Motorcycle intelligent anti-theft evaluation method and system

Abstract

The embodiment of the present application relates to the technical field of vehicle monitoring, and discloses a motorcycle intelligent anti-theft evaluation method, comprising: processing a first sensing parameter to determine whether a current parking mode matches a set mode, if yes, continuing to detect; if no, entering an active verification mode; in the active verification mode, controlling to start a second sensing component to obtain a second sensing parameter within a set time range; processing the second sensing parameter within the set time range to determine whether the corresponding motorcycle has a risk of being stolen, if yes, performing an alarm operation. The motorcycle intelligent anti-theft evaluation method in the embodiment of the present application only starts a first sensing component in an initial monitoring state, and activates a second sensing component and an alarm module only in an abnormal state, thereby reducing system standby power consumption and prolonging equipment endurance. The non-theft scene automatically returns to the initial state, avoids repeated alarms from interfering with users, and improves user experience.

Description

Technical Field

[0001] This invention relates to the field of motorcycle technology, and specifically to a method and system for intelligent anti-theft assessment of motorcycles. Background Technology

[0002] Currently, most existing motorcycle anti-theft systems rely on single, threshold-based alarm mechanisms (such as triggering an alarm when vibration exceeds a certain intensity). This results in high false alarm rates (e.g., due to wind or accidental impact), significant risk of missed alarms (e.g., inability to detect technical or probing thefts), poor user experience, and limited protection capabilities. Therefore, designing an intelligent anti-theft assessment and early warning system has become a pressing technical problem for those skilled in the art. Summary of the Invention

[0003] To address the aforementioned shortcomings, this invention discloses an intelligent anti-theft assessment method for motorcycles, which enables efficient anti-theft assessment.

[0004] The first aspect of this invention discloses a method for intelligent anti-theft assessment of motorcycles, comprising: When the motorcycle is in the initial monitoring state, the first sensing component is controlled to work to obtain the detected first sensing parameters, and the first sensing parameters are processed to determine whether the current parking mode matches the set mode. If yes, the detection continues; if no, the active verification mode is entered. In the active verification mode, the second sensing component is activated to obtain second sensing parameters within a set time range; The second sensor parameters within a set time range are identified and processed to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is initiated; otherwise, the system returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, activating the camera to acquire images, and keeping the audio-visual module in a working state.

[0005] As an optional implementation, in a first aspect of the present invention, the step of controlling the activation of a second sensing component to acquire second sensing parameters within a set time range in the active verification mode includes: In the active verification mode, a sound signal is randomly generated within a set time period to control the sound module to work and to control the activation of the second sensing component to obtain the second sensing parameters within the set time range.

[0006] As an optional implementation, in a first aspect of the present invention, the first sensing component includes a vibration sensor, a tilt sensor, or a current sensor; the first sensing parameter includes a first vibration parameter, a first tilt angle change parameter, or a current sensing parameter. The step of processing the first sensing parameters to determine whether the current parking mode matches the set mode includes: The first vibration parameter, the first tilt angle change parameter, or the current sensing parameter are processed to determine the set of parameters in the current parking mode; the set of parameters in the current parking mode is numerically compared with the set of parameters in the set mode to determine whether the current parking mode matches the set mode.

[0007] As an optional implementation, in a first aspect of the present invention, the process of identifying second sensing parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: The triaxial acceleration, triaxial angular velocity, and triaxial magnetic field data of the motorcycle frame area are collected using an IMU sensor at a sampling rate of no less than 100Hz; the motorcycle frame area includes the motorcycle body area, the motorcycle front wheel axle area, and the motorcycle rear wheel axle area. The phase difference of IMU data in the front and rear axle regions is used to quantify the degree of twisting of the motorcycle frame during illegal handling. The abnormal proportional relationship between angular acceleration and linear acceleration during illegal movement was analyzed based on the weight distribution of the motorcycle. Based on roll angle time series data, monitor abnormal posture change sequences when the side supports are retracted or the vehicle is righted. The risk of the motorcycle being stolen is determined by the phase difference of IMU data, the abnormal ratio between angular acceleration and linear acceleration during movement, and the abnormal attitude change sequence. Alternatively, vehicle location data collected by a GNSS module can be combined with geofence information of typical motorcycle parking areas to determine the risk of theft. By deploying an array of vibration sensors at key structural points of a motorcycle, the risk of the motorcycle being stolen can be determined based on the structural vibration characteristics collected by the array of vibration sensors. The key structural points of the motorcycle include the frame triangle area, the engine suspension point, and the side stand connection point.

[0008] As an optional implementation, in the first aspect of the present invention, determining whether a motorcycle is at risk of being stolen based on the structural vibration characteristics collected by the vibration sensor array includes: The vibration sensor array will collect structural vibration characteristics and match them with the wind vibration interference model. If the structural vibration characteristics are low frequency and consistent throughout the vehicle, it is determined to be natural wind vibration and there is no risk of theft. If the structural vibration characteristics are local high-frequency impact, it is determined to be man-made vibration and there is a risk of theft.

[0009] As an optional implementation, in a first aspect of the present invention, the step of quantifying the degree of distortion of the motorcycle frame during illegal lifting based on the phase difference of IMU data in the front and rear axle regions includes: The method calculates the cross-correlation function and phase delay of the Z-axis acceleration signals from two IMU units deployed at the front and rear of the motorcycle frame in real time based on IMU data. When the phase delay exceeds a threshold and the cross-correlation peak value decreases, it is determined that the frame is undergoing abnormal twisting. The method further includes: Monitoring the time correlation between the sudden disappearance of GNSS signals and the vertical acceleration pulses detected by the IMU; When the IMU detects a typical acceleration pattern before the GNSS signal is lost, the highest level alarm is immediately triggered, recording the event locally and / or broadcasting the alarm to nearby smart devices via a Bluetooth Mesh network.

