ADVENTGraphene AI e-nose

CATALOGUE 2026

Graphene AI Electronic Nose

See the invisible risk.
Smell it first.

GraphAISense MK.2E

An always-on AI electronic nose that identifies the odour fingerprints from a graphene multi-sensor chip with edge AI.

The GraphAISense MK.2E: a black metal enclosure with a sensor chamber and antenna on top, and a power button and USB ports on the front.

Performance is validated for each target substance and site condition.

What is an e-nose

An electronic nose is an AI sensor that smells and decides

The human nose tells odours apart by the combination of hundreds of olfactory receptors responding at once. An AI electronic nose applies the same principle: many sensors together create an “odour fingerprint”, and AI identifies it.

Illustration comparing a human nose and brain taking in odour molecules with an e-nose device, graphene chip and AI taking in the same molecules.
Human nose and AI e-nose
StepHuman noseAI e-nose
Take inBreathes inBuilt-in pump draws in air
ReactHundreds of olfactory receptors respond30–38 graphene sensor channels respond
DecideThe brain recognises the combinationEdge AI identifies the pattern
RememberLearns odours through experienceLearns through the odour DB and Model Packs

Why watch for odour?

  • It is the earliest signal. Accidents, fires, leaks and spoilage show up as odour before you can see them.
  • Conventional sensors react after the event starts. Temperature and gas sensors respond once something has already begun.
  • People cannot smell around the clock. An e-nose judges and records to one standard, without a break.

An e-nose is a pattern-recognition instrument, with a different purpose from an analyser that measures the absolute concentration of a specific gas. ADVENT designs and makes the sensor from graphene in-house, and the decision is made by edge AI inside the device.

How it works

Three steps from odour to data

Illustration of a grid-patterned graphene multi-sensor chip with odour molecules landing on it and signal bars of different heights rising from its cells.

Concept view of the graphene multi-sensor chip. Each cell responds differently to molecules.

  1. 01

    Air intake

    A built-in pump draws in air continuously at 0.2–1.5 L/min.

  2. 02

    Odour fingerprint

    30–38 channels each respond differently, creating an odour fingerprint.

  3. 03

    AI decision and alarm

    The substance and alarm state are decided right on the device.

What is an odour fingerprint?

The same 38 channels draw a different shape for each odour. AI reads that shape to tell substances apart.

Odour A

Odour B

CH 01CH 38

Schematic to explain the principle, not measured data.

Three-level alarm from consecutive recognition

State is set by the count of consecutive recognitions, not by one momentary response.

  • NORMALEveryday state
  • WARNINGTarget recognised 5 times in a row
  • ALARMTarget recognised 12 times in a row
Recognition
Air · substance class · Unknown
Device state
HEALTHY · DEGRADED · FAULT

Events are sent the moment the state changes. If the link drops, they are stored on the device and resent automatically on reconnection.

Three core building blocks

  • EN-MSC.3EGraphene multi-sensor chipDesigned and made in-house. 38-channel chemiresistive replaceable chip, 1-year life
  • On-deviceEdge AI recognition engineDecides on site without a network. Lightweight model, no GPU required
  • IoT / IIoTStandard monitoring integrationEthernet · LTE (option), MQTT · REST · WebSocket · OPC UA

E-Nose vs Gas Sensor

A gas sensor measures how much. An AI e-nose decides what.

Odour is the earliest signal of accidents, fires and spoilage.

Conventional single gas sensor compared with the GraphAISense AI e-nose
ItemGas sensorAI e-nose
PrincipleOne sensing film reacts to one specific gas30–38 channel graphene sensor array reacts at once
OutputOne concentration valueMulti-dimensional odour fingerprint (per-channel resistance pattern)
DecisionOnly whether a set threshold is exceededEdge AI identifies the substance + 3-level alarm state
Mixed odours and gasesHard to separate due to cross-reactivity; causes false alarmsLearns the mixture pattern itself
Unseen odoursMistaken for a known gas, or no responseSeparated and reported as Unknown/OOD
When detectedAfter concentration reaches the alarm levelFrom precursor patterns before smoke or flame
New targetsAdd a sensor for each gasAdd a Model Pack to the same device
Output and integrationMainly alarm contacts and concentration valuesRecognition, alarm and device state + standard protocols

One device tells you what is leaking or burning, and the alarm keeps working even if the network drops.

Sensing, AI recognition and on-site alarms all complete inside the device, and only the necessary events go to monitoring.

Use cases

Where it works in security, safety and industry

Tap a field to see the 12 use cases.

