Autonomous drones and IoT sensors for wildfire monitoring
An intelligent network protects the territory 24/7. IoT sensors detect the first signs of a fire and automatically dispatch the drone, which reaches the area and verifies the alarm with a thermal camera and AI. Between alarms, autonomous patrols survey vast forest areas for smoke, hotspots and thermal anomalies.
Want to estimate Docks, sensors and gateways for your area? Use the wildfire (AIB) calculator
Two layers of protection against wildfires
The forest calls the drone. The drone watches over the forest.
Fixed cameras have blind spots, ground patrols are costly and intermittent, satellites arrive hours late. By the time a wildfire is visible to the naked eye it has already gained ground. DroneBase wildfire drones integrate two complementary technologies into a single autonomous system: a network of IoT sensors that "smells" the fire during its slow-combustion phase and a thermal drone in a Dock that "sees" the hotspot from above. The software orchestrates the workflow, the Operations Centre manages events and alerts the responsible authorities.
IoT sensors for early wildfire detection
Solar-powered sensors mounted on trees detect pyrolysis gases (CO, H₂, VOC) while the wood is still burning without a flame. The alarm travels across the low-power IoT network, reaches the cloud and automatically generates a mission: the drone takes off to quickly verify the event reported by the sensor, with precise coordinates.
Autonomous patrols with a thermal drone
Even without a sensor alarm, the drone flies scheduled periodic missions from the Dock and patrols the territory with a radiometric thermal camera and an optical camera. AI analyses the live stream in real time to identify hotspots, smoke and thermal anomalies beneath the canopy: every detection becomes a georeferenced event sent to the Operations Centre.
How the autonomous wildfire detection system works
From the first sign of pyrolysis to event management, the sensor → cloud → drone → AI → Operations Centre chain closes in a few minutes. The times shown are indicative and depend on terrain, distance from the Dock and operating conditions.
The sensor smells the fire
Lightning, a cigarette, a fault: the wood burns slowly without a flame. IoT sensors distributed across the territory detect CO and H₂ at ppm level and the built-in AI filters out ambient noise (smog, agricultural smoke).
The alarm reaches the fleet
The event travels across the mesh network, the border gateway and the sensor cloud, then passes through the alarm connector to the fleet management platform with its GPS coordinates.
The drone leaves the Dock
The platform generates the verification mission with the waypoints of the sensor in alarm, interrupts any patrol in progress and the drone takes off automatically from the Dock.
Thermal camera and AI on site
The drone reaches the coordinates: the radiometric thermal camera and optical sensors confirm heat signatures, smoke or hotspots. If it is a false positive the mission closes and the drone returns to patrol.
Operations Centre and responders
Event information — coordinates, live thermal footage, weather data, site accessibility, sensor history — is made available to the Operations Centre and the responsible authorities according to the system configuration.
The drone stays on the event
The drone can keep observing the area and provide useful information on how the event evolves, with an encrypted video stream to the Operations Centre, until ground crews take over the response.
Autonomous drones for forest monitoring
Industrial hardware for extreme environments, integrated with a cloud fleet management platform: from alarm to airborne drone in under 60 seconds, with no staff on site.
DJI Dock 3
Climate-controlled drone-in-a-box station, IP56, operating from −30 °C to +50 °C, with backup battery and lightning protection. The drone lives in the Dock, recharges itself and takes off on schedule or on alarm. The Dock can be installed in a fixed position or on a vehicle with the dedicated mount, for mobile outposts in the highest-risk areas.
Thermal drone
Enterprise drone with a 640×512 radiometric thermal camera (measuring from −40 °C to +500 °C), 48 MP optical cameras with hybrid zoom up to 112×, laser rangefinder and NIR illuminator: the thermal fire drone identifies hotspots and smoke even at night and beneath vegetation cover.
Automatic missions
On an alarm from a sensor or an external system, the platform generates the verification mission: the drone interrupts the current patrol and heads to the event's GPS coordinates, with no human intervention.
Patrol rounds
Routes and schedules defined by the Operations Centre over the at-risk areas, optimised on terrain profile, sensitive zones, weather and critical hours. Forest patrolling with drones, day and night.
Thermal monitoring
AI analyses the thermal and electro-optical stream in real time: inconsistent temperature gradients, smoke, hotspots beneath the canopy. Every anomaly generates a georeferenced event.
