Shared vehicle overload early warning system and method based on user behavior analysis

Abstract

The invention discloses a shared vehicle overload early warning system and method based on user behavior analysis, and relates to the technical field of vehicle early warning, and the system comprises a passenger portrait construction module which is used for dynamically adjusting an overload alarm threshold value through user historical data; the environment self-adaptive calibration module is used for compensating the weighing error of the electronic scale caused by the gradient, bumping and temperature change of the road section where the vehicle is located in real time; and the multi-scene threshold switching module is used for presetting scene configurations, each configuration comprises a specific threshold and a compensation parameter, and non-inductive switching is carried out through GPS positioning and time conditions during operation. According to the invention, the problem that the existing shared vehicle overload detection technology has obvious shortages in the aspects of personalized and scenarized management, real-time guarantee and the like is solved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle warning technology, and specifically to a shared vehicle overload warning system and method based on user behavior analysis. Background Art

[0002] In recent years, shared mobility has rapidly grown thanks to its convenience, flexibility, and environmental friendliness. However, this has also led to an increasing number of safety risks, particularly overloading. Common shared electric bicycles and scooters often use underbody pressure sensors or under-pedal weighing devices to detect loads, but these have the following major drawbacks in actual operation: Existing systems usually artificially set a unified overload alarm threshold at the factory or in the early stages of operation, ignoring the weight differences between different users and fluctuations in the additional weight of luggage. For scenarios where lighter users carry heavy objects or heavier users carry less luggage, this threshold may generate a large number of false alarms and may also result in missed alarms.

[0003] When a vehicle is traveling on urban roads, underground garages, or even on slopes or uneven rural roads, the tilt angle and bumpy vibrations can cause significant deviations in the weighing module output. Some high-end solutions incorporate temperature, tilt, and acceleration sensors, but these often only perform simple threshold filtering or one-time calibration, making it difficult to meet the needs of real-time, continuous compensation and unable to eliminate environmental impacts with millisecond-level response.

[0004] To this end, the present invention provides a shared vehicle overload warning system and method based on user behavior analysis. Summary of the Invention

[0005] The purpose of the present invention is to provide a shared vehicle overload warning system and method based on user behavior analysis to solve the existing problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a shared vehicle overload warning system based on user behavior analysis, comprising: A passenger profile building module is used to dynamically adjust the overload alarm threshold using historical user data; Environmental adaptive calibration module, used to compensate in real time for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; The multi-scenario threshold switching module is used to preset scenario configurations. Each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime based on GPS positioning and time conditions.

[0007] A further improvement of the present invention is that the passenger portrait construction module includes a data acquisition unit and a passenger model training unit; The data acquisition unit is used to collect the weight value when the user scans the code to unlock the door. , represents the weight of the i-th weighing, and a linear displacement sensor is fixed on the inner side of the center shock absorber between the rear axle and the body of the vehicle. One end of the sensor is fixed on the shock absorber cylinder, and the other end moves synchronously with the shock absorber piston rod. The sensor output is a voltage signal After the vehicle is unlocked, read the center displacement sensor value once at no-load instant and record it as the reference stroke , with a cycle Read the current voltage , converted to travel , is the zero voltage, and the real-time stroke is obtained by exponential smoothing ; Define the smooth stroke corresponding to the i-th measuring point as , then the stroke change is .

[0008] A further improvement of the present invention is that the passenger model training unit includes, after each code scanning to unlock, first reading the previous weight value obtained after the user scanned the code to weigh himself last time, and the current weight value obtained by scanning the code to weigh himself this time from the ECU; combined with the smoothing coefficient, it is sent into the weighted moving average model, and the model automatically balances the latest measurement with the historical trend in the manner of α times the current weight value plus (1-α) times the previous weight value, thereby obtaining a predicted weight, recorded as W_pred; at the same time, a safety factor k and a smoothing factor α are obtained through a prediction parameter adaptive strategy; then, the system further calls the user's weight standard deviation; the safety factor is multiplied by the weight standard deviation to obtain a confidence tolerance; finally, the system predicts the weight W pred Add it to the confidence tolerance to generate the intelligent overload threshold; after the vehicle completes zero point calibration and measures the current "real load", the ECU only needs to compare the real load with the intelligent overload threshold Wth t A comparison is performed; if the actual load is less than the intelligent overload threshold, the vehicle is released normally; if the actual load is greater than or equal to the intelligent overload threshold, an overload alarm or power-off protection is immediately triggered.

