Unmanned guided vehicle state parameter estimation system and method

Abstract

The present invention discloses a state parameter estimation system and method for an unmanned guided vehicle, including: a sampling module for sampling the operation data of each device in the unmanned guided vehicle to obtain an operation scene set; a division module for setting a time scale, dividing the operation scene set according to the time scale, and obtaining a first number of first operation scenes; a clustering module for performing scene clustering processing on the first number of first operation scenes to obtain a second number of second operation scenes; an estimation module for respectively performing state parameter estimation on the second number of second operation scenes to obtain state parameter estimation results; and when determining that the state parameter estimation result is an abnormal event, issuing an alarm prompt. The state parameters of the unmanned guided vehicle are accurately determined, and comprehensive monitoring of the unmanned guided vehicle is achieved based on the state parameters of the unmanned guided vehicle, the accuracy of monitoring the unmanned guided vehicle is improved, and the safe and reliable operation of the unmanned guided vehicle is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned guided vehicles, and in particular to a system and method for estimating state parameters of an unmanned guided vehicle. Background Art

[0002] The state parameters of unmanned guided vehicles are key state information during the operation of the vehicle, and are also an important basis for controlling the stable and safe operation of the vehicle. State estimation is an effective means for unmanned guided vehicles to obtain state information. For complex nonlinear systems such as unmanned guided vehicles, accurate and real-time estimation of the state parameters of unmanned guided vehicles is the premise for ensuring precise control of the vehicle. At present, most common unmanned guided vehicles use various signal collectors to collect various data, and only judge whether the operation of the unmanned guided vehicle is abnormal by comparing each data separately. However, based on the separate comparison of each data, the number of comparisons is large, and the state parameters of the unmanned guided vehicle cannot be accurately determined, which affects the judgment efficiency. At the same time, it will also cause the judgment process to be incomplete, which will lead to incorrect judgment of the operation of the unmanned guided vehicle. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems in the above-mentioned technologies to a certain extent. To this end, the first purpose of the present invention is to propose an unmanned guided vehicle state parameter estimation system, which reduces the amount of data, improves the monitoring efficiency, accurately determines the state parameters of the unmanned guided vehicle, realizes comprehensive monitoring of the unmanned guided vehicle based on the state parameters of the unmanned guided vehicle, improves the accuracy of monitoring the unmanned guided vehicle, and ensures the safe and reliable operation of the unmanned guided vehicle.

[0004] The second objective of the present invention is to provide an automated monitoring method for an unmanned guided vehicle.

[0005] To achieve the above object, a first embodiment of the present invention provides a state parameter estimation system for an unmanned guided vehicle, comprising:

[0006] A sampling module is used to sample the operating data of each device in the unmanned guided vehicle to obtain an operating scenario set;

[0007] A division module, used to set a time scale, divide the operation scenario set according to the time scale, and obtain a first number of first operation scenarios;

[0008] A clustering module, configured to perform scene clustering processing on a first number of first operation scenes to obtain a second number of second operation scenes; wherein the second number is smaller than the first number;

[0009] The estimation module is used to estimate the state parameters of the second number of second operation scenarios respectively to obtain the state parameter estimation results; when it is determined that the state parameter estimation results indicate that an abnormal event has occurred, an alarm prompt is issued.

[0010] According to some embodiments of the present invention, the sampling module includes:

[0011] Establish a module for obtaining device information of each device in the unmanned guided vehicle, arranging the devices from large to small according to the importance of each device included in the device information, and then establishing a queue for each device;

[0012] The data acquisition module is used to sample the operating data of each device in the unmanned guided vehicle based on the queue to obtain an operating scenario set.

[0013] According to some embodiments of the present invention, the sampling module further includes:

[0014] The preprocessing module is used to preprocess the sampled data, remove outlier data and abnormal data in the sampled data, and obtain valid data;

[0015] The generation module is used to generate a set of running scenarios based on valid data.