A robot with the ADVENT logo patrolling inside a port terminal.
01

Patrol robot olfaction

Field trial
Where
Indoor patrol routes
Target
Abnormal gas and odour events
Result
Odour logged with location and video
02

Restricted-area anomalies

Applicable
Where
Utility tunnels, substations
Target
Patterns unlike normal (Unknown)
Result
Reports anomalies in unmanned spaces
03

Fuel and flammable vapour

Expansion
Where
Arson-risk zones, underground car parks
Target
Gasoline, diesel, kerosene vapour
Result
Early awareness of flammable vapour
  • Field trialTrials under way or being arranged with public bodies, municipalities and companies
  • ApplicableApplied with current Model Packs and standard integration (includes Integration)
  • ExpansionApplied after training on site data

Recognition performance is validated per Model Pack by target substance, background interference, concentration range and site conditions.

Expanding fields

Site data training keeps widening the range of application.

  • Confined spacesSuffocation and explosion prevention
  • Cold chainFreshness and spoilage
  • Smart buildingsIndoor air, abnormal odour
  • FarmingLivestock odour and gas
  • MobilityVehicle cabin air
  • Medical and bioResearch stage

Platform

Gathering odours from the field to give machines a sense of smell

Data from patrol robots, industrial equipment and processes, and urban and public facilities forms a physical AI olfactory data platform.

A humanoid robot sensing a stream of blue air with its nose, a concept image of giving robots a sense of smell.
  1. 01Collect and label olfactory data
  2. 02Build the odour DB (fingerprint library)
  3. 03Train and validate Model Packs
  4. 04Deploy Model Packs to field devices

The more data accumulates, the wider the recognition range. As cameras handle sight and microphones hearing, ADVENT is building the olfactory data standard for robots and AI systems.

Specifications

GraphAISense MK.2E specifications

Designed for continuous operation in security, safety and industrial settings.

Hardware

Dimensions (L×W×H)
270 × 250 × 90 mm + sensor chamber 60 mm
Multi-sensor chip
EN-MSC.3E · graphene chemiresistive
Sensing channels
38 channels (30–38 active)
Full-channel scan
2 s or less
Warm-up
15 min
Sampling
Built-in pump, forced intake · 0.2–1.5 L/min
Chip life / service
1 year / inspection and recalibration every 6 months
Protection / installation
IP40 · indoors or in a protective enclosure

Performance and operation

Power
AC 220 V, 50/60 Hz (external adapter)
Portable power
20,000 mAh · 100 W power bank
Max. consumption
75 W
Battery operation
8 h or more · charges in 3 h or less
Operating temp. / humidity
−5 to +50 ℃ / 20–98 %RH, non-condensing
Air pressure
80–107 kPa
Network
Ethernet / LTE (option)
Protocols
MQTT · REST · WebSocket · OPC UA

Main analytes

Ammonia (NH₃) · hydrogen sulfide (H₂S) · carbon monoxide (CO) · hydrocarbons (CH₄ etc.) · alcohol vapours · ketones · aldehydes

Odour detection

Smouldering gas · fuel vapour (gasoline, diesel, kerosene) · industrial gas odorants (mercaptans etc.) · alcoholic beverage vapour · wastewater evaporation · fragrances · liquid narcotics and e-cigarettes

About ADVENT

An AI olfaction platform company turning odour into data

ADVENT Co., Ltd. began in graphene nanomaterials research. We design and make graphene multi-sensor chips in-house, and with edge AI and our own olfactory database, provide continuous monitoring where the human nose cannot reach.

Founded
27 August 2024 (R&D since 2019)
Patents
KR 10-2958752 and 3 others
Certified
Corporate R&D Center · Venture Company (Innovation Growth type)
Field trials
Fire precursor, odour and port security trials with Korean public bodies and local governments
Capabilities
In-house graphene chip production · edge AI · olfactory DB design
Overseas
US entity in Delaware

Contact

Adoption and field trial enquiries

We can help with site consultations, proof-of-concept partnerships, custom Model Packs for target substances, and technical materials.

Head office
#909, 28 Digital-ro 33-gil, Guro-gu, Seoul, Republic of Korea
R&D Center
#420, 237 Yangji-ro, Sosa-gu, Bucheon-si, Gyeonggi-do
Plant
144-6, 3·1manse-ro 281beon-gil, Jangan-myeon, Hwaseong-si, Gyeonggi-do
US entity
8 The Green #22818, Dover, DE 19901, USA

We give machines the sense of smell.

© ADVENT Co., Ltd. Specifications may change for improvement. Some site photos and illustrations are staged or AI-generated for explanation.

Call us