DE102023125817A1 - Method and device for electronic animal monitoring - Google Patents

Abstract

The present invention relates to a method and apparatus for electronic animal monitoring and, more particularly, to a method and apparatus for distributed transmission of monitoring data obtained from electronic monitoring of wild animals such as vultures, pigeons and other bird species to a remote receiver.
A method according to the invention for electronic animal monitoring comprises attaching monitoring devices (10, 20, 30, 40) to at least two animals (1, 2, 3, 4), wherein the monitoring devices (10, 20, 30, 40) are configured to automatically establish a local area network (LN) for wireless short-range data communication; capturing monitoring data with at least one monitoring device (10, 20, 30, 40) and providing the monitoring data as an individually distinguishable data object (DO); segmenting the data object (DO) into at least two data segments (DO1, DO2, DO3, DO4) and distributing the data segments (DO1, DO2, DO3, DO4) to the monitoring devices (10, 20, 30, 40) forming a local area network (LN) at the time of segmentation, wherein the data segments (DO1, DO2, DO3, DO4) are uniquely assignable to the data object (DO); and transmitting the distributed data segments (DO1, DO2, DO3, DO4) by the individual monitoring devices (10, 20, 30, 40) to a remote receiver (70) by means of wireless long-range data communication. A monitoring device (10, 20, 30, 40) according to the invention is configured to carry out a method according to the invention.

Description

The present invention relates to a method and apparatus for electronic animal monitoring and, more particularly, to a method and apparatus for distributed transmission of monitoring data obtained from electronic monitoring of wild animals such as vultures, pigeons and other bird species to a remote receiver.

State of the art

Species conservation and wildlife management are primarily concerned with preserving, promoting, or regulating specific animal species in their natural environment. This requires in-depth knowledge, particularly of their specific habitats, typical animal behavior patterns, population development in a region, and a comprehensive understanding of the influence of a wide variety of factors on these data. Research into the behavior and individual movement patterns of wild animals (especially birds) is not possible through direct observation. The distances covered by the animals are often too great for this. Furthermore, this usually only allows the observation of larger groups, not of individual animals.

For several decades, so-called "wildlife radio telemetry" has been used for such purposes. The devices used for this purpose use global navigation satellite systems to determine the animals' current positions. The position data can be stored (logged) over a long period of time and transmitted to remote receivers, for example, via a cellular connection, a proprietary radio link, or via satellite link.

However, with the miniaturization of electronic cameras, it is now also possible to attach corresponding image or video cameras to animals to study their behavior. The camera data is generally stored first and only read out after the camera has been removed from the animal. The camera can be attached to the animal's fur, for example, using adhesive.

Electronic tracking devices with integrated radio transmitters are particularly well-known for larger animals such as dogs and cats. Such tracking devices usually consist of a single unit that is attached to the animal's body using a harness. This compact unit contains a power supply, sensors, a processing unit, communications technology, and antennas. The tracking devices can also be equipped with additional cameras. However, due to their size and weight, as well as their relatively limited period of use, these conventional devices are not suitable for long-term observation of wild animals, especially small migratory birds. Tracking devices usually require specially trained animals, which are then used, for example, as trained search and rescue dogs to rescue people buried in avalanches or similar situations.

For monitoring even small animals such as birds, there are specially adapted camera systems that operate autonomously (e.g., with solar cells for charging a battery) even for several months. A miniature camera can be combined with a conventional animal transmitter (a so-called "tag") in such a way that it enables photographs from the animal's respective viewing direction ("animal or bird's eye view") for in-depth observations of the animal's behavior, e.g., its feeding or mating behavior. The transmitter can be attached to the animal's back, for example, since integrated antennas provide the best radiation there.

The camera system and animal transmitter can now also be provided as fully autonomous intelligent systems ("smartTags"). The smartTags can be equipped with a camera, a variety of sensors for recording movement and/or environmental data, AI chips for image or sensor data processing, and a communication module for data transmission.