[0010] As an optional implementation, in a first aspect of the present invention, controlling the first sensing component to operate to acquire detected first sensing parameters, and processing the first sensing parameters to determine whether the current parking mode matches a set mode, includes: The current signal of the vehicle's main power circuit is monitored by a current sensor installed on the vehicle, and the current signal is matched with a set threshold. In the active verification mode, controlling the activation of the second sensing component to acquire second sensing parameters within a set time range includes: The time-domain and frequency-domain characteristics of the current signal are analyzed to generate an initial theft hypothesis. The time-domain characteristics include the rise slope, pulse shape, and duration, and the frequency-domain characteristics are obtained by performing a fast Fourier transform on the current signal. In response to a specific initial theft hypothesis, a subset of sensors is dynamically selected and activated from multiple sensors of the vehicle to form a verification chain, and auxiliary data is obtained from the sensor subset, which is the most relevant and energy-efficient combination of sensors selected according to the specific initial theft hypothesis.

[0011] As an optional implementation, in a first aspect of the present invention, generating an initial theft hypothesis includes: matching the time-domain features of the current signal with a pre-stored current behavior fingerprint database, the current behavior fingerprint database containing current patterns associated with different theft tools or methods; and identifying potential theft intent categories based on the matching results. The process of identifying and processing the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: Assign weights to different types of auxiliary data acquired from a subset of sensors; Based on the assigned weights and the auxiliary data, a confidence score for the specific initial theft hypothesis is calculated; The confidence score is compared with a preset threshold to determine whether there is a risk of theft.

[0012] A second aspect of this invention discloses a motorcycle intelligent anti-theft assessment system, characterized in that it includes: First monitoring module: When the motorcycle is in the initial monitoring state, it controls the first sensing component to work to obtain the detected first sensing parameters, and processes the first sensing parameters to determine whether the current parking mode matches the set mode. If yes, the detection continues; if no, it enters the active verification mode. Second monitoring module: used to control the activation of the second sensing component to obtain second sensing parameters within a set time range in the active verification mode; Judgment module: Used to identify and process the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is performed; if not, it returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, starting the camera to acquire images, and controlling the sound and light module to be in a working state.

[0013] A third aspect of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to execute the intelligent anti-theft assessment method for motorcycles disclosed in the first aspect of the present invention.

[0014] A fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the intelligent anti-theft assessment method for motorcycles disclosed in the first aspect of the present invention.

[0015] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In this embodiment of the invention, the intelligent anti-theft assessment method for motorcycles initially activates only the first sensing component during monitoring. The second sensing component and alarm module are only activated when an anomaly occurs, reducing system standby power consumption and extending device battery life. In non-theft scenarios, the system automatically reverts to the initial state, avoiding repeated alarms that could disturb the user and improving the user experience. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the intelligent anti-theft assessment method for motorcycles disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of a process for determining secondary risks, as disclosed in an embodiment of the present invention. Figure 3 This is another schematic diagram of the process for determining secondary risks disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a motorcycle intelligent anti-theft assessment system provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] It should be noted that the terms first, second, third, fourth, etc., in the specification and claims of this invention are used to distinguish different objects, not to describe a specific order. The terms used in the embodiments of this invention include and have, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0020] Example 1 Please see Figure 1 , Figure 1This is a flowchart illustrating the intelligent anti-theft assessment method for motorcycles disclosed in this invention. The execution entity of the method described in this embodiment is an execution entity composed of software and / or hardware. This execution entity can receive relevant information via wired or / or wireless means and can send certain instructions. It may also have certain processing and storage functions. This execution entity can control multiple devices, such as remote physical servers or cloud servers and related software, or local hosts or servers and related software that perform related operations on devices located in a certain location. In some scenarios, it can also control multiple storage devices, which may be placed in the same location as the device or in different locations. Figure 1 As shown, this motorcycle intelligent anti-theft assessment method includes the following steps: S101: When the motorcycle is in the initial monitoring state, control the first sensing component to work to obtain the detected first sensing parameters, and process the first sensing parameters to determine whether the current parking mode matches the set mode. If yes, continue the detection; if no, enter the active verification mode. S102: In the active verification mode, control the activation of the second sensing component to obtain the second sensing parameters within a set time range; S103: The second sensor parameter within the set time range is identified and processed to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is performed; otherwise, the system returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, activating the camera to acquire images, and controlling the sound and light module to be in a working state.

[0021] This invention improves the accuracy and reliability of anti-theft identification by setting up multi-level sensor verification, initial pattern matching and dynamic parameter identification within a set time window, which greatly reduces the false alarm rate caused by environmental interference (such as wind or slight collisions) and improves the accuracy of theft risk assessment.

[0022] In practice, active verification only occurs when there is no match, avoiding excessive alarms triggered at the first sign of trouble, and reducing user interference and waste of system resources. By setting up alarms through multiple channels, information pushes allow users to remotely monitor vehicle anomalies in real time, camera images facilitate subsequent evidence collection and tracking, and the continuous operation of the audio-visual modules can effectively deter theft and buy time for response.

[0023] In this embodiment of the invention, the first sensing component operates normally, while the second sensing component only activates in case of an anomaly, reducing the continuous operation of high-power modules and extending the standby time and battery life of the anti-theft system. In this embodiment of the invention, the alarm operation is only performed after risk is confirmed, avoiding unnecessary energy consumption and adapting to the limited capacity of motorcycle onboard power supplies.

[0024] More preferably, in the active verification mode, controlling the activation of the second sensing component to acquire second sensing parameters within a set time range includes: In the active verification mode, a sound signal is randomly generated within a set time period to control the sound module to work and to control the activation of the second sensing component to obtain the second sensing parameters within the set time range.

[0025] Upon detecting an initial anomaly, an alarm is randomly triggered, and the behavior following the alarm is further monitored. This is followed by further theft risk detection, as the owner's anticipated behavior is not affected, but a thief's behavior will be psychologically impacted. Further behavioral determination is then achieved by monitoring specific behavioral patterns.

[0026] The active verification mode introduces a new linkage mechanism that combines random sound signal triggering with second-sensor parameter acquisition. By actively interfering with the vehicle's operation, the system verifies the vehicle's status and can more accurately distinguish between intentional theft and other forms of theft. For example, slight vehicle shaking caused by wind and the owner's lack of response to sound signals will be mitigated by the sound signal triggered by the thief touching the vehicle. This abnormal parameter fluctuation is detected by the second-sensor component, significantly improving the accuracy of theft risk assessment and preventing false alarms caused by natural factors.

