An unmanned hangar anti-theft system, method, storage medium and computer program product

Abstract

The application discloses an unmanned vehicle warehouse anti-theft system and method, a storage medium and a computer program product. The method comprises the following steps: collecting multi-source sensing data inside and outside an unmanned vehicle warehouse; determining a threat level according to the multi-source sensing data; and performing a response action corresponding to the threat level according to the threat level. The application realizes the transformation from passive defense to active intelligent protection. Through a hierarchical response mechanism, differentiated disposal from local warning to asset tracking can be performed according to the threat degree. The safety, reliability and intelligent level of the unmanned vehicle warehouse are comprehensively improved.

Description

Technical Field

[0001] This invention belongs to the field of anti-theft technology for drone hangars, specifically relating to an anti-theft system, method, storage medium, and computer program product for drone hangars. Background Technology

[0002] With the widespread adoption of drone technology, drone hangars have become essential equipment for convenient storage and rapid deployment of drones. However, when drone hangars are placed in unattended areas, they become easy targets for theft. Traditional mechanical locks or simple electronic alarms have the following shortcomings: passive protection, only alarming after damage has occurred; high false alarm rate, susceptible to environmental interference; and limited response methods, lacking effective deterrence and tracking capabilities. Summary of the Invention

[0003] To address the problems described in the background art, this invention proposes an anti-theft system, method, storage medium, and computer program product for unmanned aerial vehicle (UAV) hangars.

[0004] An anti-theft system for unmanned aerial vehicle hangars, which achieves one of the objectives of this invention, includes: The multi-sensor fusion sensing module is used to collect multi-source sensing data inside and outside the UAV hangar; the multi-source sensing data refers to the raw signals or data streams after preliminary preprocessing collected and output by all activated sensors in the multi-sensor fusion sensing module, and these data come from different types of sensors. The control and processing module is used to determine the threat level based on the multi-source sensing data; The graded response module is used to execute response actions corresponding to the threat level.

[0005] Furthermore, it also includes an identity authentication module, which sends an authorization window signal to the control processing module after the user's identity authentication is successful. The control processing module pauses the determination of the threat level for a preset time period after receiving the authorization window signal, and automatically resumes the determination of the threat level after the preset time period ends.

[0006] Furthermore, the identity authentication module is used to verify operator permissions, including a biometric recognition unit (fingerprint, face), a key linkage unit (Bluetooth / UWB key), or a dynamic authorization unit (one-time password). When a user successfully authenticates via any method, the module generates an authorization window signal with a preset validity period (e.g., 30 seconds). Upon receiving the authorization window signal, the control processing module pauses threat level determination to avoid false alarms triggered by legitimate operations (such as picking up or placing a drone). After the validity period expires, the threat level determination function automatically resumes.

[0007] Furthermore, the multi-sensor fusion sensing module includes at least two of the following: a vibration sensor, a displacement sensor, a capacitive proximity sensor, an in-cabin status sensor, and a sound recognition unit, wherein the in-cabin status sensor is a pressure sensor or a photoelectric sensor.

[0008] Furthermore, to achieve comprehensive monitoring of the drone hangar, the sensors in the multi-sensor fusion sensing module are arranged as follows: Vibration sensors are installed near the locks on the drone hangar doors and at weak points in the cabin walls to detect specific frequency vibration signals generated by violent damage.

[0009] The displacement sensor is installed at the junction of the hatch and the cabin. It uses a high-precision laser or inductive displacement sensor to monitor whether the hatch has undergone unauthorized displacement.

[0010] The capacitive proximity sensor is installed on the inside of the hangar shell. Its sensing electrodes cover the lock area and the perimeter of the door, forming an electrostatic field to detect the capacitance change caused when a human body or metal tool approaches within 10cm.

[0011] The in-cabin status sensor is installed below or on the surface of the drone parking platform inside the drone hangar. When using a pressure sensor, it is embedded in the four corners or center of the parking platform; when using a through-beam photoelectric sensor, the transmitter and receiver are installed on opposite sides of the parking platform, allowing the light path to pass through the underside of the drone fuselage. This sensor is used to detect whether the drone is in position.

