PROTRACK

**Product and the concrete problem it solves.** PROTRACK attacks one of the most acute and fast-worsening problems in modern airborne operations: what a drone or aircraft does when its satellite navigation fails. GPS/GNSS jamming and spoofing have moved from exotic electronic-warfare edge cases to everyday conditions across Ukraine, the Middle East, and increasingly civil airspace, and most unmanned and manned platforms degrade badly the moment their satellite fix is corrupted. PROTRACK's answer is a software layer that lets a platform navigate and geo-reference the world using the one sensor it almost always already carries — a camera. Its flagship module, LOCATOR, provides GNSS-free positioning by continuously testing GPS and inertial (INS) telemetry, detecting when it is drifting or corrupted, and correcting position from imagery instead. A companion module, FLIGHT, performs airborne geo-positioning: it detects and identifies user-defined stationary and moving targets in live or recorded video, plots their real-world coordinates on a map, and can generate an in-mission ortho-mosaic (an aerial map stitched on the fly). The pitch is not a new drone but a resilience-and-exploitation layer that can be added to platforms that already exist.

**Core technology and how it actually works.** The technical heart of PROTRACK is "video anchoring" — the accurate, per-pixel matching of real-time and offline video to geographic maps, from which precise real-world coordinates are extracted directly from the footage. Around that sits a proprietary convolutional-neural-network analytics stack for object detection, classification and tracking, event detection (smoke, fire, leaks), change detection, and feature extraction/image matching, working across daytime, thermal, IR, and SWIR imagery so the capability survives into night and degraded-visibility conditions. Two design choices give the approach commercial leverage. First, it is hardware-agnostic software: PROTRACK states it requires no specialized hardware and runs on standard PCs and commercial single-board computers (Linux or Windows), integrating with platform controllers and C5I/C2 systems rather than demanding a bespoke navigation box. Second, it is architected as a family of interoperable products — LOCATOR (positioning), FLIGHT (geo-reference and mapping), OMNIUM C2 (a scalable command-and-control application for terrain dominance and wide-area situational awareness), plus specialized tools such as Foresight (long-range coordinate extraction across day/thermal/IR/SWIR), SeeOn (wide-area PTZ surveillance), WideEye (orientation for ground and low-altitude platforms), and Musaic (automatic ortho-rectified GeoTIFF mosaic construction). The unifying idea is that a camera plus PROTRACK's algorithms can substitute for, or cross-check, satellite navigation and deliver map-grade coordinates from imagery.

**Market, customers, and go-to-market.** PROTRACK sells into an unusually broad set of buyers for a company its size, precisely because navigation-denial is a shared failure mode across civil and military users. Its stated markets are UAV/drone manufacturers, defense and security, transportation, commercial services, infrastructure and civil services, and homeland security. The go-to-market is software-and-integration led: because the product installs on standard compute and plugs into existing controllers and C5I systems, PROTRACK can be embedded by drone OEMs and system integrators, or sold to end users who want to harden fleets they already operate. That model lowers the adoption barrier — operators do not have to redesign an airframe to gain GPS-denied resilience — but it also means PROTRACK's commercial fate is tied to winning design-ins and integration relationships rather than selling a marquee hardware platform. The company is headquartered in Jerusalem (68 Kanfey Nesharim St.) and positions itself as a worldwide vendor of computer-vision, GNSS-free positioning, and video-analytics technologies.

**Traction, funding, and third-party validation.** Here calibration matters: PROTRACK is a quiet, opaque company, and much of what would normally anchor a diligence memo is simply not disclosed. There is no publicly reported venture funding round, revenue figure, headcount, or named customer roster, and the company's official pages carry no founding year, named founders, or deployment counts. Several third-party aggregators list a founding year of 1998, but this should be treated as unverified — there are at least two unrelated non-Israeli companies also called "Protrack" (an Indian vehicle-fleet GPS-tracking firm and a consumer GPS-tracker supplier), and aggregator profiles frequently conflate them. What is verifiable is that PROTRACK is a real, currently operating Israeli company: it maintains an active product website and product suite, appears as an Israeli company on Startup Nation Central's Finder, is listed in unmanned-systems supplier directories, and was cited in 2026 Israeli-tech coverage as a computer-vision firm developing GNSS-independent aerial positioning and actively seeking an AI partner for visual-language grounding. The honest read is a technically credible niche vendor with fielded, commercialized products but a thin public evidence trail on scale.

