DroneShield is a broad, layered counter-UAS portfolio, and this page will not pretend otherwise. Its own material describes a modular layered defence built from “Passive RF Detection & Disruption”, “Optical Sensors”, “Radars” and “Edge Computing”, across products including RfRecon, RfPatrol Mk2, DroneSentry, DroneSentry-X Mk2, the DroneGun family and the DroneSentry-C2 command layer. It detects and defeats. GOTEAM does one half of that, deliberately: it detects and identifies, and stops there.
This is not an “we are passive, they emit” page. DroneShield’s RF detection is passive too, and says so plainly: RfPatrol Mk2 is described as a “completely passive non-emitting wearable UAS detection device”, and RfRecon as “ultra-wideband passive detection across the RF spectrum”. Both companies ship sensors that broadcast nothing. The difference is not emission. It is what the passive sensor is listening to.
An RF sensor detects a radio link. A thermal sensor detects an aircraft. Those are different physical observables, and they fail in different places. RF detection requires the target to transmit — that is what the technique is, not a criticism of any implementation of it. GOTEAM’s detector reads heat and shape, which an airframe produces whether or not it is talking to anyone: as GOTEAM puts it, it “detects the aircraft’s shape and heat, not its radio link, so fibre-tethered, pre-programmed or fully autonomous drones that transmit nothing are just as visible.” The cost of that is equally physical: a camera needs line of sight and an atmosphere it can see through, and a radio does not.
The rows below are written at the level of the technique — passive RF sensing versus passive thermal imaging — wherever a vendor-specific fact is not published. That framing is both safer and more durable: a spec sheet changes, and the physics of a receiver that needs a transmitter does not. Every DroneShield capability marked here is drawn from DroneShield’s own published material; see the note at the foot of the page.
Almost every row restates one fact: GOTEAM observes the aircraft, not its radio. That single choice is what makes a radio-silent drone visible, removes the dependency on a signature library, and puts a picture of the target in front of the operator — and it is also what costs GOTEAM everything in the third table.
| Capability | GOTEAM | Passive RF detection |
|---|---|---|
| Detects a drone that transmits nothing — fibre-optic-controlled, pre-programmed, fully autonomous | ✓detects heat and shape, not a link | ✗a receiver needs a transmitter — true of the technique, not one product |
| Detection does not depend on a signal library staying current against new or modified radios | ✓a new airframe is still a warm object in the sky | ◐DroneShield’s RfAI-3 “senses unknown drones without catalog dependency” and RfRecon is “designed to detect and classify RF activity across a broad spectrum without relying on predefined signal catalogs” — but it still needs the drone to transmit |
| Produces an image of the target the operator can look at and judge | ✓annotated clip and zoom-crop per confirmed track | ◐not from the RF layer; DroneShield sells optical sensors as a separate part of the portfolio |
| Discriminates drone from bird and clutter on the sensor itself | ✓on-edge neural classification, plus track confirmation before a target is declared | ✗birds do not transmit; the question does not arise at the RF layer. DroneShield answers it in a separate optical product: DroneOptID is used in “confirming that an object is a drone and not a bird” |
| Every target in the field of view resolved on every frame, at the same latency | ✓fixed stare, no gimbal to point — “one drone or twenty, at the same latency” | — |
| Detection, classification and tracking run entirely on the device, fully offline | ✓needs only power and a field of view; store-and-forward when the link drops | —“Edge Computing” is listed as a portfolio capability; offline behaviour not documented at this level |
| Detection engine licensable as software, onto a third party’s thermal core and compute | ✓camera-agnostic; Jetson or x86 with Intel Arc | — |
| Cannot be jammed, spoofed or direction-found by the target | ✓nothing is emitted and nothing is being listened for | ◐a passive receiver is also undetectable, but its input is a channel the adversary controls and can simply not use |
These are the rows where a tick in both columns would be accurate and useless. Both products genuinely do all of them. They do them from different physics, with different failure modes — read the cells, not the mark.