When using smartTags to monitor vultures, for example, a typical recurring behavior pattern is that the animals track down carcasses (or carrion) of large game animals in order to subsequently eat them. An onboard AI can therefore be specifically trained to detect that the birds are currently feeding by evaluating recorded motion sensor data. Since behavioral researchers are interested in learning exactly which carcasses are involved, the camera can capture corresponding images of the dead animal. These can then be classified according to specific criteria using the onboard AI. The classification result can then be transmitted to a remote receiver via the communication module. Since vultures, in particular, typically reside in remote areas, data transmission usually has to take place via a satellite connection. For long-term monitoring, Robust and particularly energy-efficient IoT transmission methods are suitable, for example mioty/TS-UNB.

However, there are also scenarios where not only the classified information (typically the type of animal and the GPS position of the location) needs to be transmitted, but also the captured image from which, for example, the animal species was derived during classification or, in particular, could not be clearly derived. Such image files typically range in size from several kB to several MB. However, the large data volumes push traditional IoT protocols to their limits, as these typically assume data sizes in the range of only a few bytes (approximately 10-245 bytes). However, large amounts of data can certainly be transmitted using such protocols. This typically involves fragmenting the data into smaller packets, which can then be transmitted serially as individual packets to a remote receiver and subsequently reassembled there. However, such a process usually takes a considerable amount of time (up to several minutes), places a significant strain on the transmitter's battery ("continuous transmission"), and is also very error-prone (e.g., connection interruption or failure, battery empty, etc.). It could also happen that during a satellite overpass, it is not possible to transmit all data packets completely, and therefore it is necessary to wait until the transmitter is overpassed again by a corresponding satellite. This significantly increases the transmission latency.

Disclosure of the invention

It is therefore an object of the present invention to provide an improved method and apparatus for electronic animal monitoring which avoids or at least significantly reduces the problems occurring in the prior art.

These objects are achieved according to the invention by the features of the independent patent claims. Advantageous embodiments of the invention are contained in the respective subclaims.

A first aspect of the present invention relates to a method for monitoring animals, comprising attaching monitoring devices to at least two animals, wherein the monitoring devices are configured to automatically establish a local network for wireless short-range data communication; capturing monitoring data with at least one monitoring device and providing the monitoring data as an individually distinguishable data object; segmenting the data object into at least two data segments and distributing the data segments to the monitoring devices forming a short-range network at the time of segmentation, wherein the data segments are uniquely assignable to the data object; and transmitting the distributed data segments by the individual monitoring devices to a remote receiver by means of wireless long-range data communication.

The main idea of the present invention is therefore an intelligent distribution of the amount of data to be transmitted among different monitoring devices. For this to succeed, at least two such monitoring devices must be available for distribution. Accordingly, the monitoring device must be attached to at least two animals, although for redundancy reasons, more than one monitoring device can be attached to one animal. The animals can in particular be birds, preferably birds from a common flock. However, the monitoring devices can also be attached to different animals across species. Within a common habitat, for example, different scavengers such as hyenas and vultures could be equipped with corresponding monitoring devices.

The monitoring devices should be configured to automatically establish a local network for wireless short-range data communication. Since said monitoring devices are intended for autonomous use with the greatest possible energy efficiency, the establishment and control of a local network for wireless short-range data communication should be carried out entirely locally. For this purpose, it is particularly preferred that the short-range network be established using a Bluetooth connection.

The monitoring devices are intended to enable the acquisition of monitoring data and the provision of the monitoring data as an individually distinguishable data object. The data objects can be, for example, sensor data, image data, classification data obtained from the image data, etc. These can each form individual data objects or be linked to form a data array. The data object preferably comprises an image or video together with a unique identification feature.

Preferably, the at least one monitoring device for capturing monitoring data comprises an automatic, preferably by means of Artificial intelligence (AI) controls a camera for capturing images or videos. The AI (or a corresponding AI chip or machine learning system) can be specifically trained to automatically recognize certain animal behaviors. The camera can then be controlled accordingly to capture images or videos based on the situation. The camera's energy consumption is minimized by only capturing images when a behavioral situation previously determined by the AI can actually be observed.

