Ryze Tello

Top Five Best Drones for Beginners

If you’re buying your first-ever drone, there are four things to consider

  1. Stabilisation features. As not yet a very experienced pilot, you’d want a gliding drone with GPS stabilization, which automatically maintains a steady flight path and minimizes drift. Models with even a basic gyroscope and accelerometer also add extra “fineness” to your flight and footage. 
  1. Safety controls like automatic return-to-home, hoover on lost signal, geo-fencing, and obstacle avoidance also reduce the odds of drone crash or loss several days after purchasing. 
  1. Battery life. The cheapest beginner drones have a flight time of 10 minutes, which is hardly enough to really enjoy the experience or hone your skills, not mention capture some decent shots. Pay some extra money for a starter drone with a 1500mAh 4S battery. 
  1. User-friendly piloting app. Spend as much time on checking out the piloting app as reviewing the drone itself. You’d want software with intuitive controls and a convenient smartphone app. The must-have features for beginners are altitude hold, headless mode, and auto-hovering. 

Here are five models that fit the above description to the dot. 

DJI NEO

DJI NEO is arguably the most affordable videography drone on the market right now, with a going price of under $200. At 135 g (4.8 oz), the drone can be flown in every country without any UAV authorization. (Although you should still maintain common sense safety rules like avoiding people overflight or cruising in restricted zones).

But for its tiny size, DJI NEO offers some mighty fine filming features: 4K video, 12MP stills, presets for auto-shooting from different dynamic angles, vertical video recording, and voice-activated control. 

Ryze Tello 

If you’re looking for a cheap, mostly indoor beginner drone, Ryze Tello can be loads of fun. It’s ridiculously simple to control using the app and flight controller combo. And you can add FVI goggles for extra giggles if you’re training your racing skills. On the downside, the flight time is just 13 minutes and the max flight distance is 100 m/328 ft, so it’s definitely not the best drone for filming outdoorsy escapades.  


Potensic ATOM SE

Foldable ATOM SE boasts build quality. It’s sturdy, durable, but still lightweight (250 grams/8.8 ounces) to not fall under commercial drone regulations. The 4 km / 13,123 ft transmission range gives you ample room to shoot some great stills with a Sony 12MP CMOS sensor camera with a 118° FOV and a vertical +20° to -90° adjustment angle. 

ATOM SE can stay in the air for about 30 minutes at low altitudes, with no wind conditions. But the flight duration and quality drop a lot if you’re out on a windy day. 


Parrot ANAFI Ai 

If you have extra cash to spare for a more professional drone, ANAFI Ai hooks you up with a roster of incredible features for autonomous photogrammetry. The 48 MP camera with a 14 EV dynamic range and 6x zoom shoots the crispest images, even at high flight speeds. 

The video is of stellar quality too: 4K video, including in P-Log and HDR10 4K videos up to 30 fps. And those planning some UAV mapping missions would appreciate specialized photogrammetry flight modes available in the drone app and 1-click flight plan creation.


HappyModel BNF Crux35

As a beginner racing drone, we can full-heartedly recommend HappyModel BNF Crux35. Retailing for under $150, the FPV drone touts an EX1404  high-efficiency motor and HQPROP 3.5-inch three-blade propeller for a robust thrust-to-weight ratio. With a 4S 850mAh battery, you’ll get about 16 minutes of flight time (but that’s all subject to how fast you choose to go). 

The FPV camera isn’t amazing (although you can easily replace it with a better alternative). This is a good starter racing drone that won’t hurt you too much if you crash it. 

Discover even more drone recommendations in Bavovna’s UAV directory

Remote Operated Vehicles

How Remote Operated Vehicles (ROVs) Generate Efficiency Underwater 

A remotely operated vehicle (ROV) is an autonomous, unmanned submersible used in underwater missions. Highly maneuverable, low-energy, and equipped with advanced robotics and AI, ROVs streamline underwater inspections and provide new knowledge of the global oceans. 

Here’s how remote operation vehicles make waves (pun intended) in maritime operations, the oil and gas industry, sea farming, and oceanography. 

Top 3 Remotely Operated Vehicle Use Cases 

Top-of-the-line ROVs have 4K resolution, wide lens cameras, and auto-color correction for crisp underwater imagery. Strong LED lights give visibility in murky environments. Some models may be tethered for communication with the operator, while others use AI-powered autonomous navigation.

Depending on the use case, a remotely operated vehicle can feature a robotic arm for contact manipulations, imaging sonar for 3D mapping and obstacle detection, plus onboard data processing units. 

These characteristics make the following underwater operations possible. 