[0016] According to some embodiments of the present invention, the generating module comprises:

[0017] Aggregation module for:

[0018] Based on the functional characteristics of the unmanned guided vehicle components included in the valid data, the basic functional categories of the unmanned guided vehicle components are classified and aggregated to form the basic functional categories of the unmanned guided vehicle components as the functional layer; the basic functional categories include architecture, assets and configuration roles;

[0019] Based on the interaction features of the unmanned guided vehicle components included in the valid data, the basic association categories of the unmanned guided vehicle components are classified and aggregated to form the basic association categories of the unmanned guided vehicle components as the association layer; the basic association categories include the configuration role interaction relationship definition and the role behavior influence model of each component;

[0020] Based on the data features of the unmanned guided vehicle constituent elements included in the valid data, the basic logical categories of the unmanned guided vehicle constituent elements are classified and aggregated to form the basic logical categories as the logical layer; the basic logical categories include the interaction logic between the architecture, assets and configuration roles;

[0021] The combination module is used to combine based on the functional layer, the association layer and the logic layer to obtain a set of operating scenarios.

[0022] According to some embodiments of the present invention, the combined module comprises:

[0023] The first determination module is used to determine the leaf nodes of the operation element tree corresponding to the function layer, the association layer and the logic layer and the elements corresponding to the leaf nodes respectively;

[0024] The second determination module is used to determine all elements in the path where the element corresponding to each leaf node is located and combine them to obtain an operation scenario set.

[0025] According to some embodiments of the present invention, the clustering module includes:

[0026] An extraction module, configured to perform feature extraction on the first number of first operation scenarios respectively, and determine a feature vector for each operation scenario;

[0027] Compute module for:

[0028] Calculating similarities between all first running scenarios based on a similarity measurement method;

[0029] Based on the K-means clustering algorithm, clustering is performed according to the similarities between all the first operation scenarios to obtain a second number of second operation scenarios.

[0030] According to some embodiments of the present invention, the estimation module is also used to evaluate the continuity between two adjacent second operating scenarios, obtain an evaluation value, and determine whether it is less than a preset threshold; when it is determined that the evaluation value is less than the preset threshold, it indicates that there is missing data and a prompt message is issued.

[0031] According to some embodiments of the present invention, the algorithm for obtaining the evaluation value by the estimation module includes:

[0032] Calculate the evaluation value S of the coherence between the first second operation scenario and the second second operation scenario:

[0033]

[0034] Among them, α is the weight coefficient of the spatial layout of the first second operating scene and the second second operating scene for the evaluation value of coherence; H1 is the range of the spatial layout of the first second operating scene; H2 is the range of the spatial layout of the second second operating scene; W0 is the similarity of the operating parameters of the first second operating scene and the second second operating scene; W1 is the operating parameter of the first second operating scene; W2 is the operating parameter of the second second operating scene; β is the weight coefficient of the scene switching duration between the first second operating scene and the second second operating scene for the evaluation value of coherence, α+β=1; T1 is the actual scene switching duration between the first second operating scene and the second second operating scene; T0 is the standard scene switching duration between the first second operating scene and the second second operating scene.

[0035] According to some embodiments of the present invention, a detection module is further included, which is used to randomly select two second operation scenarios before the estimation module respectively estimates the state parameters of the second number of second operation scenarios, analyze the scene clustering results after clustering processing, and obtain an analysis value. When it is determined that the analysis value is less than a preset analysis threshold, it indicates that the scene clustering processing is unqualified and the scene clustering processing needs to be re-performed;

[0036] Calculate the analysis value D of the scene clustering result after clustering processing:

[0037]

[0038] Among them, H a is the number of sub-operation scenarios included in the a-th second operation scenario; Z is each sub-operation scenario in the a-th second operation scenario; c a is the average value of the distances between the sub-operation scenarios included in the a-th second operation scenario; H b is the number of sub-operation scenarios included in the b-th second operation scenario; V is each sub-operation scenario in the b-th second operation scenario; c b is the average value of the distances between the sub-operation scenarios included in the b-th second operation scenario; S(a, b) is the Euclidean distance between the cluster centers of the a-th second operation scenario and the b-th second operation scenario.