[0027] The emission of random sound signals has a deterrent effect. When a thief is committing a theft, the sudden sound disrupts their rhythm, causing them to panic and abandon their theft, thus achieving a proactive preventative effect compared to traditional passive alarms. Using random sound signals as a trigger allows the second sensing component to collect more targeted and effective parameters, avoiding the continuous collection of large amounts of invalid data within a set time range. Simultaneously, the linkage between sound signals and sensor acquisition can quickly filter out valuable parameter information, reducing the computational load of data processing and improving the overall system response speed.

[0028] Within a set time range, the identification of the second sensor parameter and the triggering of the newly added random sound signal form a dual verification logic of passive monitoring and active triggering, further reducing the false alarm space. This upgrades the theft risk assessment from single-parameter analysis to a closed-loop verification of active stimulation and feedback identification, resulting in a qualitative improvement in anti-theft accuracy. The newly added sound signal in this embodiment not only assists in parameter acquisition but also triggers deterrence in advance. Together with subsequent alarm operations (information push, image acquisition, and audio-visual alarms), it forms a proactive deterrent and multi-response protection system, extending the time before theft occurs and providing users with more opportunities for remote response and on-site handling.

[0029] The sound module only activates in active verification mode, and its timing, combined with the set data acquisition from the second sensor component, does not add excessive power consumption. Simultaneously, accurate parameter recognition reduces energy waste from invalid alarms, maintaining the system's low power consumption advantage while ensuring anti-theft performance.

[0030] The above features, through an innovative linkage design of active sound triggering and sensor parameter acquisition, solve the technical pain points of traditional anti-theft systems, such as passive monitoring being susceptible to environmental interference, high false alarm rate, and delayed deterrence. Without increasing the system's energy consumption too much, it achieves a triple improvement in the accuracy of theft risk identification, the deterrent effect of anti-theft, and the system response efficiency.

[0031] More preferably, the first sensing component includes a vibration sensor, a tilt sensor, or a current sensor; the first sensing parameter includes a first vibration parameter, a first tilt angle change parameter, or a current sensing parameter. The step of processing the first sensing parameters to determine whether the current parking mode matches the set mode includes: The first vibration parameter, the first tilt angle change parameter, or the current sensing parameter are processed to determine the set of parameters in the current parking mode; the set of parameters in the current parking mode is numerically compared with the set of parameters in the set mode to determine whether the current parking mode matches the set mode.

[0032] The first sensing component of this invention includes three types of sensors: vibration, tilt, and current, which correspond to three core parking risk scenarios: vehicle vibration triggering abnormal attitude deviation circuit. It can adapt to the monitoring needs of different vehicle models and different parking environments (such as open-air parking and garage parking), and solves the technical pain point that a single sensor cannot fully cover parking abnormalities.

[0033] The first set of sensing parameters corresponds one-to-one with the sensor type (vibration parameters, tilt angle change parameters, and current parameters), ensuring that the collected parameters accurately reflect the state of the corresponding monitoring dimension. This provides reliable data support for subsequent pattern matching and avoids interference from invalid parameters. By comparing parameter set values, the abstract parking pattern matching is transformed into a specific analysis of parameter value differences. For example, the actual vibration amplitude, tilt angle change, and current value are quantitatively compared with set thresholds, avoiding errors caused by subjective judgment and making the matching results more objective and repeatable.

[0034] By establishing corresponding parameter sets for the parameter characteristics of different sensors, it is possible to accurately identify the difference between slight vibrations during normal parking, the dormant current of the compliant parking posture circuit, and the current of the illegally raised tilt angle circuit due to abnormal prying vibration. This significantly reduces the misjudgment rate caused by environmental interference (such as wind or slight vehicle shaking) or normal operation, and reduces unnecessary triggering of active verification modes.

[0035] Processing and comparing parameters for a single sensor type eliminates the need for multi-sensor data fusion, simplifying the initial monitoring process, improving pattern matching response speed, and ensuring rapid identification and proactive verification of abnormal states. The lower energy consumption for parameter acquisition and processing of a single sensor component, combined with the logic of activating a high-power secondary sensor only in case of anomalies, further reduces overall system energy consumption, extends the anti-theft system's standby time, and is compatible with the limited power supply capabilities of motorcycle onboard power supplies.

[0036] This invention, through precise multi-sensor selection and quantitative parameter comparison, ensures effective filtering of invalid interference during the initial monitoring phase. It only enters active verification mode upon confirming a genuine anomaly, forming a progressive protection system with subsequent active verification and multi-method alarms, thus improving the overall reliability of the anti-theft solution from the source. The multi-sensor selection design allows the anti-theft system to be customized according to user needs and vehicle configuration. For example, vibration sensors can be prioritized for vehicles frequently parked outdoors, while current sensors can be used for models with easily tampered circuits.

[0037] More preferably, the step of identifying and processing the second sensing parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: S1021: Collect triaxial acceleration, triaxial angular velocity, and triaxial magnetic field data of the motorcycle frame area using an IMU sensor at a sampling rate of not less than 100Hz; the motorcycle frame area includes the motorcycle body area, the motorcycle front wheel axle area, and the motorcycle rear wheel axle area. S1022: Quantify the degree of distortion of the motorcycle frame during illegal lifting based on the phase difference of IMU data in the front and rear axle regions; S1023: Analyze the abnormal proportional relationship between angular acceleration and linear acceleration during illegal movement based on the weight distribution of the motorcycle; S1024: Based on roll angle time series data, monitor abnormal posture change sequences when the side supports are retracted or the vehicle is righted. S1025: Determine whether the corresponding motorcycle is at risk of being stolen based on the phase difference of IMU data, the abnormal proportional relationship between angular acceleration and linear acceleration during movement, and the abnormal attitude change sequence; Alternatively, vehicle location data collected by a GNSS module can be combined with geofence information of typical motorcycle parking areas to determine the risk of theft. By deploying an array of vibration sensors at key structural points of a motorcycle, the risk of the motorcycle being stolen can be determined based on the structural vibration characteristics collected by the array of vibration sensors. The key structural points of the motorcycle include the frame triangle area, the engine suspension point, and the side stand connection point.