Asset Inspections 

ROVs can replace divers on tedious and dangerous underwater missions. In the oil & gas industry, offshore platform operators use ROVs to inspect for structural corrosion, cracks, and other deformations. Some models also have specialized ultrasonic gauges to detect thickness loss in pipelines or rinsers or perform weld seam inspections for signs of fatigue.

Petrobras uses a fleet of ROVs to carry out inspection, maintenance, and repair of underwater equipment along the Brazilian coast. TAQA Netherlands also deployed uncrewed surface vessels for integrity inspections of shallow water infrastructure in the North Sea.

Similarly, ROVs have become the go-to tool for routine inspections in the maritime industry. Offering faster turn-around and lower costs, remote-operated vehicles can be used for hull, rudder, propeller, and anchor inspections. Norwegian Coastguard recently enlisted a Blueye Robotics’ X3 ROV to run hull inspections. 

Underwater Pipeline Surveying

Underwater pipeline leaks cost oil and gas operators millions in revenue loss and regulatory fines. ROVs can survey up to 25 km (15.5 miles) of underwater pipelines daily, helping operators manage maintenance. Specialized models with NDT tools can also detect early-stage pipe thinning for preventive strategies. And micro-ROVs can inspect pipelines from the inside through access points as small as 19-inch manholes.

Total Energies recently completed a pilot pipeline inspection program in the North Sea with Freedom AUV. The vehicle inspected over 120 km (75 miles) of submarine pipelines and 60km (37 miles) of near-shore pipelines for structural defects, with all data acquired in a single pass. 

Sustainability Initiatives 

ROVs have also become a staple in oceanography, helping scientists learn about marine life, pollution levels, and oceanic ecosystem trends. Besides data acquisition, they help drive meaningful change and offset environmental impacts.

Norwegian seafood producer Mowi increased its ROV deployments by 40% over the last year to support new initiatives for improved fish welfare. ROVs help remove dead fish from net pens, segregate jellyfish, and clean up harmful algae blooms. Thanks to the innovation, Mowi Scotland achieved a 96% superior-grade fish at harvest, while improving its animal welfare standards.

Chevron, in turn, deployed a pair of mini-ROVs — one providing visuals to the operator and the second doing the scrubbing — to clean underwater platform structures from unwanted marine growth. The compact ROV minimizes diver use and contributes to cleaner global waters.

Remote-operated vehicles also have dozens of other already feasible deployment scenarios and even more possible ones with greater adoption of AI-powered navigation systems. Some ROV use cases are currently limited by poor connectivity in the seas. At Bavovna, aim to solve this problem with AI. Compatible with both UAVs and ROVs, our INS brings the power of AI sensor fusion and autonomous navigation in GPS-denied environments. Learn more about our AI hybrid-INS system.

inertial navigation unit

How to Compensate for Drone Inertial Navigation Unit Deficiencies 

An internal navigation system (INS) provides extra situational awareness to UAVs, complementing other navigation units like GNSS, SLAM, or LIDAR-based systems. Working independently of satellites, an INS is great for countering GPS signal loss or targeted UAV jamming. But most commercial internal navigation units in drones aren’t without some critical shortcomings. 

Shortcomings of Commercial UAV Inertial Navigation Units

Most top-of-the-shelf drones are pre-furbished with internal navigation units that still require external data sources like a ground station or fusion with other sensors for autonomous navigation in GPS-denied environments. More advanced UAVs may feature better units, but they’re still susceptible to a range of issues. 

Sensitivity to Environmental Conditions 

MEMS-based inertial navigation units exhibit performance degradation under high temperatures, rapid temperature cycling, composite stress, high vibrations, and electromagnetic interference (e.g., from nearby equipment or transmission towers). High-end internal navigation systems can run smooth at a -40°C to 85°C (-40°F to 185°F) range and maintain very low bias residual errors. But they also come at a premium price tag and may not be compatible with all drone models. 

Sensor Drift 

Most INS require initial initialization and regular calibration to compensate for the inevitable drift, leading to cumulative velocity or displacement errors during flights in GPS-denied environments. 

The common types of sensor drift in drone internal navigation units include: 

  • Zero-offset (bias) drift occurs when sensor readings are inaccurate due to ongoing errors.
  • Integration drift accumulates over time due to different errors in gyroscope and accelerometer readings. 
  • Scale factor errors due to quick temperature cycling or equipment wear. 
  • Noise-induced drift (random walk) is caused by random interferences and manifests as unpredictable errors. 

Without regular recalibrations, INS errors pile up, jeopardizing autonomous flights. 

High Power Consumption

More advanced internal navigation units with ring laser gyroscopes are more power-hungry. While FOGs offer higher accuracy, even the most lightweight models consume 1.3 W against an average of 3 mA among MEMS-based systems. 

To compensate for INS drift, many companies also combine IMU readings with computer vision systems to deliver greater accuracy. With sensor fusion done on the edge, power consumption goes up massively, impeding UAV flight time. 

