Motion abnormity detection method and device

Abstract

The invention discloses a motion anomaly detection method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of collecting running state data generated by a target vehicle within a preset time period; wherein the operation state data comprises torque data and acceleration data; fitting the torque data and the acceleration data to obtain a Gaussian distribution diagram; wherein the Gaussian distribution diagram comprises a plurality of operation state points corresponding to the operation state data; and identifying abnormal state points from the operation state points based on the Gaussian distribution map, and detecting whether the motion of the target vehicle is abnormal or not according to the abnormal state points. According to the embodiment, the motion state can be automatically detected in the running process of the target vehicle, the target vehicle is maintained or repaired in advance, the time cost loss is reduced, and the sorting efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method and apparatus for detecting motion abnormalities. Background Technology

[0002] In logistics warehousing scenarios, the "nanny wolf" (also known as a sorting wall) is an important piece of automated warehousing equipment. It consists of a baggage feeding station, barcode scanning equipment, sorting carts, compartments, and sorting racks. It can automatically distribute goods requiring categorization into corresponding compartments in a random arrangement according to demand, helping to improve sorting efficiency. Goods enter from the baggage feeding station, are scanned, and then transported to the appropriate compartment by the sorting cart. However, the movement of the sorting cart mainly relies on the cooperation of gears and chains. Wear and tear or jamming of related parts can cause abnormal movement of the sorting cart, ultimately leading to equipment failure. Currently, sorting cart failures are mainly detected and maintained after the fact. By the time a failure occurs, it may have already resulted in losses in sorting efficiency, and equipment maintenance may take even longer. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for detecting motion anomalies, which can automatically collect the operating status data of a target vehicle within a preset time period, perform fitting processing on the above data to generate a Gaussian distribution map, and detect and determine whether there are any anomalies in the motion of the target vehicle based on the Gaussian distribution map, thereby realizing automatic motion status detection during the operation of the target vehicle, pre-maintenance or repair of the target vehicle, reducing time cost losses and improving sorting efficiency.

[0004] To achieve the above objectives, according to one aspect of the present invention, a method for detecting motion anomalies is provided, comprising: Collect operational status data of the target vehicle within a preset time period; the operational status data includes torque data and acceleration data; The torque and acceleration data are fitted to obtain a Gaussian distribution map; the Gaussian distribution map includes operating state points corresponding to multiple operating state data. Abnormal state points are identified from the operating state points based on Gaussian distribution maps, and the movement of the target vehicle is detected to be abnormal based on the abnormal state points.

[0005] Optionally, the torque and acceleration data are fitted to obtain a Gaussian distribution plot, including: Multiple probability distribution units are constructed based on torque data, acceleration data, and preset target influencing variables; A Gaussian distribution plot is generated based on multiple probability distribution units.

[0006] Optionally, abnormal state points are identified from the running state points based on the Gaussian distribution map, including: In a Gaussian distribution diagram, determine whether the running state point falls within the preset distribution range corresponding to any probability distribution unit; If a running status point does not fall within the preset distribution range corresponding to any probability distribution unit, the running status point is regarded as an abnormal status point.

[0007] Optionally, in response to a running state point not falling within a preset distribution range corresponding to any probability distribution unit, the running state point is considered an abnormal state point, including: Determine the probability distribution center corresponding to each probability distribution unit; Calculate the distance between the running status point and the probability distribution center corresponding to each probability distribution unit. If the distance exceeds the preset distance standard deviation, the running status point is regarded as a candidate anomaly. Determine the torque value corresponding to the candidate abnormal point. If the torque value corresponding to the candidate abnormal point exceeds the preset normal torque value, the candidate abnormal point is regarded as an abnormal state point.

[0008] Optionally, detecting whether the target vehicle's movement is abnormal based on abnormal state points includes: Count the number of abnormal status points and the number of operational status points; Calculate the ratio between the number of abnormal state points and the number of running state points to obtain the proportion of abnormal state points; If the proportion of abnormal state points exceeds a preset percentage threshold or the number of abnormal state points exceeds a preset quantity threshold, it is determined that the movement of the target vehicle is abnormal.

[0009] Optionally, after detecting whether the target vehicle's movement is abnormal based on the abnormal state points, the method further includes: In response to the detection of abnormal movement of the target vehicle, maintenance suggestion information is generated; Maintenance recommendations are sent to maintenance personnel so that they can perform pre-maintenance on the target vehicle based on these recommendations.