[0038] This invention employs a sampling rate of at least 100Hz to collect triaxial acceleration, angular velocity, and magnetic field data, covering key areas such as the vehicle body and front and rear axles. This allows for precise capture of subtle dynamic changes caused by theft activities (such as lifting, prying, and righting), avoiding the loss of crucial features due to insufficient sampling rate and providing high-density, high-fidelity data support for subsequent analysis. Simultaneous data collection from multiple areas overcomes the limitations of single-location monitoring, comprehensively reflecting the overall vehicle status and preventing thieves from evading detection through localized operations.

[0039] Specifically, by using the phase difference of IMU data from the front and rear axles, the system can accurately identify the torsional deformation of the frame caused by illegal handling, effectively distinguishing between the slight shaking of a normally parked vehicle and the structural deformation caused by theft, and further eliminating interference from the natural environment (such as wind or minor vehicle collisions).

[0040] By combining motorcycle weight distribution analysis with anomalies in the ratio of angular acceleration to linear acceleration, it is possible to accurately determine whether a vehicle has been illegally moved (e.g., pushed away, carried), avoiding misjudgments based on the vehicle's own weight and improving the accuracy of identifying mobile theft. Based on roll angle time-series data, it can accurately capture abnormal posture sequences such as side stand retraction and vehicle righting. These actions are typical pre-theft behaviors, enabling early warning of theft risks and buying time for subsequent loss mitigation.

[0041] Risk assessment is based on three core features: the degree of frame torsion, the ratio of acceleration during movement, and the sequence of attitude changes. This forms a three-dimensional verification system that integrates structural deformation, movement status, and attitude changes, significantly reducing the probability of misjudgment based on a single feature and ensuring the accuracy and reliability of theft risk assessment.

[0042] This invention combines geofence information from typical motorcycle parking areas (such as garages and residential parking spaces) with GNSS location data to determine whether a vehicle has exceeded a safe zone. This allows for rapid identification of theft involving illegally driven vehicles, enabling real-time location-based monitoring. It is particularly suitable for scenarios where vehicles are stolen and then driven away from their parking areas. The GNSS path focuses on macroscopic changes in vehicle position, complementing the microscopic dynamic monitoring of the IMU path. This covers both static lifting and dynamic driving scenarios, enhancing the comprehensiveness of the anti-theft solution.

[0043] With full coverage of key structural points, the system captures typical vibration characteristics of theft. Vibration sensors are placed at key structural points such as the frame triangle, engine suspension point, and side support connection. These locations are high-frequency areas where thieves pry and disassemble vehicle parts. The system can accurately collect specific vibration signals generated by theft operations and avoid vibration interference from non-critical areas.

[0044] Array-based data fusion improves the accuracy of vibration feature recognition. By comparing data from multiple nodes of the sensor array, it can accurately distinguish the vibration characteristics of theft operations (such as picking locks and disassembling parts) from environmental vibrations (such as vibrations from passing vehicles and wind). This enables accurate identification of local theft behaviors (such as disassembling parts) and fills the gap in the monitoring of local component theft using IMU and GNSS.

[0045] The IMU path, GNSS path, and vibration sensor array path target three core theft scenarios: static lifting / dynamic removal of partial components due to abnormal posture. This forms a comprehensive, blind-spot-free anti-theft monitoring system, solving the technical pain point of traditional anti-theft solutions where a single identification path cannot cover all theft scenarios. The multi-path selectable design of this invention allows the anti-theft system to be customized according to user scenarios (such as open-air parking or garage parking) and vehicle configuration requirements, enhancing the flexibility and practicality of the technical solution.

[0046] This identification and processing step provides accurate triggering criteria for subsequent alarm operations, ensuring that alarms are only activated when a genuine risk of theft is confirmed. This avoids user interference and energy waste caused by invalid alarms. It forms a complete protection loop of monitoring-identification-alarm with the initial monitoring mode matching and the active verification parameter collection, significantly improving the overall anti-theft performance.

[0047] Different identification paths can be flexibly activated according to the needs of the scenario. For example, in areas with weak GNSS signals, such as garages, the IMU and vibration sensor array paths can be activated first. In open areas, the GNSS path can be combined to achieve dual verification, ensuring identification accuracy while rationally allocating system resources and reducing unnecessary energy consumption.

[0048] More preferably, determining whether a motorcycle is at risk of being stolen based on the structural vibration characteristics collected by the vibration sensor array includes: The vibration sensor array will collect structural vibration characteristics and match them with the wind vibration interference model. If the structural vibration characteristics are low frequency and consistent throughout the vehicle, it is determined to be natural wind vibration and there is no risk of theft. If the structural vibration characteristics are local high-frequency impact, it is determined to be man-made vibration and there is a risk of theft.

[0049] This invention establishes a differentiated judgment standard between natural wind vibration and vibration caused by theft, based on two core dimensions: frequency and vibration range (uniform throughout the vehicle / localized impact). Natural wind vibration exhibits low-frequency and uniform characteristics throughout the vehicle, while vibrations generated by theft (such as lock picking or component disassembly) are localized high-frequency impacts. The two characteristics are significantly different, enabling accurate differentiation.

[0050] A targeted matching wind vibration interference model is used. A wind vibration interference model is pre-built as a reference benchmark. The structural vibration characteristics collected by the sensor array are accurately matched with the model. This filters out the most common wind vibration interference in the natural environment from the source, solving the core pain point of false alarms caused by wind vibration in traditional anti-theft systems and improving the accuracy of theft risk assessment.

[0051] This technology adds a targeted interference filtering mechanism to the vibration sensor array's identification path, forming a closer synergy with IMU multi-dimensional identification and GNSS geofence identification. The vibration sensor array focuses on the accurate identification of localized theft operations while effectively avoiding wind vibration interference, complementing other identification paths and further improving the reliability of anti-theft coverage across all scenarios.

[0052] In the active verification mode, the vibration feature identification logic can quickly eliminate wind-induced vibration interference. Only when human-induced vibration is confirmed will theft risk be determined and an alarm be triggered, avoiding invalid alarms from being transmitted to subsequent stages, reducing user interference, and reducing energy waste caused by alarm operations. It forms a double filter with the initial monitoring mode matching logic, further improving the practicality of the overall anti-theft solution.