[0012] The sound recognition unit is installed near the control circuit board inside the drone hangar. Its microphone faces out of the hangar or is connected to the outside through a sound guide hole to collect ambient sound signals.

[0013] The placement and number of the aforementioned sensors can be adjusted according to the specific structure of the drone hangar, but all should ensure comprehensive coverage of the doors, locks, and internal assets.

[0014] Furthermore, the control processing module extracts the time-domain peak value or frequency-domain main frequency of the multi-source sensing data as features, compares the extracted features with a preset intrusion feature template, and generates a preliminary sensing event to characterize potential threats based on the comparison results.

[0015] The preset intrusion feature template refers to a set of feature vectors that are pre-collected by collecting signal samples of typical intrusion actions (such as knocking, cutting, and picking locks), extracting their time-domain peak values ​​or frequency-domain dominant frequency features, and then associating them with the corresponding event types. During comparison, the similarity between the real-time features and the template is calculated, and a match is made if the similarity exceeds a threshold.

[0016] Furthermore, the time-domain peak value or frequency-domain dominant frequency of the multi-source sensing data is extracted as a feature. The extraction method includes: For the extraction of time-domain peak values, the maximum amplitude of the filtered signal is calculated within a preset time window.

[0017] To extract the dominant frequency in the frequency domain, a Fast Fourier Transform (FFT) is performed on the filtered signal, and the frequency component with the largest amplitude after the transform is taken as the dominant frequency.

[0018] Furthermore, the preset intrusion feature template is set by pre-collecting signal samples of typical intrusion actions and extracting the aforementioned features. The real-time extracted features are compared with the template for similarity, such as using Euclidean distance or cosine similarity. When the similarity exceeds a preset threshold, a preliminary perception event is generated.

[0019] Furthermore, the method by which the control processing module determines the threat level includes: comparing the chronological order of the timestamps of each initial sensing event with a preset intrusion behavior chain, and determining whether the installation locations corresponding to the sensor identifiers in each event belong to the same preset spatial partition; when the time sequence matches and the locations belong to the same preset spatial partition, the threat level is determined to be level two; otherwise, the threat level is determined to be level one. A preset spatial partition refers to a logical area identifier pre-assigned to a sensor based on its installation location in the drone hangar. This includes, but is not limited to, lock areas, door areas, and hangar top areas. Sensors belonging to the same logical area are considered spatially associated.

[0020] Furthermore, the preset intrusion behavior chain refers to the sensor trigger sequence corresponding to a typical intrusion scenario. For example, the sequential triggering of a proximity sensor, vibration sensor, and displacement sensor corresponds to lock picking; the sequential triggering of a vibration sensor and sound recognition unit corresponds to cutting. The system can preset one or more intrusion behavior chain templates. When multiple preliminary sensing event sequences match multiple templates simultaneously, as long as any template is matched and the spatial location is associated, the threat level is determined to be level two.

[0021] Furthermore, the determination process involves checking whether the installation locations corresponding to the sensor identifiers in each event belong to the same preset spatial partition. For example, a vibration sensor located in the lock area and a displacement sensor located in the hatch area both belong to the same intrusion logic area; while sensors on the hangar roof belong to different areas. When the time sequence matches and the locations belong to the same preset spatial partition, the threat level is determined to be level two; otherwise, it is determined to be an isolated or low-confidence event, triggering only a level one warning.

[0022] Furthermore, the method for determining whether the installation locations corresponding to sensor identifiers in each event belong to the same preset spatial partition includes: pre-assigning a preset spatial partition identifier for each sensor to identify the area where the sensor is located, such as a lock area, a hatch area, or a hangar roof area; extracting the spatial partition identifiers corresponding to the sensor identifiers in each preliminary sensing event; comparing whether all spatial partition identifiers are the same; if they are the same, it is determined that they belong to the same preset spatial partition; otherwise, it is determined that they do not belong to the same spatial partition. For example, the pre-assigned spatial partition identifier for the vibration sensor is identifier number 1, which represents the lock area, and the pre-assigned spatial partition identifier for the displacement sensor is identifier number 2, which represents the hatch area. The two partition identifiers are different, so they do not belong to the same preset spatial partition; while the spatial partition identifier for the sensor on the hangar roof is identifier number 3, which represents the hangar roof area, and is also different from other areas. When the time sequence matches and the spatial partition identifiers corresponding to each event are the same, the threat level is determined to be level two; otherwise, it is level one.