[0010] According to a second aspect of the present invention, a motion anomaly detection device is provided, comprising: The data acquisition module is used to collect the operating status data of the target vehicle within a preset time period; the operating status data includes torque data and acceleration data. The fitting module is used to fit the torque and acceleration data to obtain a Gaussian distribution map; the Gaussian distribution map includes operating state points corresponding to multiple operating state data. The detection module is used to identify abnormal state points from the running state points based on the Gaussian distribution map, and to detect whether there are any abnormalities in the movement of the target vehicle based on the abnormal state points.

[0011] According to a third aspect of the present invention, an electronic device is provided, comprising: One or more processors; Memory, used to store one or more programs. When one or more programs are executed by one or more processors, the one or more processors implement the methods of any of the above embodiments.

[0012] According to a fourth aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method of any of the above embodiments.

[0013] According to a fifth aspect of the present invention, a computer program product is provided, including a computer program that, when executed by a processor, implements the method of any of the above embodiments.

[0014] One embodiment of the above invention has the following advantages or beneficial effects: It collects operating status data of the target vehicle within a preset time period; wherein the operating status data includes torque data and acceleration data; it performs fitting processing on the torque data and acceleration data to obtain a Gaussian distribution map; wherein the Gaussian distribution map includes multiple operating status points corresponding to the operating status data; based on the Gaussian distribution map, it identifies abnormal status points from the operating status points, and detects whether the target vehicle's movement is abnormal based on the abnormal status points. This embodiment can automatically collect operating status data of the target vehicle within a preset time period, perform fitting processing on the above data to generate a Gaussian distribution map, and detect and judge whether the target vehicle's movement is abnormal based on the Gaussian distribution map, thereby realizing automatic motion status detection during the target vehicle's operation, allowing for pre-emptive maintenance or repair of the target vehicle, reducing time costs and improving sorting efficiency.

[0015] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the main flow of the motion anomaly detection method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the main flow of a motion anomaly detection method according to an optional embodiment of the present invention; Figure 3 This is a schematic diagram of the main modules of a motion anomaly detection device according to an embodiment of the present invention; Figure 4 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied; Figure 5 This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0017] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0018] It should be noted that the acquisition, storage, and application of personal information involved in the embodiments of the present invention comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0019] In logistics warehousing scenarios, the "nanny wolf" (also known as a sorting wall) is an important piece of automated warehousing equipment. It consists of a baggage feeding station, barcode scanning equipment, sorting carts, compartments, and sorting racks. It can automatically distribute goods requiring categorization into corresponding compartments in a random arrangement according to demand, helping to improve sorting efficiency. Goods enter from the baggage feeding station, are scanned, and then transported to the appropriate compartment by the sorting cart. However, the movement of the sorting cart mainly relies on the cooperation of gears and chains. Wear and tear or jamming of related parts can cause abnormal movement of the sorting cart, ultimately leading to equipment failure. Currently, sorting cart failures are mainly detected and maintained after the fact. By the time a failure occurs, it may have already resulted in losses in sorting efficiency, and equipment maintenance may take even longer.

[0020] In view of this, according to one aspect of the present invention, a method for detecting motion abnormalities is provided.

[0021] Figure 1 This is a schematic diagram of the main flow of a motion anomaly detection method according to an embodiment of the present invention. Figure 1 As shown, the motion abnormality detection method according to an embodiment of the present invention includes the following steps S101 to S103.

[0022] Step S101: Collect the operating status data of the target vehicle within a preset time period; wherein, the operating status data includes torque data and acceleration data.

[0023] The target vehicle is a work vehicle requiring motion status monitoring and anomaly detection, used to complete material handling, sorting, transportation, or other predetermined tasks during actual operation, such as a distribution vehicle. Operational status data is a comprehensive set of data reflecting the target vehicle's power output and motion response characteristics during operation, used to reflect the vehicle's kinematic and dynamic performance under different operating conditions, including but not limited to torque and acceleration data. Torque data is the torque information output by the target vehicle's power system to the transmission components during drive; acceleration data refers to the information on the acceleration changes of the target vehicle along a preset direction of motion during operation.

[0024] Collecting operational status data of the target vehicle within a preset time period can be achieved by deploying torque and acceleration sensors on the vehicle to collect corresponding torque and acceleration data in real time during vehicle operation and recording them synchronously according to timestamps. Alternatively, internal monitoring parameters can be read from the target vehicle's control system or drive control unit to obtain torque and acceleration data calculated or fed back by the control system in real time, thereby forming a complete set of operational status data within the preset time period.

