Method and equipment for preventing accidental operation of scooter and medium

Abstract

The invention discloses a scooter accidental operation prevention method and device and a medium, and relates to the technical field of intelligent safety, and the method comprises the steps: collecting a multi-source operation state data set, extracting a speed change feature, an acceleration change feature, a posture inclination feature, a braking force feature, an environment approaching feature and a temperature feature, and calculating a real-time risk score value. Comparing with a historical scene threshold value to obtain a current risk level; a control parameter set is obtained based on the current risk level and the multi-source operation state data set, and intervention is carried out; collecting execution process data to generate an event sample, and marking and characterizing to obtain an event feature data set; and deducing a potential risk event chain by adopting an event chain prediction model based on the event feature data set, dynamically reconstructing an intervention window to obtain intervention optimization parameters, and updating an accident prevention control parameter set through a multi-objective optimization strategy to realize continuous optimization. The operation safety and the control response accuracy of the scooter can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent safety technology, and in particular to a method and device for preventing accidental operation of a scooter, and a medium. BACKGROUND

[0002] In recent years, with the rapid growth of urban micro-travel demand, scooters have been widely used as a flexible and environmentally friendly short-distance transportation tool around the world. In order to improve the safety of scooters, related technologies have been continuously developed, gradually introducing multi-source sensor data collection, real-time control strategies, and intelligent auxiliary braking, etc. to reduce the risk of accidents caused by driver misoperation, sudden obstacles or complex environmental conditions. Some advanced anti-accident control methods attempt to combine speed, acceleration, attitude angle, braking force and surrounding environment perception data for comprehensive analysis, and adjust the braking and driving actuators through specific control parameters, thereby achieving active intervention to a certain extent.

[0003] The existing related technologies still have the following shortcomings: first, they rely on fixed threshold triggering intervention, lack dynamic prediction based on future potential risk event chains and intervention window reconstruction, the response timing is not accurate enough, and it is difficult to balance safety and comfort; second, the anti-accident control parameter optimization goal is single, lacks multi-objective comprehensive optimization considering safety, comfort, energy consumption and thermal management, resulting in insufficient adaptability under different working conditions, and the intervention effect is difficult to continuously optimize. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a method for preventing accidental operation of a scooter, which solves the problems of lack of dynamic intervention mechanism based on event chain prediction and single anti-accident control parameter optimization goal in the existing method for preventing accidental operation of a scooter.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a method for preventing accidental operation of a scooter, comprising,

[0008] Collecting a multi-source running state data set during the running of the scooter, extracting multi-source running state features from the multi-source running state data set, and obtaining a risk score value;

[0009] Obtaining a current risk level through the risk score value, and obtaining a control parameter set based on the multi-source running state data set and the current risk level;

[0010] The command is executed by controlling the parameter set, execution process data is acquired, event samples are obtained through the execution process data, the event samples are subjected to dangerous result labeling and feature processing, and event feature data sets are acquired;

[0011] Based on the event feature data set, an event chain prediction model is used to obtain a potential risk event chain, an intervention window is reconstructed according to the potential risk event chain, and intervention optimization parameters are acquired;

[0012] According to the intervention optimization parameters and the control parameter set, a multi-objective optimization strategy is used to calculate and update the anti-accident control parameter set, and the updated control parameter is acquired;

[0013] The updated control parameter is transmitted to the scooter through a wireless communication mode, so that the scooter completes the continuous optimization of the anti-accident operation according to the updated control parameter.

[0014] As a preferred scheme of the method for preventing the scooter from operating accidentally, the multi-source running state data set during the running of the scooter is collected, multi-source running state features are extracted from the multi-source running state data set, and a risk score value is acquired, and the specific steps are as follows,

[0015] The speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and brake component temperature data during the running of the scooter are collected, and the multi-source running state data set is obtained;

[0016] The speed change feature, acceleration change feature, attitude tilt feature, braking force feature, environmental proximity feature and temperature feature are obtained through the multi-source running state data set, and the multi-source running state feature is obtained;

[0017] Based on the multi-source running state feature, the real-time risk score value is calculated by setting the weight.

