A radar monitoring station for animal migration
The radar monitoring station addresses the operational challenges of existing systems by using a 3-6 mm wavelength radar and computer-processed data to effectively monitor and identify animal behaviors and migration patterns, improving data collection and reducing human resource requirements.
Patent Information
- Application Number
- PCT/PL2023/050113
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-12-29
- Publication Date
- 2025-07-03
AI Technical Summary
Existing pest control systems for monitoring animal migration, particularly land animals like rodents, are difficult to operate, costly, and lack effective data on animal behavior, especially in complex environments with obstacles.
A radar monitoring station using a radar transceiver operating in 3-6 mm wavelengths, combined with computer-processed radar and video data, identifies animal positions, sizes, and behaviors, and is equipped with sensors and communication modules to provide comprehensive animal monitoring.
Enables detailed data collection on animal behavior and migration patterns, even in obstructed areas, providing species identification and event detection, enhancing operational efficiency and reducing human resource needs.
Smart Images

Figure PL2023050113_03072025_PF_FP_ABST
Abstract
Description
[0001] A RADAR MONITORING STATION FOR ANIMAL MIGRATION
[0002] The present application relates to a radar monitoring station for animal migration. The monitoring station disclosed in the present application finds its use in pest control or in monitoring a local wildlife ecosystem.
[0003] Distributed systems for pest control exist in the prior art, but they are difficult to operate and suffer from many drawbacks regarding their maintenance, expressed in high operating costs and in the need to allocate a significant human resources. The effectiveness of known systems is difficult to measure as their data describing animal behavior is minimal or non-existent.
[0004] CN109164445A discloses a FM (Frequency modulation) stepped-frequency waveform design method of a distributed insect radar. The radar transmitting FM stepped-frequency waveform of n FM stepped-frequency systems is adopted in the distributed insect radar; n is an integer, which is greater than or equal to 2; for each radar, the frequency point of a frequency stepping sub-band signal transmitted inthe single frequency band is staggered in time; each radar is designed with three kinds of distance waveforms, namely the shortdistance waveform, the middle-distance waveform and the long-distance waveform; and the coverage ranges of the three kinds of distance waveforms are joined. The invention further provides a use method of the distributed insect radar, having two working modes. By adoptionof the FM stepped-frequency waveform design method and use method of the distributed insect radar in the invention, effective detection on a target in an airspace can be realized; and cross interference between radars is inhibited.
[0005] CN109116348A discloses a long-distance track simulation method facing insect migration, take-off and cruising. The method is characterized by determining the take-off area of insects based on the condition that a background atmospheric field is set to be reasonable; then, based on the verified directional flight strategy existing in the insects and the migration track speed and the direction of the insects observed by a radar, determining the speed and the head orientation of the insects during a migration process; and using a Lagrange particle diffusion model to simulate the migration trackof the insects in the background atmospheric field. An effective method for accurately simulating the track of insect migration and predicting an immigration area is provided. Compared with an existing insect migration track simulation method, by using the method of the invention, based on the background atmospheric field, the take-off area of the insect migration is determined, and an accurate insect flight characteristic is added to cruising track simulation so as to accurately simulate the track of the insect migration and predict the possible immigration area. Simultaneously, the solution disclosed provides an effective means for researching an flying insect migration behavior. CN116310470A discloses a migrant fly and bird classification method based on dual-polarization weather radar data. According to the method, the reflectivity factor Z, the velocity spectrum width W, the differential reflectivity ZDR, the differential phase shift psi DP and the correlation coefficient rho HV in the dualpolarization weather radar data serve as feature input, migrating insects and birds are classified based on a classifier, the result shows that the feature composed of the five parameters can effectively improve the accuracy of quantizing the migrated bird biomass and the migrated insect biomass, and the method has the advantages of being high in practicability and the like. The method is helpful for monitoring biological migration conditions, analyzing the influence of meteorological factors on biological activities and researching biological interaction.