[0053] More preferably, the step of quantifying the degree of torsion of the motorcycle frame during illegal handling based on the phase difference of IMU data from the front and rear axle regions includes: The method calculates the cross-correlation function and phase delay of the Z-axis acceleration signals from two IMU units deployed at the front and rear of the motorcycle frame in real time based on IMU data. When the phase delay exceeds a threshold and the cross-correlation peak value decreases, it is determined that the frame is undergoing abnormal twisting. The method further includes: Monitoring the time correlation between the sudden disappearance of GNSS signals and the vertical acceleration pulses detected by the IMU; When the IMU detects a typical acceleration pattern before the GNSS signal is lost, the highest level alarm is immediately triggered, recording the event locally and / or broadcasting the alarm to nearby smart devices via a Bluetooth Mesh network.

[0054] Specifically, this method precisely quantifies the frame's torsional state, pinpoints the core characteristics of illegal lifting, and calculates the cross-correlation function and phase delay using the Z-axis acceleration signals from the front and rear axle IMU units. This transforms the physical deformation of the frame torsion into quantifiable parameters, enabling accurate detection of illegal lifting activities. Compared to traditional vibration sensors that can only detect vibration intensity, this method can directly identify the structural deformation of the frame caused by stress, avoiding misinterpreting normal collisions and shaking as theft.

[0055] A double judgment condition of setting phase delay exceeding a threshold and cross-correlation peak decreasing is provided, which further improves the accuracy of distortion recognition, eliminates local signal differences caused by uneven road surfaces and slight vehicle displacement, and ensures that risk judgment is triggered only when the frame has abnormal distortion.

[0056] For the static theft scenario where a thief lifts and moves the motorcycle (such as loading it into a carriage or transporting it to a hidden place), the method can accurately identify through the frame distortion feature, solves the technical pain points that traditional GNSS relies on position change and a single vibration sensor cannot distinguish lifting and moving from vibration, and realizes comprehensive coverage of two core theft scenarios: static lifting and moving and dynamic driving away.

[0057] In specific implementation, monitoring the temporal correlation between the sudden disappearance of GNSS signals and IMU vertical acceleration pulses can accurately determine whether GNSS signal loss is caused by natural interference or active shielding by thieves (such as placing a signal jammer). When the typical IMU acceleration mode (such as lifting, shaking) appears before GNSS loss, the behavior can be locked as theft, avoiding misjudgment or missed judgment caused by natural signal interruption.

[0058] For high-risk theft scenarios involving signal shielding and vehicle movement, the highest-level alarm is activated immediately, and events are recorded through local storage to provide key data for subsequent forensics and tracking; meanwhile, the Bluetooth Mesh network is used to broadcast alarms to surrounding intelligent terminals, realizing dual responses of local evidence retention and surrounding linkage, expanding the alarm range, and improving on-site deterrence and the recovery probability of the stolen vehicle.

[0059] Only when the high-risk conditions of GNSS signal loss and typical IMU acceleration are met, the highest-level alarm is triggered, which avoids excessive response to natural signal interruption and normal vehicle movement, reasonably allocates system resources, reduces unnecessary energy consumption, and simultaneously reduces ineffective interference to users.

[0060] In the active verification mode, through IMU phase difference analysis and GNSS-IMU time correlation monitoring, the normal state and high-risk theft behavior can be quickly distinguished, significantly improving the response speed and loss stop capability of the overall anti-theft scheme.

[0061] By utilizing the broadcast characteristic of Bluetooth Mesh network, it breaks the limitation that traditional anti-theft systems can only push alarms to the user terminal, realizes linkage response of surrounding intelligent terminals, forms a regional anti-theft network, which is especially suitable for crowded areas such as residential areas and parking lots, improves the discovery probability of theft behaviors and enhances the practicability of the anti-theft scheme.

[0062] Specifically, the system continuously monitors the quality of GNSS signals. A sudden, unexpected and severe signal loss, especially when the vehicle is stationary or moving at a low speed, is itself a huge red alarm.

[0063] Under normal circumstances, the signal disappears when the vehicle enters a tunnel or underground parking garage. At this time, the IMU (Inertial Measurement Unit) detects normal driving posture and acceleration, which has a reasonable logical correlation with the signal loss. In the case of theft, the signal disappears when the vehicle is stationary and the GNSS signal is suddenly lost. At the same time, the IMU detects a typical vertical acceleration pulse (the vehicle is rapidly lifted) or a horizontal acceleration pattern (the vehicle is pushed / pulled onto the trailer).

[0064] The unexpected disappearance of the GNSS signal and the strong correlation between the IMU detecting transport acceleration are impossible for thieves to fake. While jammers can block radio waves, they cannot eliminate the change in gravitational acceleration caused by lifting the vehicle. The received alarm beacon can be rebroadcast using its own energy and antenna at higher power or on a different channel. This expands the alarm range. The device could be a vehicle of the same model as the motorcycle or a Bluetooth-enabled smart terminal.

[0065] The system monitors the time correlation between the sudden disappearance of the GNSS signal (entering the metal cargo box) and the vertical acceleration pulse detected by the IMU (being lifted onto the truck); when a typical lift-and-drop acceleration pattern is detected before the GNSS signal is lost, the highest level alarm is immediately triggered, even if the vehicle can no longer communicate via cellular network at this time, the event is recorded locally and the alarm is broadcast to surrounding vehicles via Bluetooth Mesh network.

[0066] More preferably, the step of controlling the first sensing component to operate, acquiring detected first sensing parameters, and processing the first sensing parameters to determine whether the current parking mode matches the set mode includes: The current signal of the vehicle's main power circuit is monitored by a current sensor installed on the vehicle, and the current signal is matched with a set threshold. In the active verification mode, controlling the activation of the second sensing component to acquire second sensing parameters within a set time range includes: S102a: Analyze the time-domain and frequency-domain characteristics of the current signal to generate an initial theft hypothesis, wherein the time-domain characteristics include the rise slope, pulse shape, and duration, and the frequency-domain characteristics are obtained by performing a fast Fourier transform on the current signal; S102b: In response to a generated specific initial theft hypothesis, dynamically select and activate a subset of sensors from multiple sensors of the vehicle to form a verification chain, and obtain auxiliary data from the sensor subset, wherein the sensor subset is the most relevant and energy-efficient combination of sensors selected according to the specific initial theft hypothesis.