[0018] As a preferred scheme of the method for preventing the scooter from operating accidentally, the current risk level is acquired through the risk score value, and the specific steps are as follows,

[0019] The real-time risk score value data is obtained through the real-time risk score value, and the historical scene threshold value data is obtained through the historical scene threshold value;

[0020] The real-time risk score value data and the historical scene threshold value data are compared in value, a risk comparison result is acquired, and the current risk level is determined according to the risk comparison result.

[0021] As a preferred scheme of the method for preventing the scooter from operating accidentally, the control parameter set is obtained based on the multi-source running state data set and the current risk level, and the specific steps are as follows,

[0022] The current risk level is matched with the multi-source operation state data set, and a risk feature fusion weight is calculated;

[0023] The risk feature fusion weight is combined with the corresponding feature value in the multi-source operation state data set, and a link reliability coefficient is calculated;

[0024] Based on the link reliability coefficient and the current risk level, a risk level threshold is determined, and based on the multi-source operation state data, the risk feature fusion weight, the link reliability coefficient and the risk level threshold, a control parameter set is obtained.

[0025] As a preferred scheme of the method for preventing accidental operation of the scooter, wherein: the control parameter set is executed to obtain execution process data, the event feature data set is obtained by performing dangerous result labeling and feature processing on the event sample, and the specific steps are as follows,

[0026] The control parameter set is sent to the control left and right brakes, front and rear drive modules and buffer devices for intervention operation, and execution process data is obtained;

[0027] The time sequence information of the execution process data is read, the execution process time sequence data set is obtained, and the feature change segment related to the abnormal operation or potential risk event is extracted for feature label matching to obtain the event sample;

[0028] According to the event sample, dangerous result labeling and feature processing are performed to obtain an event feature data set. As a preferred scheme of the method for preventing accidental operation of the scooter, wherein:

[0029] Based on the event feature data set, a potential risk event chain is obtained, and an intervention window is reconstructed according to the potential risk event chain to obtain an intervention optimization parameter, and the specific steps are as follows,

[0030] Based on the event feature data set, the potential risk event chain is obtained by step-by-step calculation according to a preset time window;

[0031] According to the potential risk event chain, the intervention window is dynamically reconstructed to obtain the intervention optimization parameter.

[0032] As a preferred scheme of the method for preventing accidental operation of the scooter, wherein: according to the intervention optimization parameter and the control parameter set, a multi-objective optimization strategy is adopted to calculate and update the anti-accident control parameter set, and the updated control parameter is obtained, and the specific steps are as follows,

[0033] The intervention optimization parameter and the control parameter set are parameter mapped and weight synthesized to obtain an initial solution of the anti-accident control parameter set;

[0034] The initial solution based on the anti-accident control parameter set is an optimization variable, the comprehensive optimization target value is calculated according to a multi-objective optimization strategy, and the candidate anti-accident control parameter set is iteratively solved, so that the candidate anti-accident control parameter set is obtained.

[0035] The candidate anti-accident control parameter set is subjected to constraint checking and stability testing, and is fused with the control parameter set for consistency, so that the updated control parameter is obtained.

[0036] As a preferred scheme of the method for preventing the scooter from being operated by accident, the updated control parameter is transmitted to the control system through wireless communication for continuous optimization of the anti-accident operation, and the specific steps are as follows,

[0037] Based on the updated control parameter, the updated control parameter is transmitted to the scooter through wireless communication, and the updated control parameter is used in the subsequent collection and control process, so that the continuous optimization of the anti-accident operation effect is realized.

[0038] In a second aspect, the application provides a computer device comprising a memory and a processor, the memory storing a computer program, wherein the computer program is executed by the processor to implement any step of the method for preventing the scooter from being operated by accident according to the first aspect of the application.

[0039] In a third aspect, the application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement any step of the method for preventing the scooter from being operated by accident according to the first aspect of the application.

[0040] The application has the following beneficial effects: the future potential risk event chain is deduced through the event chain prediction model based on the event feature data set, the intervention window is reconstructed in advance to improve the accuracy of intervention response, the anti-accident control parameter set is adaptively improved in multiple working conditions through parameter mapping and weight synthesis of the intervention optimization parameter and the control parameter set, the anti-accident control parameter set is updated based on the multi-objective optimization strategy of safety, comfort, energy consumption and thermal management, and the continuous optimization of the intervention effect and the balanced improvement of the comprehensive performance are realized. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0042] Fig. 1 The flowchart of the method for preventing the scooter from being operated by accident.