IoT sensors for early wildfire detection
The wildfire sensor network turns the forest into a capillary IoT platform: sensors that detect pyrolysis gases within the first 60 minutes of a 2×2 metre fire, before flames are visible. Four modules working together, with a 15-year lifetime, no lithium batteries and no maintenance.
Wildfire sensors
Solar-powered trunk-mounted sensors that detect pyrolysis CO, H₂ and VOC at ppm level. Built-in AI (Bosch BME688) for reliable fire recognition and false-positive rejection. They also measure temperature, humidity and atmospheric pressure to monitor forest health.
- 100 m radius to detect 60 min of a 2×2 m fire
- 15 years maintenance-free, solar powered
- No lithium batteries: supercapacitors
- IP67, from −40 °C to +85 °C, 100% humidity
- LoRaWAN bands EU868 / AS923 / NA902-928
- Installed at 3 m height on the tree
Mesh gateway
Extends the LoRaWAN network well beyond a single hop: sensor messages pass from gateway to gateway in self-healing multi-hop mode up to the border gateway. Auto-configuration, automatic failover and over-the-air updates. Solar powered, maintenance-free.
- 2–6 km between mesh gateways, multi-hop
- Typically 100 sensors per gateway
- Automatic network configuration
- Self-healing with automatic failover
- Compatible with third-party LoRaWAN sensors
- Solar, 10–15 years maintenance-free
Border gateway
Closes the loop to the cloud. Installed in a forest lodge or near settlements, it receives messages from the mesh gateways and forwards them to the sensor cloud platform. Triple redundant connectivity — 4G/LTE-M, Ethernet and satellite — for coverage even in areas with no mobile signal.
- 20 mesh gateways managed per border gateway
- 4G/LTE-M with 2G/GPRS fallback
- Integrated Ethernet adapter
- Satellite uplink for remote areas
- Compatible with Ku/C-band satellite terminals
- PoE or solar power (optional)
Sensor cloud
Centralised hub that collects sensor data through the mesh and border gateways. Managed from a web app (site management) and a mobile app for field deployment. Multi-site, scalable, accessible wherever there is an Internet connection, with dedicated credentials and role-based permissions.
- Web app for site management (browser)
- Mobile app for deployment (smartphone)
- Multi-site, geographically distributed
- Real-time environmental data
- Automatic alarm notifications
- Native connector to the fleet management platform
Sensor and gateway technical data according to the IoT network manufacturer's datasheets.
When the sensor detects a fire, the drone responds
The sensor cloud and the fleet management platform are linked by a native alarm connector. When a sensor detects pyrolysis gases, the alarm travels mesh gateway → border gateway → cloud → connector → fleet platform in a matter of seconds.
Actual coverage depends on configuration, terrain, Dock position, IoT network, operating conditions and airspace constraints.
The platform receives the precise GPS coordinates of the event, interrupts the current autonomous patrol and sends the Dock's drone to the exact spot for thermal verification, fully automatically. The same mechanism integrates external alarm systems: motion sensors, fire protection systems, dispatch systems, distributed acoustic sensors.
The result is a wildfire system in which a single Dock station, together with the ground sensor network, can survey and protect extensive territories through the combined action of AI patrols and IoT detection.
Automatic trigger
Sensor alarm → drone airborne in under 60 seconds.
Thermal verification
AI and thermal camera to confirm the event and rule out false positives.
Scheduled AI patrols
Autonomous patrols across the whole protected area, independent of the sensors.
Multi-site scalability
More Docks and more sensors for larger territories, from a single platform.
Fleet platform integrations
Pre-built connectors let the drone talk to the systems the authority or company already uses; custom connectors can be built for specific use cases.
Alarms
Wildfire sensors, motion sensors, gunshot detection, dispatch systems: the drone responds automatically to the alarm.
Video streaming
Drone and Dock streams to the control room's Video Management Systems, for real-time security operations.
Data processing
AI software for high-resolution maps, 3D models, orthomosaics and territory-specific reports.
UTM
Air traffic management systems for situational awareness and regulatory-compliant BVLOS operations.
Detect & Avoid
Third-party systems for monitoring surrounding aircraft, in support of BVLOS requirements.