[0006] In the prior art weather radar was used for long distance observation of flying animals like insects or birds. In the prior art unsolved problem exists relating to observation of migration of land animals, in particular rodents.
[0007] The invention relates to the subject matter of claim 1 .
[0008] The present application discloses a radar monitoring station for animal migration comprising: a radar transceiver operating in a range of wavelengths from 3 mm to 6 mm adapted to transmit and receive signals reflected from a monitored area, a computer-implemented processing means adapted to process signals from the radar unit to identify animals within the range of the monitoring station while the radar transceiver is at least partially pointing downward.
[0009] Preferably, the computer-implemented processing means comprises an Al or machine learning module adapted to process radar signals in the form of a cloud of points to identify animal position and size and / or behavioral pattern of the animal within the range of the monitoring station.
[0010] Preferably, the computer-implemented processing means is adapted to identify animal position and size and / or behavioral pattern of the animal within the range of the monitoring station using a residual image which is a result of subtracting the current radar image of the monitored area from a previous radar image of the monitored area.
[0011] Preferably, the computer-implemented processing means is adapted to identify the position and size and / or the behavioral pattern of the animal within the range of the monitoring station, using a residual image which is a result of subtracting the current radar image of the monitored area from a reference radar image of the monitored area.
[0012] Preferably, the radar monitoring station is further provided with a video camera covering the monitored area and collecting images in a visible light spectrum and / or in an infrared light spectrum to generate a video data stream representing the monitored area.
[0013] Preferably, the computer-implemented processing means is adapted to combine the processed radar image of the monitored area with the video data representing the monitored area to obtain a multimodal, composite and spatially aligned image of the monitored area.
[0014] Preferably, the computer-implemented processing means are further adapted to process signals from the radar unit to identify an alarm or a request for one or more events comprising in particular at least one of: class, order, family, genus or species of the animal, number of animals, duration of the visit of the animal within the range of the radar monitoring station, path of the animal within the range of the monitoring station, behavioral classes, direction from which the animal entered the monitored area. Preferably, the monitoring station is further provided with a set of sensors, and the computer-implemented processing means are further adapted to process signals from the set of sensors in order to identify an alarm or a request related to: opening of the monitoring station, closing of the monitoring station, movement of the monitoring station, biting, hitting, vandalism, filling or adding of bait, water entering or leaving the monitoring station.
[0015] Preferably, the monitoring station comprises one or more environmental sensors to detect: temperature, humidity, brightness within the monitored area.
[0016] Preferably, the radar monitoring station is further provided with a communication module adapted to establish a communication channel within a communication network.
[0017] The radar monitoring station makes it possible to collect a large number of valuable data describing the behavior of the monitored population of land animals, in particular rodents. Such data include identification of the animal species, recording of the exact time and direction of movement of the animal, identification of individual behaviors such as passing, feeding, fighting for resources such as food. In particular, the present invention is advantageous in low visibility conditions or for monitoring the area of complicated geometry, or covered with obstacles, where using a radar operating in a wavelength from 1 mm to 20 mm allows detection of animals hidden from view in the visible or infrared light.
[0018] The radar monitoring station is preferably used in a confined space such as warehouses, workshops, production facilities, hospitals, shops, etc. Alternatively, it can be used in the open file such as animal farms, or loans, driveways, yards or gardens.
[0019] The present invention is described in preferred embodiments in reference to drawings where:
[0020] Fig. 1 shows schematically the radar monitoring station according to the preferred embodiment of the invention;
[0021] Fig. 2 shows the radar monitoring station according to the another embodiment of the invention;
[0022] Fig. 3 shows the radar monitoring station according to the yet another embodiment of the invention.
[0023] In the preferred embodiment as shown in Fig. 1, an animal migration radar monitoring station 100 comprises a radar transceiver operating in a range of wavelengths from 3 mm to 6 mm and adapted to transmit and receive signals reflected from a monitored area, computer-implemented processing means adapted to process signals from the radar unit to identify animals within the range of the monitoring station, while the radar transceiver is at least partially directed downwardly, i.e. toward a surface of the earth. In Fig. 1 the radar monitoring station is pointing directly down monitoring the area with the obstacles B, where rodents R are following paths P.