[0025] Step S102: Fit the torque data and acceleration data to obtain a Gaussian distribution map; wherein, the Gaussian distribution map includes operating state points corresponding to multiple operating state data.

[0026] A Gaussian distribution plot is a probabilistic distribution representation used to describe the statistical distribution relationship between torque and acceleration data. It characterizes the clustering properties of operating state data in the feature space using probability density methods, reflecting the typical motion patterns of a target vehicle under different operating conditions. Operating state points are two-dimensional or multi-dimensional feature vectors composed of torque and acceleration data collected at the same time or within the same sampling period. These points are represented by specific locations on the Gaussian distribution plot, characterizing the operating state of the target vehicle at the corresponding time.

[0027] In this embodiment, the fitting process can be based on a pre-trained Gaussian mixture model. The parameters of the Gaussian mixture model are estimated iteratively, causing the torque and acceleration data to form multiple probability distribution units in the feature space. These probability distribution units are then superimposed to generate a Gaussian distribution map. Alternatively, the torque and acceleration data can be standardized, and the joint probability density can be fitted based on a preset number of Gaussian distribution functions. The mean and variance of each Gaussian distribution function are determined using maximum likelihood estimation or Bayesian estimation, thereby forming a Gaussian distribution map that reflects the overall distribution characteristics of the operating status data.

[0028] Step S103: Identify abnormal state points from the running state points based on the Gaussian distribution map, and detect whether there is any abnormality in the movement of the target vehicle based on the abnormal state points.

[0029] In this embodiment, the probability density of each operating state point in the Gaussian distribution map can be determined. If the probability density of a certain operating state point under any Gaussian distribution cell is lower than a preset probability threshold, the operating state point is identified as deviating from the normal operating distribution and thus determined as an abnormal state point. Alternatively, the distance between the operating state point and the distribution center of each Gaussian distribution cell can be calculated. If the distance from the operating state point to the nearest distribution center exceeds a preset standard deviation, the operating state point is identified as not being covered by the normal distribution and thus identified as an abnormal state point.

[0030] When detecting whether the target vehicle's movement is abnormal based on the aforementioned abnormal state points, the proportion of abnormal state points among all operating state points within a preset time period can be statistically analyzed. If the proportion exceeds a preset threshold, it is determined that the target vehicle's movement is abnormal. Alternatively, by combining the changing characteristics of torque and acceleration data corresponding to the abnormal state points, the continuity or concentration of the abnormal state points in the time dimension can be assessed. If the abnormal state points show a clustering trend during continuous operation, it is determined that the target vehicle has potential movement abnormalities and a maintenance prompt is triggered.

[0031] This embodiment collects operational status data of a target vehicle within a preset time period. This operational status data includes torque and acceleration data. The torque and acceleration data are fitted to obtain a Gaussian distribution map. This Gaussian distribution map includes multiple operational status points corresponding to the various operational status data. Based on the Gaussian distribution map, abnormal status points are identified from these operational status points, and the movement of the target vehicle is detected based on these abnormal status points. This embodiment can automatically collect operational status data of a target vehicle within a preset time period, perform fitting processing on the data to generate a Gaussian distribution map, and detect and determine whether the movement of the target vehicle is abnormal based on the Gaussian distribution map. This enables automatic motion status detection during the operation of the target vehicle, allowing for pre-emptive maintenance or repair, reducing time costs and improving sorting efficiency.

[0032] Optionally, the torque data and acceleration data are fitted to obtain a Gaussian distribution map, including: constructing multiple probability distribution units based on torque data, acceleration data and preset target influencing variables; and generating a Gaussian distribution map based on the multiple probability distribution units.

[0033] When fitting torque and acceleration data, a joint feature space is first constructed based on the torque and acceleration data collected from the target vehicle within a preset time period. Preset target influencing variables are introduced to characterize external influencing factors that are difficult to collect directly; these target influencing variables participate in the modeling as latent variables. Then, multiple probability distribution units are constructed by clustering and probabilistic modeling of the joint feature space. Each probability distribution unit corresponds to a typical operating state, and its parameters can be jointly determined by torque data, acceleration data, and target influencing variables. Further, multiple probability distribution units are weighted and superimposed to form a Gaussian distribution map that comprehensively reflects the distribution characteristics of the target vehicle's operating state. In another implementation, the torque and acceleration data can be normalized and noise suppressed first. Then, the joint probability density is fitted based on a preset number of Gaussian distribution models. The distribution center, covariance parameter, and weight coefficients of each probability distribution unit are determined through iterative estimation, thereby generating a Gaussian distribution map. The number of probability distribution units is adaptively adjusted through the model to suit different operating scenarios.