[0023] Furthermore, when the threat level is level two, if the in-cabin status sensor in the multi-sensor fusion perception module detects that the UAV has changed from an in-situ state to a missing state, the threat level is updated from level two to level three. Updating the threat level from level two to level three requires prior confirmation of a level two intrusion event. When the threat level is level two, the control processing module continuously monitors the output signals of the in-cabin status sensors (pressure or photoelectric sensors). If the UAV is detected to have changed from an in-situ state to a missing state within a preset time window (e.g., 10 seconds), the threat level is immediately updated from level two to level three, and a corresponding tracking response is triggered.

[0024] Internal status sensors refer to sensors installed on the parking platform inside the drone hangar to detect whether the drone is in position. These are specifically pressure sensors or through-beam photoelectric sensors. Pressure sensors output a level signal based on weight changes, while photoelectric sensors output a level signal based on the on / off state of an optical path. Local visual cues refer to non-auditory alerts performed by the drone hangar itself, typically including controlling the external LED strip to flash at a specific color or frequency. Local audible and visual alarms refer to high-intensity deterrent actions performed by the drone hangar itself, including activating a built-in high-decibel siren. Furthermore, before determining the threat level, the control processing module also includes filtering the multi-source sensing data to filter out noise signals outside a preset frequency range.

[0025] Furthermore, when the drone storage is a vehicle-mounted drone storage, the control processing module is also used to load preset scenario modes according to the vehicle status to reduce the false alarm rate; the vehicle status can be obtained through the vehicle bus, including but not limited to: engine off status, vehicle locked status, driving status, maintenance mode status, etc.

[0026] Furthermore, the preset scenario mode refers to a set of configuration parameters adaptively loaded according to the vehicle status. Each mode corresponds to a list of activated sensors and judgment parameters, such as vibration threshold and time window, including the following three types: (1) Guardian mode: When the vehicle is turned off and locked, the system automatically enters the guard mode. In this mode, all sensors are activated and high-sensitivity judgment parameters are used. For example, the vibration threshold is set to a low value of 0.1g and the time window is set to a short value of 2 seconds to achieve comprehensive security monitoring of the drone hangar.

[0027] (2) Service mode: When the vehicle is in maintenance, the system enters service mode. In this mode, the local sound and light alarm function is turned off to avoid unnecessary alarms caused by maintenance operations, but the remote notification function is retained, such as pushing reminders to the owner's APP. At the same time, the sensors can still collect data but do not trigger the local deterrence response. The maintenance status can be triggered by the user manually switching or by detecting signals such as the engine hood being opened.

[0028] (3) Transportation mode: When the vehicle is in motion (e.g., the speed is greater than 5 km / h), the system automatically enters the transportation mode. In this mode, the sensitivity of external motion detection sensors (e.g., capacitive proximity sensors) is turned off or significantly reduced, while the trigger thresholds of vibration sensors and displacement sensors are increased (e.g., the vibration threshold is increased to 0.5g) to avoid continuous false alarms caused by normal vibrations and wind noise during vehicle operation. However, the monitoring function of in-cabin status sensors (e.g., UAV presence detection) is retained to prevent the UAV from accidentally falling off during transportation.

[0029] Furthermore, the graded response module performs response actions corresponding to the threat level, including: when the threat level is level one, activating local visual alerts; when the threat level is level two, activating local audible and visual alarms, and triggering vehicle horn or hazard lights via vehicle bus when the drone library is a vehicle-mounted drone library; when the threat level is level three, sending a remote locking command to the drone and receiving the location information transmitted back by the drone.

[0030] Furthermore, the local visual cues in the Level 1 response include controlling the LED light strip outside the hangar to emit blue or white light at a gentle frequency; the local audible and visual alarms in the Level 2 response include sending a command to the vehicle via the CAN bus to trigger the built-in high-decibel siren and red strobe lights. In Level 3, a remote locking command is sent to the drone, instructing the drone to transmit its GPS location information back to the cloud platform at a high frequency.