[0039] To achieve the above object, a second embodiment of the present invention provides an automated monitoring method for an unmanned guided vehicle, comprising:

[0040] Sampling the operating data of each device in the unmanned guided vehicle to obtain an operating scenario set;

[0041] Setting a time scale, dividing the operation scenario set according to the time scale, and obtaining a first number of first operation scenarios;

[0042] Performing scenario clustering processing on the first number of first operating scenarios to obtain a second number of second operating scenarios; wherein the second number is smaller than the first number;

[0043] State parameter estimation is performed on each of the second number of second operation scenarios to obtain state parameter estimation results; when it is determined that the state parameter estimation result indicates that an abnormal event has occurred, an alarm prompt is issued.

[0044] The present invention proposes a system and method for estimating state parameters of an unmanned guided vehicle. The system samples the operation data of each device in the unmanned guided vehicle, constructs an operation scenario set, segments and clusters the operation scenario set, realizes integrated processing of the operation scenarios, and makes judgments based on the operation scenarios of the unmanned guided vehicle, thereby reducing the amount of data, improving monitoring efficiency, accurately determining the state parameters of the unmanned guided vehicle, realizing comprehensive monitoring of the unmanned guided vehicle based on the state parameters of the unmanned guided vehicle, improving the accuracy of monitoring the unmanned guided vehicle, and ensuring the safe and reliable operation of the unmanned guided vehicle.

[0045] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0046] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0048] Figure 1 is a block diagram of a state parameter estimation system for an unmanned guided vehicle according to an embodiment of the present invention;

[0049] Figure 2 is a block diagram of a sampling module according to one embodiment of the present invention;

[0050] Figure 3 is a flow chart of a method for estimating state parameters of an automated guided vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0052] like Figure 1 As shown, the first embodiment of the present invention proposes a state parameter estimation system for an unmanned guided vehicle, comprising:

[0053] A sampling module is used to sample the operating data of each device in the unmanned guided vehicle to obtain an operating scenario set;

[0054] A division module, used to set a time scale, divide the operation scenario set according to the time scale, and obtain a first number of first operation scenarios;

[0055] A clustering module, configured to perform scene clustering processing on a first number of first operation scenes to obtain a second number of second operation scenes; wherein the second number is smaller than the first number;

[0056] The estimation module is used to estimate the state parameters of the second number of second operation scenarios respectively to obtain the state parameter estimation results; when it is determined that the state parameter estimation results indicate that an abnormal event has occurred, an alarm prompt is issued.

[0057] Working principle of the above technical solution: In this embodiment, the components in the unmanned guided vehicle include motors, sensors (such as infrared sensors, magnetic sensors, laser sensors, etc.), controllers (such as control center modules), batteries, etc. Determine the parameters that need to be sampled. Including voltage, current, active power, reactive power, frequency, temperature, humidity, vibration. Determine the sampling frequency, that is, the number of samples per second or per minute. This depends on the dynamic characteristics of the system and the required data resolution. Sensors and measuring equipment are installed at key locations of each component of the unmanned guided vehicle for real-time collection of operating data. Ensure the accuracy and reliability of the sampling equipment, and calibrate and maintain it regularly. Collect the operating data of each component in real time based on a data acquisition system (such as a SCADA system). The data contains a timestamp, sampling point location, and corresponding parameter values.

[0058] In this embodiment, the time scale includes seconds, minutes, hours, or even longer time scales. The corresponding time scale is selected according to the monitoring requirements. For example, 5 seconds can be selected. The operation scene set is segmented according to the time scale into each first operation scene. The division of each operation scene is realized to facilitate subsequent processing.

[0059] In this embodiment, scene clustering processing is performed on each independent first operating scenario to obtain a second number of second operating scenarios. In order to improve data processing efficiency, scene clustering processing is performed to classify some of the same type of first operating scenarios to obtain a second number of second operating scenarios.

[0060] In this embodiment, the second operating scenario is identified based on a pre-trained scene recognition model. Automatically identify and classify different operating scenarios, reduce the workload of manual identification, and improve the accuracy and efficiency of identification. The method for obtaining a scene recognition model includes: determining sample scenario data; training a deep learning model, such as a convolutional neural network (CNN), through sample scenario data so that the model can learn the characteristics and differences between various operating scenarios, and output state parameter estimation results. The state parameter estimation results include the operating status of each device, the connection status of each device, the coordinated working status, etc. When it is determined that the state parameter estimation result is an abnormal event, an alarm prompt is issued. Abnormal events include abnormal operating status of a device, abnormal connection status of adjacent devices, abnormal coordinated working status, etc.