Using AI to Improve Inertial Navigation Unit Data Processing

INS performance can be massively improved using supervised machine learning and reinforcement learning techniques. Sensor fusion algorithms can compensate for accumulated errors and reduce noise-induced disruptions. 

Bavovna developed an AI-powered hybrid navigation kit, featuring a low-SWAP, EMI-shileded onboard unit, and a fine-tuned sensor fusion model, trained for each drone model. The entire system, including our internal navigation unit and AI-powered flight controller, weighs only 1.7lb (800gr), making it compatible with a range of UAV models. The max current consumption is 10A, preventing excessive battery drainage. 

Unlike other systems, Bavovna can maintain an ultra-long range, field-tested range of 155 miles (250km) for fully autonomous flights without any GPS reliance. The accuracy rate can be as high as 99.98%, thanks to on-edge, continuous compensation for error rates. With Bavovna hybrid INS, you can fly regular and FPV missions without worrying about signal loss, targeted interferences, or harsh temperatures. 

Learn more about our AI navigation solution for UAVs. 

Best Long Range Drones

Best Long Range Drones 2025

If you’re looking for a reliable drone for tactical, security, mapping, or surveying missions, you’ll need a model with a long-range. The newest fixed-wing VTOLs boast an impressive range of up to 1500 miles and an extended flight time of up to an entire day.

Long Range Drones: Summary Table

ModelRangeFlight timeKey features
Spirit-X500 km (311 miles)500 km (310 miles), at 150 km (93 miles) an hourZero emission hybrid electrical VTOL Integrated hydrogen fuel cell power system Payload capacity of up to 150 kg (330 lbs.) or 1,470 liters / 52 ft 3 Autonomous piloting mode 
Raybird Up to 2500 km (1553 miles)Up to 28 hoursTactical unmanned aerial system EW threat protectionUp to 5kg in payload capacity Catapult mechanical launcher available  
Tekever AR3 Long-range VTOL100 km (62 miles8 to 16 hours Long-range, high endurance UAV Dual-side looking SAR Integration with a range of sensor optics and payloads All-terrain retreival
Trinity PRO 100 km (62  miles)90 minElectrical, vertical takeoff, fixed-wing UAV Quantum-Skynode autopilot Native compatibility with 5 cameras Portable base station to enable high-precision PPK processing
Fly Dragon FDG24 240km (150 miles)210 minElectrical, vertical takeoff, fixed-wing UAV Reinforced carbon fiber composite airframe IPX4 water ingress rating 25m/s average cruising speed 

Gadfin Spirit-X

Gadfin is one of the top innovators in the long-range fixed-wing drone market. Spirit-X, one of its latest releases, boasts an impressive range of  500 km (311 miles), thanks to a hybrid powertrain, fueled by hydrogen fuel cells. The underbelly box can fit up to 100 kg (220 lb) in cargo or can be used to attach heavy-weight payloads. Foldable low wing, in turn, allows takeoff in landing with limited surface area, making it ideal for urban drone deliveries or rescue missions.  

Key characteristics:

  • 150 km/h (93 mph) cruising speed
  • Fold-out wings and carbon fiber fuselage 
  • Fixed tricycle wheeled landing gear
  • Distributed Electric Propulsion (DEP) redundancy system 

Raybird


Ukrainian startup Skyeton built a compact, but mighty UAS for military, security, and surveillance missions. Assembled in just 25 minutes, Raybird can stay in the air for 28 hours. The digital data link stays strong for up to 120 km (75 miles), although the range can be extended to up to 2500 km (1553 miles). A customizable, hot-swap payload bay can accommodate any camera, SAR, night vision, and laser target designator gear. 

Key characteristics:

  • 18-28 hour flight time
  • 5kg payload capacity 
  • 25 min assembly time 
  • 4500 m max altitude 

Read our full review of Raybird UAS

Tekever AR3 

Portuguese Tekever produces a great lineup of modular, long-range VTOLs.  AR3 is one of its best models because it’s packed to the brim with amazing technologies: a Synthetic Aperture Radar, common GCS with A4, A5, and A3, a recovery parachute, a customizable payload bay, and an optional BVLOS data link to the boot. It’s a marvelous model for a roster of land and sea-based missions.