[0031] Furthermore, before determining the threat level, the control processing module also includes filtering the multi-source sensing data to filter out noise signals outside a preset frequency range.

[0032] A method for preventing theft of unmanned aerial vehicle hangars to achieve the second objective of this invention includes: Collect multi-source sensing data from inside and outside the drone hangar; The threat level is determined based on the multi-source sensing data; Execute the response action corresponding to the threat level.

[0033] A non-transitory computer-readable storage medium for achieving the third objective of the present invention, wherein a computer program is stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the anti-theft method for unmanned aerial vehicle hangars.

[0034] A computer program product for achieving the fourth objective of the present invention includes a computer program / instruction that, when executed by a processor, implements the steps of the drone hangar anti-theft method.

[0035] The beneficial effects of this invention include: This invention addresses the problems of passive protection, high false alarm rates, and limited response methods in existing drone hangar security systems by constructing a multi-sensor fusion perception and hierarchical response system. Utilizing multiple sensors such as vibration, displacement, proximity, cabin status, and sound recognition, it collects multi-source perception data from inside and outside the drone hangar in real time. Standardized preliminary perception events are generated by extracting time-domain peak or frequency-domain dominant frequency features and comparing them with preset intrusion templates. Based on this, the control processing module performs spatiotemporal correlation analysis on the timestamps of each event and the sensor installation locations. Only when multiple sensor events match a preset intrusion behavior chain in time sequence and belong to the same preset spatial zone is it confirmed as a genuine intrusion and a level two threat is output, significantly reducing false alarms caused by environmental interference or isolated noise. Furthermore, when the drone is a vehicle-mounted drone, it automatically switches between preset scenario modes such as guard, service, and transportation based on the vehicle status, adaptively adjusting sensor activation types and judgment parameters, further improving reliability and intelligence in different usage scenarios. In addition, the identity authentication module provides an authorization window during legitimate operation, suspending threat judgment and avoiding false alarms caused by normal use. This invention transforms passive defense into proactive sensing and tiered intervention, significantly enhancing anti-theft capabilities. Furthermore, when the drone hangar is a vehicle-mounted drone hangar, it achieves deep integration with vehicle security, providing comprehensive, reliable, and low-false-alarm intelligent protection for vehicle-mounted drone hangars. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the structure of an embodiment of the system described in this invention; Figure 2 This is a flowchart illustrating an embodiment of the method described in this invention. Detailed Implementation

[0037] The following detailed embodiments are provided to explain the technical solutions of the present invention, so that those skilled in the art can understand the present invention. The scope of protection of the present invention is not limited to the following specific embodiments. Any modifications or improvements made by those skilled in the art that incorporate the technical solutions of the present invention but differ from the following detailed embodiments are also within the scope of protection of the present invention.

[0038] This invention provides an anti-theft system for drone hangars, such as... Figure 1 As shown, it includes: a multi-sensor fusion perception module, deployed inside and outside the drone hangar, for collecting multi-source perception data from inside and outside the drone hangar; a control processing module, for determining the threat level based on the multi-source perception data; and a graded response module, for executing response actions corresponding to the threat level.

[0039] In one embodiment, the system further includes an identity authentication module, which sends an authorization window signal to the control processing module after the user's identity authentication is successful. The control processing module pauses the determination of the threat level for a preset time period after receiving the authorization window signal, and automatically resumes the determination of the threat level after the preset time period ends.

[0040] In one embodiment, the identity authentication module is used to verify operator permissions, including a biometric recognition unit (such as fingerprint or face), a key linkage unit (Bluetooth / UWB key), or one-time dynamic authorization. When a user successfully authenticates through any method, the module generates an authorization window signal with a preset validity period of 30 seconds. Upon receiving the authorization window signal, the control processing module pauses threat level determination to avoid triggering false alarms for legitimate operations such as picking up or placing drones. After the validity period expires, the threat level determination function automatically resumes.

[0041] In one embodiment, a user interaction module is also included, integrated into the vehicle's human-machine interface and / or the user's mobile terminal, for status visualization, alarm management, etc.