[0024] Preferably, the radar monitoring station faces straight down to monitor an area directly below the radar monitoring station, in alternative embodiments the radar station is placed on a wall and faces at an angle to the monitoring area. Such alternative embodiments are depicted on fig. 2 and 3 where the radar monitoring station 100 is at least partially directed downwardly. Fig. 3 is showing the radiation emitted from the radar monitoring station can reach walls of a confined space, in this embodiment the computer-implemented processing means of the radar monitoring station are adapted to take into consideration and remove or ignore reflections from the walls W made of radar non permissible material like reinforced concrete. This allows a large area to be monitored even if there are obstacles in the visible or infrared light that would shadow and hide some parts of the monitored area. Due to the characteristics of the electromagnetic emission in the range of 5 mm, obstacles such as boxes, do not obscure the view, it is possible to observe the migration of animals.
[0025] The radar module preferably used is a coherent linear frequency modulated continuous wave (LFMCW) radar, which is a short-range device that is particularly effective in a highly reflective environment, since this is a ground surveillance radar, i.e. a radar that is at least partially directed downwards. Any other radar module suitable for such applications can be used, provided it produces an output in the form of a cloud of points.
[0026] The raw signal from the radar unit has the form of a cloud of points defined by the spatial coordinates x, y, z. Such signal is initially filtered by the mean shift algorithm and introduced into the machine learning module adapted to detect the movement of animals.
[0027] The computer-implemented processing means comprise an Al or machine learning module adapted to process the radar signals in the form of a cloud to identify the position and size and / or behavioral pattern of the animal within the range of the monitoring station.
[0028] The radar pulses are typically emitted at a frequency of 5 to 15 Hz, which allows identification of the animal moving at a speed of 0 to 7 m / s, which is typical of a rodent. The radar emission of relatively low power in the spectrum of wavelengths around 5 mm provides an effective range of up to 20 m.
[0029] The computer-implemented processing means is adapted to identify the position and size and / or behavioral pattern of the animal within the range of the monitoring station using a residual image which is a result of subtracting the current radar image of the monitored area from a previous radar image of the monitored area or a reference radar image of the monitored area.
[0030] The radar monitoring station is further provided with a video camera 105 which covers the monitored area and collects images in a visible light spectrum and / or in an infrared light spectrum to produce a video data stream or a still image representing the monitored area. The computer implemented processing means are adapted to combine the processed radar image of the monitored area with the video data or a still image representing the monitored area to obtain a multimodal, composite and spatially aligned image of the monitored area.
[0031] The computer-implemented processing means is further adapted to process signals from the radar unit to identify an alarm or a request for one or more events comprising in particular at least one of: class, order, family, genus or species of the animal, number of animals, duration of the visit of the animal within the range of the radar monitoring station, path of the animal within the range of the monitoring station, behavioral classes, direction from which the animal entered the monitored area. Additionally the computer processing means are adapted to identify the above parameters using image recognition algorithms applied on the video stream provided by the camera in order to augment the identification process or to run it autonomously if the radar signals allows to identify the location of the animal only.
[0032] The monitoring station is further provided with a set of sensors and the computer-implemented processing means is further adapted to process signals from the set of sensors in order to identify an alarm or a request related to: opening of the monitoring station, closing of the monitoring station, movement of the monitoring station, biting, hitting, vandalism, filling or adding of bait, water entering or leaving the monitoring station.
[0033] The monitoring station includes one or more environmental sensors to detect: temperature, humidity, brightness within the monitored area. The radar monitoring station is further provided with a communication module 120 adapted to establish a communication channel within a communication network, preferably a LoRa, GSM or satellite communication module. The communication channel is either wireless, fiber optic or cable depending on the infrastructure in which the radar station is placed.