[0043] Fig. 2Flowchart for event chain prediction of event feature dataset.

[0044] Fig. 3 Flowchart for reconstructing intervention window and obtaining intervention optimization parameter.

[0045] Fig. 4 Flowchart for intervention optimization parameter and control parameter set mapping and optimization. DETAILED DESCRIPTION

[0046] In order to make the above objectives, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0047] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the concept of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0048] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0049] Reference Figs. 1-4 For one embodiment of the present application, the embodiment provides a method for preventing accidental operation of a scooter, comprising the following steps:

[0050] S1: Collecting a multi-source running state data set during the running process of the scooter, extracting multi-source running state features from the multi-source running state data set, and obtaining a risk score value.

[0051] Collecting speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and braking component temperature data during the running process of the scooter to obtain a multi-source running state data set.

[0052] Further, the speed data collected during the running process of the scooter, the acceleration data collected by the acceleration sensor, the attitude angle data collected by the attitude angle sensor, the braking force data collected by the braking force sensor, the buffer rod position data collected by the buffer rod position sensor, the environmental distance data collected by the environmental distance sensor and the braking component temperature data collected by the temperature sensor are merged according to the same time stamp to obtain a multi-source running state data set.

[0053] The speed change feature, the acceleration change feature, the attitude tilt feature, the braking force feature, the environment proximity feature and the temperature feature are obtained through the multi-source running state data set.

[0054] Further, the speed data, the acceleration data, the attitude angle data, the braking force data, the buffer rod position data, the environment distance data and the braking component temperature data are collected in time sequence and differential calculation is performed to obtain the change rate sequence. The mean value, the variance, the peak value, the root mean square value and the change amplitude and other feature parameters are extracted from the change rate sequence to form the speed change feature, the acceleration change feature, the attitude tilt feature, the braking force feature, the environment proximity feature and the temperature feature, and the multi-source running state feature is constructed.

[0055] Based on the multi-source running state feature, the real-time risk score value is calculated by setting the weight.

[0056] Further, the multi-source running state feature data is read, the correlation between the historical event feature data and the accident result is calculated and the preset weight parameter is obtained after normalization processing, the multi-source running state feature data is matched with the preset weight parameter one by one to form the matching result of the multi-source running state feature data and the weight parameter, the matching result of the multi-source running state feature data and the weight parameter is weighted, a weighted feature value set is obtained, and the real-time risk score value R s is calculated through the weighted feature set, and the expression is:

[0057]

[0058] wherein f i represents the feature value of the i-th feature of the multi-source running state feature data, w i represents the weight of the i-th feature of the multi-source running state feature data, σ link represents the link difference measure, γ represents the link difference suppression coefficient, and σ(x) represents the normalization function.

[0059] It should be noted that the link difference suppression coefficient γ is calibrated through regression analysis based on historical multi-link running data and corresponding risk levels, the real risk level is obtained by statistical analysis of historical running data and accident records to minimize the error between the risk score value and the real risk level, and the link difference measure σ link reflects the difference degree between a plurality of key data acquisition links, and is obtained by measuring the difference between the physical quantities obtained from two links at the same time t k through the normalized root mean square error, and the expression is:

[0060]

[0061] wherein, represents the value of link 1 at the jth feature, represents the value of link 2 at the jth feature, n represents the number of features, σ j represents the standard deviation of the jth feature in the historical sample.

[0062] S2: Obtain the current risk level through the risk score value, and obtain the control parameter set based on the multi-source running state data set and the current risk level.

[0063] Obtain the current risk level by comparing the real-time risk score value with the historical scene threshold value.

[0064] Further, obtain real-time risk score value data through the real-time risk score value, obtain historical scene threshold value data through the historical scene threshold value, the historical scene threshold value includes a low risk threshold value and a high risk threshold value, compare the real-time risk score value data with the historical scene threshold value data, obtain a risk comparison result, determine the current risk level according to the risk comparison result, and obtain current risk level data L r .

[0065] For example, when the current risk level is 1, it means that the real-time risk score value is less than the low risk threshold value, when the current risk level is 2, it means that the real-time risk score value is greater than or equal to the low risk threshold value and less than the high risk threshold value, and when the current risk level is 3, it means that the real-time risk score value is greater than or equal to the high risk threshold value.