[0042] In one embodiment, the multi-sensor fusion sensing module includes at least two of the following: a vibration sensor, a displacement sensor, a capacitive proximity sensor, an in-cabin status sensor, and a sound recognition unit, wherein the in-cabin status sensor is a pressure sensor or a photoelectric sensor.

[0043] In one embodiment, to achieve comprehensive monitoring of the drone hangar, the sensors in the multi-sensor fusion sensing module are arranged as follows: Vibration sensors are installed near the locks on the drone hangar doors and in weak points on the cabin walls, such as hinge connections, to detect specific frequency vibration signals generated by violent damage such as knocking, cutting, and prying.

[0044] Displacement sensors are installed at the junction of the hatch and the hull. They are high-precision laser or inductive displacement sensors used to monitor whether the hatch has been moved without authorization (such as being pried open).

[0045] The capacitive proximity sensor is installed on the inside of the hangar shell. Its sensing electrodes cover the lock area and the perimeter of the door, forming an electrostatic field to detect the capacitance change caused when a human body or metal tool approaches within 10cm.

[0046] The in-cabin status sensor is installed below or on the surface of the drone parking platform inside the drone hangar. When using a pressure sensor, it is embedded in the four corners or center of the parking platform; when using a through-beam photoelectric sensor, the transmitter and receiver are installed on opposite sides of the parking platform, allowing the light path to pass through the underside of the drone fuselage. This sensor is used to detect whether the drone is in position.

[0047] The sound recognition unit is installed near the control circuit board inside the drone hangar. Its microphone faces out of the hangar or is connected to the outside through a sound guide hole. It is used to collect ambient sound signals (such as the sound of an electric drill or breaking glass) to identify acoustic features related to theft.

[0048] The placement and number of the aforementioned sensors can be adjusted according to the specific structure of the drone hangar (such as roof-mounted or trunk-embedded), but all should ensure comprehensive coverage of the doors, locks, and internal assets.

[0049] In one embodiment, the control processing module extracts the time-domain peak value or frequency-domain main frequency of the multi-source sensing data as features, compares the extracted features with a preset intrusion feature template, and generates a preliminary sensing event to characterize potential threats based on the comparison result.

[0050] In one embodiment, the time-domain peak value or frequency-domain dominant frequency of the multi-source sensing data is extracted as a feature. The extraction method includes: For the extraction of time-domain peak values, the maximum amplitude of the filtered signal is calculated within a preset time window of 0.5 seconds.

[0051] To extract the dominant frequency in the frequency domain, a Fast Fourier Transform (FFT) is performed on the filtered signal, and the frequency component with the largest amplitude after the transform is taken as the dominant frequency.

[0052] In one embodiment, the preset intrusion feature template is set by pre-collecting signal samples of typical intrusion actions such as knocking, cutting, and lock picking, and then extracting the aforementioned features. The features extracted in real time are compared with the template for similarity, such as using Euclidean distance or cosine similarity. When the similarity exceeds a preset threshold of 80%, a preliminary perception event is generated.

[0053] In one embodiment, the method for the control processing module to determine the threat level includes: comparing the chronological order of the timestamps of each preliminary sensing event with a preset intrusion behavior chain, and determining whether the installation location corresponding to the sensor identifier in each event belongs to the same preset spatial partition; when the time sequence matches and the location belongs to the same preset spatial partition, the threat level is determined to be level two; otherwise, the threat level is determined to be level one.

[0054] In one embodiment, the preset intrusion behavior chain refers to the sensor trigger sequence corresponding to a typical intrusion scenario. For example, the sequential triggering of a proximity sensor, vibration sensor, and displacement sensor corresponds to lock picking; the sequential triggering of a vibration sensor and sound recognition unit corresponds to cutting. The system can preset one or more intrusion behavior chain templates. When multiple preliminary sensing event sequences match multiple templates simultaneously, as long as any template is matched and the spatial location is associated, the threat level is determined to be level two.