[0034] In a preferred embodiment of the invention, the radar monitoring station is powered from a grid, alternatively it may be provided with a battery pack, fuel cell, generator or other source of electrical power.
Claims
Claims1 . A radar monitoring station (100) for animal migration, comprising a radar transceiver operating in a wavelength range of 3 mm to 6 mm adapted to transmit and receive signals reflected from a monitored area, a computer-implemented processing means adapted to process radar signals to identify animals within the range of the monitoring station, while the radar transceiver is at least partially directed downwardly.
2. A radar monitoring station according to claim 1, wherein the computer-implemented processing means comprises an Al or machine learning module adapted to process radar signals in the form of a cloud of points to identify the position and size and / or behavioral pattern of the animal within the range of the monitoring station.
3. The radar monitoring station of claim 2, wherein the radar pulses are emitted 5 to 15 times per second.
4. The radar monitoring station according to claim 1 or 2, wherein the computer-implemented processing means is adapted to identify the position and size and / or behavioral pattern of the animal within the range of the monitoring station using a residual image which is a result of subtracting the current radar image of the monitored area from a previous radar image of the monitored area.
5. A radar monitoring station according to claim 1 or 2, wherein the computer-implemented processing means is adapted to identify the position and size and / or behavioral pattern of the animal within the range of the monitoring station, using a residual image which is a result of subtracting the current radar image of the monitored area from a reference radar image of the monitored area.
6. The radar monitoring station of any one of claims 1 to 4, further comprising a video camera covering the monitored area and collecting images in a visible spectrum of light and / or in an infrared spectrum of light to produce a video data stream representing the monitored area.
7. The radar monitoring station according to claim 5, wherein the computer-implemented processing means are adapted to combine a processed radar image of the monitored area with the video data representing the monitored area to obtain a multimodal, composite and spatially aligned image of the monitored area.
8. The radar monitoring station according to any one of claims 1 to 6, wherein said computer- implemented processing means are further adapted to process signals from said radar unit to identify an alarm or a request for one or more events comprising in particular at least one of: class, order, family, genus or species of the animal, number of animals, duration of the visit of the animal within the range of the radar monitoring station, path of the animal within the range of the monitoring station, behavioral classes, direction from which the animal entered the monitored area.
9. The radar monitoring station according to any of claims 6 to 8 wherein said computer implemented processing means are further adopted to adapted to process signals from a camera to identify an alarm or a request for one or more events comprising in particular at least one of: class, order, family, genus or species of the animal, number of animals, duration of the visit of the animal within the range of the radar monitoring station, path of the animal within the range of the monitoringstation, behavioral classes, direction from which the animal entered the monitored area, using image recognition algorithms applied on the video stream provided by the camera in order to augment the identification process or to run it autonomously, in a case the radar signals allows to identify the location of the animal only.
10. The radar monitoring station according to any of claims 1 to 9, wherein the monitoring station is further provided with a set of sensors and the computer-implemented processing means are further adapted to process signals from the set of sensors in order to identify an alarm or a request related to: opening of the monitoring station, closing of the monitoring station, movement of the monitoring station, biting, hitting, vandalism, filling or adding of bait, water entering or leaving the monitoring station.
11. A radar monitoring station according to any one of claims 1 to 10, wherein the monitoring station comprises one or more environmental sensors to detect: temperature, humidity, brightness within the monitored area.
12. The radar monitoring station according to any one of claims 1 to 11 , wherein the radar monitoring station is further provided with at least one communication module adapted to establish a communication channel within a communication network, selected from a group comprising: LoRa, GSM or satellite communication module.
Citation Information
Patent Citations
Long-distance track simulation method facing insect migration, take-off and cruising
CN109116348A
FM (Frequency modulation) stepped-frequency waveform design method and use method of distributed insect radar
CN109164445A
Model insect and bird classification method based on dual-polarization weather radar data
CN116310470A
Living body radar system, identification method, and feature database establishment method
EP3851871A1
Pest detector to identify a type of pest using machine learning
US20210153489A1