**Founders and team background.** The leadership and founding team of the Israeli PROTRACK entity are not publicly documented in the sources reviewed, and the founder/CEO names that surface in web searches (e.g., for "Protrack GPS Technology") belong to the unrelated non-Israeli namesake and should not be attributed here. This opacity is itself a diligence finding: the company's breadth of fielded products implies a competent computer-vision and geospatial engineering bench, but team depth, key-person concentration, and reserve/veteran pedigree — usually central to assessing an Israeli defense-adjacent vendor — cannot be confirmed from public material and would require direct engagement.

**Competitive dynamics.** PROTRACK sits inside a crowded and strategically hot "assured PNT / GPS-denied" landscape, and it competes along two different axes. (1) Vision-based navigation peers — most directly the Israeli company Sightec (software-first visual autonomy, NavSight) and the US firm Shield AI (Hivemind visual/AI navigation) — pursue the same "use the camera when GPS dies" thesis; PROTRACK's differentiation is its emphasis on video-to-map anchoring and coordinate extraction (geo-referencing and target geo-location), not merely relative visual odometry. (2) RF/hardware resilience players such as infiniDome (anti-jam GNSS protection) and Regulus Cyber (spoofing detection) solve the denial problem by protecting the satellite signal rather than replacing it — adjacent and often complementary rather than head-to-head. Above both sit incumbent avionics suppliers (Honeywell, Collins Aerospace) advancing vision-aided and alternative-PNT navigation, and defense primes (Elbit, IAI) that can build organic vision-aided navigation and geospatial exploitation into their own platforms. PROTRACK's edge is a mature, multi-product, hardware-agnostic software suite that spans positioning, mapping, and analytics; its vulnerability is scale — better-capitalized rivals and primes can out-invest and potentially absorb the category.

**Defense, security, and resilience dual-use relevance.** Unlike many records where dual-use is an adjacency, PROTRACK's dual-use is core and explicit: the company markets the identical stack to civilian, security, surveillance, and military users, and the underlying capability — flying, navigating, and geo-locating targets without a satellite fix — is simultaneously a commercial resilience feature and a front-line military requirement. GPS-denied navigation, camera-based target geo-location for C5I hand-off, and rapid in-mission mapping map directly onto contested-airspace ISR, counter-UAS support, base and border protection, and terrain dominance (OMNIUM C2), while the same functions serve infrastructure inspection, emergency response, and delivery in civil GPS-outage conditions. The strategic weight is high because navigation denial is now a defining feature of modern conflict and a growing civil-aviation hazard, and because software that restores positioning on existing, low-cost compute is exactly the kind of asymmetric, scalable resilience allied forces are prioritizing.

**Growth stage, trajectory, and key diligence risks.** PROTRACK reads as a long-operating, self-sustaining niche vendor with a broad, fielded product suite rather than a venture-scaled growth company — a profile that argues for genuine technical maturity but limited visibility into its commercial trajectory. The key diligence risks are: (1) **opacity** — no disclosed funding, revenue, headcount, customers, leadership, or firmly verified founding year, compounded by same-name confusion with non-Israeli firms; (2) **scale and capital** — small, apparently bootstrapped footprint against venture-backed rivals (Sightec, Shield AI) racing the same market; (3) **consolidation** — primes and larger autonomy vendors folding vision-aided navigation into their stacks; (4) **unbenchmarked claims** — accuracy, range, and coverage are company-stated, not independently validated in public; (5) **technical envelope** — visual positioning depends on usable imagery and reference maps, so feature-poor terrain (open water, uniform desert, night without thermal reference) can constrain performance; and (6) **procurement/concentration** — long defense sales cycles and likely customer concentration typical of niche vendors. The bull case is a technically credible, hardware-agnostic GPS-denied-navigation capability aimed squarely at one of the decade's defining operational problems; the bear case is a small, opaque vendor that struggles to scale or defend the category against better-funded competitors.