Aquatic drone with sonar for the recognition of marine species using artificial intelligence

WO2026159385A1PCT designated stage Publication Date: 2026-07-30PICÓN SAEZ ANDRÉS FRANCISCO
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
PICÓN SAEZ ANDRÉS FRANCISCO
Filing Date
2026-01-23
Publication Date
2026-07-30

Smart Images

  • Figure ES2026070038_30072026_PF_FP_ABST
    Figure ES2026070038_30072026_PF_FP_ABST
Patent Text Reader

Abstract

The aquatic drone with sonar for the recognition of marine species using artificial intelligence comprises a floating hull (1), motors and a rudder system powered by a power supply (2), a control unit (3) with a communications module (8), and a positioning system (4), and also comprises a sonar system (5), formed by a transmitter and one or more receivers with a display unit, and wherein the control unit (3) comprises an artificial intelligence module, trained to locate, analyse and classify marine species based on the acoustic signals from the sonar and the environmental, spatial and temporal information captured by the drone.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] DESCRIPTION

[0002] Aquatic drone with sonar for marine species recognition using artificial intelligence

[0003] TECHNICAL SECTOR

[0004] This application concerns an aquatic drone with sonar, an autonomous vehicle, which incorporates a series of improvements for use as an assistant in fishing operations, generally marine, by recognizing species using artificial intelligence. It also performs useful functions in other fields such as cartography, geology, marine biology, fisheries, etc.

[0005] STATE OF THE ART

[0006] The marine ecosystem faces increasing pressure due to intensive fishing practices, pollution, and climate change. Traditional methods, such as indiscriminate fishing with little selectivity in gear, generate negative impacts, including:

[0007] - Incidental catches of non-target species, many of which have no economic value but do have ecological value.

[0008] - Discarding of non-marketable fish, which affects biodiversity and contributes to the waste of marine resources.

[0009] - Alteration of natural behaviors of species due to intensive artificial lighting.

[0010] The invention offers an ecological alternative that significantly reduces these effects, promoting more selective and responsible fishing, aligned with international marine conservation goals.

[0011] The applicant is unaware of any device similar to the invention or that offers similar advantages.

[0012] BRIEF EXPLANATION OF THE INVENTION The invention relates to an aquatic drain with sonar for the recognition of marine species using artificial intelligence, according to the claims. Its embodiments improve and solve problems of the art.

[0013] This drone allows the selection of fished species, avoiding damage to species without economic value, in addition to optimizing fishing resources and reducing hydrocarbon consumption, since it is not necessary to move the vessel until the school of fish is detected and the figure of an auxiliary vessel in the process is eliminated.

[0014] The underwater device with sonar for marine species recognition using artificial intelligence, in its most complete form, offers a significant improvement to fishing techniques. Powered by electric batteries, it further reduces greenhouse gas emissions, contributing to the fight against climate change. It is an environmentally friendly alternative to traditional methods, aligning with the principles of sustainability and marine conservation, while promoting more selective and responsible fishing.

[0015] Among others, it offers the following advantages:

[0016] 1. Efficiency: Reduction of time and resources in locating schools of fish.

[0017] 2. Sustainability and profitability: More selective fishing, minimizing discards and incidental catches, as well as saving resources and additional maintenance and labor in handling an extra vessel.

[0018] 3. Energy savings: Lower fuel consumption as it is powered by electric charging batteries and has a lower weight and greater maneuverability.

[0019] 4. Reduced environmental impact: Bioluminescence simulation is less disruptive than traditional light-emitting boats and produces fewer hydrocarbon emissions. 5. Adaptability: The drone can operate in diverse aquatic, marine, or lake environments, facilitating fishing even in hard-to-reach areas.

[0020] This approach combines technological innovation with sustainability, modernizing fishing, especially purse seine fishing, and adapting it to current environmental demands.

[0021] The most basic version of the AI-powered marine species recognition sonar-equipped underwater drone comprises a floating housing, motors, and a rudder system powered by a power source. It also includes a control unit with a communications module and a positioning system. Furthermore, it comprises a sonar system consisting of a transmitter and one or more receivers with indicator units that convert the received acoustic signal into an electrical signal, which is displayed on the control unit. This signal is transmitted directly to the drone's central module, which includes integrated AI software with machine learning technology designed to locate, analyze, and classify marine species based on the sonar's acoustic signals and the environmental, spatial, and temporal parameters captured by the drone's interface and sensors.

[0022] Likewise, the drone will use as a data source some marine species pre-programmed from global databases such as “FishBase”.

[0023] The power source is preferably a battery rechargeable by solar cells or by a small wind turbine.

[0024] Ideally, the drone has environmental sensors to measure water salinity, temperature, or the direction of ocean currents, among other variables.

[0025] One preferred embodiment of the drone has a lighting system that simulates natural bioluminescence.

[0026] The rest of the memoir describes other specific achievements.

[0027] DESCRIPTION OF THE FIGURES

[0028] A section of drawings is included, which depicts the following for illustrative purposes only:

[0029] Figure 1: Diagram of an example drone.

[0030] Figure 2: Schematic top view of an example of purse seine fishing where the drone example from Figure 1 is applied.

[0031] Figure 3: Shows an example of a drone capturing information from a school of fish.

[0032] Figure 4: Schematic of an example control unit. MODES OF IMPLEMENTATION OF THE INVENTION

[0033] Next, a brief description is given of one way of carrying out the invention, as an illustrative and non-limiting example thereof.