[0066] Among them, the low risk threshold value and the high risk threshold value are obtained by clustering analysis or quantile statistics on the distribution of real-time risk score values in the historical running data set.

[0067] Obtain the corresponding control parameter set through the current risk level and the multi-source running state data set.

[0068] Further, obtain the numerical change amplitude of the speed change feature, the acceleration change feature, the attitude tilt feature, the braking force feature, the environmental proximity feature and the temperature feature in the multi-source running state data set, compare the numerical change amplitude with the risk level threshold interval determined based on the historical data, filter the related features according to the falling risk level interval, determine the feature combination corresponding to the current risk level, calculate the risk feature fusion weight, and the expression is:

[0069]

[0070] Among them, w i represents the risk feature fusion weight of the ith feature, r i represents the stability factor of the ith feature, represents the current risk level L rThe corresponding feature importance coefficient k represents the temperature coefficient.

[0071] It should be noted that, The feature importance coefficient k is determined by normalizing the correlation of each feature with the accident result in the historical event feature data set r The feature importance coefficient k is determined by normalizing the correlation of each feature with the accident result in the historical event feature data set i The stability factor of the i-th feature is represented as:

[0072]

[0073] where σ i The standard deviation of the i-th feature in the sliding time window calculated from the multi-source running state data set is represented as ε, and ε is a positive number to prevent the denominator from being zero.

[0074] The link reliability coefficient η link is calculated by combining the risk feature fusion weight with the corresponding feature value in the multi-source running state data set, and the expression is:

[0075]

[0076] where T represents the number of sampling points in the sampling time window, w i The risk feature fusion weight of the i-th feature is represented as f i (t) represents the i-th feature value collected at time t, The statistical reference value of the i-th feature value in the historical stable running state is represented as F max The theoretical maximum difference of all features after normalization is represented as D

[0077] It should be noted that the statistical reference value is obtained by calculating the mean or median of the historical running data in the stable interval.

[0078] Based on the link reliability coefficient and the current risk level, the risk level threshold is determined, and the expression is:

[0079]

[0080] where T low The low risk threshold in the risk level threshold is represented as T high The high risk threshold in the risk level threshold is represented as T The high risk threshold reference in the historical scenario threshold is represented as T The low risk threshold reference in the historical scenario threshold is represented as T linkdenotes the link reliability coefficient, k denotes the link reliability coefficient weight, δ(L r ) denotes the current risk level hysteresis compensation term, ε denotes the gap amount, and clip(x, a, b) denotes controlling the value in the interval [a, b].

[0081] It should be noted that η link By comparing the characteristic changes of the link in the running process with the historical stable characteristic reference value, the average normalized difference degree of the characteristic deviation is calculated, and the result of 1 minus the difference degree is taken as the link reliability coefficient. The current risk level hysteresis compensation term δ(L r ) is set by balancing between avoiding frequent switching and response timeliness after statistical calculation of the risk score value fluctuation range at different risk level switching in the historical multi-source running state data set and risk level change record. The link reliability coefficient weight k is a fixed value obtained by establishing a weighted evaluation function between safety index, comfort index and energy consumption index by using a multi-objective optimization method in the historical multi-source running state data set and risk event occurrence record, and searching for parameters with the minimum risk misjudgment rate and response delay as the target. The gap amount ε is the minimum difference value of the high risk threshold and the low risk threshold in the stable running state by analyzing the historical multi-source running state data set and risk level determination record, and selecting the minimum difference value of the stable sample as a fixed value under the premise of ensuring that the risk level switching does not overlap.

[0082] The risk level threshold and the brake force feature, the buffer rod position feature and the temperature feature in the multi-source running state data set are used to determine the brake and buffer parameters. The brake and buffer parameters are used as brake and buffer parameters. Through the risk feature fusion weight, the link reliability coefficient, the risk level threshold and the brake and buffer parameters, a control parameter set is constituted.

[0083] S3: Execute the command through the control parameter set, obtain the execution process data, obtain the event sample through the execution process data, and perform dangerous result labeling and feature processing on the event sample to obtain an event feature data set.

[0084] The control parameter set is sent to the control left and right brakes, front and rear drive modules and buffer devices for intervention operation.