[0034] The above embodiments enable the effective characterization of complex motion conditions using probability distribution units without directly acquiring external influencing factors. This improves the accuracy of fitting torque and acceleration data, provides a stable and reliable distribution benchmark for subsequent abnormal state point identification, and thus enhances the accuracy of target vehicle motion anomaly detection and early warning capabilities.

[0035] Optionally, identifying abnormal state points from the running state points based on the Gaussian distribution map includes the following steps: in the Gaussian distribution map, determining whether the running state point falls within the preset distribution range corresponding to any probability distribution unit; in response to the running state point not falling within the preset distribution range corresponding to any probability distribution unit, the running state point is regarded as an abnormal state point.

[0036] In this embodiment, each operating state point is first mapped to the feature space corresponding to a Gaussian distribution map, and its corresponding preset distribution range is determined based on the distribution parameters of each probability distribution unit. This preset distribution range characterizes the reasonable fluctuation range of torque and acceleration data of the target vehicle under normal operating conditions. For each operating state point, it is determined whether it falls within the preset distribution range corresponding to any probability distribution unit. If the operating state point is not covered by any probability distribution unit, it indicates that the operating state point significantly deviates from the normal operating mode in a statistical sense, and thus the operating state point is considered an abnormal state point. In another embodiment, a corresponding confidence interval can be set for each probability distribution unit, and the probability density value of the operating state point under each probability distribution unit can be calculated. If the probability density of the operating state point under all probability distribution units is lower than a preset probability threshold, it is determined that the operating state point does not belong to the normal distribution range, and thus it is identified as an abnormal state point. Furthermore, the distance information between the operating state point and the distribution center of each probability distribution unit can be combined to perform a secondary screening of abnormal state points to improve the stability of anomaly identification.

[0037] This embodiment can accurately distinguish between normal and abnormal states, effectively reducing the risk of misjudgment caused by a single threshold judgment, and making the identification of abnormal state points more consistent with the actual motion characteristics of the target vehicle.

[0038] Figure 2 This is a schematic diagram of the main flow of a motion anomaly detection method according to an optional embodiment of the present invention, as shown below. Figure 2 As shown, in response to a running status point not falling within the preset distribution range corresponding to any probability distribution unit, the running status point is regarded as an abnormal status point, including: determining the probability distribution center corresponding to each probability distribution unit; calculating the distance between the running status point and the probability distribution center corresponding to each probability distribution unit, and in response to the distance exceeding a preset distance standard deviation, the running status point is regarded as a candidate abnormal point; determining the torque value corresponding to the candidate abnormal point, and in response to the torque value corresponding to the candidate abnormal point exceeding a preset normal torque value, the candidate abnormal point is regarded as an abnormal status point.

[0039] In the process of treating operating state points as anomalous state points, the probability distribution center corresponding to each probability distribution unit is first determined. The probability distribution center is used to characterize the typical torque and acceleration data combination of the target vehicle under the corresponding operating conditions. Then, for the operating state point to be judged, the distance between it and the probability distribution center corresponding to each probability distribution unit is calculated, and the distance is compared with a preset distance standard deviation. If the distance exceeds the preset distance standard deviation corresponding to the probability distribution unit, it indicates that the operating state point significantly deviates from the normal distribution range, thus identifying it as a candidate anomalous point. Further, the torque value corresponding to the candidate anomalous point is extracted and compared with a preset normal torque value. If the torque value exceeds the preset normal torque value, it is determined that the candidate anomalous point has not produced a matching motion response under high load conditions, and is ultimately regarded as an anomalous state point. In another implementation, after determining the candidate anomalous points, the deviation of acceleration data can be introduced as an auxiliary judgment condition. By simultaneously satisfying the joint constraint of distance exceeding the limit and abnormal torque increase, anomalous state points can be screened out; or, based on the continuous occurrence characteristics of operating state points in the time dimension, multiple consecutive candidate anomalous points can be merged for judgment to reduce the impact of occasional noise on the anomalous identification results.

[0040] By combining statistical distribution characteristics and kinematic constraints, abnormal state points can be stratified and screened, effectively distinguishing between real mechanical anomalies and short-term fluctuations, improving the accuracy and reliability of anomaly identification, and providing a more targeted basis for predictive maintenance of target vehicles.