[0061] The beneficial effects of the above technical solution are as follows: based on sampling the operating data of each device in the unmanned guided vehicle, an operating scenario set is constructed, the operating scenario set is segmented and clustered, and integrated processing of the operating scenarios is realized; based on judging the operating scenarios of the unmanned guided vehicle, the amount of data is reduced, the monitoring efficiency is improved, the state parameters of the unmanned guided vehicle are accurately determined, and comprehensive monitoring of the unmanned guided vehicle is achieved based on the state parameters of the unmanned guided vehicle, thereby improving the accuracy of monitoring the unmanned guided vehicle and ensuring the safe and reliable operation of the unmanned guided vehicle.

[0062] like Figure 2 As shown, according to some embodiments of the present invention, the sampling module includes:

[0063] Establish a module for obtaining device information of each device in the unmanned guided vehicle, arranging the devices from large to small according to the importance of each device included in the device information, and then establishing a queue for each device;

[0064] The data acquisition module is used to sample the operating data of each device in the unmanned guided vehicle based on the queue to obtain an operating scenario set.

[0065] The working principle and beneficial effects of the above technical solution are as follows: the importance of each component is marked, and the importance of each component included in the equipment information is arranged from large to small, and then a queue for each component is established to realize the division of monitoring priority of each component in the unmanned guided vehicle, and the operation data of each component in the unmanned guided vehicle is sampled based on the queue to obtain an operation scenario set.

[0066] According to some embodiments of the present invention, the sampling module further includes:

[0067] The preprocessing module is used to preprocess the sampled data, remove outlier data and abnormal data in the sampled data, and obtain valid data;

[0068] The generation module is used to generate a set of running scenarios based on valid data.

[0069] The working principle and beneficial effects of the above technical solution are as follows: preprocessing the sampled data to remove outlier data and abnormal data in the sampled data to obtain valid data; avoiding the influence of outlier data and abnormal data on the accuracy of the generated operation scenario set, thereby improving the accuracy of the data.

[0070] According to some embodiments of the present invention, the generating module comprises:

[0071] Aggregation module for:

[0072] Based on the functional characteristics of the unmanned guided vehicle components included in the valid data, the basic functional categories of the unmanned guided vehicle components are classified and aggregated to form the basic functional categories of the unmanned guided vehicle components as the functional layer; the basic functional categories include architecture, assets and configuration roles;

[0073] Based on the interaction features of the unmanned guided vehicle components included in the valid data, the basic association categories of the unmanned guided vehicle components are classified and aggregated to form the basic association categories of the unmanned guided vehicle components as the association layer; the basic association categories include the configuration role interaction relationship definition and the role behavior influence model of each component;

[0074] Based on the data features of the unmanned guided vehicle constituent elements included in the valid data, the data are classified and aggregated to form basic logical categories of the unmanned guided vehicle constituent elements as a logical layer; the basic logical categories include the interaction logic between the architecture, assets and configuration roles;

[0075] The combination module is used to combine based on the functional layer, the association layer and the logic layer to obtain a set of operating scenarios.

[0076] The working principle of the above technical solution: Architecture: refers to the overall structure and layout of the unmanned guided vehicle system, such as power distribution, network topology, etc. Assets: refers to the various devices and resources in the unmanned guided vehicle system. Configuration role: refers to the role or function played by different devices or components in a specific unmanned guided vehicle system, such as main power supply, backup power supply, etc. The functional layer mainly describes the functional characteristics of different components in the unmanned guided vehicle. These functional characteristics determine how the equipment in the unmanned guided vehicle responds to specific operating instructions and realizes the expected functions. For example, some components may be responsible for energy conversion and transmission, while others may be responsible for control and protection to ensure the stable operation of the entire unmanned guided vehicle. The functional layer is the core of the unmanned guided vehicle, which contains the basic functions and performance indicators of all components in the unmanned guided vehicle. The information of the functional layer usually includes but is not limited to the rated voltage, current, power, efficiency and other parameters of the device.