Missions and logs
Import of detailed missions and management of assets, batteries and flight logs across systems.
Wildfire monitoring with autonomous patrols
The system does not necessarily wait for an alarm: the drone can patrol the territory preventively. The platform schedules repeat missions over the area of interest, with routes optimised on terrain profile, risk zones, weather conditions and operating schedule.
During each patrol the AI algorithms analyse the thermal and electro-optical stream in real time, looking for hotspots, thermal anomalies, smoke and other events detectable by the available sensors, including people in restricted areas. Every anomaly generates a georeferenced event sent to the Operations Centre.
The system learns over time: every false positive labelled by the operator refines the model on the specific territory and progressively reduces noise.
- Scheduled missions by day and by night, with no staff on site
- Radiometric thermal camera and optical camera on the same flight
- Georeferenced events with attached video and frames
- Automatic patrol interruption on a sensor alarm
A wildfire system designed for large territories
The integration of sensor network, fleet platform and Dock is not a sum of products: each component amplifies the others. The sensors extend the drone's eye, the drone verifies the sensors' alarms, the cloud orchestrates everything in real time.
Extended coverage per Dock
In the reference configuration (1 border gateway, 20 mesh gateways, ~2,000 sensors) a single Dock covers up to 70 km². Scale by adding Docks and gateways.
Early detection
IoT sensors detect the fire during slow combustion, 30–60 minutes before flames are visible: the decisive time advantage for acting within the Golden Hour.
Double AI verification
First filter: on-board AI in the sensor. Second filter: computer vision on the thermal drone. False positives reduced to operationally manageable levels.
Scheduled AI patrols
Autonomous 24/7 patrols over the area; AI continuously analyses the thermal stream and spots anomalies independently of the sensors.
Resilient connectivity
Self-healing LoRaWAN mesh network, 4G/LTE/Ethernet/satellite border gateway, fleet on redundant 4G/5G/Ethernet: operational even without mobile coverage.
Maintenance-free sensors for 15 years
Solar-powered sensors with supercapacitors, no lithium batteries, over-the-air updates: low operating costs over the long term.
End-to-end security
AES-256 encryption across the whole chain, from sensor to Operations Centre; platforms designed for GDPR, ISO 27001 and NIS requirements, with immutable audit logs.
Scalable cloud-native
From a single site to entire regional territories: multi-tenant Operations Centre, unified dashboard, continuous updates.
Regulatory framework
Designed for the requirements of EU Civil Protection, the EU Forest Strategy, PNIEC (Italy's national energy and climate plan) and the EASA Specific category. Support in checking eligibility for PNRR (Italy's recovery plan) calls and regional wildfire (AIB) funds.
An incipient fire detected during pyrolysis is a manageable fire. A fire detected when flames are visible from kilometres away is already out of scale. The integration of sensors, drone and Operations Centre does not fight the fire: it stops it from becoming an emergency.
DroneBase Aerial Surveillance DivisionFrom remote mountains to industrial sites in forested areas
The Dock operates where traditional surveillance cannot reach: alpine ridges, isolated forest perimeters, solar farms, infrastructure crossing wooded areas. A single autonomous station complements patrols, lookout towers and fixed cameras, with no operators on site.
Integrated system vs traditional methods
A comparison between the integrated drone + IoT sensors + Operations Centre system and traditional wildfire surveillance techniques.
| Parameter | Traditional methods | DroneBase system |
|---|---|---|
| Detection during slow combustion | Not possible | 2–30 minutes via IoT sensors |
| Verification response time | 15–30 minutes (ground patrols) | Indicative: under 2 minutes from take-off |
| Coverage per station | Fixed cameras with blind spots | Up to 70 km² (reference configuration) |
| Automatic alarm verification | Manual, by an operator | Drone with AI and thermal camera |
| False alarms | Frequent (operator fatigue) | Reduced by double AI verification |
| Operational continuity | Staff shifts | 24/7, 365 days |
| Maintenance required | Daily | Maintenance-free sensors for 15 years |
| Cost per km² protected | High | Lower, to be quantified in the project |
| Forest health monitoring | None | Included (temperature, humidity, pressure) |
| Environmental impact | Built infrastructure | Solar sensors, lithium-free |
Wildfire drones for public authorities and Civil Protection
The system helps regions, provinces, municipalities, mountain communities, parks, Civil Protection, fire brigades, forestry consortia and managers of large estates or infrastructure in forested areas to proactively protect citizens, natural heritage and assets from forest fires. Six direct benefits, four structural qualities.