[0009] A further improvement of the present invention is that the prediction parameter adaptive strategy in the passenger model training unit includes collecting the false alarm rate FPR and the false negative rate FNR of the previous cycle in each update cycle, and adjusting the safety factor k according to the following formula: ; in, represents the safety factor learning rate, Indicates setting the maximum acceptable false alarm rate threshold. Indicates setting the maximum acceptable false negative rate threshold; Calculate the most recent M prediction residuals , residual variance , if the residual variance Greater than the set target residual variance , then increase α, otherwise, decrease α, then the α update formula is: , Represents the smoothing coefficient learning rate.

[0010] The present invention is further improved in that the environment adaptive calibration module includes an error model establishment unit and a weight calibration unit; the error model establishment unit includes a tilt sensor installed on the vehicle body to detect the tilt angle of the vehicle in real time. At the same time, the acceleration sensor is used to continuously monitor the acceleration changes caused by the bumps and record them as the current acceleration. The electronic scale itself continues to output the original weight signal at a fixed sampling frequency. The system first reads the current inclination angle and then calculates the weight of the weight according to the current inclination angle. The vertical component loss is obtained; then the system reads the current acceleration and obtains the vibration compensation interference by multiplying it with the vibration compensation coefficient; the weight calibration unit adds the vertical component loss and the vibration compensation interference to obtain the environmental error compensation value, and adds it to the original weight reported by the electronic scale to obtain the current net load.

[0011] The present invention is further improved in that the multi-scene threshold switching module includes setting numbers S1, S2, and S3 for each scene, number S1 corresponds to urban road conditions, number S2 corresponds to suburban road conditions, and number S3 corresponds to night mode; and calculating the threshold increment ratio of the corresponding scene , obtain the environmental error compensation value eec, and set the scene trigger condition; the geo-fence polygon corresponding to each scene is represented by a vertex list as ,in, Represents a pair of longitude and latitude coordinates, and the polygon is closed. A horizontal ray is emitted from the test point (x, y) to the east, and the number of intersections between the ray and each edge of the polygon is counted. If the number of intersections is odd, the point is inside the polygon, and if the number of intersections is even, the point is outside the polygon. For each edge of the polygon, determine whether the latitude of the horizontal ray intersects with the edge, calculate the intersection coordinates, and if the intersection coordinates are greater than the longitude of the starting point, it is considered a valid intersection, and then the geographic fence is obtained. If the GPS coordinates fall within the geographic fence of a certain scene, and the current time or temperature meets the time period or temperature conditions of the scene, it is determined to be the scene. If multiple scenes are met at the same time, the one with the highest priority is selected according to the priority table. If no scene matches, it falls back to the default mode, which is represented as , eec=0.

[0012] A further improvement of the present invention is that the multi-scenario threshold switching module is also equipped with a threshold gradient strategy, which maintains consistency in multiple determinations during the threshold transition period through the following formula: ; in, Indicates the elapsed time from the start of the switch, TS indicates the total transition time; when the switch starts, set the internal variable , accumulated in the next judgment cycle , recalculate After the judgment is completed, When TS is reached, let , and exit the transition state.

[0013] A further improvement of the present invention is that the multi-scene threshold switching module also includes a priority table, which includes counting the number of times each scene S is touched within the time window TS. When two or more scenes fall into the same moment at the same time, the scene priorities are sorted from high to low according to the number of times touched.