In the application example mentioned above with the smart tags attached to vultures, a data object can, for example, summarize and bundle all available information on a single such behavioral situation. Specifically, a corresponding data object could thus include the exact time and/or location as an example of a unique identifying feature, an image of the discovery situation, the type of animal carcass identified by classification using image recognition, and/or other environmental data such as temperature and humidity. Furthermore, the data object can also combine or bundle data from multiple monitoring devices. A corresponding combined data array can then potentially comprise several MB, with the size being able to be adapted to the actual information requirements of the remote recipient using efficient compression algorithms. However, the data object can also be simply an image or video recorded by a camera.

According to the invention, the data object is then segmented into at least two data segments and the data segments are distributed to the monitoring devices forming a short-range network at the time of segmentation. For this purpose, a corresponding short-range network must have been automatically established beforehand, and the individual monitoring devices must have identified themselves as active participants in the network. This identification can also provide an indication of possibly limited (e.g., low battery level, limited satellite connection, little free storage space) or completely excluded (e.g., minimal battery level, no satellite connection, no free storage space) participation in the method, so that these indications can be taken into account accordingly during segmentation as part of a distribution plan. However, a method according to the invention provides for segmentation into at least two data segments and for at least one data segment to be subsequently distributed to at least one further monitoring device for transmission.

Preferably, the data object can be segmented based on static rules or adaptively based on at least one feature of the monitoring devices forming a short-range network at the time of segmentation. A static rule would, for example, be a fixed segmentation of the data object into a number of data segments corresponding to the number of monitoring devices forming a short-range network at the time of segmentation, without taking specific characteristics of the individual monitoring devices into account. In contrast, adaptive segmentation should be able to take these characteristics into account accordingly.

Preferably, the data object is segmented based on size and/or content. With size-based segmentation, only the size of the individual data segments is specified. For example, a specification or rule for size-based segmentation could include that all data segments should have a uniform size (e.g., 10 kB each). With content-based segmentation, on the other hand, the content of images, for example, is taken into account during segmentation. If, for example, a specific type of animal carcass is identified in an image through classification and image recognition, the edge of the image or the area around the location where it was found may not be important in the image data. These areas can then be excluded from the outset when segmenting the data object containing the image data, or they can only be encoded in the data segments with low image quality or detail.

The original data object is distributed to at least two monitoring devices after segmentation according to the invention. This is in contrast to the prior art, where even the fragmented data objects are reserved for long-range data transmission in only one monitoring device. It should be noted that the term "segmentation" is used to distinguish it from the term "fragmentation," whereby the term "segmentation" should be understood more broadly in terms of content. While segmentation should also include simple fragmentation (in the sense of simply breaking a data object down into individual fragments), methods that go beyond pure fragmentation can also be applied. For example, the generated data segments can be provided with error correction across the data segments. The data object can therefore also first be placed in a suitable transport container (e.g., using a lossless and fault-tolerant communication device) for segmentation. compression method), which enables error-redundant decomposition into individual data packets as interconnected data segments. The individual data segments can then be individually fragmented within the monitoring devices for further transmission using an IoT protocol, also according to conventional transmission methods according to the state of the art. The segmentation according to the invention thus primarily relates to the distribution of data objects in the short-range network. The subsequent transmission of the distributed data segments by the individual monitoring devices to a remote receiver can then be carried out according to the state of the art or using any other transmission protocol suitable for the required applications. Wireless long-range data communication preferably takes place via a satellite connection (SAT) or an existing terrestrial telecommunications network (e.g., GSM, 3G, 4G, 5G, WiFi).

Preferably, the distributed data segments are transmitted in a decentralized manner by the individual monitoring devices based on individual criteria of the respective monitoring devices. Individual criteria can be, for example, a low charge level of a power supply battery, poor reception conditions, or another ongoing monitoring process. In particular, after receiving the data segments for transmission to a remote receiver, the individual monitoring devices can behave like a conventional monitoring device according to the state of the art.

A second aspect of the present invention relates to a monitoring device for monitoring animals, comprising a camera for recording images or videos; at least one sensor for recording movement and/or environmental data; at least one KI chip for image or sensor data processing; and a communication module for data transmission by means of short-range and long-range data communication; wherein the monitoring device is configured to carry out a method according to the invention for monitoring animals.

The advantages of the device according to the invention result directly from the above description of the method and apply accordingly to the device described.