[0055] In one embodiment, the determination of whether the installation locations corresponding to the sensor identifiers in each event belong to the same preset spatial partition is as follows: for example, a vibration sensor is located in the "lock area" and a displacement sensor is located in the "door area," both belonging to the same intrusion logic area; while the sensor on the top of the hangar belongs to a different area. When the time sequence matches and the location belongs to the same preset spatial partition, the threat level is determined to be level two; otherwise, it is level one (isolated or low-confidence event, triggering only an early warning).

[0056] In one embodiment, the method for determining whether the installation location corresponding to the sensor identifier in each event belongs to the same preset spatial partition includes: pre-assigning a preset spatial partition identifier for each sensor to identify the area where the sensor is located, including: a lock area, a door area, or a hangar top area; extracting the spatial partition identifier corresponding to the sensor identifier in each preliminary sensing event; comparing whether all spatial partition identifiers are the same; if they are the same, it is determined that they belong to the same preset spatial partition; otherwise, it is determined that they do not belong to the same spatial partition. When the time sequence matches and the spatial partition identifiers corresponding to each event are the same, the threat level is determined to be level two; otherwise, it is level one.

[0057] In one embodiment, when the threat level is level two, if the in-cabin status sensor in the multi-sensor fusion perception module detects that the UAV has changed from an in-situ state to a missing state, the threat level is updated from level two to level three. Updating the threat level from level two to level three requires prior confirmation of a level two intrusion event. When the threat level is level two, the control processing module continuously monitors the output signals of the in-cabin status sensors (pressure or photoelectric sensors). If it detects that the UAV has changed from an in-situ state to a missing state within a 10-second time window, it immediately updates the threat level from level two to level three and triggers a corresponding tracking response.

[0058] In one embodiment, before determining the threat level, the control processing module further includes filtering the multi-source sensing data to filter out noise signals outside a preset frequency range.

[0059] In one embodiment, the graded response module performs response actions corresponding to the threat level, including: when the threat level is level one, activating local visual alerts; when the threat level is level two, activating local audible and visual alarms, and triggering vehicle horn or hazard lights via vehicle bus when the drone library is a vehicle-mounted drone library; when the threat level is level three, sending a remote lock command to the drone and receiving the location information transmitted back by the drone.

[0060] In one embodiment, the local visual cue in the first-level response includes: controlling the LED light strip outside the hangar to emit blue or white light at a gentle frequency of 1 second on and 2 seconds off; the local audible and visual alarm in the second-level response includes: sending a command to the vehicle via the CAN bus to trigger a built-in high-decibel siren of at least 120dB and a red strobe light with a frequency of 5Hz. In the third-level response, remote locking commands such as no-fly or data erasure are sent to the drone, and the drone is instructed to transmit its GPS location information back to the cloud platform at a high frequency of once every 10 seconds.

[0061] In one embodiment, before determining the threat level, the control processing module further includes filtering the multi-source sensing data to remove noise signals outside a preset frequency range. The preset frequency range is determined by pre-collecting sample signals of typical intrusion actions and environmental noise and performing spectral analysis. Specific methods include: (1) Collect the following two types of signal samples respectively: Intrusion action signals: Simulate violent sabotage behavior, including: using metal tools to knock on locks, using chainsaws or angle grinders to cut the hull, and using crowbars to pry open hatches. Each action is repeated multiple times, with a sampling frequency of 1kHz.

[0062] Environmental noise signals: Collect various environmental noises that the vehicle may encounter, including: wind and rain sounds simulating different wind speeds, resonance noise when the vehicle is driving at different road surfaces and speeds, nearby construction noise, and noise from pedestrian traffic.

[0063] (2) Perform a Fast Fourier Transform (FFT) on the above sample signals to obtain the power spectral density distribution of each signal and the frequency range of each type of signal. For example, the main frequency of the knocking lock is 100Hz-150Hz, and the main frequency of the cutting action is 50Hz-100Hz.

[0064] (3) Frequency range determination: Based on the above power spectral density distribution analysis, the passband of the filter is set to 50Hz to 150Hz. This range can effectively retain the characteristic signals of the intrusion action while filtering out most of the environmental noise.