[0014] On the other hand, the present invention provides a shared vehicle overload warning method based on user behavior analysis, comprising the following steps: S1. Dynamically adjust the overload alarm threshold using user historical data; S2, real-time compensation for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; S3, preset scene configuration, each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime through GPS positioning and time conditions.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention first uses a passenger profile construction module, leveraging historical weight measurement data and changes in center shock absorber travel, to accurately model each user's weight distribution. This then combines exponentially weighted moving average prediction with confidence tolerance calculation based on historical standard deviations to dynamically adjust the overload alarm threshold. This allows for rapid response to actual weight changes while effectively suppressing occasional measurement jitter, significantly reducing false alarm and missed alarm rates. 2. A linear displacement sensor is placed inside the rear axle's center shock absorber. After hardware and software filtering, it obtains real-time smoothed stroke values, which are then combined with weight change to determine occupant status. This secondary verification mechanism effectively identifies abnormal load behaviors such as people standing on the pedals or heavy objects being attached, further improving the reliability and safety of overload detection. 3. Through the environmental adaptive calibration module, the vehicle's inclination angle and vertical acceleration signals are collected separately, and the interference of slope and bumps is quickly eliminated, ensuring the high-precision stability of the electronic scale readings under various road conditions and temperature changes, and avoiding weight measurement deviations caused by environmental factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a framework diagram of the shared vehicle overload warning system based on user behavior analysis of the present invention; Figure 2 This is a flow chart of the shared vehicle overload warning method based on user behavior analysis of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0018] The term "and / or" is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0019] Example 1 Figure 1 The framework diagram of the shared vehicle overload warning system based on user behavior analysis disclosed in this embodiment is shown, including: A passenger profile building module is used to dynamically adjust the overload alarm threshold using historical user data; Environmental adaptive calibration module, used to compensate in real time for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; The multi-scenario threshold switching module is used to preset scenario configurations. Each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime based on GPS positioning and time conditions.

[0020] The passenger portrait construction module includes a data acquisition unit and a passenger model training unit; The data acquisition unit is used to collect the weight value when the user scans the code to unlock the door. , represents the weight of the i-th weighing, and a linear displacement sensor is fixed on the inner side of the center shock absorber between the rear axle and the body of the vehicle. One end of the sensor is fixed on the shock absorber cylinder, and the other end moves synchronously with the shock absorber piston rod. The sensor output is a voltage signal , which is linearly related to the relative displacement of the piston rod; after the vehicle is started / unlocked, the center displacement sensor value is read once at no-load instant and recorded as the reference stroke , with a cycle Read the current voltage , converted to travel , is the zero voltage, and the real-time stroke is obtained by exponential smoothing ; Define the i-th measurement point, such as after the first weight measurement after unlocking and before the second code scanning, the corresponding smooth stroke is , then the stroke change is If many people or heavy objects are on board, the center shock absorber will be compressed extra. Significantly larger than the small fluctuations caused by road bumps. If more people or heavy objects are on the vehicle, the center shock absorber will be compressed additionally. Significantly larger than the tiny fluctuations caused by road bumps; simultaneously obtain timestamp, geographic location, and ambient temperature.

[0021] The sensor output is connected to the multi-channel analog-to-digital converter (ADC) inside the ECU. The sampling accuracy is recommended to be at least 12 bits and the sampling rate is above 100Hz to capture road bumps and load differences. The ECU is configured with differential input and performs hardware filtering (typically 50Hz anti-aliasing filtering), followed by software filtering (low-pass, sliding average) to ensure smooth displacement signals.

[0022] The passenger model training unit includes the following steps: after each code scan to unlock, first read the previous weight value obtained after the user's last weight measurement by scanning the code, and the current weight value obtained by scanning the code for weight measurement; combined with the smoothing coefficient, it is sent into the weighted moving average model. The model automatically balances the latest measurement with the historical trend by adding (1-α) times the previous weight value, thereby obtaining a predicted weight, recorded as W_pred; at the same time, a safety factor k and a smoothing coefficient α are obtained through a prediction parameter adaptive strategy. The smoothing coefficient α here is regularly issued by the cloud and can be adjusted online according to measurement noise and weight fluctuations to ensure that the prediction can quickly respond to the user's actual weight changes and minimize occasional measurement jitter; then, the system further calls the user's weight standard deviation, which is calculated based on the user's past multiple weight measurement data to quantify the range of normal fluctuations in the user's weight (including luggage) each time he gets on the bus; the safety factor is multiplied by the weight standard deviation to obtain a confidence tolerance; finally, the system predicts the weight W pred Add it to the confidence tolerance to generate the intelligent overload threshold; after the vehicle completes zero point calibration and measures the current "real load", the ECU only needs to compare the real load with the intelligent overload threshold Wth t A comparison is performed; if the actual load is less than the intelligent overload threshold, the vehicle is released normally; if the actual load is greater than or equal to the intelligent overload threshold, an overload alarm or power-off protection is immediately triggered.