Key characteristics:

  • Up to 90 km/h (55mph) speed 
  • 4 kg (8.8 lbs) payload capacity 
  • 16-hour max flight time 
  • 25 kg (55lbs) max takeoff weight

Trinity PRO

Trinity PRO eVTOL from Quantum Systems is a ‘nerdy’ sidekick for long-range mapping missions. Boasting a Quantum-Skynode autopilot and Linux-based mission computer, this UAV can be easily upgraded with extra AI capabilities and downstream payload integrations. A native GNSS module and complementary QBase 3D software make Trinity PRO one of the best long-range drones for land surveying and mapping missions

Key characteristics:

  • 90min flight time
  • 700ha area coverage
  • 18m/s wind tolerance
  • 5.75 kg (12.7 lbs) max takeoff weight 
  • IP 55 rating 

Fly Dragon FDG24 

Chinese Fly Dragon produces robust, affordable long-range fixed-wing drones with four rotors for smooth takeoff and landing. It can carry up to 1.6 kg (3.5 lb) in payloads for up to 3.5 hours, making it a great choice for different industrial use cases — building inspection, crop monitoring, site surveying, or emergency mission patrols. A sturdy airframe made of carbon fiber composites and an IPX4 ingress rating guarantee the gear’s long service life. 

Key characteristics:

  • 25m/s cruising speed 
  • 240 km max range 
  • 210-minute flight time 
  • 10 kg/22lbs max take-off mass
  • Level 5 wind resistance 

Discover even more long-range UAVs from our directory

thermal drone

How to Choose a Thermal Drone

A thermal drone is great gear for a variety of missions — from surveillance and security patrolling to industrial asset inspection. And there’s been plenty of new model releases over the last year. 

But with a sharp price tag, the “cheapest” models start at $6K, so you don’t really want to wing it (pun intended). Learn how to choose the right thermal drone from our quick guide. 

Important Thermal Drone Features To Consider 

To choose an all-around pleaser, check each option against these criteria: 

Thermal Camera Settings 

Many Electro-optical (EO)/Infrared (IR) camera payloads for drones are available. Look for systems with a resolution of at least 640 x 512 px for crisp images. 

A broad spectral band in the IR sensor is advantageous for better scene recognition and performance under various weather conditions (e.g., fog, rain, snow). Check if you can set custom isotherm ranges manually to further tune your equipment for the use case.

The best thermal imaging drones combine thermal sensors with an RGB camera for an impeccable shooting experience.

Radiometric Functionality

Thermal drones with radiometric features capture precise temperature readings, rather than differences in ranges. This allows you to calculate precise data in measured area (min, max, average °F/ °C), giving a broader read of trends. For example, you can evaluate heat distribution across industrial pipelines to measure thermal efficiency.

Some thermal drones can be auto-programmed to focus on specific temperature ranges for streamlined data collection. For example, you can set a custom range to inspect temp variations in cold storage rooms to detect heat loss.

Gimbal Stabilization

A solid gimbal reduces blur in thermal footage, especially models with gyroscopic stabilization. It also provides extra angular velocity to track fast-moving targets 

(e.g., if you’re using a thermal drone for border security tasks). 

Flight Time and Range

Most enterprise quadcopters can stay airborne for 30 to 50 minutes, depending on weather and load. Larger fixed-wing and VTOLs like Albatross UAV can cruise for up to 4 hours at 20 m/s (and it can be equipped with a thermal camera payload). 

Range matters for covering larger areas. Advanced drone transmission systems can sustain a steady range of up to 6-9 miles (10-15 km).

For extended operating time, you can choose a tethered thermal drone, which can hover for days when connected to a power source. With Bavovna’s AirTower Mode, tethered drones can operate fully autonomously even in GPS-denied environments.

Environmental Durability

The best thermal drones boast exceptionally high endurance, including wind resistance up to 23 knots, IP55 rating against water and dust damage, and built-in redundancies. For safe missions, look for models with redundant properrels, INS components, and motors.

Best Thermal Drones for 2025 

Need some recs? Here are the top picks from Bavovna’s team:

  • Skydio X10. Measuring just 31.1” x 25.6” x 5.7”, Skydio X10 can stay in the air for up to 40 minutes with a max speed of 45mph. The hybrid imaging system combines a narrow 64MP RGB camera, a 48MP telephoto one, and a radiometric thermal camera with  640 x 512 px resolution and under 30mk sensitivity. 
  • Autel EVO Max 4T. With an IP43 rating and an extra pair of hot-swappable batteries, Evo Max 4T is a reliable companion for a range of missions. Equipped with a hybrid RGB/thermal camera, this UAV can muster an impressive temperature Measurement Range of -20°C to 550°C. GPS-denied navigation mode is a great bonus.
  • Inspired Flight IF1200. IF1200 model from Inspired Flight is sturdy and robust. It can lift up to 19.1 lbs in payloads while staying in the air for 35 to 43 minutes. It’s compatible with the Gremsy VIO F1 thermal camera, featuring a 4K zoom sensor, a 640×512 radiometric sensor, and an integrated 2400m laser rangefinder. Thermal sensitivity range is ≤ 50 mk, giving you crisp imagery under any flight conditions. 