[0067] This invention, by monitoring the current signal of the main power circuit and matching it with a set threshold, can accurately detect illegal operations of vehicle circuits by thieves, such as stealing electricity by connecting external devices, tampering with the circuit to start the vehicle, and damaging the power system. It fills the gap in the monitoring of circuit theft scenarios by traditional vibration and tilt sensors, and realizes dual monitoring of parking posture and circuit status.

[0068] The current signal threshold matching logic is simple and efficient, quickly distinguishing between the dormant current of normal parking and the current fluctuations of abnormal operation. This avoids misjudgments caused by minor circuit interference and provides a reliable triggering basis for subsequent active verification. The current sensor only needs to collect the current signal from the main power circuit, without continuously waking up high-power modules, resulting in extremely low energy consumption. This adapts to the limited power supply capacity of motorcycle onboard power supplies and ensures 24 / 7 operation during the initial monitoring phase, without affecting the standby time of the anti-theft system due to energy consumption issues.

[0069] Multi-dimensional current feature analysis accurately generates initial theft hypotheses. By extracting the time-domain features (rise slope, pulse shape, duration) and frequency-domain features (obtained through fast Fourier transform) of the current signal, different types of circuit operations can be accurately distinguished. For example, the current signal of electricity theft shows a steady rise and a continuous high current in the time domain, while the signal of tampering with the circuit to start the vehicle has the characteristics of pulse fluctuation and specific frequency domain peaks.

[0070] By analyzing multi-dimensional features, a single current anomaly is transformed into a specific initial theft hypothesis, such as an electricity theft hypothesis or a circuit tampering hypothesis. This upgrades the process from anomaly identification to behavior prediction, providing a clear direction for subsequent accurate verification. Triggering subsequent verification based on the initial theft hypothesis avoids the resource waste of indiscriminately activating all sensors in traditional active verification modes, making the verification process more targeted and significantly improving the response speed and identification accuracy of the active verification phase.

[0071] On-demand sensor combinations maximize verification accuracy. The system dynamically selects the most relevant subset of sensors based on a specific initial theft hypothesis. For example, for the electricity theft hypothesis, voltage sensors and power monitoring modules are prioritized; for the circuit tampering-starting hypothesis, IMU sensors and GNSS modules are activated, ensuring that verification data is highly correlated with the theft hypothesis and improving the reliability of verification results. The construction of the verification chain enables sensors to work collaboratively, forming a closed loop of current anomaly prediction and multi-sensor cross-verification, avoiding the limitations of single-sensor verification and further reducing the false positive rate.

[0072] Select the most relevant and energy-efficient subset of sensors to minimize energy consumption during the active verification phase while ensuring verification accuracy. For example, prioritize waking up low-power voltage sensors rather than high-power cameras to avoid unnecessary energy consumption, extend the overall battery life of the anti-theft system, and adapt to the power supply limitations of motorcycle onboard power supplies.

[0073] From initial current signal threshold matching during monitoring, to current feature analysis to generate theft hypotheses during the active verification phase, and then to dynamic sensor subset verification, a complete protection chain is formed for circuit-related theft scenarios. This chain complements the monitoring paths of IMU, GNSS, and vibration sensor arrays, further improving the overall anti-theft coverage. The selection of dynamic sensor subsets avoids the waste of resources from waking up all sensors, allowing the system's computing power and energy consumption to be concentrated on processing sensor data most relevant to the current theft hypothesis. This improves the response speed of theft risk identification, ensuring that theft can be quickly detected and alarms triggered.

[0074] By employing a hierarchical logic that triggers active verification through current threshold matching, generates theft hypotheses through feature analysis, and verifies and confirms risks using a subset of sensors, invalid interference is filtered out at each level. Alarms are only triggered when a genuine theft risk is confirmed, significantly reducing invalid interference to users and improving the practicality and user acceptance of the anti-theft system.

[0075] More preferably, generating the initial theft hypothesis includes: matching the time-domain features of the current signal with a pre-stored current behavior fingerprint database, the current behavior fingerprint database containing current patterns associated with different theft tools or methods; and identifying potential theft intent categories based on the matching results. The process of identifying and processing the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: Assign weights to different types of auxiliary data acquired from a subset of sensors; Based on the assigned weights and the auxiliary data, a confidence score for the specific initial theft hypothesis is calculated; The confidence score is compared with a preset threshold to determine whether there is a risk of theft.

[0076] This invention, through matching the time-domain features of current signals with a pre-stored current behavior fingerprint database, transforms abstract current anomalies into specific theft tool / method classifications (such as short-circuit activation of external power theft devices, forced disassembly of power supplies, etc.). This overcomes the limitations of traditional techniques that rely solely on thresholds to determine anomalies but cannot identify the type of anomaly, achieving a precise upgrade from anomaly perception to intent prediction.

[0077] The fingerprint database is built based on current patterns of different theft tools and methods, specifically covering various circuit-related theft scenarios. This avoids missed detections due to different theft methods and significantly improves the fit and reliability of initial theft assumptions. In other words, it doesn't just look at the magnitude of the current, but also analyzes the changing state of the current, extracting multi-dimensional features in the time and frequency domains as intermediate judgment parameters.

[0078] Specifically, current timing patterns include: rising slope, duration, and pulse shape. A steep rising edge and short pulse may indicate an attempt to start the motor (starter motor engaging). A stepped rising edge may indicate that circuits are being gradually activated (e.g., lights first, then the ECU).

[0079] Current spectrum characteristics: Specific frequency components in the current signal are analyzed using FFT. 50 / 60Hz power frequency interference: Possible cause of jump-starting using substandard AC equipment. Specific switching frequency: Corresponds to the electronic control characteristics of a known car theft tool.

[0080] Current loop correlation: Combine vehicle CAN bus data to determine if the current path is legal. If there is main current but no valid wake-up frame on the CAN bus: it is determined to be an illegal connection that bypasses the vehicle control system.

[0081] Load impedance estimation: Based on the relationship between voltage drop and current rise, estimate the type of load being connected. If estimated as a low-impedance resistive load: it may be a direct short circuit or connected to a headlight. If estimated as an inductive load: it may be attempting to drive a starter motor or oil pump.