[0085] Further, the intervention allocation ratio is calculated through the control parameter set, the current risk level and the multi-source running state data set, and the expression is:

[0086]

[0087] Wherein, π u denotes the intervention allocation ratio, u denotes the channel type identifier, s u,0 denotes the allocation reference coefficient of channel u, su,1 represents the distribution sensitive coefficient of the channel u, L r represents the current risk level, v represents the channel traversal variable in the denominator summation process.

[0088] It should be noted that s u,0 represents the basic distribution amount of the channel u when the risk level is zero, which is obtained by regression analysis on historical multi-source running state data set and actual intervention effect data, s u,1 represents the distribution amount of the channel u with the risk level L r the rate of change of increase, which is obtained by statistical regression on historical risk level and intervention distribution ratio, the channel type identifier includes left and right brakes brk, front and rear drive modules drv, and buffer devices buf, and the channel traversal variable is used to uniformly weight and sum the distribution reference coefficients and the distribution sensitive coefficients of the three channels.

[0089] The intervention control target value is obtained by multiplying the control parameter in the parameter set and the intervention distribution ratio corresponding to the parameter, and the intervention control target value is sent to the left and right brakes, the front and rear drive modules and the buffer devices to complete the intervention operation of the left and right brakes, the front and rear drive modules and the buffer devices.

[0090] When the intervention operation is performed, the execution process data is obtained.

[0091] Further, in the execution of the intervention control process, the actual output of the displacement of the left brake, the right brake, the front drive module, the rear drive module and the buffer device is recorded in real time by the sensor, the actuator output data is obtained, the speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and brake component temperature generated in the execution process are collected, the execution intervention running data is obtained, the execution intervention running data and the actuator output data are time synchronized to form the execution process data.

[0092] The event sample is obtained through the execution process data.

[0093] Further, by executing the process data, reading the time series information of speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and brake component temperature recorded in the execution process data, obtaining an execution process time series data set, based on the execution process time series data set, statistically analyzing the speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and brake component temperature in the execution process time series data set, detecting the change rate, deviation value and out-of-limit situation, when any feature appears mutation in continuous period (for example, the speed decreases by more than 3 m / s in 1 s, or the attitude angle deviates by more than 15°), exceeds the normal range (for example, the brake component temperature exceeds 80℃) or is inconsistent with the actuator output data, the continuous period is divided into a feature change segment, a feature change segment set is formed, the feature change segment set is matched with the execution process time series data set for feature label, and an event sample is generated.

[0094] According to the event sample, a dangerous result labeling and feature processing is performed to obtain an event feature data set.

[0095] Further, the feature information of speed, acceleration, attitude angle, braking force, buffer rod position, environmental distance and brake component temperature in the event sample is read to form an event sample feature set, based on the event sample feature set, the actual dangerous result situation is combined to label the dangerous result of each event sample, forming an event sample feature set with dangerous result labeling, and the event sample feature set with dangerous result labeling is normalized, time window is divided and feature vector is extracted to form an event feature data set.

[0096] S4: Based on the event feature data set, a potential risk event chain is obtained, and according to the potential risk event chain, an intervention window is reconstructed to obtain an intervention optimization parameter.

[0097] Based on the event feature data set, the potential risk event chain is obtained by step-by-step calculation according to the preset time window.

[0098] Further, based on the event feature data set, an event feature time series is formed by aligning the time stamp and resampling according to a uniform sampling period, by combining the brake response time and the historical accident duration and comparing the accuracy and timeliness of risk prediction under different time windows, the best interval is selected as the preset time window, the event feature time series is cut off by sliding with the preset time window to form an event chain prediction input sequence, the event chain prediction input sequence is input into an event chain prediction model for forward calculation to obtain an event chain prediction intensity sequence, and the event chain cumulative risk value expression is calculated in the form of discrete window integral as follows:

[0099]

[0100] wherein λ[m] represents the intensity of the event chain prediction intensity sequence at the mth sampling point, x i [m] represents the ith feature value of the event chain prediction input sequence at the mth sampling point, d represents the feature dimension, θ i represents the feature weight coefficient of the event chain prediction model, the value range is [0, 1], μ represents the baseline intensity, Δt represents the sampling interval, M represents the number of sampling points in the preset time window, R chain represents the event chain cumulative risk value.