Discover more UAV companies in our directory

uav mapping

Primer on UAV Mapping

Collecting aerial data used to be a daunting challenge. Tape measures, foot patrols with a theodolite, or cost-inhibitive helicopter flights. 

Unmanned aerial vehicle (UAV) mapping changed the aerial data collection game, bringing extra speed, lower costs, and greater precision. 

New to the concept? Here are the essentials you need to know about UAV mapping. 

UAV Mapping and Surveying Use Cases 

Fixed-wing UAVs, VTOL drones, and enterprise quadcopters can stay in the air for up to an hour (and sometimes more), giving surveying teams ample time to perform various geodesic tasks. The best mapping drones also include specialized payloads for high-precision data collection like thermal cameras, multispectral sensors, magnetometers, gas detectors, and LiDAR systems for 3D scanning. 

Here’s how businesses use UAVs for surveying and mapping tasks:

  • Topographic mapping. Drones help create HD orthomosaics and 3D models for cadastral surveying, allotment planning, and a host of other civil surveying use cases. Swiss Canton of Valais used a WingtraOne mapping drone to conduct mountain village land surveys in 3 days, instead of 2 weeks with conventional methods. 
  • Mining exploration. Using drones, operators can assess resources and plan excavations based on geospatial information. Rugged, in-door models also help assess sub-terrain corridors to ensure safe and effective operations. WACO S.R.L. used the Elios 3 drone to inspect dangerous rock detachments inside its quarry (Italy), providing teams with valuable operational data. 
  • Urban planning. City planners rely on UAVs to collect visual data for 3D modeling, land classification, and smarter resource allocation. Thanks to automated route planning and high-precision data capture, drones substantially reduce the cost and fieldwork hours. A surveying team in Weinan City, China, used drones to collect oblique cityscape imagery with greater efficiency. Based on this data, a comprehensive 3D model was created with an accuracy level of up to 5 cm.
  • Road construction surveying. For large-scale transportation projects, UAVs provide seamless data capture for large-area linear maps, reducing the complexities of planning, monitoring, and documenting new construction projects. The Norwegian Public Roads Administration uses drones to survey underway projects more effectively. Mapping a 3-mile road with a drone can take just 2.5 hours and $270 in labor costs vs 6 days and $5,200 with terrestrial laser scanners. 

Shortcomings of UAV Mapping 

Although UAV mapping comes with a slew of benefits, it’s still a tedious process, susceptible to different disruptions. 

Weather can be a major factor as lighter, commercial models are inoperable in high winds, heavy rain, or snowfall. Fog, in turn, can cause sensor interference, increasing the risks of collisions and data capture accuracy. 

Terrain type can also aggravate the UAV’s technical limitations, leading to signal loss, mapping errors, and scrambled navigation. Dense vegetation, large water bodies, mountains, sudden elevation changes, and high-rise buildings, limit GNSS/GPS signal propagation. In such cases, it’s worth looking into a solution for GPS-denied navigation

Regulations. Many UAV aerial mapping use cases require BVLOS permissions, which may be hard to obtain in certain jurisdictions due to bureaucratic red tape. Drone operators must also comply with privacy requirements, as well as other rules related to flying over people and close to restricted areas. 

Data accuracy. Although drone technology has made major leaps, technical limitations still remain. Automated, on-device data processing can impact footage quality and accuracy. Also, discrepancies between vertical measurements can vary significantly without Ground Control Points (GCPs). On average, you need to place 12 GCPs for small to medium sites 

(7 and 39 ha) and up to 18 for the large sites (342 ha). This adds extra workload for field teams. 

Improving UAV Mapping with Bavovna Hybrid INS 

Overcome the challenges of GPS signal obstruction with Bavovna Hybrid INS Navigation Kit. Measuring just 150 x 134 x 73 mm, Bavovna kit enables reliable, long-range navigation with AI-powered sensor fusion. Custom-trained for each drone and a variety of operational scenarios, Bavovna helps operators fly mapping missions without getting held back by signal propagation delays, interferences, or drift bias. 

Discover Bavovna Hybrid INS Navigation Kit

Inertial Reference Unit?

What is an Inertial Reference Unit (IRU)?

An internal reference unit (IRU) is a three-axis system that provides precise attitude, velocity, and navigation information to the vehicles using internal sensors. Unlike basic sensor arrays, an IRU applies sensor fusion technologies to correct errors and estimate PNT with higher precisions, enabling seamless navigation in GPS-denied environments.

Components of Inertial Reference Unit (IRU)

IRU processes data from internal sensors to provide navigation, stabilization, and autonomous functionality to piloted vehicles. Depending on the sensor combination, an IRU can continuously measure angular acceleration, linear velocity, attitude (roll, pitch), position,  platform azimuth, magnetic and true heading, body angular rates, and more. 