[0034] The aquatic drone with sonar for marine species recognition using artificial intelligence, as shown in the embodiment, consists of a floating housing (1), generally hydrodynamic for easy movement across the water's surface. The drone's total weight depends on whether it is used in fresh or salt water, thus controlling its buoyancy. Motors and a rudder system allow for controlled movement of the housing (1). The housing (1) may be shaped like a catamaran hull.

[0035] The drone comprises a power source (2), generally a solar-rechargeable battery, although it may include a small wind turbine as an alternative or complement, or be charged solely by cable when on the associated fishing vessel. The battery may be removable. It is possible, but less preferred, for it to have an internal combustion engine or fuel cell.

[0036] The drone has a control unit (3) that connects to the motors and the rudder, and a positioning system (4) that can be relative to the vessel (for example, by triangulation to two antennas on the vessel) or total (geopositioning, for example, GPS or Galileo).

[0037] The drone has a sonar system (5), consisting of a transmitter and one or more receivers. The sonar system (5) allows it to detect objects with acoustic characteristics different from water, primarily fish.

[0038] The control unit (3), with direct transmission to the drone's central module, comprises integrated artificial intelligence software designed to locate, analyze, and classify marine species. This allows for the identification of detected species before capture, significantly reducing bycatch and unwanted discards. It relies on the size detected by the sonar system (5) and other factors such as whether or not the fish belong to a school, the depth at which they are located, etc. Other potentially relevant factors, depending on the species, include position (in terms of sea or ocean region, distance from the coast), time of year, ocean temperature, and salinity readings from the drone's own sensors. (6)

[0039] Preferably, the drone has sensors (6) to measure water salinity, temperature and / or the direction of ocean currents.

[0040] The training of the artificial intelligence module can utilize a global database of fish species such as “FishBase”.

[0041] The underside of the housing (1) preferably incorporates a lighting system (7), typically LED, that simulates natural bioluminescence. This lighting system (7) attracts fish to the surface in a controlled and less disruptive manner than traditional methods, such as light boats. This capability improves fishing operations by facilitating the precise location of schools of fish without the need for intensive or energy-intensive methods. The lighting system (7) is located at the bottom of the housing (1), submerged, emitting flashes in a wavelength range of 470-490 nm, blue-green light. The flashes are intermittent, and the control unit (3) or its own microcontrollers adjust the intensity, power, or frequency of the flashes according to the specific species, as well as according to the ambient light (measured by photometers or based on the time of day). The system can be switched on and off manually, based on the photometer readings.

[0042] The control unit (3) is associated with a communications module (8) for communication with the associated vessel, either directly or via a land-based system. The communications module (8) will have sufficient range, preferably several kilometers, to allow it to operate away from the vessel.

[0043] The control unit (3) will be able to recreate realistic three-dimensional scenarios from the drone's captured surrounding environment using the echoes generated by the sonar system (5) and the artificial intelligence module.

[0044] In operation, the user programs and trains the control unit (3) according to the fishing area, using historical data or migration patterns. The drone is then taken to the area, either on board or tethered by a line to the stern of the vessel. Once the fishing area is reached, the drone is deployed, using its sonar to detect schools of fish while measuring environmental parameters such as temperature, depth, and currents. When the drone detects a school, the acoustic data is processed by the artificial intelligence module to estimate the species, the size of the school, and the approximate horizontal and vertical distance. If the school and species match the fishing targets, the drone approaches to a preset distance, usually a range between a maximum and minimum value, and activates its lighting system (7), simulating bioluminescence to attract the fish to the surface.

[0045] Simultaneously, the drone transmits the location of the school of fish and the environmental conditions to the vessel, where it can be observed on a command unit.

[0046] From that moment on, the vessel carries out its fishing work (Figure 2), while the drone tracks the school to inform the vessel of changes, activate or deactivate the lighting system (7), and other actions to control the behavior of the school.

[0047] Fishing methods will depend on the species, and may include purse seine fishing or other methods.

[0048] Once the work is completed, the drone returns to the vessel, either autonomously or controlled from the command unit.

[0049] All the information gathered during the operation (temperature, currents, species detected, seabed shape, presence of rocks, shipwrecks, etc.) can be used as a fisheries monitoring tool by fisheries inspectors. It also allows for adjustments to future fishing operations, identifying patterns and improving the drone's location and operational efficiency.

Claims

CLAIMS 1- An aquatic drone with sonar for recognizing marine species using artificial intelligence, comprising a floating housing (1), motors and a rudder system powered by a power supply (2), a control unit (3) with a communications module (8), and a positioning system (4) characterized in that it comprises a sonar system (5), formed by a transmitter and one or more receivers with an indicator unit, and in that the control unit (3) comprises an artificial intelligence module, trained to locate, analyze and classify marine species from the acoustic signals of the sonar and the environmental, spatial and temporal information captured by the drone. 2- Aquatic drone with sonar for the recognition of marine species by means of artificial intelligence, according to claim 1, characterized in that the power source (2) is a rechargeable battery powered by solar cells or a small wind turbine. 3- Aquatic drone with sonar for the recognition of marine species by means of artificial intelligence, according to claim 1, characterized in that it has sensors (6) to measure the salinity of the water, the temperature or the direction of the marine currents. 4- Aquatic drone with sonar for the recognition of marine species by means of artificial intelligence, according to claim 1, characterized in that it has a lighting system (7) that simulates natural bioluminescence.