[0041] Optionally, detecting whether the target vehicle's movement is abnormal based on abnormal state points includes: counting the number of abnormal state points and the number of running state points; calculating the ratio between the number of abnormal state points and the number of running state points to obtain the percentage of abnormal state points; and determining that the target vehicle's movement is abnormal in response to the percentage of abnormal state points exceeding a preset ratio threshold or the number of abnormal state points exceeding a preset quantity threshold.

[0042] First, the identified abnormal state points are statistically analyzed within a preset time period, and the total number of all operational state points within the corresponding time period is also counted. The ratio of the number of abnormal state points to the total number of operational state points is calculated to obtain the abnormal state point percentage, which reflects the frequency of abnormal operational states during the overall operation of the target vehicle. Further, the abnormal state point percentage is compared with a preset percentage threshold, or the number of abnormal state points is compared with a preset quantity threshold. If either comparison result meets the anomaly determination condition, it is determined that the target vehicle's movement deviates from the normal operating range within the preset time period, thus determining that the target vehicle's movement is abnormal. In another implementation, abnormal state points can be statistically analyzed segmented according to time sequence, calculating the abnormal state point percentage within different time windows, and combining this with the continuous occurrence characteristics of abnormal state points for a comprehensive judgment. For example, when abnormal state points continuously exceed a preset threshold in multiple adjacent time windows, it is determined that the target vehicle has a persistent movement anomaly, thereby improving the stability and reliability of anomaly determination.

[0043] The above embodiments can transform discrete abnormal state points into overall quantitative indicators, enabling a macroscopic assessment of the target vehicle's motion state, avoiding misjudgments caused by individual abnormal points, and improving the robustness of anomaly detection results.

[0044] Optionally, after detecting whether the target vehicle's movement is abnormal based on the abnormal state point, the method further includes: generating maintenance suggestion information in response to detecting that the target vehicle's movement is abnormal; and sending the maintenance suggestion information to maintenance personnel so that the maintenance personnel can perform pre-maintenance on the target vehicle based on the maintenance suggestion information.

[0045] After detecting abnormal movement in the target vehicle, a maintenance suggestion generation process is automatically triggered based on the anomaly detection results. First, the number and proportion of abnormal state points, along with the torque and acceleration data characteristics corresponding to these points, are analyzed to determine the possible types of motion anomalies in the target vehicle. Based on this analysis, maintenance suggestions are generated, indicating potential risks such as component wear, motion stagnation, or abnormal power transmission. Then, the maintenance suggestions are sent to the maintenance personnel's terminal via a pre-defined information interaction interface, enabling them to perform targeted pre-maintenance operations on the target vehicle before a serious malfunction occurs. In another implementation, maintenance suggestions can be prioritized based on the distribution characteristics of abnormal state points under different operating conditions. These suggestions can be dynamically adjusted based on the target vehicle's historical maintenance records and operating load. Simultaneously, the maintenance suggestions can be synchronized to the equipment management system to trigger maintenance work order generation or maintenance resource scheduling processes, thereby achieving automated linkage of the maintenance process.

[0046] The above methods can directly transform motion anomaly detection results into actionable maintenance recommendations, achieving closed-loop management from anomaly identification to maintenance decision-making. This reduces operational losses caused by sudden failures, improves the operational reliability and maintenance efficiency of the target vehicle, and enhances the practical application value of predictive maintenance.

[0047] According to a preferred embodiment of the present invention, a method for detecting motion anomalies in broadcast vehicles based on kinematic analysis is proposed. This method starts from the structure and motion mechanism of the broadcast vehicle, models and analyzes easily obtainable and representative operational status data, and introduces external influencing factors that are difficult to collect directly as latent variables. This allows for the predictive detection of potential motion anomalies without increasing additional sensing costs. This approach is not only applicable to broadcast vehicle equipment but also has broad applicability in the field of automated logistics and warehousing equipment.

[0048] Broadcasting vehicles typically operate on a chain-driven track powered by gears, with the motion generated by the torque output of a drive motor. From a kinematic perspective, changes in motor torque directly affect the broadcasting vehicle's acceleration, which in turn affects other motion parameters such as speed. Therefore, under normal operating conditions, torque and acceleration data should exhibit a relatively stable statistical relationship. Based on this kinematic foundation, this embodiment models the relationship between torque and acceleration data to assess the degree of deviation between predicted and actual observations. When the deviation exceeds a preset range, a potential motion anomaly is considered to exist.

