Why Planespect

What drone inspection delivers today

Several vendors — including Donecle, Mainblades, and others — now offer autonomous drone scanning for aircraft exterior inspection. ^{1, 2} A typical system completes a full narrowbody scan in under one hour using laser- or LiDAR-assisted navigation, AI image analysis, digital defect mapping, and 3D positioning. The output covers general visual inspections, paint quality, regulatory markings, and lightning-strike assessment — replacing manual processes that once required 6–12 hours and multiple technicians.

These systems have particular strength in paint evaluation. Automated algorithms map rivet rash and peeling across the aircraft skin and produce global paint-wear reports in roughly 45 minutes on a narrowbody. ³ Operators use these reports to schedule repainting for fuel savings and to monitor livery condition across the fleet. 3D sensors support high-accuracy dent measurement (down to 0.1 mm) and clear traceability for MRO quality gates. ⁴

This is valuable work. But the output remains tied to the aircraft's visible surface. Every defect is a coordinate on a photo or a 3D mesh — not a location inside the aircraft's structural architecture.

What Planespect adds

Planespect starts from the same rapid drone-based image acquisition but adds a structural interpretation layer and a multi-modal sensing architecture. Three capabilities distinguish it from surface-only scanning.

OEM-supported structural mapping, developed in collaboration with Embraer, aligns drone imagery with CAD-based structural models. Each defect is placed in its exact aircraft zone — load path, fatigue-prone area, or primary structure — rather than appearing as an isolated annotation on an image. Risk scoring then combines visual severity with the zone's engineering importance.

Model-wide learning exploits the consistency of structural models across an entire aircraft type. By aggregating findings from multiple airframes, the platform builds fleet-wide degradation baselines for every structural zone. When a new finding appears — say a dent near a fuselage splice — Planespect compares it against how similar findings in the same zone have progressed across dozens of other airframes to forecast whether it will remain stable or grow toward allowable limits. Systematically codifying experienced inspectors' accumulated expertise sharpens these forecasts with judgment that data alone cannot provide.

Modular multi-modal sensing extends the drone beyond visible light. The Phase 1 payload includes an optional radiometric thermal camera that captures subsurface evidence — delamination, debonding, moisture ingress — invisible to RGB alone. ⁸ Phase 2 adds pulsed eddy current for quantitative wall-thickness confirmation on flagged zones ⁹ and a hyperspectral pilot for objective coating-chemistry assessment. ^{10, 11} Each modality enters the platform through a bounded evaluation with defined pass/fail criteria, building on the visual core rather than replacing it.

Together, these capabilities convert a single inspection snapshot into a forward-looking maintenance plan. Operators gain not only objective records for warranty claims and lease transitions but also forecasts of when and where the next maintenance action will be needed — enabling data-driven paint programs, predictive zone-level scheduling, and measurable fuel-savings potential.

Side-by-side comparison

OEM structural mapping

Planespect's collaboration with Embraer integrates detailed CAD-based structural models directly into the inspection workflow. Every drone-captured image is aligned with the aircraft's internal architecture, placing each finding in its exact structural zone — primary load path, fatigue-critical spar, or high-stress fastener row — rather than only on a visible surface mesh.

Once a defect is located within its zone, the system applies risk scoring that combines two inputs: the visual severity of the defect (crack length, dent depth, corrosion area) and the engineering importance of the zone. A minor scratch in a low-load fairing receives different priority than the same scratch near a wing-root attachment point. The resulting report links each finding to the relevant Aircraft Maintenance Manual (AMM) references — allowable damage limits, repair procedures, and task-card mappings specific to that zone — eliminating the manual cross-referencing that traditionally consumes hours of engineering time.

Existing drone inspection tools also position defects on a 3D model, but their mapping relies on visible surface features — frames, stringers, ribs — recognizable in the imagery. ⁵ This provides good traceability and repeatability for general visual inspections and paint evaluation. It does not, however, incorporate the deeper OEM structural dataset or the zone-importance weighting that Planespect derives from Embraer's engineering data.

Because Planespect maintains a consistent structural model across the entire E-Jet narrowbody family, the same framework applies unchanged to every airframe of the same type. This consistency is the prerequisite for model-wide learning.

Model-wide learning and predictive intelligence

Because Planespect maintains a single structural model across every airframe of an Embraer narrowbody type, the platform can aggregate data beyond individual inspections.

Each inspection adds time-stamped findings to a longitudinal record for every structural zone. Over repeated scans on multiple airframes, the system calculates zone-specific wear rates, identifies repeat problem areas — for example, fastener-line cracking in a particular fuselage bay — and quantifies progression trends. Bayesian methods continuously refine risk predictions by combining fresh drone imagery with the accumulated fleet baseline: a finding that took 18 months to reach repair threshold on 30 other airframes sets a concrete expectation for the current one. The consistent OEM structural model ensures that data from one aircraft is directly comparable to every other aircraft of the same type, producing reliable fleet-level statistics without manual re-alignment.