Rapid alerts
The system detects the incipient fire within minutes of pyrolysis and immediately dispatches the drone for thermal verification. Authorities can act before the flames spread.
Operational efficiency
Automated 24/7 monitoring of vast forest areas: fewer manual patrols and less exposure of staff to night shifts in rugged terrain.
Cost containment
Preventing large-scale fires reduces spending on suppression, aerial assets, clean-up and environmental restoration. The return is measurable on the ground, from the first event intercepted.
Climate adaptation
Preventing and managing outbreaks contributes to a more resilient territory, reducing the CO₂ emissions linked to large wildfires.
Habitat conservation
Early detection protects wildlife and natural habitats, from Natura 2000 sites to national and regional protected areas.
Reduced maintenance
Solar-powered sensors operating for 10–15 years with no maintenance or batteries; over-the-air firmware updates. Install once, operate for over a decade.
Simple
Distributed, scalable architecture supporting a large number of low-cost sensors.
Efficient
Overcomes the power and connectivity constraints typical of remote forests.
High-performing
Spots incipient fires before they flare into visible flames.
Decisive
No servicing needed for at least 15 years from sensor installation.
Who needs continuous forest protection
Public administration
Regions, provinces, municipalities and mountain communities responsible for public forest land and at-risk wildland-urban interface areas.
Civil Protection and wildfire services (AIB)
Continuous support to wildfire operational structures with monitoring, automatic alerting and live hotspot visualisation.
Forestry
Forestry consortia, sawmills and forest management companies with extensive areas of vulnerable timber.
Strategic infrastructure
Power grids, gas pipelines, railways and motorways crossing wooded areas: potential targets and potential ignition sources.
Solar farms
Utility-scale plants in semi-arid areas surrounded by vegetation, where a nearby fire compromises high-value assets.
Tourism and hospitality
Holiday villages, eco-resorts, mountain hotels and farm stays in forested areas: protection of guests, assets and continuity.
Parks and protected areas
National parks, regional reserves, Natura 2000 sites: where loss of vegetation cover is irreversible ecological damage.
Wildland-urban interface
Municipalities and settlements adjacent to wooded areas: protection of homes, schools, hospitals and evacuation routes.
Military bases and sensitive sites
Defence installations and communication centres in remote forest areas, where traditional verification is limited.
Insurance and risk management
Insurers covering forest, agricultural and residential assets in fire-prone areas: real-time monitoring for premiums and claims.
Precision agriculture
Vineyards, olive groves, orchards and large estates: protection of crops and adjacent forest areas, with microclimate data.
Industrial sites and storage
Timber yards, sawmills, paper mills and biomass plants: where a small outbreak can quickly become a serious event.
Operations Centre for 24/7 wildfire monitoring
The sensors smell, the drone sees, the software orchestrates: the Operations Centre supervises the system, video-verifies events and manages escalation to the responsible authorities. DroneBase can deliver the service in managed mode, with no pilots or authorisation procedures on the customer's side.
Certified 24-hour staffing
Video verification and escalation from a Type C Operations Centre, compliant with UNI 10891 and EN 50518, licensed for Category 2 (Italy).
Operations under ENAC LUC
Missions, including BVLOS, are authorised under the DroneBase LUC certification (EU Reg. 2019/947): CONOPS, SORA and NOTAM included in the service.
Quality and continuity
DroneBase is ISO 9001:2015 certified and a DJI Super Gold Dealer; contractual SLA of up to 99.9% uptime, with maintenance and updates included in the managed service.
Performance certified by independent tests and validated in over 30 industrial sectors. Information security certifications are held by the platform providers; the LUC and aeronautical responsibility are held by DroneBase; the Operations Centre qualifications are held by the partner Operations Centre.
The same Dock + Operations Centre ecosystem is used in drone video surveillance of infrastructure and large sites.
Frequently asked questions about wildfire drones
How are drones used for wildfire monitoring?