[0082] Identifying potential theft intent categories based on matching results allows for more targeted selection of subsequent sensor subsets. For example, when a fingerprint of an external power theft device is matched, the voltage sensor and power monitoring module are activated first; when a fingerprint of a short circuit activation is matched, the IMU and GNSS are linked to monitor vehicle movement, avoiding the waste of computing power and energy caused by indiscriminately activating sensors and improving the efficiency of proactive verification.

[0083] By assigning differentiated weights to different types of auxiliary data—for example, current-related auxiliary data has a higher weight than vibration data for the circuit tampering hypothesis, and IMU acceleration data has a higher weight than current data for the vehicle movement hypothesis—risk assessment is made more closely aligned with the core characteristics of the current theft intent, avoiding the one-sidedness of a single data dimension. The weighting mechanism fully considers the correlation between different sensor data and theft hypotheses, enabling confidence scores to accurately reflect the true degree of risk and solving the problems of misjudgment and missed judgment caused by traditional one-size-fits-all judgments.

[0084] By calculating a confidence score for the initial theft hypothesis, the risk of theft is transformed into a quantifiable numerical indicator, replacing vague yes / no judgments and achieving precise risk grading. For example, a score above a high threshold triggers the highest level alarm, a score between high and low thresholds initiates enhanced monitoring, and a score below the low threshold returns to the initial state, making the response strategy more flexible. The logic of comparing the confidence score with the preset threshold is simple and efficient, quickly outputting risk assessment results to ensure timely identification of theft while avoiding overreaction to low-risk anomalies.

[0085] From matching current behavior fingerprint databases to generate accurate theft hypotheses, to weighted calculation of confidence scores to confirm risks, and then to triggering graded alarms based on the scores, a complete intelligent protection chain is formed for circuit theft scenarios. It works in deep collaboration with the monitoring paths of IMU, GNSS, and vibration sensor arrays to further improve the coverage of all scenarios and all types of theft.

[0086] Fingerprint database matching makes sensor subset selection more accurate, and the weighted scoring mechanism avoids invalid alarms. The combination of the two greatly reduces unnecessary sensor wake-ups and data processing, ensuring anti-theft accuracy while minimizing system power consumption and adapting to the limited power supply capabilities of motorcycle onboard power supplies. In this embodiment of the invention, the intelligent anti-theft assessment method for motorcycles initially activates only the first sensing component during monitoring. The second sensing component and alarm module are only activated when an anomaly occurs, reducing system standby power consumption and extending device battery life. In non-theft scenarios, the system automatically reverts to the initial state, avoiding repeated alarms that could disturb the user and improving the user experience.

[0087] Example 2 Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of the intelligent anti-theft assessment system for motorcycles disclosed in an embodiment of the present invention. Figure 4 As shown, the intelligent anti-theft assessment system for motorcycles may include: First monitoring module 21: When the motorcycle is in the initial monitoring state, it controls the first sensing component to work to obtain the detected first sensing parameters, and processes the first sensing parameters to determine whether the current parking mode matches the set mode. If yes, it continues to detect; if no, it enters the active verification mode. Second monitoring module 22: used to control the activation of the second sensing component to obtain second sensing parameters within a set time range in the active verification mode; Judgment module 23: Used to identify and process the second sensing parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is performed; if not, it returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, starting the camera to acquire images, and controlling the sound and light module to be in a working state.

[0088] In this embodiment of the invention, the intelligent anti-theft assessment method for motorcycles initially activates only the first sensing component during monitoring. The second sensing component and alarm module are only activated when an anomaly occurs, reducing system standby power consumption and extending device battery life. In non-theft scenarios, the system automatically reverts to the initial state, avoiding repeated alarms that could disturb the user and improving the user experience.

[0089] Example 3 Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device can be a computer, a server, etc. Of course, in certain cases, it can also be a mobile phone, tablet computer, monitoring terminal, or other smart device, as well as an image acquisition device with processing capabilities. Figure 5 As shown, the electronic device may include: Memory 510 storing executable program code; Processor 520 coupled to memory 510; The processor 520 calls the executable program code stored in the memory 510 to execute some or all of the steps in the motorcycle intelligent anti-theft assessment method in Embodiment 1.

[0090] This invention discloses a computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps in the intelligent anti-theft assessment method for motorcycles in Embodiment 1.

[0091] This invention also discloses a computer program product, wherein when the computer program product is run on a computer, the computer performs some or all of the steps in the intelligent anti-theft assessment method for motorcycles in Embodiment 1.

[0092] This invention also discloses an application publishing platform, which is used to publish computer program products. When the computer program products are run on a computer, the computer performs some or all of the steps in the motorcycle intelligent anti-theft assessment method in Embodiment 1.

[0093] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0097] In the embodiments provided by this invention, it should be understood that B corresponding to A means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0098] Those skilled in the art will understand that some or all of the steps in the various methods of the embodiments described can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0099] The above provides a detailed description of the intelligent anti-theft assessment method, system, electronic device, and storage medium for motorcycles disclosed in the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for intelligent anti-theft assessment of motorcycles, characterized in that, include: When the motorcycle is in the initial monitoring state, the first sensing component is controlled to work to obtain the detected first sensing parameters, and the first sensing parameters are processed to determine whether the current parking mode matches the set mode. If yes, the detection continues; if no, the active verification mode is entered. The process of controlling the first sensing component to work to obtain the detected first sensing parameters and processing the first sensing parameters to determine whether the current parking mode matches the set mode includes: monitoring the current signal of the vehicle's main power circuit through a current sensor installed on the vehicle, and matching the current signal with a set threshold. In the active verification mode, controlling the activation of the second sensing component to acquire second sensing parameters within a set time range; the step of controlling the activation of the second sensing component to acquire second sensing parameters within a set time range in the active verification mode includes: The time-domain and frequency-domain characteristics of the current signal are analyzed to generate an initial theft hypothesis. The time-domain characteristics include the rise slope, pulse shape, and duration, and the frequency-domain characteristics are obtained by performing a fast Fourier transform on the current signal. In response to a specific initial theft hypothesis, a subset of sensors is dynamically selected and activated from multiple sensors of the vehicle to form a verification chain, and auxiliary data is obtained from the sensor subset to verify the initial theft hypothesis, wherein the sensor subset is the most relevant and energy-efficient combination of sensors selected according to the specific initial theft hypothesis; The second sensor parameters within a set time range are identified and processed to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is initiated; otherwise, the system returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, activating the camera to acquire images, and keeping the audio-visual module in a working state.