[0101] It should be noted that the feature dimension d is determined by the number of independent feature types contained in the event feature data set, for example, when the event feature data set contains speed change feature, acceleration change feature, posture tilt feature, braking force feature, environmental proximity feature and temperature feature, d = 6, θ i is obtained by supervised learning training on the historical event feature data set, and the value range is set to [0, 1] because the importance of different features needs to be normalized in the event chain prediction model, and the sum of all feature weight coefficients is used for probability or proportion operation, so as to avoid that a certain feature factor value is too large to dominate the prediction result. The baseline intensity μ is obtained by statistical calculation on the average occurrence intensity of the same risk event in the historical event feature data set under the condition of no intervention.

[0102] The event chain cumulative risk value is matched with the event chain prediction intensity sequence one by one, the distribution of the historical event chain cumulative risk value is statistically analyzed, and the frequency of accident occurrence in different risk value intervals is analyzed combined with the labeled accident sample. When the accident probability shows an obvious upward trend in a certain value interval, the starting point of the value interval is taken as the event chain trigger threshold, and the event chain trigger threshold is analyzed. When the event chain cumulative risk value is greater than or equal to the event chain trigger threshold and the corresponding event chain prediction intensity sequence value is continuously increasing in the time dimension, the event chain prediction intensity sequence meeting the condition is sorted in chronological order to form a potential risk event chain arranged in chronological order and outputted, which is used for subsequent intervention window dynamic reconstruction and accident prevention control strategy adjustment.

[0103] According to the potential risk event chain, the intervention window is dynamically reconstructed, and the intervention optimization parameter is obtained.

[0104] Further, based on the time distribution and cumulative risk value of the events in the potential risk event chain, the starting time and duration of the intervention window are adjusted to complete the dynamic reconstruction of the intervention window, and the expression is:

[0105]

[0106] wherein W new represents the length of the dynamically reconstructed intervention window, Wbase represents the initial set intervention window length, a represents the intervention adjustment sensitivity coefficient, and the value range is [0, 1], R cmn represents the cumulative risk value of the potential risk event chain, G represents the risk score critical value, and R max represents the theoretical maximum value of the cumulative risk value, T event represents the time when the first high-risk event in the potential risk event chain is expected to occur, T now represents the current time.

[0107] It should be noted that the intervention adjustment sensitivity coefficient a is determined by regression analysis of historical intervention effect and risk response efficiency. If a < 0, the intervention direction will be reversed, which violates the risk suppression goal. If a > 1, the intervention amplitude will exceed the system safety margin, which may cause over-braking or false triggering of intervention, thereby reducing comfort and safety. The risk score critical value G is determined by statistical analysis of historical risk event data and safety specification requirements, which is the minimum risk score critical value required to trigger the anti-accident control action, and is determined after experimental calibration and safety margin correction.

[0108] The dynamically reconstructed intervention window is matched with the risk characteristics of the potential risk event chain to generate intervention optimization parameters.

[0109] S5: According to the event feature data set, a multi-objective optimization strategy is used to calculate and update the anti-accident control parameter set to obtain the updated control parameters.

[0110] The intervention optimization parameters and the control parameter set are parameter mapped and weight synthesized to obtain the initial solution of the anti-accident control parameter set.

[0111] Further, through the intervention optimization coefficient, a parameter mapping relationship matrix of the intervention optimization parameters and the control parameter set is established, the intervention optimization parameters are mapped to the corresponding parameter dimensions of the control parameter set to form the mapped control parameter set, and the initial solution of the anti-accident control parameter set is obtained by weighted synthesis according to the preset weight coefficient.

[0112] Through the initial solution of the anti-accident control parameter set, the anti-accident control parameter set is adjusted based on the multi-objective optimization strategy of safety, comfort, energy consumption and thermal management to obtain the updated control parameters.

[0113] Further, the speed change feature, acceleration change feature, attitude tilt feature, braking force feature, environmental proximity feature and temperature feature in the event feature data set are used to form the event feature input set, and the event feature input set is substituted into the multi-objective optimization strategy of safety, comfort, energy consumption and thermal management to calculate the performance index values of each optimization target, and the expression is:

[0114]

[0115] wherein J total denotes the comprehensive optimization target value, x ∈ {s, c, e, h} represents safety, comfort, energy consumption and thermal management respectively, denotes the target value obtained by normalizing the event feature data set, z x denotes the reference baseline value of each target, φ x denotes the adaptive weight, P cand denotes the candidate anti-accident control parameter set, P ori denotes the original anti-accident control parameter set, μ denotes the parameter update smoothing coefficient.