[0023] The ECUs (electronic control units) on shared mobility vehicles typically have limited resources and are not equipped to run complex deep learning models.

[0024] Real-time requirements: Threshold update and judgment must be completed at the moment of unlocking (millisecond level), and long calculation delays cannot be tolerated.

[0025] The prediction parameter adaptive strategy in the passenger model training unit includes collecting the false alarm rate FPR and false negative rate FNR of the previous cycle in each update cycle, and adjusting the safety factor k according to the following formula: ; in, represents the safety factor learning rate, Indicates setting the maximum acceptable false alarm rate threshold. Indicates setting the maximum acceptable false negative rate threshold; when When the false negative rate is too high, the first term pushes k to increase and raises the threshold to reduce the false negative rate. when When the false positive rate is too high, the second term pushes k to decrease, lowering the threshold to reduce false positives.

[0026] α affects the prediction's confidence in the latest and historical weight measurements: a larger α results in a faster response to new weight measurements but may be affected by noise; a smaller α results in a smoother prediction but is slower to update and may miss sudden changes in weight.

[0027] Calculate the most recent M prediction residuals , residual variance , if the residual variance Greater than the set target residual variance , then increase α, otherwise, decrease α, then the α update formula is: , Represents the smoothing coefficient learning rate.

[0028] As k increases, the threshold becomes higher, the alarm becomes more robust, and the number of missed alarms decreases (FNR↓), but the number of false alarms increases (FPR↑); as k decreases, the alarm becomes more sensitive, and the number of false alarms decreases (FPR↓), but the number of missed alarms increases (FNR↑).

[0029] The environment adaptive calibration module includes an error model building unit and a weight calibration unit; The error model establishment unit includes a tilt sensor installed on the vehicle body to detect the tilt angle of the vehicle in real time. At the same time, the acceleration sensor is used to continuously monitor the acceleration changes caused by the bumps and record them as the current acceleration. The electronic scale itself continues to output the original weight signal at a fixed sampling frequency. The system first reads the current inclination angle and then calculates the weight of the weight according to the current inclination angle. The vertical component loss is obtained; then the system reads the current acceleration and obtains the vibration compensation interference by multiplying it with the vibration compensation coefficient; the weight calibration unit adds the vertical component loss and the vibration compensation interference to obtain the environmental error compensation value, and adds it to the original weight reported by the electronic scale to obtain the current net load.

[0030] The vibration compensation coefficient is obtained through experiments and calibration under different types of bumpy roads. It is used to convert the impact of instantaneous acceleration into a corresponding weight compensation value. The multi-scene threshold switching module includes setting numbers S1, S2, and S3 for each scene, where number S1 corresponds to urban traffic conditions, number S2 corresponds to suburban traffic conditions, and number S3 corresponds to night mode; and calculating the threshold increment ratio of the corresponding scene. , obtain the environmental error compensation value eec, and set the scene trigger condition; the geo-fence polygon corresponding to each scene is represented by a vertex list as ,in, Represents a pair of longitude and latitude coordinates, and the polygon is closed. A horizontal ray is emitted from the test point (x, y) to the east, and the number of intersections between the ray and each edge of the polygon is counted. If the number of intersections is odd, the point is inside the polygon, and if the number of intersections is even, the point is outside the polygon. For each edge of the polygon, determine whether the latitude of the horizontal ray intersects with the edge, calculate the intersection coordinates, and if the intersection coordinates are greater than the longitude of the starting point, it is considered a valid intersection, and then the geographic fence is obtained. If the GPS coordinates fall within the geographic fence of a certain scene, and the current time or temperature meets the time period or temperature conditions of the scene, it is determined to be the scene. If multiple scenes are met at the same time, the one with the highest priority is selected according to the priority table. If no scene matches, it falls back to the default mode, which is represented as , eec=0; the trigger condition for S1 is that the GPS is in the city center area, the temperature is 0-40℃, and the time is unlimited; the trigger condition for S2 is that the GPS is in the suburban fence, the temperature is -10-35℃, and the time is unlimited; the trigger condition for S3 is 21:00-06:00 local time.