[0049] In the actual data analysis process, visual analysis of the torque, speed, and acceleration data collected by the broadcast vehicle during continuous operation reveals that there are no obvious linear or simple nonlinear relationships between torque and speed data, or between torque and acceleration data. This is because the motion state of the broadcast vehicle is not only affected by the motor torque, but also by the combined effects of various external factors such as resistance, gravity, track condition, and the position of the broadcast vehicle, and these external factors change dynamically with the operating conditions. Combining kinematic theory and actual data distribution, it can be found that abnormal states usually manifest as the broadcast vehicle being unable to effectively convert large torques into corresponding accelerations under specific conditions due to track jamming, gear wear, etc., resulting in significantly insufficient acceleration under high torque conditions. This characteristic constitutes the core basis for motion anomaly detection in this embodiment.

[0050] From a modeling perspective, the relationship between torque and acceleration data can be described by a conditional probability distribution. Torque and acceleration data are explicit observable variables, while external influencing factors (such as track conditions, gear status, and load information) participate in the modeling as latent variables that are difficult to collect directly. Although the latent variables themselves cannot be precisely characterized, their influence manifests as distribution regions of varying densities and shapes within the feature space formed by the torque and acceleration data. Therefore, this embodiment employs a Gaussian mixture model to fit the joint distribution of torque and acceleration data. Multiple Gaussian distribution units are used to characterize the data patterns under different operating conditions, thereby indirectly reflecting the influence of latent variables on the motion state.

[0051] The mathematical representation of the Gaussian mixture model is as follows: ; in, The mean of each module, Indicates a Gaussian distribution. Gaussian distribution sample The probability of [the probability]. A Gaussian mixture model consists of k Gaussian distribution units, each of which is a two-dimensional Gaussian distribution. Each Gaussian distribution unit is one of them. The weight, and The number of Gaussian distribution units can be determined by `calinski_harabasz_score`, and the parameters of each Gaussian distribution unit can be calculated using the EM algorithm. For a set of data, each Gaussian distribution unit in the Gaussian mixture model corresponds to a set of latent variables related to the relationship between torque and acceleration. The Gaussian mixture model combined with the EM algorithm can effectively bypass the estimation of the latent variables themselves, instead maximizing the expectation of the latent variables in a given data scenario. For example, for the records generated by a broadcast van over a continuous operating period, a Gaussian distribution map is obtained by fitting a Gaussian mixture model. All data can be represented by four Gaussian distribution units. Each solid ellipse represents a range of one standard deviation from the distribution center, and each dashed ellipse represents a range of two standard deviations from the distribution center. Points outside three standard deviations for any Gaussian distribution unit are considered outliers. Based on kinematic characteristics, only points with torque greater than the torque of all normal points are considered. For the records generated by a broadcast van over a continuous operating period, the number or proportion of outliers can be used as a standard for judging potential faults. If it exceeds a certain threshold, it indicates that the corresponding broadcast van may have potential anomalies and needs maintenance.

[0052] In practical applications, for a single broadcast vehicle, records generated over a continuous operating period are collected, including the corresponding torque and acceleration. Sometimes, velocity can be collected first, and acceleration can be indirectly calculated. A Gaussian mixture model is used to fit the relationship between torque and acceleration, resulting in a Gaussian distribution map. Outliers are identified after fitting, and those with torque values ​​greater than all normal points are selected as the final anomalies. The number of anomalies or their proportion of all records in the current data is calculated. If the number or proportion of anomalies exceeds a certain threshold, the corresponding broadcast vehicle is considered to have potential anomalies, and maintenance is recommended.

[0053] This embodiment enables effective modeling and prediction of abnormal movement of distribution vehicles under conditions of limited data dimensions and complex operating conditions. It reduces reliance on human experience, detects potential fault risks in advance, and improves the overall operational stability and maintenance efficiency of the logistics sorting system.

[0054] According to a second aspect of the present invention, a motion abnormality detection device is provided.

[0055] Figure 3 This is a schematic diagram of the main modules of a motion anomaly detection device according to an embodiment of the present invention. Figure 3 As shown, the motion abnormality detection device 300: The data acquisition module 301 is used to acquire the operating status data of the target vehicle within a preset time period; wherein, the operating status data includes torque data and acceleration data; The fitting module 302 is used to fit the torque data and acceleration data to obtain a Gaussian distribution map; wherein, the Gaussian distribution map includes operating state points corresponding to multiple operating state data; The detection module 303 is used to identify abnormal state points from the running state points based on the Gaussian distribution map, and to detect whether the movement of the target vehicle is abnormal based on the abnormal state points.