This data-driven analysis is complemented by systematic collection of inspectors' tacit knowledge. Through structured interviews, Planespect codifies the expertise that veteran technicians apply but rarely articulate: subtle visual cues, pattern recognition heuristics, and judgment calls developed over decades of hands-on work. As Michael Polanyi observed, "we can know more than we can tell." ⁷ With an aging inspection workforce, embedding this knowledge in a searchable database before it is lost to retirement turns individual experience into a durable fleet resource.

The practical outcomes are direct. Maintenance planners receive prioritized work lists that concentrate engineering resources on the highest-risk zones, with progression forecasts that show which findings need attention now and which can safely wait until the next scheduled check. Paint-program decisions draw on objective, fleet-wide wear data that supports fuel-savings calculations for operators and lessors. For lease transitions and warranty claims, the platform produces auditable, OEM-aligned condition reports covering each zone's history and projected trajectory.

Drone inspection vendors already provide operator-level fleet traceability and paint-quality benchmarking across a narrowbody fleet. ^{3, 6} Planespect extends this by embedding OEM structural models and fleet data aggregation into the analysis. The result is a shift from reacting to visible defects on individual aircraft to anticipating type-level issues before they affect dispatch reliability or residual value.

Beyond visible light

Current drone inspection platforms rely exclusively on RGB cameras and 3D sensors. Detection is limited to what is visible on the painted surface: cracks, dents, peeling, discoloration. Defects beneath intact paint — early corrosion, coating delamination, moisture trapped in lap joints — remain invisible until they progress far enough to produce surface symptoms.

Planespect's modular payload architecture addresses this limitation in two stages.

Phase 1: Thermal subsurface triage. A radiometric thermal camera, carried alongside the RGB core on the same drone, captures infrared contrast during the standard scan path. In the controlled hangar environment, delamination, disbonds, and moisture pockets produce measurable thermal signatures under passive conditions without external heating. The AI triage layer fuses RGB and thermal findings: a paint anomaly that also shows thermal contrast beneath the surface receives higher priority than one that does not. Lightweight drone-ready thermal payloads such as the DJI Zenmuse H30T add negligible weight and no additional scan time. ⁸

Phase 2: Confirmatory metrology and spectral assessment. For zones flagged by visual and thermal triage, Phase 2 introduces two targeted capabilities. Pulsed eddy current provides quantitative wall-thickness measurement through paint on conductive metal skins, turning a visual flag into a measured thickness value. ⁹ Snapshot hyperspectral imaging in the visible and near-infrared range (400–1000 nm) captures spectral signatures that distinguish paint composition, UV degradation, and early chemical indicators of substrate corrosion beneath intact coatings. ^{10, 11} Both enter the platform as bounded pilots on high-value zones, building ground-truth data for the AI layer before committing to full-surface deployment.

Thermal, eddy current, and spectral sensing are not yet standard features of commercial aircraft drone inspection platforms, though some vendors are exploring thermal integration. By designing for multi-modal data from the start, Planespect can deepen its analysis as sensor payloads mature — without changing the core inspection workflow or the structural data model that ties every finding to OEM engineering data.

References

1. Donecle (2025). "Learn: Automated Aircraft Inspection FAQ." donecle.com/learn
2. Mainblades (2025). "Frequently Asked Questions." mainblades.com/faq
3. Donecle (2019). "Evaluate the Paint Quality of Your Aircraft in 45 Minutes." donecle.com
4. Donecle (2021). "Just One Hour: Drone-Based Automatic Dent Inspection of a Rafale Aircraft." donecle.com
5. Donecle (2025). "Solution." donecle.com/solution
6. Donecle (2021). "Regional Jet Center Selects Donecle's Automatic Drone for Aircraft Inspections." donecle.com
7. Polanyi, M. (1966). The Tacit Dimension. University of Chicago Press.
8. DJI (2025). "Zenmuse H30 Series — Multi-Sensor Payload with Thermal Imaging." dji.com/zenmuse-h30-series
9. Voliro (2025). "Drone-Enabled Pulsed Eddy Current (PEC) Sensor." voliro.com/pec-payload
10. Cubert (2025). "Ultris X20 Plus — Hyperspectral Camera for UAV Aerial Mapping." cubert-hyperspectral.com/products
11. Resonon (2025). "Pika L — Lightweight VNIR Hyperspectral Camera." 400–1000 nm, 281 channels. resonon.com/Pika-L