[0031] The multi-scenario threshold switching module also implements a threshold gradient strategy, which maintains consistency across multiple decisions during threshold transitions using the following formula: ; in, Indicates the elapsed time from the start of the switch, TS indicates the total transition time; when the switch starts, set the internal variable , accumulated in the next judgment cycle , recalculate After the judgment is completed, When TS is reached, let , and exit the transition state.

[0032] The multi-scene threshold switching module also includes a priority table, which includes counting the number of times each scene S is touched within the time window TS. When two or more scenes fall into two or more scenes at the same time, the scene priorities are sorted from high to low according to the number of times touched.

[0033] The more times a scenario is touched, the more frequently it's used, and the greater its impact on the user experience. When multiple scenarios are encountered simultaneously, you can prioritize those with lower user experience impact by eliminating those with higher impact. Prioritize these scenarios from highest to lowest number of times touched, allowing the system to execute policies more stably and consistently in high-frequency scenarios. Number of times touched can be aggregated and counted by day, week, or month.

[0034] The thresholds, weights and other setting values ​​may be set by default according to the present invention, or may be set by an operator.

[0035] Example 2 Figure 2 The flowchart of the shared vehicle overload warning method based on user behavior analysis of the present invention is shown. Based on the same inventive concept as Example 1, the present invention provides a shared vehicle overload warning method based on user behavior analysis, including the following steps: S1. Dynamically adjust the overload alarm threshold using user historical data; S2, real-time compensation for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; S3, preset scene configuration, each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime through GPS positioning and time conditions.

[0036] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0038] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0039] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0040] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A shared vehicle overload warning system based on user behavior analysis, characterized by: include: A passenger profile building module is used to dynamically adjust the overload alarm threshold using historical user data; Environmental adaptive calibration module, used to compensate in real time for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; The multi-scenario threshold switching module is used to preset scenario configurations. Each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime based on GPS positioning and time conditions.

2. The shared vehicle overload warning system based on user behavior analysis according to claim 1 is characterized by: The passenger portrait construction module includes a data acquisition unit and a passenger model training unit; The data acquisition unit is used to collect the weight value when the user scans the code to unlock the door. , represents the weight of the i-th weighing, and a linear displacement sensor is fixed on the inner side of the center shock absorber between the rear axle and the body of the vehicle. One end of the sensor is fixed on the shock absorber cylinder, and the other end moves synchronously with the shock absorber piston rod. The sensor output is a voltage signal After the vehicle is unlocked, read the center displacement sensor value once at no-load instant and record it as the reference stroke , with a cycle Read the current voltage , converted to travel , is the zero voltage, and the real-time stroke is obtained by exponential smoothing ; Define the smooth stroke corresponding to the i-th measuring point as , then the stroke change is .

3. The shared vehicle overload warning system based on user behavior analysis according to claim 2 is characterized by: The passenger model training unit includes the following steps: after each code scan to unlock, first read the previous weight value obtained after the user scanned the code to weigh the car, and the current weight value obtained by scanning the code to weigh the car this time; combine the smoothing coefficient and send it into the weighted moving average model, and the model automatically balances the latest measurement with the historical trend in the manner of α times the current weight value plus (1-α) times the previous weight value, so as to obtain the predicted weight, which is recorded as W_pred; at the same time, the safety factor k and the smoothing factor α are obtained through the prediction parameter adaptive strategy; then, the system further calls the user's weight standard deviation; multiplies the safety factor by the weight standard deviation to obtain the confidence tolerance; finally, the system predicts the weight W pred Added with the confidence tolerance to generate the intelligent overload threshold; After the vehicle completes zero point calibration and measures the current "real load", the ECU only needs to compare the real load with the intelligent overload threshold Wth t A comparison is performed; if the actual load is less than the intelligent overload threshold, the vehicle is released normally; if the actual load is greater than or equal to the intelligent overload threshold, an overload alarm or power-off protection is immediately triggered.