[0056] Optionally, the fitting module 302 is also used for: Multiple probability distribution units are constructed based on torque data, acceleration data, and preset target influencing variables; A Gaussian distribution plot is generated based on multiple probability distribution units.

[0057] Optionally, the detection module 303 is also used for: In a Gaussian distribution diagram, determine whether the running state point falls within the preset distribution range corresponding to any probability distribution unit; If a running status point does not fall within the preset distribution range corresponding to any probability distribution unit, the running status point is regarded as an abnormal status point.

[0058] Optionally, the detection module 303 is also used for: Determine the probability distribution center corresponding to each probability distribution unit; Calculate the distance between the running status point and the probability distribution center corresponding to each probability distribution unit. If the distance exceeds the preset distance standard deviation, the running status point is regarded as a candidate anomaly. Determine the torque value corresponding to the candidate abnormal point. If the torque value corresponding to the candidate abnormal point exceeds the preset normal torque value, the candidate abnormal point is regarded as an abnormal state point.

[0059] Optionally, the detection module 303 is also used for: Count the number of abnormal status points and the number of operational status points; Calculate the ratio between the number of abnormal state points and the number of running state points to obtain the proportion of abnormal state points; If the proportion of abnormal state points exceeds a preset percentage threshold or the number of abnormal state points exceeds a preset quantity threshold, it is determined that the movement of the target vehicle is abnormal.

[0060] Optionally, the motion anomaly detection device 300 further includes a generation module, which is used for: In response to the detection of abnormal movement of the target vehicle, maintenance suggestion information is generated; Maintenance recommendations are sent to maintenance personnel so that they can perform pre-maintenance on the target vehicle based on these recommendations.

[0061] It should be noted that the specific implementation details of the motion abnormality detection device of the present invention have been described in detail in the motion abnormality detection method above, so the details will not be repeated here.

[0062] According to a third aspect of the present invention, an electronic device is provided, comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in the first aspect of the present invention.

[0063] According to a fourth aspect of the present invention, a computer-readable medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method provided in the first aspect of the present invention.

[0064] According to a fifth aspect of the present invention, a computer program product is provided, including a computer program that, when executed by a processor, implements the method provided in the first aspect of the present invention.

[0065] Figure 4 An exemplary system architecture 400 is shown, to which the motion anomaly detection method or motion anomaly detection device of the present invention can be applied.

[0066] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, and 403, a network 404, and a server 405. Network 404 serves as the medium for providing communication links between terminal devices 401, 402, and 403 and server 405. Network 404 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0067] Users can use terminal devices 401, 402, and 403 to interact with server 405 via network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, and 403, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0068] Terminal devices 401, 402, and 403 can be various electronic devices with displays that support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0069] Server 405 can be a server that provides various services, such as a backend management server that supports shopping websites browsed by users using terminal devices 401, 402, and 403 (for example only). The backend management server can analyze and process data such as received motion anomaly detection requests, and feed back the processing results (such as motion anomaly detection results - for example only) to the terminal device.

[0070] It should be noted that the motion anomaly detection method provided in this embodiment of the invention is generally run by server 405, and correspondingly, the motion anomaly detection device is generally set in server 405.

[0071] It should be understood that Figure 4 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0072] The following is for reference. Figure 5 It shows a schematic diagram of the structure of a computer system 500 suitable for implementing a terminal device of the present invention. Figure 5 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0073] like Figure 5As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the system 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0074] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. Drive 510 is also connected to I / O interface 505 as needed. Removable media 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 510 as needed so that computer programs read from them can be installed into storage section 508 as needed.

[0075] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is run by the central processing unit (CPU) 501, it performs the functions defined above in the system of this invention.

[0076] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0077] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more operable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually operate substantially in parallel, and they may sometimes operate in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0078] The modules described in the embodiments of the present invention can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor may be described as including a data acquisition module, a fitting module, and a detection module. The names of these modules do not necessarily limit the module itself; for example, a data acquisition module may also be described as "a module for acquiring operational status data of a target vehicle within a preset time period."

[0079] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include: collecting operating status data generated by a target vehicle within a preset time period; wherein the operating status data includes torque data and acceleration data; performing fitting processing on the torque data and acceleration data to obtain a Gaussian distribution map; wherein the Gaussian distribution map includes multiple operating status points corresponding to the operating status data; identifying abnormal status points from the operating status points based on the Gaussian distribution map, and detecting whether the movement of the target vehicle is abnormal based on the abnormal status points.