In two complementary ways. On alarm: when an IoT sensor detects pyrolysis gases, the drone takes off automatically from the Dock and reaches the coordinates to verify the event with a thermal camera and AI. On patrol: the drone flies scheduled missions over the territory looking for hotspots, smoke and thermal anomalies, even without any alarm.
How does early fire detection with IoT sensors work?
The sensors, mounted on trees at around 3 m height and solar powered, measure the CO, H₂ and VOC produced by pyrolysis, i.e. the slow combustion of wood before a flame appears. AI built into the sensor distinguishes the gas signature of a fire from ambient noise such as smog or agricultural smoke. The alarm travels via LoRaWAN through the mesh gateways to the border gateway and the cloud.
What happens when a sensor detects a possible fire?
The sensor cloud sends the event, with its GPS coordinates, to the fleet management platform through an alarm connector. The platform generates a verification mission, interrupts any patrol in progress and launches the drone in under 60 seconds. The drone verifies the situation with its thermal camera and optical sensors; if the event is confirmed, the information is made available to the Operations Centre and the responsible authorities, otherwise the mission closes.
Can a drone fly automatic patrols over a forest?
Yes. The platform schedules repeat missions with routes and times defined by the Operations Centre, optimised on terrain, risk zones and weather conditions. During the patrol the AI analyses the thermal and electro-optical stream in real time and generates a georeferenced event for every anomaly. Beyond-visual-line-of-sight (BVLOS) operations are subject to a site assessment and are carried out under the DroneBase LUC.
How is the thermal camera used to spot a fire?
The 640×512 radiometric thermal camera measures the temperature of every point in the image from −40 °C to +500 °C. Computer vision algorithms look for temperature gradients inconsistent with the surroundings, hotspots beneath the canopy and heat signatures compatible with an incipient fire, even at night, and cross-check them with the optical camera and visible smoke.
How much territory can an autonomous drone system monitor?
In the reference configuration, with 1 border gateway, 20 mesh gateways and around 2,000 sensors, a single Dock covers up to 70 km². Actual coverage depends on configuration, terrain, Dock position, extent of the IoT network, operating conditions and airspace constraints: it is defined in the project. The wildfire (AIB) calculator provides a first estimate of Docks, sensors and gateways.
What is the advantage of integrating ground sensors and drones?
The sensors detect the fire during slow combustion, 30–60 minutes before flames are visible, but they cannot see it; the drone sees and verifies from above, but it cannot be everywhere. Integrated, the sensors extend the drone's perception deep into the forest and the drone verifies the sensors' alarms, reducing false positives through double AI verification.
Can the system operate 24 hours a day?
Yes. The Dock is climate-controlled, IP56 and operates from −30 °C to +50 °C; the drone recharges automatically between missions; the sensors are solar powered with supercapacitors. Staffing is guaranteed by the 24-hour Operations Centre, with a contractual SLA of up to 99.9% uptime in the managed service.
How are alarms from the sensors handled?
Every alarm is georeferenced and includes the sensor's history. The Operations Centre receives the event together with the drone's live thermal footage, weather data and site accessibility, and manages escalation to the fire brigade, Civil Protection or the regional coordination centre according to the configuration agreed with the authority.
Can several Docks be installed to monitor very large territories?
Yes. The architecture is scalable: more Docks, more gateways and more sensors are managed by the same cloud platform and the same multi-tenant Operations Centre, with a unified dashboard. The multi-site configuration is sized in the project according to the area to be covered.
Why DroneBase
Since 2012, Europe’s reference point for professional UAV technology
DJI Super Gold Dealer, the only one in Europe. Headquarters, warehouse, service centre and demo area in Rimini, Italy: you see the system working before you buy it, and this is where it comes back for maintenance and updates.
- ISO 9001 certified — certified quality in sales, training and service
- Training with certified instructors — DroneBase Academy, pilot certificates and STS scenarios
- Technical and regulatory consulting on complex scenarios, ENAC and EASA
- R&D for complex projects and tailor-made integrations
- Full support, before and after the sale, with a dedicated technical contact
Let's talk about your territory
Every territory requires a specific configuration of aerial coverage, sensor network, connectivity and operational management: DroneBase designs the system architecture with a feasibility study and coverage simulation on real maps. Fill in the form: one of our specialists will take charge of your request and get back to you with technical and commercial information and, if needed, to arrange a consultation or a demo. No obligation.