2. The intelligent anti-theft assessment method for motorcycles as described in claim 1, characterized in that, In the active verification mode, controlling the activation of the second sensing component to acquire second sensing parameters within a set time range includes: In the active verification mode, a sound signal is randomly generated within a set time period to control the sound module to work and to control the activation of the second sensing component to obtain the second sensing parameters within the set time range.

3. The intelligent anti-theft assessment method for motorcycles as described in claim 1, characterized in that, The first sensing component includes a vibration sensor, a tilt sensor, or a current sensor; the first sensing parameter includes a first vibration parameter, a first tilt angle change parameter, or a current sensing parameter. The step of processing the first sensing parameters to determine whether the current parking mode matches the set mode includes: The first vibration parameter, the first tilt angle change parameter, or the current sensing parameter are processed to determine the set of parameters in the current parking mode; The parameters in the current parking mode are compared with those in the setting mode to determine whether the current parking mode and the setting mode match.

4. The intelligent anti-theft assessment method for motorcycles as described in claim 1, characterized in that, The process of identifying and processing the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: The triaxial acceleration, triaxial angular velocity, and triaxial magnetic field data of the motorcycle frame area are collected using an IMU sensor at a sampling rate of no less than 100Hz; the motorcycle frame area includes the motorcycle body area, the motorcycle front wheel axle area, and the motorcycle rear wheel axle area. The phase difference of IMU data in the front and rear axle regions is used to quantify the degree of twisting of the motorcycle frame during illegal handling. The abnormal proportional relationship between angular acceleration and linear acceleration during illegal movement was analyzed based on the motorcycle's weight distribution. Based on roll angle time series data, monitor abnormal posture change sequences when the side supports are retracted or the vehicle is righted. The risk of the motorcycle being stolen is determined by the phase difference of IMU data, the abnormal ratio between angular acceleration and linear acceleration during movement, and the abnormal attitude change sequence. Alternatively, vehicle location data collected by a GNSS module can be combined with geofence information of typical motorcycle parking areas to determine the risk of theft. By deploying an array of vibration sensors at key structural points of a motorcycle, the risk of the motorcycle being stolen can be determined based on the structural vibration characteristics collected by the array of vibration sensors. The key structural points of the motorcycle include the frame triangle area, the engine suspension point, and the side stand connection point.

5. The intelligent anti-theft assessment method for motorcycles as described in claim 4, characterized in that, The step of determining whether a motorcycle is at risk of being stolen based on the structural vibration characteristics collected by the vibration sensor array includes: The vibration sensor array will collect structural vibration characteristics and match them with the wind vibration interference model. If the structural vibration characteristics are low frequency and consistent throughout the vehicle, it is determined to be natural wind vibration and there is no risk of theft. If the structural vibration characteristics are local high-frequency impact, it is determined to be man-made vibration and there is a risk of theft.

6. The intelligent anti-theft assessment method for motorcycles according to claim 4, characterized in that, The method of quantifying the degree of distortion of the motorcycle frame during illegal handling based on the phase difference of IMU data from the front and rear axle regions includes: The method calculates the cross-correlation function and phase delay of the Z-axis acceleration signals from two IMU units deployed at the front and rear of the motorcycle frame in real time based on IMU data. When the phase delay exceeds a threshold and the cross-correlation peak value decreases, it is determined that the frame is undergoing abnormal twisting. The method further includes: Monitoring the time correlation between the sudden disappearance of GNSS signals and the vertical acceleration pulses detected by the IMU; When the IMU detects a typical acceleration pattern before the GNSS signal is lost, the highest level alarm is immediately triggered, recording the event locally and / or broadcasting the alarm to nearby smart devices via a Bluetooth Mesh network.

7. The intelligent anti-theft assessment method for motorcycles according to claim 1, characterized in that, The process of generating an initial theft hypothesis includes: matching the time-domain features of the current signal with a pre-stored current behavior fingerprint database, which contains current patterns associated with different theft tools or methods; and identifying potential theft intent categories based on the matching results. The process of identifying and processing the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen includes: Assign weights to different types of auxiliary data acquired from a subset of sensors; Based on the assigned weights and the auxiliary data, a confidence score for the specific initial theft hypothesis is calculated; The confidence score is compared with a preset threshold to determine whether there is a risk of theft.

8. A motorcycle intelligent anti-theft assessment system, characterized in that, include: The first monitoring module is used to control the first sensing component to operate when the motorcycle is in the initial monitoring state, to obtain the detected first sensing parameters, and to process the first sensing parameters to determine whether the current parking mode matches the set mode. If yes, the detection continues; if no, the active verification mode is entered. The control of the first sensing component to operate to obtain the detected first sensing parameters and to process the first sensing parameters to determine whether the current parking mode matches the set mode includes: monitoring the current signal of the vehicle's main power circuit through a current sensor installed on the vehicle, and matching the current signal with a set threshold. The second monitoring module is used to control the activation of a second sensing component to acquire second sensing parameters within a set time range in the active verification mode; the control of activating the second sensing component to acquire second sensing parameters within a set time range in the active verification mode includes: The time-domain and frequency-domain characteristics of the current signal are analyzed to generate an initial theft hypothesis. The time-domain characteristics include the rise slope, pulse shape, and duration, and the frequency-domain characteristics are obtained by performing a fast Fourier transform on the current signal. In response to a specific initial theft hypothesis, a subset of sensors is dynamically selected and activated from multiple sensors of the vehicle to form a verification chain, and auxiliary data is obtained from the sensor subset to verify the initial theft hypothesis, wherein the sensor subset is the most relevant and energy-efficient combination of sensors selected according to the specific initial theft hypothesis; Judgment module: Used to identify and process the second sensor parameters within a set time range to determine whether the corresponding motorcycle is at risk of being stolen. If so, an alarm operation is performed; if not, it returns to the initial monitoring state. The alarm operation includes one or more of the following: sending information to the user terminal, starting the camera to acquire images, and controlling the sound and light module to be in a working state.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes a computer to perform the intelligent anti-theft assessment method for motorcycles according to any one of claims 1 to 7.

Images