The most common IRU sensor components include: 

  • Gyroscopes: Measure rotation rate around three axes —  roll, pitch, and yaw
  • Accelerometers: Mesure linear acceleration to determine changes in speed and position 
  • Magnetometer: Measures the orientation relative to the Earth’s magnetic north to determine heading direction  

All collected raw data is processed on the edge to optimize output accuracy with Kalman Filtering — a probabilistic data fusion algorithm that corrects for noise and drift to provide more accurate PNT data to the vehicle. IRU can also perform dead reckoning calculations to enable navigation in GPS-denied environments. Processed data is immediately transmitted to the flight controller through SPI, I2C, or CAN for real-time course correction. 

IRUs are the key component of many autopilot systems in commercial aviation and maritime vessels. 

Inertial Reference Unit (IRU) vs Internal Measurement Unit (IMU): What’s the Difference?

An internal reference unit and an internal measurement unit are the main components of modern internal navigation systems (INS). But each serves a slightly different purpose. 

  • An IMU only aggregates raw internal sensor data without applying any extra computations. 
  • An IRU collects and processes raw data to create precise position, velocity, and orientation outputs. 

In most cases, the IMU plays a supporting role. It provides extra inputs for redundancy or slight course correction when GPS is used. IMU data can also be combined with visual inputs from cameras or LiDAR to enhance simultaneous localization and mapping (SLAM) algorithmic outputs. Or, in the case of Bavovna, used as part of an AI-powered navigation solution

IRU, in contrast, may work in conjunction or independently from other external systems (e.g., GNSS) to provide a higher degree of autonomy and reliability. But that also comes at a higher price tag. Honeywell’s internal reference systems, used in many private and commercial jets, have a starter price of $300,000. Each also weighs anywhere between 9 and 40 pounds.  This inhibits IRU’s usage for UAVs. 

Bringing the Power of IRU to Drone Navigation 

The main advantage of IRU is on-edge data processing which enables more advanced, autonomous aerial navigation scenarios. But that comes at a ‘cost’ of larger hardware size because commercial IRUs include large processors, memory storage, and sometimes redundant system components. 

IMUs, in turn, are optimized for lightweight applications like drones, which come at a ‘cost’ of limited processing powers. At Bavovna, we’ve decided to solve this problem with the help of AI. Rather than trying to enable continuous on-device processing, we’re pre-training AI algorithms locally on the vehicle flight data. Effectively, we’re helping UAVs develop ‘memory’ and then use it autonomously in the field. 

For example, we’ve successfully trained a model for AirTower flights — fully autonomous ascending and descending to set height, hoovering, and returning to a designated landing point without any GNSS connectivity with a 0.5% positioning error. Similarly, we trained other drone models to fly more complex trajectories with 98% accuracy. 

Discover how Bavovna enables high-precision aerial navigation with an AI-powered INS solution

uav jamming

How to Counter UAV Jamming

Uncrewed aerial vehicles (UAVs) have become the crux of military operations and security patrols. But every drone reconnaissance or patrol mission comes with the almost imminent danger of equipment loss as UAV jamming has become the norm in contested territories. 

How UAV Jamming Works 

All UAV models have one inherent weakness — communication links. So attackers use various techniques to disrupt communication between a drone and its operator or navigation systems by causing signal interference and/or overwhelming signal receivers. The usual targets for jamming are 900 MHz, 2.4 GHZ, and 5.8 GHz frequencies, as well as 4G and 5G frequencies. 

Common UAV jamming techniques include: 

  • Radiofrequency (RF) jamming pollutes the same frequency drone operators use with powerful noise signals or rapidly switching frequencies to meddle with more advanced drone models. 
  • GPS jamming or spoofing: Jammers overwhelm the UAV’s drone receiver, causing it to lose its position data. GPS spoofing is a more advanced technique, used to mislead the onboard systems drone about its location and force it to head elsewhere. 
  • Broadband jamming aims to overpower the UAV’s communication systems by flooding the airspace with high electromagnetic noise to disrupt the UAV’s control link. 
  • Narrowband jamming targets specific coms frequencies, used by drone operators to neutralize the communications. Unlike broadband jamming, there are fewer ‘collateral impacts’ on other airspace users. 

GPS jamming, in particular, is on the rise, especially in the areas of ongoing military conflicts. But the problems also ‘spill’ to other regions. Since the Russian invasion of Ukraine, GNSS jamming and spoofing have increased substantially across the eastern Mediterranean, Baltic Sea, and Arctic regions, according to the European Union Aviation Safety Agency. 