[0077] Configuration role interaction relationship definition: clarifies the connection mode and mutual influence between different configuration roles, such as the connection between power supply and load, the action of motor in case of failure, etc. Role behavior impact model: describes the impact of the behavior of a certain configuration role on other roles or the entire system, such as the impact of a certain device failure on other devices or the system as a whole. The association layer mainly describes the interaction and dependency between different components in the unmanned guided vehicle. These interaction relationships define how the elements work together to form a complete unmanned guided vehicle. For example, the correct association needs to be established between sensors and actuators to ensure that the system's automated control can accurately respond to environmental changes. The association layer reflects the relationship between devices in the unmanned guided vehicle and the way they work together. This includes the physical connection between devices, interface protocols, communication mechanisms, etc., to ensure the accurate transmission and processing of information between different devices.

[0078] Interaction logic: refers to the logical relationship between different elements in the unmanned guided vehicle system based on their functions and interaction characteristics, such as power balance between power supply and load, response logic of each device in case of failure, etc. The logic layer, also known as the interaction logic layer, mainly describes the data logic relationship between different components in the unmanned guided vehicle. These logical relationships enable the vehicle controller to process the input information and make decisions according to preset rules or algorithms. The logic layer involves the software part of the unmanned guided vehicle, including the description of programs, algorithms and logical rules. This layer of information enables the unmanned guided vehicle to implement complex control logic, such as temperature control, energy optimization, etc. These logics are usually implemented through programming languages ​​and may involve advanced technologies such as real-time operating systems and database management systems.

[0079] The role of the combined module is to combine the elements of these layers on the basis of the functional layer, the association layer and the logic layer to form a complete unmanned guided vehicle system. This combination process involves modularization, standardization and interoperability, ensuring the flexibility and scalability of the unmanned guided vehicle control system. The combined module is the integrated part of the unmanned guided vehicle system, which ensures that different functional modules can be correctly assembled into a complete system. This involves multiple aspects such as interface matching, communication coordination and data integration between modules. The design of the combined module usually needs to follow a certain architectural style and interoperability principles to ensure the integrity and reliability of the system.

[0080] Beneficial effects of the above technical solution: The basic functional categories, basic association categories and basic logic categories of the constituent elements of the unmanned guided vehicle constitute the core framework of the unmanned guided vehicle. These categories provide a solid foundation for the design and implementation of the unmanned guided vehicle control system through detailed analysis and definition of the functional layer, association layer and logic layer. The combination module is the key part of combining these basic elements into a complete system. Through such layering and modularization, an efficient, reliable and maintainable unmanned guided vehicle control system can be realized, which facilitates the accurate determination and description of the operating scenario set.

[0081] According to some embodiments of the present invention, the combined module comprises:

[0082] The first determination module is used to determine the leaf nodes of the operation element tree corresponding to the function layer, the association layer and the logic layer and the elements corresponding to the leaf nodes respectively;

[0083] The second determination module is used to determine all elements in the path where the element corresponding to each leaf node is located and combine them to obtain an operation scenario set.

[0084] The working principle and beneficial effects of the above technical solution: Constructing the operation element tree: First, construct corresponding operation element trees for the functional layer, association layer and logic layer respectively. These tree structures reflect the hierarchical relationship and dependency relationship between the components of the unmanned guided vehicle. Determine the leaf node: In the constructed operation element tree, determine the leaf node of each tree. Leaf nodes usually represent the most specific and indivisible components or functions of the unmanned guided vehicle. Determine the elements corresponding to the leaf node: For each leaf node, determine the specific elements of the corresponding unmanned guided vehicle components or functions. These elements are specific parameters, configuration information, interaction rules, etc. of the equipment. Determine the path where the elements are located: Starting from each leaf node, trace back to the root node (or a specified upper node), and determine the path where the elements corresponding to the leaf node are located. This path represents the hierarchical relationship of the unmanned guided vehicle components from the whole to the part, from the abstract to the specific. Extract all elements in the path: Extract the elements of all the nodes passed along the determined path. These elements together constitute a complete description of the unmanned guided vehicle components or functions corresponding to the leaf node. Combine to form an operation scenario: Combine all the extracted elements to form a complete operation scenario. This scenario describes the operation status and behavior mode of the unmanned guided vehicle system under certain conditions. Construct an operation scenario set: Repeat the above process, operate each leaf node in the functional layer, association layer, and logic layer, and finally construct an operation scenario set containing multiple operation scenarios. This scenario set covers the main operation status and behavior mode of the unmanned guided vehicle system under different conditions, and obtains an accurate and comprehensive operation scenario set.