The inventive approach to solving the technical problem mentioned at the outset is therefore based on the fact that instead of a conventional serial transmission of data from a single transmitter (i.e. from a monitoring device set up for remote data transmission), the data transmission is parallelized with the aid of other cooperative transmitters in a local network.

The parallelization of data transmission will be illustrated below using a practical example of vultures equipped with smart tags (or conventional tags without AI technology) as a monitoring device.

Vultures are known to operate in loose flocks, staying relatively close to each other, within a few meters, especially for feeding. This flocking behavior can be exploited for improved tag-based monitoring of vultures.

For the method according to the invention, at least two of the birds in such a flock should be equipped with transmitters. In addition to the modules intended for recording monitoring data, these transmitters should also have a suitable near-field communication module (e.g., Bluetooth) to establish a local near-field communication network with the other transmitters in the flock. This can be, in particular, a so-called ad hoc network or, more precisely, an Extreme Edge Network (EEN).

One of the vultures (vulture A) can now autonomously capture an image or video, process it within the surveillance device, and, if necessary, decide whether the captured image should be transmitted to a remote receiver. According to the invention, however, this transmission no longer takes place directly from this transmitter to a corresponding satellite; instead, a so-called "cooperative backhauling" approach is used. Satellite-based IoT networks, in particular, can be used for remote data transmission.

The image captured by Vulture A can first be compressed. Conventional standard image compression methods such as JPG or HEIC can be used for this purpose. The current HEIC standard is preferred, as it typically allows for very small file sizes.

The captured image is then segmented. This can, for example, be a simple fragmentation, meaning the entire file is split into several file parts. The number and size of the fragments can depend on the size of the local network (i.e., how many participants are visible and active). were registered using the method according to the invention) and the performance of the individual tags (e.g. battery, transmission power, processing power, etc.).

Once segmentation is complete, Vulture A can distribute the individual data segments to the other visible tags in the area. Each of the other tags can then transmit one or more data segments to a remote receiver, for example, via a separate satellite connection. Acceptance and successful transmission of the individual data segments is confirmed to the initiating network node, in this case Vulture A. At the remote receiver, the data fragments can then be reassembled into the original file (data object).

The fragmentation and compression steps can also be reversed. This would mean that the entire image is first divided into smaller sub-images, each of which is then compressed separately. This variant has the advantage that when the images are merged, a missing sub-packet represents exactly one missing piece in the image, while the rest of the image can be displayed without any problems. Additionally, an error protection or error correction method can be added for the individual data fragments. For example, a checksum or the application of raptor codes (fountain codes) can be used to reconstruct individual, incorrectly transmitted sub-packets. In a segmentation process according to the invention, these steps can be combined in any way.

In order to assemble the data, the remote receiver must know (e.g., by appropriately indexing or marking the individual data segments or packets) that they represent part of a larger file or data object. All cooperative senders in the local network can also use a common ID when transmitting the individual data segments. The remote receiver therefore does not need to know in advance which senders were responsible for dividing the fragments into their respective transmissions. From the remote receiver's perspective, all fragments are received by the "same" sender. From the receiver's perspective, this means there is no difference compared to a serial transmission by just one sender according to the state of the art. The "special" ID can, for example, be a common group ID that is used exclusively for distributed transmission across multiple senders. The "normal" or unique ID of the senders can still be used for all other transmissions.

A method according to the invention offers significant advantages over the prior art. Parallelizing data transmission reduces its effective latency. This is further reduced by increasing the probability of data segments being transmitted during a satellite overpass. The method also enables more robust, secure, and stable data transmission (e.g., in the event of a transmitter failure). Using suitable fragmentation/coding methods or a suitable transport container, fault-tolerant transmission can be achieved, allowing at least parts of the data object to be evaluated despite transmission errors or packet loss. Finally, the transmission energy required for long-range data transmission is distributed across multiple monitoring devices, significantly reducing the load on each individual monitoring device compared to the prior art. This increases the service life of a monitoring device according to the invention, and existing batteries for self-sufficient energy supply do not need to be recharged as quickly.

The following is a concrete example scenario for a practical application of a method according to the invention or corresponding monitoring devices. This case study is not intended to limit the present invention.