Drone jamming is also on the rise all across the US-Mexican border, where drug traffickers rely on jammers to thwart drones, deployed by the border security forces. Given the ease and low cost of obtaining UAV jamming tools (a device can cost a couple hundred dollars), the question of protection becomes equally important for civilian and military use. 

Levels of GPS interference, recorded on March 23, 2024. Source: GPS World 

How to Counter UAV Jamming 

Various anti-jamming options have emerged to protect the ‘weak link’ in UAV devices—

However, the best protection is removing the underlying vulnerability. GNSS/GPS technology isn’t just susceptible to targeted jamming. Signal also degrades due to natural magnetic interferences—complex geological terrain, high-rise urban structures, and the natural levels of emissions, produced by various equipment.  This complicates drone use in mining, telecom, oil and gas, and many other industries. 

Internal navigation systems (INS) have emerged as an alternative UAV navigational technology. Advances in AI sensor data fusion make INS as reliable as GPS-only navigation. 

Bavovna has developed a low-energy, external navigation system, compatible with most commercial drone models. The devices process data from onboard sensors (accelerometer, gyroscope, compass, etc) and optional external systems (e.g., computer vision camera or LiDAR) with the help of pre-trained AI algorithms to provide reliable navigation in GPS-denied environments. Positioned in an EMI-protected case (which successfully passed EMF tests), Bavovna hybrid INS offers staunch protection against UAV jamming. For extra security, our system can be integrated with SIGINT RF modules. 

Boasting a longer range compared to other internal navigation systems for drones, our system maintains a 98% average accuracy rate, even when flying complex trajectories. All thanks to fine-tuned AI models, trained on live vehicle data, which can compensate for individual sensor deficiencies. 

Discover how Bavovna is securing drone operations with a hybrid INS system

GPS Denied Navigation

GPS-Denied Navigation: 3 Best Solutions

Since 1993, the GPS has been tightly integrated into our daily lives. From recording bank transactions to guiding transatlantic flights, the technology generates about $1 billion a day in economic impact

But just like any other system, GPS has its fair share of limitations. Challenging terrain, signal jamming, and spoofing can render it useless. And that happens quite a lot. Drone use cases in mining are severely limited by naturally occurring magnetic interference. Thousands of commercial flights get affected every year by targeted or incidental GPS signal jamming. Not to mention countless security and military operations, where signal spoofing is the name of the game. 

Source: FT

Soundly, alternative technologies exist for GPS-denied navigation — and here are the top 3 solutions. 

Hybrid INS Powered by AI 

Internal navigation systems (INS) rely on data gyroscopes and accelerometers to estimate the vehicle’s current position, in relation to its last known point. The problem, however, is that many off-the-shelf systems lack accuracy, especially over a longer range. 

Bavovna is changing that with its AI-driven inertial navigation unit. Compact, low-power, and EMI-shielded, Bavovna brings AI sensor fusion technology to UAV navigation. The onboard unit can process data from any number of sensors—accelerometer, gyroscope, compass, barometer,  vector airflow, ultrasonic, infrared, or optical flow sensors—to deliver high-precision navigation in GPS-denied environments. The endpoint positioning error is just under 0.5% even when flying complex routes. Our solution is fine-tuned on live flight data from your vehicle, ensuring unbeatable reliability and durability in a variety of conditions.

For instance, our latest deployment on Aurelia X6 Max Pro-D allows performing fully autonomous air tower missions—vertically take-off, hoover, and land, without any maps or additional correction from GPS or RTK.  With Bavovna’s AI kit, you can safely establish communication relays, perform terrain reconnaissance, perform security monitoring, and fly a range of other missions without worrying about GPS signal degradation, jamming, or loss.  

Quantum Positioning Systems 

The Royal Navy is looking to another emerging technology to improve INS—quantum computing.  Atoms exhibit quantum behavior changes in response to the smallest amount of motion when cooled near absolute zero. These changes can be measured and used to obtain positioning, navigation, and timing (PNT) data. The catch? Cooling down atoms requires huge, power-hungry equipment. 

Aquark Technologies may have found the answer to this quandary. The quantum startup develops miniature quantum systems. Its compact cold atom navigation system uses lasers to bring the temperature down to (-273.15C), which makes it possible to collect motion data on an atomic level and use it for navigation. Aquark Technologies has been successfully tested on a Royal Navy patrol vessel in October 2024. 

Silicon Photonic Optical Gyroscopes (SiPhOG)

Fiber-optic and ring laser gyros offer the best accuracy, but they are also too expensive for many commercial applications. MEMS gyroscopes are way cheaper but lack precision. ANELLO Photonics wants to close this gap with its SiPhOG technology. 