[0065] It should be noted that this frequency range can be adaptively adjusted based on the specific sensor model, installation location, and operating environment. For example, if the vehicle frequently travels on bumpy roads, the low-frequency cutoff frequency can be appropriately increased to 80Hz; if the drone hangar is installed in a high-noise environment, the passband range can be appropriately narrowed. The adjustment method follows the same sample acquisition and spectrum analysis process as described above.

[0066] In one embodiment, when the drone storage is a vehicle-mounted drone storage, the control processing module is further configured to load a preset scenario mode based on the vehicle status to reduce the false alarm rate; the vehicle status can be obtained through the vehicle bus, including but not limited to: engine off status, vehicle locked status, driving status, maintenance mode status, etc.

[0067] In one embodiment, the preset scenario mode refers to a set of configuration parameters adaptively loaded according to the vehicle state. Each mode corresponds to a list of activated sensors and judgment parameters, such as vibration threshold and time window, including the following three types: (1) Guardian mode: When the vehicle is turned off and locked, the system automatically enters the guard mode. In this mode, all sensors are activated and high-sensitivity judgment parameters are used. For example, the vibration threshold is set to a low value of 0.1g and the time window is set to a short value of 2 seconds to achieve comprehensive security monitoring of the drone hangar.

[0068] (2) Service mode: When the vehicle is in maintenance, the system enters service mode. In this mode, the local sound and light alarm function is turned off to avoid unnecessary alarms caused by maintenance operations, but the remote notification function is retained, such as pushing reminders to the owner's APP. At the same time, the sensors can still collect data but do not trigger the local deterrence response. The maintenance status can be triggered by the user manually switching or by detecting signals such as the engine hood being opened.

[0069] (3) Transportation mode: When the vehicle is in motion (e.g., the speed is greater than 5 km / h), the system automatically enters the transportation mode. In this mode, the sensitivity of external motion detection sensors (e.g., capacitive proximity sensors) is turned off or significantly reduced, while the trigger thresholds of vibration sensors and displacement sensors are increased (e.g., the vibration threshold is increased to 0.5g) to avoid continuous false alarms caused by normal vibrations and wind noise during vehicle operation. However, the monitoring function of in-cabin status sensors (e.g., UAV presence detection) is retained to prevent the UAV from accidentally falling off during transportation.

[0070] Through adaptive switching of the aforementioned scenario modes, the system can intelligently adjust security strategies according to the actual vehicle usage scenario. In Guardian mode, high-sensitivity detection is only activated when the vehicle is stationary with the engine off and locked, avoiding false triggers due to vibration during driving or maintenance. In Service mode, local audible and visual alarms are proactively disabled to prevent unnecessary alarms caused by normal operation by maintenance personnel. In Transportation mode, the vibration threshold is increased and the sensitivity of the proximity sensor is reduced to avoid continuous false alarms caused by normal wind noise and road bumps during vehicle operation. Thus, by differentially suppressing the main interference sources under different vehicle states, the false alarm rate is significantly reduced across all scenarios while ensuring the actual intrusion detection rate.

[0071] This invention provides a method for preventing theft in drone hangars, including: Collect multi-source sensing data from inside and outside the drone hangar; The threat level is determined based on the multi-source sensing data; Execute the response action corresponding to the threat level.

[0072] In one embodiment, the system further includes user authentication, including biometric identification (such as fingerprints or faces), key linkage (Bluetooth / UWB keys), or one-time dynamic authorization. Once the user successfully authenticates using any method, the module generates an authorization window signal with a preset validity period of 30 seconds. Upon receiving the authorization window signal, the control processing module pauses threat level determination to prevent legitimate operations such as picking up or placing drones from triggering false alarms. After the validity period expires, the threat level determination function automatically resumes.

[0073] In one embodiment, multiple sensors are used to collect multi-source sensing data inside and outside the UAV hangar; the multiple sensors include at least two of vibration sensors, displacement sensors, capacitive proximity sensors, cabin status sensors and sound recognition units, wherein the cabin status sensor is a pressure sensor or a photoelectric sensor.