The initial situation can be such that three vultures are equipped with transmitters (smart tags) as surveillance devices according to the invention. These tags include, among other things, sensors for motion detection (accelerometers), a camera for image capture, a GPS receiver, Bluetooth for near-field data communication, and a mioty transmitter for direct communication with a satellite network. A microprocessor and a microcontroller are used to evaluate the data.

Given this initial situation, the following scenario could occur: The vultures find a large animal carcass and settle down to feed. The AI processors detect that the vultures are feeding and trigger the capture of an image with one or more cameras. The images can then be stored locally, along with the GPS position, in non-volatile memory on the individual smartTags. In the background, an ad hoc communication network is established between the tags using Bluetooth. This can then be used to communicate information about the status of the individual smartTags to each participant in the network (e.g., battery level, current utilization, or availability to perform other tasks).

The captured images are then evaluated by the AI. For example, the evaluation results in a carcass classification of "zebra" for one tag and "springbok" for another. However, the classification of the third tag is unsuccessful. Therefore, this image is to be sent to a server for human evaluation. Due to the data exchange that has already taken place between the tags, Tag 3 "knows" that the other two tags have a functioning satellite module available and are not currently performing any other tasks, meaning they are ready to transmit data. Tag 3 then segments the data to be sent into three equal parts. This means that the image file is processed byte by byte and split at 1/3 and 2/3 of the file size (size-based segmentation). Metadata is also appended to the data: This can be a unique identifier for the entire package and an index, information about the location and time of the image capture, etc. The prepared data segments are transmitted to Tag 1 and Tag 2 via Bluetooth.

All tags then begin transmitting data via satellite (this does not have to happen simultaneously, and can, in particular, occur independently of one another). After reception by a satellite network, the individual data segments are available at the ground station of a remote receiver (gateway). Based on the metadata, the server can reassemble these data segments into an overall image or file and make them available for further analysis.

Further preferred embodiments of the invention result from the features mentioned in the subclaims.

The various embodiments of the invention mentioned in this application can be advantageously combined with one another, unless otherwise stated in individual cases.

Brief description of the drawings

• 1 eine schematische Darstellung eines Verfahren zur Überwachung von Tieren gemäß Stand der Technik,
• 2 eine schematische Darstellung zur Segmentierung und Verteilung eines beispielhaften Datenobjekts bei einem erfindungsgemäßen Verfahren zur Überwachung von Tieren,
• 3 eine schematische Darstellung zur Übertragen der Datensegmente eines beispielhaften Datenobjekts bei einem erfindungsgemäßen Verfahren zur Überwachung von Tieren, und
• 4 eine schematische Darstellung einer beispielhaften Ausführungsform einer erfindungsgemäßen Überwachungsvorrichtung.

The invention is explained below in exemplary embodiments with reference to the accompanying drawings. It shows:

• 1 a schematic representation of a method for monitoring animals according to the state of the art,
• 2 a schematic representation of the segmentation and distribution of an exemplary data object in a method according to the invention for monitoring animals,
• 3 a schematic representation for transmitting the data segments of an exemplary data object in a method according to the invention for monitoring animals, and
• 4 a schematic representation of an exemplary embodiment of a monitoring device according to the invention.

Detailed description of the drawings

1 shows a schematic representation of a method for monitoring animals 1 according to the prior art. The exemplary method according to the prior art essentially comprises attaching a monitoring device 10 to an animal 1, capturing monitoring data with the monitoring device 10 and providing the monitoring data as an individually distinguishable data object DO; fragmenting the data object DO into at least two data fragments; and transmitting the at least two data fragments by the monitoring device 10 to a remote receiver 70 by means of wireless long-range data communication. In the example shown, the wireless long-range data communication takes place via a satellite 60. The monitoring device 10 can comprise a camera 12 for recording images or videos in order to capture monitoring data.

As already described at the beginning of the prior art, this has the disadvantage that the transmission protocols used are not suitable for transmitting large amounts of data, as is typically required for image transmissions, and therefore the data object must be fragmented into individual data packets for serial transmission. However, due to the typically low transmission rates, the transmission of the fragments is very time-consuming and error-prone. In addition, a lot of energy is required for the transmission, which is only available to a limited extent due to the autonomous energy supply of the monitoring device 10. In the case of low energy reserves, it may be necessary to wait for the battery of the monitoring device 10 to be charged via a solar cell. In addition, further delays in the transmission can occur due to a poor or non-existent communication connection with the satellite 60.