[0080] The computer program product provided in this embodiment of the invention includes a computer program that, when executed by a processor, implements the motion anomaly detection method in this embodiment of the invention.

[0081] According to the technical solution of the present invention, the following advantages or beneficial effects are achieved: Operating status data of the target vehicle generated within a preset time period is collected; wherein, the operating status data includes torque data and acceleration data; the torque data and acceleration data are fitted to obtain a Gaussian distribution map; wherein, the Gaussian distribution map includes multiple operating status points corresponding to the operating status data; abnormal status points are identified from the operating status points based on the Gaussian distribution map, and the abnormal status points are used to detect whether the target vehicle's movement is abnormal. This embodiment can automatically collect operating status data of the target vehicle generated within a preset time period, perform fitting processing on the above data to generate a Gaussian distribution map, and detect and determine whether the target vehicle's movement is abnormal based on the Gaussian distribution map, thereby realizing automatic motion status detection during the target vehicle's operation, allowing for pre-emptive maintenance or repair of the target vehicle, reducing time costs and improving sorting efficiency.

[0082] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0083] It should be noted that the acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

Claims

1. A method for detecting motion abnormalities, characterized in that, include: Collect operating status data of the target vehicle within a preset time period; wherein, the operating status data includes torque data and acceleration data; The torque data and the acceleration data are fitted to obtain a Gaussian distribution map; wherein, the Gaussian distribution map includes multiple operating state points corresponding to the operating state data; Based on the Gaussian distribution map, abnormal state points are identified from the operating state points, and the movement of the target vehicle is detected to be abnormal based on the abnormal state points.

2. The method according to claim 1, characterized in that, The torque data and acceleration data are fitted to obtain a Gaussian distribution map, including: Multiple probability distribution units are constructed based on the torque data, the acceleration data, and the preset target influencing variables; The Gaussian distribution map is generated based on the plurality of probability distribution units.

3. The method according to claim 2, characterized in that, Based on the Gaussian distribution map, abnormal state points are identified from the operating state points, including: In the Gaussian distribution diagram, it is determined whether the running state point falls within the preset distribution range corresponding to any of the probability distribution units; If the running state point does not fall within the preset distribution range corresponding to any of the probability distribution units, the running state point is regarded as an abnormal state point.

4. The method according to claim 3, characterized in that, In response to the fact that the running state point does not fall within the preset distribution range corresponding to any of the probability distribution units, the running state point is regarded as an abnormal state point, including: Determine the probability distribution center corresponding to each probability distribution unit; Calculate the distance between the running state point and the probability distribution center corresponding to each probability distribution unit, and in response to the distance exceeding a preset distance standard deviation, designate the running state point as a candidate anomaly point; The torque value corresponding to the candidate abnormal point is determined. In response to the torque value corresponding to the candidate abnormal point exceeding the preset normal torque value, the candidate abnormal point is regarded as an abnormal state point.

5. The method according to claim 1, characterized in that, Detecting whether the target vehicle's movement is abnormal based on the abnormal state points includes: Count the number of abnormal state points and the number of running state points; Calculate the ratio between the number of abnormal state points and the number of operating state points to obtain the percentage of abnormal state points; In response to the proportion of abnormal state points exceeding a preset percentage threshold or the number of abnormal state points exceeding a preset quantity threshold, it is determined that the movement of the target vehicle is abnormal.

6. The method according to claim 1, characterized in that, After detecting whether the target vehicle's movement is abnormal based on the abnormal state points, the method further includes: In response to the detection of abnormal movement of the target vehicle, maintenance suggestion information is generated; The maintenance recommendation information is sent to the maintenance personnel so that the maintenance personnel can perform pre-maintenance on the target vehicle based on the maintenance recommendation information.

7. A device for detecting abnormal motion, characterized in that, include: The data acquisition module is used to collect the operating status data of the target vehicle within a preset time period; wherein, the operating status data includes torque data and acceleration data; The fitting module is used to perform fitting processing on the torque data and the acceleration data to obtain a Gaussian distribution map; wherein, the Gaussian distribution map includes multiple operating state points corresponding to the operating state data; The detection module is used to identify abnormal state points from the operating state points based on the Gaussian distribution map, and to detect whether the movement of the target vehicle is abnormal based on the abnormal state points.

8. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-6.

Images