[0116] wherein z x is the reference baseline value of the corresponding target obtained by statistical analysis of the historical event feature data set in the corresponding target dimension, P cand is a plurality of anti-accident control parameter combinations obtained after the event feature data set is input into the multi-objective optimization strategy based on safety, comfort, energy consumption and thermal management, P ori is the initial control parameter combination determined based on historical operation data, factory safety calibration value and regulatory requirements at the initial stage of the scooter operation, φ x denotes the adaptive weight based on the current risk level L r and the link reliability coefficient η link calculated, and the expression is:

[0117]

[0118] wherein, denotes the current risk level L r , the priority score value of the i-th optimization target, denotes the current risk level L r , the priority score value of the u-th optimization target.

[0119] It should be noted that the priority score value is the degree of influence of each optimization target on the probability of accident occurrence under different risk levels.

[0120] Based on the comprehensive optimization target value, the anti-accident control parameter set P ori is adjusted, the historical event feature data set is statistically analyzed, and the value of the corresponding target at the set quantile point is extracted as the reference baseline value. The deviation of each optimization target value from the reference baseline value is calculated, and the anti-accident control parameter set P ori is weighted and corrected in combination with the adaptive weight to obtain the updated control parameter P new , so that the subsequent intervention execution link can achieve a balance between safety, comfort, energy consumption and thermal management.

[0121] S6: Transmit the updated control parameters to the control system through wireless communication mode, and continuously optimize the anti-accident operation effect.

[0122] Based on the updated control parameters, write back to the scooter control system through wireless communication mode.

[0123] Further, the updated control parameters are input into the wireless communication data packaging process to generate wireless transmission data packets, and the wireless transmission data packets are encrypted through the encryption coding process and sent to the scooter control system receiver. The scooter control system receiver obtains the updated control parameters through the decryption and decoding process.

[0124] In the subsequent collection and control process, the updated control parameters are used to continuously optimize the anti-accident operation effect.

[0125] Further, in the subsequent scooter running process, the updated control parameters are continuously used as the core input to participate in the collection and real-time control decision of multi-source running state data. The collected multi-source running state data and the updated control parameters are combined to generate a real-time risk score value, which is dynamically compared with a historical scene threshold to obtain a current risk level. Then, according to the current risk level, the left and right brakes, the front and rear drive modules, and the buffer device are intervened and operated, and the execution process data is recorded and converted into an event feature data set. Through a multi-objective optimization strategy, the anti-accident control parameter set is iteratively updated, so that the control strategy continuously adapts to the changing operating conditions and driving characteristics, and the anti-accident operation effect is continuously optimized and stably improved.

[0126] The embodiment also provides a computer device suitable for the method of preventing the scooter from operating accidentally, which includes a memory and a processor. The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the method of preventing the scooter from operating accidentally as proposed in the above embodiment.

[0127] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0128] The embodiment also provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method for preventing the scooter from being operated by accident according to the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk or an optical disk.

[0129] To sum up, the application achieves the following effects. The event chain prediction model based on the event feature data set is used to deduce a future potential risk event chain, so that an intervention window is reconstructed in advance to improve the accuracy of intervention response. The intervention optimization parameters are mapped and combined with the control parameter set to improve the adaptability of the anti-accident control parameter set in multiple working conditions. The anti-accident control parameter set is updated based on a multi-objective optimization strategy of safety, comfort, energy consumption and thermal management, so that the intervention effect is continuously optimized and the comprehensive performance is balanced.

[0130] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for preventing accidental operation of a scooter, characterized in that: include, Collect multi-source operational status datasets during the operation of scooters, extract multi-source operational status features from the multi-source operational status datasets, and obtain risk score values; The current risk level is obtained through the risk score, and the set of control parameters is obtained based on the multi-source operating status dataset and the current risk level. Commands are executed by controlling the parameter set to obtain execution process data. Event samples are obtained from the execution process data. Dangerous results are labeled and characterized on the event samples to obtain an event feature dataset. Based on the event feature dataset, obtain the potential risk event chain, reconstruct the intervention window according to the potential risk event chain, and obtain the intervention optimization parameters; Based on the intervention optimization parameters and control parameter sets, a multi-objective optimization strategy is used to calculate and update the anti-accident control parameter set, and obtain the updated control parameters. The updated control parameters are transmitted to the scooter via wireless communication, enabling the scooter to continuously optimize its operation to prevent accidents.