2 shows a schematic representation for the segmentation and distribution of an exemplary data object DO in a method according to the invention for monitoring animals 1, 2, 3, 4. The method shown for monitoring animals comprises attaching monitoring devices 10, 20, 30, 40 to four animals 1, 2, 3, 4, wherein the monitoring devices 10, 20, 30, 40 are used for automatically setting up a local network LN for wireless short-range data communication. nication are set up; a recording of monitoring data with at least one monitoring device 10, 20, 30, 40 and provision of the monitoring data as an individually distinguishable data object DO. The figure shows in particular a segmentation of the data object DO into four data segments DO1, DO2, DO3, DO4 and a distribution of the data segments DO1, DO2, DO3, DO4 to the four monitoring devices 10, 20, 30, 40 forming a short-range network LN at the time of segmentation, wherein the data segments DO1, DO2, DO3, DO4 should be unambiguously assignable to the data object DO. Also shown is a satellite 60 for wireless long-range data communication.

For example, animals 1, 2, 3, and 4 are supposed to be vultures from a common flock. However, monitoring devices 10, 20, 30, and 40 can also be attached to different animals across species. The short-range network LN shown is preferably established via Bluetooth connection.

For example, the DO data object is an image together with a unique identifier (DOI). The image in the figure is supposed to show a dead animal that the monitored vultures are approaching to feed on. A unique identifier (DOI) can be, for example, the time of capture, the location (GPS coordinates), an image number, or a combination thereof.

The segmentation of the data object DO can be performed based on static rules or adaptively based on at least one feature of the monitoring devices 10, 20, 30, 40 forming the shown short-range network LN at the time of segmentation. A static rule can, for example, consist of dividing the data object for segmentation by the number (or an integer multiple thereof) of monitoring devices 10, 20, 30, 40 forming the short-range network LN at the time of segmentation. Each monitoring device 10, 20, 30, 40 can then be provided to transmit packets of data segments of equal size. Adaptive segmentation based on at least one feature first requires knowledge of the corresponding feature of the monitoring devices 10, 20, 30, 40. For example, the feature can include the charge level of the battery of the individual monitoring devices 10, 20, 30, 40. Assuming that the transmission conditions for all monitoring devices 10, 20, 30, 40 in a swarm are approximately the same, it may be useful to segment them in such a way that more data packets are allocated to a monitoring device 10, 20, 30, 40 with a high battery charge level than to a monitoring device 10, 20, 30, 40 with only a low battery charge level. In this respect, the transmission effort can be optimally adapted to the available resources through adaptive segmentation.

The segmentation of the DO data object can be performed, in particular, based on size and/or content. Content-based means that, for example, based on a previous image analysis for classification, individual image parts can be weighted so that the particularly relevant image parts (e.g., those showing the animal carcass) are preferentially segmented. This allows for increased transmission reliability by adding additional redundancy or optimal information distribution.

3 shows a schematic representation for transmitting the data segments DO1, DO2, DO3, DO4 of an exemplary data object DO in a method according to the invention for monitoring animals 1, 2, 3, 4. The situation shown essentially corresponds to that of 2 However, in particular, a transmission of the distributed data segments DO1, DO2, DO3, DO4 by the individual monitoring devices 10, 20, 30, 40 to a remote receiver 70 by means of wireless long-range data communication is shown. In the example shown, this is carried out via a satellite 60. However, the wireless long-range data communication can also or additionally be implemented via a terrestrial telecommunications network.

Preferably, the transmission of the distributed data segments DO1, DO2, DO3, DO4 is carried out in a decentralized manner by the individual monitoring devices 10, 20, 30, 40 based on individual criteria of the respective monitoring devices 10, 20, 30, 40. This means that after the transmission of the data segments DO1, DO2, DO3, DO4 to the individual monitoring devices 10, 20, 30, 40, these can independently transmit the received data segments DO1, DO2, DO3, DO4 to the remote receiver 70 via wireless long-range data communication. In particular, they can first wait for the battery to be recharged via an optional solar cell, for example, due to a low charge level of a battery used for power supply.