[0074] The multi-source sensing data includes the following data: (1) Vibration data; it is collected by micro-vibration sensors and reflects the mechanical vibration waveforms generated by physical impacts, knocks, cutting, etc. on the hangar shell or locks. The sampling frequency is typically 1kHz. (2) Displacement data; collected by high-precision displacement sensors, reflecting the relative positional change between the hatch and the cabin, with an accuracy of up to 0.1 mm; (3) Proximity sensing data; collected by a capacitive proximity sensor, reflecting the change in electrostatic capacitance caused when a human or metal object approaches the hangar shell; (4) Cabin status data; collected by pressure sensors or photoelectric sensors, reflecting the status signal of whether the UAV is placed in the designated location in the hangar, such as in place / missing; (5) Ambient sound data; collected by the sound recognition unit, reflecting the sound wave signals in the environment around the hangar, with a typical sampling frequency of 44.1kHz.

[0075] These data were collected in parallel over time and came from different installation locations in space, together forming a multi-source information set that reflects the physical environment and asset status inside and outside the hangar.

[0076] This invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the various steps of the method described in this invention.

[0077] This invention also provides a non-transitory computer-readable storage medium storing a computer program. The computer program includes program instructions that, when executed by a processor, implement the various steps of the method described in this invention, which will not be elaborated further here.

[0078] The computer-readable storage medium can be the data transmission apparatus or the internal storage unit of a computer device provided in any of the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium can also be the external storage device of the computer device, such as the plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device.

[0079] Furthermore, the computer-readable storage medium may include both internal storage units and external storage devices of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data that is to be output or has already been output.

[0080] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0082] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0084] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A drone hangar anti-theft system, characterized in that, include: The multi-sensor fusion sensing module is used to collect multi-source sensing data inside and outside the drone hangar; The control and processing module is used to determine the threat level based on the multi-source sensing data; The graded response module is used to execute response actions corresponding to the threat level.

2. The anti-theft system for unmanned aerial vehicle hangars as described in claim 1, characterized in that, It also includes an identity authentication module, which sends an authorization window signal to the control processing module after the user's identity is successfully authenticated. The control processing module suspends the determination of the threat level for a preset time period after receiving the authorization window signal.

3. The anti-theft system for unmanned aerial vehicle hangars as described in claim 1, characterized in that, The control processing module extracts the time-domain peak value or frequency-domain main frequency of the multi-source sensing data as features, compares the extracted features with the preset intrusion feature template, and generates preliminary sensing events to characterize potential threats based on the comparison results.

4. The anti-theft system for unmanned aerial vehicle hangars as described in claim 1, characterized in that, The method for the control processing module to determine the threat level includes: comparing the order of the timestamps of each preliminary sensing event with a preset intrusion behavior chain, and determining whether the installation location corresponding to the sensor identifier in each event belongs to the same preset spatial partition; when the time sequence matches and the location belongs to the same preset spatial partition, the threat level is determined to be level two; otherwise, the threat level is determined to be level one.

5. The anti-theft system for unmanned aerial vehicle hangars as described in claim 4, characterized in that, When the threat level is level two, if the cabin status sensor in the multi-sensor fusion perception module detects that the UAV has changed from an in-situ state to a missing state, the threat level will be updated from level two to level three.

6. The anti-theft system for unmanned aerial vehicle hangars as described in claim 1, characterized in that, When the drone library is a vehicle-mounted drone library, the control processing module is also used to load a preset scenario mode according to the vehicle status in order to reduce the false alarm rate.

7. The anti-theft system for unmanned aerial vehicle hangars as described in claim 1, characterized in that, The graded response module performs response actions corresponding to the threat level, including: when the threat level is level 1, activating local visual alerts; when the threat level is level 2, activating local audible and visual alarms, and triggering vehicle horn or hazard lights via vehicle bus when the drone library is a vehicle-mounted drone library; when the threat level is level 3, sending a remote lock command to the drone and receiving the location information transmitted back by the drone.

8. A method for preventing theft from unmanned aerial vehicle (UAV) hangars as described in claim 1, characterized in that, include: Collect multi-source sensing data from inside and outside the drone hangar; The threat level is determined based on the multi-source sensing data; Execute the response action corresponding to the threat level.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the anti-theft method for unmanned aerial vehicle hangars as described in claim 8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the anti-theft method for drone hangars as described in claim 8.

Images