2. The method for preventing accidental operation of a scooter as described in claim 1, characterized in that: The process involves collecting a multi-source operational status dataset during the scooter's operation, extracting multi-source operational status features from this dataset, and obtaining a risk score. The specific steps are as follows: Data on speed, acceleration, attitude angle, braking force, buffer bar position, environmental distance, and braking component temperature of the scooter during operation are collected to obtain a multi-source operating status dataset. By using a multi-source operational status dataset, we obtain velocity change features, acceleration change features, attitude tilt features, braking force features, environmental proximity features, and temperature features to obtain multi-source operational status features. Based on the characteristics of multi-source operation status, a real-time risk score is calculated by setting weights.

3. The method for preventing accidental operation of a scooter as described in claim 2, characterized in that: The specific steps for obtaining the current risk level through a risk score are as follows: Real-time risk score data is obtained through real-time risk score values, and historical scenario threshold data is obtained through historical scenario threshold values. The real-time risk score data is compared with the historical scenario threshold data to obtain the risk comparison result, and the current risk level is determined based on the risk comparison result.

4. The method for preventing accidental operation of a scooter as described in claim 3, characterized in that: The specific steps for obtaining the control parameter set based on the multi-source operational status dataset and the current risk level are as follows: The current risk level is matched with the multi-source operational status dataset to calculate the risk feature fusion weight; The link reliability coefficient is calculated by combining the risk feature fusion weights with the corresponding feature values ​​in the multi-source operational status dataset. Based on the link reliability coefficient and the current risk level, the risk level threshold is determined. Based on multi-source operational status data, risk feature fusion weights, link reliability coefficient, and risk level threshold, a set of control parameters is obtained.

5. The method for preventing accidental operation of a scooter as described in claim 4, characterized in that: The steps involve executing commands through a control parameter set, obtaining execution process data, acquiring event samples from the execution process data, labeling and characterizing the event samples to obtain an event feature dataset. The control parameter set is sent to the left and right brakes, front and rear drive modules and buffer devices to intervene and obtain execution process data. Read the time series information of the execution process data, obtain the execution process time series dataset, and extract the feature change segments related to abnormal operations or potential risk events for feature label matching to obtain event samples; Based on the event samples, dangerous results are labeled and characterized to obtain an event feature dataset.

6. The method for preventing accidental operation of a scooter as described in claim 5, characterized in that: The process involves obtaining potential risk event chains based on event feature datasets, reconstructing intervention windows based on these chains, and obtaining intervention optimization parameters. The specific steps are as follows: Based on the event feature dataset, potential risk event chains are obtained by performing step-by-step calculations according to a preset time window. The intervention window is dynamically reconstructed based on the potential risk event chain to obtain intervention optimization parameters.

7. The method for preventing accidental operation of a scooter as described in claim 6, characterized in that: The steps involve calculating and updating the anti-accident control parameter set using a multi-objective optimization strategy based on the intervention optimization parameters and control parameter set, and obtaining the updated control parameters. By mapping the intervention optimization parameters to the control parameter set and synthesizing the weights, the initial solution of the anti-accident control parameter set is obtained; Using the initial solution of the anti-accident control parameter set as the optimization variable, the comprehensive optimization objective value is calculated and iteratively solved according to the multi-objective optimization strategy to obtain the candidate anti-accident control parameter set; The candidate set of control parameters for accident prevention is constrained and its stability is checked, and then it is fused with the control parameter set to obtain the updated control parameters.

8. The method for preventing accidental operation of a scooter as described in claim 7, characterized in that: The updated control parameters are transmitted to the scooter via wireless communication for continuous optimization to prevent accidental operation. The specific steps are as follows: Based on the updated control parameters, the data is transmitted to the scooter via wireless communication. The updated control parameters are then used in subsequent data acquisition and control processes to continuously optimize the anti-accident operation effect.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the method for preventing accidental operation of a scooter as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the method for preventing accidental operation of the scooter as described in any one of claims 1 to 8.

Images