| GOTEAM does it by… | DroneShield does it by… | |
|---|---|---|
| Operating without emitting | Passive EO/IR. No radar pulses, no RF. Nothing to license, nothing for an adversary to direction-find, no interference with airport, telecom or first-responder radios. | Passive RF reception. RfPatrol Mk2 is a “completely passive non-emitting” device and RfRecon does “ultra-wideband passive detection across the RF spectrum”. A receiver transmits nothing either. |
| Finding the drone | Looking at it. Sixteen-bit thermal frames, sky and cloud suppression, a neural detector trained on tens of thousands of hand-verified thermal drone frames, then a tracker that only confirms a target once it persists and moves consistently. | Hearing it. SDR scanning across the ISM bands and other commonly used control frequencies, with a detection engine DroneShield describes as “Continuously updated with DroneShield’s proprietary AI/ML RfAI detection engine”; its RfAI-3 generation “senses unknown drones without catalog dependency”. |
| Using AI in the detection chain | On-edge computer vision. A purpose-trained network proposes targets a handful of pixels across; behaviour cues down-weight birds; static-clutter learning suppresses persistently hot fixed objects while keeping a hovering target. | On-signal AI. DroneShield describes RfAI-3 as “AI-powered detection without reliance on signal matching”, with RfPatrol “Continuously updated with DroneShield’s proprietary AI/ML RfAI detection engine to identify evolving signals across key RF bands”. |
| Telling the operator | A live map track plus the footage. Range, bearing and elevation per confirmed track, an annotated clip and zoom-crop thumbnails, on web, iOS or an installable PWA — and out to third-party C2 over SAPIENT. | A command layer, and the same open standard. DroneSentry-C2 in Enterprise and Tactical variants, described as “Counter-UAS software powered by superior AI and data precision” where operators “detect, track, and defeat threats” — and RfRecon that “interoperates with ATAK-CIV, SAPIENT, and third-party command-and-control (C2) platforms”, so both products speak SAPIENT to the same C2. |
| Deploying in the field | Fixed or mobile node. Vehicle-mounted, tripod or rooftop; vehicle DC, mains or solar with battery backup; a sealed IP67 edge box; field-deployable in hours. | Body-worn to fixed site. A wearable dismounted detector, vehicle-mounted DroneSentry-X Mk2, and fixed DroneSentry installations — a wider spread of form factors than GOTEAM offers. |
Where GOTEAM is stronger. It sees the drone that is not talking. Fibre-optic spools, pre-programmed routes and fully autonomous flight are the direction the threat is moving, and each of them removes the signal an RF sensor exists to hear. GOTEAM’s input is the airframe: warm motors, a hard-edged shape against a cool sky, a track that persists and moves like an aircraft. That input does not have a library, cannot be turned off by the adversary, and produces something an operator can actually look at before acting.
Where DroneShield is stronger. Everything that follows from radio rather than light. RF reaches past the horizon of an optical sensor, works with no line of sight and in weather that closes an optical path, and it finds the operator as well as the aircraft — which is often what a police or base-security customer actually needs. DroneShield also sells the half of the problem GOTEAM refuses to sell: disruption. And it sells it alongside RF, optical, radar and a command layer from one vendor, which is a procurement argument a single-layer detection product has no answer to.
Use both — that is GOTEAM’s own published position. GOTEAM states plainly that it does not replace RF and acoustic sensors but complements them, adding the thermal/optical layer and fusing through an open API and SAPIENT. The two techniques fail in opposite conditions: RF loses the silent drone, optics loses the obscured one. A site that runs only one has chosen which kind of attack it will miss.
DroneShield capabilities and all quoted phrases on this page are taken from DroneShield’s own published material as of September 2026 — droneshield.com (home), its multi-mission portfolio page, its dismounted-products page (RfPatrol Mk2, RfRecon) and its blog posts on RfAI-3 (29 July 2026), persistent RF awareness and DroneOptID. Last reviewed: October 2026. ◐ = partial — present but narrower than the other column. — = not documented publicly: it marks a capability DroneShield’s published material does not address at this level of detail. It is not a claim that the capability is absent. GOTEAM marks reflect capability published on this site and in /llms.txt; specifications beyond that are shared under NDA after vetting. Rows written at the level of the technique — that a passive receiver needs a transmitter, that an imager needs line of sight — are properties of the physics and apply to every product of that class, GOTEAM’s included. DroneShield, RfPatrol, RfRecon, DroneSentry and DroneGun are trademarks of DroneShield Ltd; this is an independent capability comparison, not an endorsement, and every product here evolves — verify specifics against current documentation.
Upload a thermal or EO clip to FlyBox and a GOTEAM field device processes it on real hardware, streaming back an annotated result with per-frame detections and range and bearing.