4 shows a schematic representation of an exemplary embodiment of a monitoring device 10 according to the invention. The animal 1 shown is said to be a pigeon The split monitoring device 10 for monitoring animals 1, which is connected to the pigeon here, for example, via a harness, comprises a camera 12 for recording images or videos; at least one sensor 14 for recording movement and/or environmental data; at least one KI chip 16 for image or sensor data processing; and a communication module 18 for data transmission by means of short-range and long-range data communication; wherein the monitoring device 10 is configured to carry out a method according to the invention. A monitoring device 10 according to the invention can further comprise a solar cell or power supply 19. In particular, a solar cell 19 can be used to charge and maintain the charge level of a battery of the power supply 19.

List of reference symbols

1

first animal

2

second animal

3

third animal

4

fourth animal

10, 20, 30, 40

monitoring device

12, 22, 32, 42

camera

14, 24, 34, 44

Sensor (e.g. GPS, position or acceleration sensor)

16, 26, 36, 46

AI chip

18, 28, 38, 48

Communication module (e.g. Bluetooth, WiFi, satellite)

19

Solar cells or energy supply (e.g. battery)

60

Satellites

70

remote recipient

LN

local network

DO

Data object

DO1, DO2, DO3, DO4

Data segments

DOI

Identification feature (“data object identifier”)

Claims (10)

A method for monitoring animals, comprising: Attaching monitoring devices (10, 20, 30, 40) to at least two animals (1, 2, 3, 4), wherein the monitoring devices (10, 20, 30, 40) are configured to automatically establish a local area network (LN) for wireless short-range data communication; Acquiring monitoring data with at least one monitoring device (10, 20, 30, 40) and providing the monitoring data as an individually distinguishable data object (DO); Segmenting the data object (DO) into at least two data segments (DO1, DO2, DO3, DO4) and distributing the data segments (DO1, DO2, DO3, DO4) to the monitoring devices (10, 20, 30, 40) forming a local area network (LN) at the time of segmentation, wherein the data segments (DO1, DO2, DO3, DO4) are uniquely assignable to the data object (DO); Transmitting the distributed data segments (DO1, DO2, DO3, DO4) by the individual monitoring devices (10, 20, 30, 40) to a remote receiver (70) by means of wireless long-range data communication. Procedure according to Claim 1 , wherein the animals are birds, preferably birds of a common flock, and/or wherein the monitoring devices are attached to different animals across species. Procedure according to Claim 1 or 2 , whereby the short-range network (LN) is established via Bluetooth connection. Method according to one of the preceding claims, wherein the at least one monitoring device (10, 20, 30, 40) for acquiring monitoring data comprises a camera (12, 22, 32, 42) for recording images or videos, which is controlled automatically, preferably by means of artificial intelligence, AI. Method according to one of the preceding claims, wherein the data object (DO) comprises an image or video together with a unique identification feature (DOI). Method according to one of the preceding claims, wherein the segmentation of the data object (DO) is carried out on the basis of static rules or adaptively on the basis of at least one feature of the monitoring devices (10, 20, 30, 40) forming a short-range network (LN) at the time of segmentation. Procedure according to Claim 6 , whereby the segmentation of the data object (DO) is size and/or content-based. Method according to one of the preceding claims, wherein the transmission of the distributed data segments (DO1, DO2, DO3, DO4) is carried out in a decentralized manner by the individual monitoring devices (10, 20, 30, 40) on the basis of individual criteria of the respective monitoring devices (10, 20, 30, 40). A method according to any one of the preceding claims, wherein the wireless long-range data communication is via a satellite link or a terrestrial telecommunications network. A monitoring device (10, 20, 30, 40) for monitoring animals (1, 2, 3, 4), comprising: a camera (12, 22, 32, 42) for recording images or videos; at least one sensor (14, 24, 34, 44) for recording movement and/or environmental data; at least one AI chip (16, 26, 36, 46) for image or sensor data processing; and a communication module (18, 28, 38, 48) for data transmission by means of short-range and long-range data communication; characterized in that the monitoring device (10, 20, 30, 40) is configured to carry out a method according to one of the preceding claims.

Images