4. The shared vehicle overload warning system based on user behavior analysis according to claim 3 is characterized by: The prediction parameter adaptive strategy in the passenger model training unit includes collecting the false alarm rate FPR and false negative rate FNR of the previous cycle in each update cycle, and adjusting the safety factor k according to the following formula: ； in, represents the safety factor learning rate, Indicates setting the maximum acceptable false alarm rate threshold. Indicates setting the maximum acceptable false negative rate threshold; Calculate the most recent M prediction residuals , residual variance , if the residual variance Greater than the set target residual variance , then increase α, otherwise, decrease α, then the α update formula is: , Represents the smoothing coefficient learning rate.

5. The shared vehicle overload warning system based on user behavior analysis according to claim 1 is characterized by: The environment adaptive calibration module includes an error model establishment unit and a weight calibration unit; the error model establishment unit includes a tilt sensor installed on the vehicle body to detect the vehicle's tilt angle in real time. At the same time, the acceleration sensor is used to continuously monitor the acceleration changes caused by the bumps and record them as the current acceleration. The electronic scale itself continues to output the original weight signal at a fixed sampling frequency. The system first reads the current inclination angle and then calculates the weight of the weight according to the current inclination angle. The vertical component loss is obtained; then the system reads the current acceleration and obtains the vibration compensation interference by multiplying it with the vibration compensation coefficient; the weight calibration unit adds the vertical component loss and the vibration compensation interference to obtain the environmental error compensation value, and adds it to the original weight reported by the electronic scale to obtain the current net load.

6. The shared vehicle overload warning system based on user behavior analysis according to claim 2 is characterized by: The multi-scene threshold switching module includes setting numbers S1, S2, and S3 for each scene, where number S1 corresponds to urban traffic conditions, number S2 corresponds to suburban traffic conditions, and number S3 corresponds to night mode; and calculating the threshold increment ratio of the corresponding scene. , obtain the environmental error compensation value eec, and set the scene trigger condition; the geo-fence polygon corresponding to each scene is represented by a vertex list as ,in, Represents a pair of longitude and latitude coordinates, and the polygon is closed. A horizontal ray is emitted from the test point (x, y) to the east, and the number of intersections between the ray and each edge of the polygon is counted. If the number of intersections is odd, the point is inside the polygon, and if the number of intersections is even, the point is outside the polygon. For each edge of the polygon, determine whether the latitude of the horizontal ray intersects with the edge, calculate the intersection coordinates, and if the intersection coordinates are greater than the longitude of the starting point, it is considered a valid intersection, and then the geographic fence is obtained. If the GPS coordinates fall within the geographic fence of a certain scene, and the current time or temperature meets the time period or temperature conditions of the scene, it is determined to be the scene. If multiple scenes are met at the same time, the one with the highest priority is selected according to the priority table. If no scene matches, it falls back to the default mode, which is represented as , eec=0.

7. The shared vehicle overload warning system based on user behavior analysis according to claim 1 is characterized by: The multi-scenario threshold switching module also implements a threshold gradient strategy, which maintains consistency across multiple decisions during threshold transitions using the following formula: ； in, Indicates the elapsed time from the start of the switch, TS indicates the total transition time; when the switch starts, set the internal variable , accumulated in the next judgment cycle , recalculate After the judgment is completed, When TS is reached, let , and exit the transition state.

8. The shared vehicle overload warning system based on user behavior analysis according to claim 7 is characterized by: The multi-scene threshold switching module also includes a priority table, which includes counting the number of times each scene S is touched within the time window TS. When two or more scenes fall into two or more scenes at the same time, the scene priorities are sorted from high to low according to the number of times touched.

9. A shared vehicle overload warning method based on user behavior analysis, for implementing a shared vehicle overload warning system based on user behavior analysis as described in any one of claims 1 to 8, characterized in that: The following steps are involved: S1. Dynamically adjust the overload alarm threshold using user historical data; S2, real-time compensation for electronic scale weight measurement errors caused by the slope, bumps and temperature changes of the road section where the vehicle is located; S3, preset scene configuration, each configuration contains specific thresholds and compensation parameters, and is seamlessly switched during runtime through GPS positioning and time conditions.

Images