A silicon photonic integrated circuit is used to manufacture the waveguides on-chip, allowing the company to achieve Fiber Optic Gyro performance with a standard silicon manufacturing process. Its INS system has a drift rate of less than 0.5° per hour and demonstrated strong performance in GPS-denied environments. It maintains accuracy within 0.1 m over distances of 0.8 km without GPS, even in orchard environments with limited GPS signals.

Navigation technology moves quickly and better alternatives to GPS are emerging every day. Many also boast high customization like Bavovna’s AI navigation kit, allowing multiple deployment scenarios across different hardware — fixed wing, tilt wing, VTOLs, multi-copters, and FPV drones. Contact us for a free demo!

What is INS?

What is an INS? Definition, Types, and Latest Innovation

Internal navigation system (INS) uses motion and rotation sensors and an onboard computer to determine the vehicle’s position, orientation, and movement speed without using visual references.  

Originally developed in the MIT Instrumentation Laboratory for a B-29 bomber in the 1950s, INS has become a staple for self-contained navigation for aerospace, maritime, and automotive industries. 

How Does an Inertial Navigation System Work?

INS uses dead reckoning to determine the vehicle’s current position by using its last known coordinates as the starting point for comparison. It then provides real-time position and navigation data by correlating changes in starting point, speed, and direction with new sensor inputs.

Most modern internal navigation systems include an inertial measuring unit (IMU) — a sensor subsystem that provides raw data inputs like altitude, position, orientation, angular rate, and linear velocity.

Source: ResearchGate 

Inertial measuring units (IMUs) typically feature the following sensors:

  • Accelerometers to calculate changes in velocity and position
  • Gyroscopes for angular velocity estimation to detect rotational motion
  • Magnetometers to determine movement direction relative to the Earth’s magnetic field
  • Barometers/Altimeters to measure atmospheric pressure for altitude calculations.

Common Types of Internal Navigation Systems 

INSs differ significantly in hardware configuration—each having different tradeoffs in accuracy, cost, and application feasibility. The common INS types are:

  • Strapdown Inertial Navigation Systems (SINS) have sensors strapped directly to the vehicle. They’re lightweight and easy to implement, ideal for drones and light robotics. But SINSs require high computation power to handle sensor noise due to vehicle motion.
  • Gimbaled Inertial Navigation Systems (GINS) use gimbals to ensure greater reference stability. But they’re heavier, more complex, and susceptible to mechanical wear.
  • Fiber Optic Gyro-based Inertial Navigation Systems (FOG INS) leverage fiber optic gyroscopes for precise rotation measurement. FOG is more immune to vibration and environmental interference but costlier. 
  • MEMS-based Inertial Navigation Systems feature accelerometers and gyroscopes, based on Micro Electro Mechanical Systems. They are cost-effective and compact but have lower accuracy than FOG or RLG systems.
  • Ring Laser Gyro-based Inertial Navigation Systems (RLG INS) use ring laser gyroscopes for precise motion measurement. They boast high durability and are vibration-immune, but come at a premium price.
  • GNSS-Aided Inertial Navigation System typically features a 3-axis gyroscope, a 3-axis accelerometer, and a GNSS receiver (and sometimes a 3-axis magnetometer) for navigation. Each contributes different coordinates for high accuracy. The problem? If the GPS is down or lagging, navigation becomes unreliable—and that’s a major limit industries aim to solve.

Inertial Navigation System vs GPS: What’s the Difference? 

INS is a self-contained system that doesn’t require external connectivity (e.g., satellite or wireless networks) to guide the vehicle. As such, it’s less prone to magnetic interferences or targeted attacks, especially with EMI shielding.

GPS, in turn, is a satellite-based system that provides positioning data only when and where there’s an unobstructed line of sight to satellites. This makes it unsuitable for underwater navigation, UAV or aircraft flights in contested environments, or autonomous driving through tunnels or underground shafts.

Due to GPS’s vulnerability to signal loss, interference, and jamming in contested environments, many organizations use INS over GPS. Recent advances in sensor fusion and AI improved internal navigation system accuracy and connectivity range.

Getting more from your INS with AI

Bavovna developed an Al-enhanced INS for uncrewed vehicles that delivers 98% accuracy over a long range and supports fully autonomous flights in GPS-denied environments.

We’ve developed a compact strap-down model with an IMU and AI-powered flight control, weighing only 800g. For navigation, we apply sensor fusion to accelerometer, gyroscope, compass, barometric pressure, and airflow data, with the option to connect more sensors. Each AI model is custom-trained for your uncrewed vehicle on at least 100 hours of live flight data to ensure top accuracy during autonomous flights.

In the field, our hybrid INS system can maintain under 0.85-meter deviation of single-point positioning without GPS, RTK, and optical navigation at 500 meters altitude with 18 m/sec wind. Learn more about Bavovna AI Navigation Kit.