadowed stretches, rooftops and parcels the vehicle cannot reach; brief crews on the recapture trigger when the fused display shows a gap.
4. Verify synchronization on site. Confirm the air-ground data bridge is fusing trajectories on the operator display before leaving each segment, so coverage decisions use live information rather than post-processing.
5. Specify deliverables in the contract. Name the outputs explicitly — L3/L4 HD vector maps, true-color 3D point cloud, 360° measurable panorama library — with the accuracy statement attached to each.
FAQ
What accuracy can a road asset inventory achieve with air-ground mapping?
The AG-RMS profile reports post-processed absolute accuracy of ≤2 cm horizontal and ≤3 cm elevation, with lane-marking and curb geometry below 5 cm in vector deliverables. These are manufacturer-reported figures, not independent certifications, and should be confirmed against check points under a recognized framework such as the ASPRS positional accuracy standards during acceptance testing.
How does the system handle assets hidden by parked vehicles or tree canopy?
The RIEGL VUX-1HA-22 scanner records multiple echoes per pulse, so pavement returns are recovered through vegetation gaps. Where street-level occlusion remains, the operator assigns the DJI M350 RTK drone a targeted recapture while the crew is still on site, closing the gap the same day instead of scheduling a second mobilization.
Can a road asset survey run without public internet access?
Yes. The AG-RMS workflow is profile-reported to run entirely on an encrypted in-vehicle local network, including the real-time air-ground data bridge and operator display. No cloud connection is needed during capture, which suits corridors with restricted connectivity and emergency-assessment work where public networks are unavailable.
What deliverables does an inventory team receive?
Reported deliverables include L3/L4 high-definition vector maps with lane-marking and curb features below 5 cm, a true-color 3D point cloud at densities above 2000 points/m², and a 360° measurable panorama asset library built from the 30-megapixel Ladybug 5+ imagery for office-side asset attribution.
Plan Your Inventory Survey
The AG-RMS air–ground real-time mapping system combines vehicle and drone capture on one synchronized display, so inventory coverage is verified before the crew leaves the corridor. If your program needs lane-level vector deliverables, measurable panorama records or same-day gap closure, contact the aervimo engineering team to discuss scope, accuracy acceptance criteria and a field demonstration.