[0085] According to some embodiments of the present invention, the clustering module includes:

[0086] An extraction module, configured to perform feature extraction on the first number of first operation scenarios respectively, and determine a feature vector for each operation scenario;

[0087] Compute module for:

[0088] Calculating similarities between all first running scenarios based on a similarity measurement method;

[0089] Based on the K-means clustering algorithm, clustering is performed according to the similarities between all the first operation scenarios to obtain a second number of second operation scenarios.

[0090] The working principle of the above technical solution: In this embodiment, the first number of first operation scenarios are respectively subjected to feature extraction to determine the feature vector of each operation scenario, including: extracting the features of the operation scenario based on the feature extraction algorithm, the features including electrical parameters (such as voltage, current, power, etc.), device status (such as open, close, fault, etc.), configuration information, time information, etc. Quantify each feature into digital variables by feature encoding, and then combine these digital variables into a feature vector. In this way, each operation scenario can be abstracted into a feature form that can be calculated and compared. For example, if a certain operation scenario involves the interaction of equipment, sensors and controllers in an unmanned guided vehicle control system, the features of these elements (such as model, status, output value, etc.) can be extracted and formed into a feature description. For this operation scenario, a feature vector can be established, which contains the attribute information of each element (such as device type, sensor reading, controller status, etc.), so that this operation scenario can be quantitatively described by a set of numbers. Different operation scenarios in the unmanned guided vehicle system are abstracted into feature vectors, so as to facilitate further analysis, comparison and processing.

[0091] In this embodiment, a similarity measurement method such as Euclidean distance is used to calculate the similarity between all first operation scenarios. The similarity is calculated for the feature vectors between each pair of first operation scenarios to obtain a similarity matrix. This matrix records the similarity values ​​between each operation scenario and can be used as the input of the subsequent clustering algorithm. Select the K value, which represents the number of clusters we want to obtain, that is, the number of second operation scenarios. The selection of this value should be based on the actual application requirements and the understanding of the operation scenarios of the elements of the unmanned guided vehicle. Initialize the cluster center and randomly select K from the feature vectors of the first operation scenario as the initial cluster center. Assign the scene to the cluster. For each first operation scenario, calculate the similarity (or distance) between its feature vector and all cluster centers, and assign it to the cluster with the highest similarity (or the smallest distance). Update the cluster center, recalculate the average value (or other summary statistics) of the feature vectors of all members of each cluster, and use this average value as the new cluster center. Iterate and converge, repeat the above steps until a certain stop condition is met (such as the change in the cluster center is less than a certain threshold, or the maximum number of iterations is reached). After K-means clustering, a second number of second operation scenarios (i.e., clusters) are obtained, each cluster representing a class of similar operation scenarios of the components of the unmanned guided vehicle. These clusters can be used for further analysis, identification, or prediction.

[0092] The beneficial effect of the above technical solution is: to achieve data integration and classification of a first number of first operating scenarios to obtain a second number of second operating scenarios.

[0093] According to some embodiments of the present invention, the estimation module is also used to evaluate the continuity between two adjacent second operating scenarios, obtain an evaluation value, and determine whether it is less than a preset threshold; when it is determined that the evaluation value is less than the preset threshold, it indicates that there is missing data and a prompt message is issued.

[0094] The working principle and beneficial effects of the above technical solution: After acquiring the data, the estimation module evaluates the continuity between the two adjacent second operating scenarios to determine whether the collected data is comprehensive, thereby avoiding the problem of incomplete data collection. When it is determined that the evaluation value is less than the preset threshold, it indicates that there is missing data, and a prompt message is issued to facilitate the handling of abnormal situations, such as checking the collection equipment, determining whether data transmission is lost, etc., to facilitate timely processing, ensure the comprehensiveness of the collected data, and facilitate comprehensive monitoring of the unmanned guided vehicle.

[0095] According to some embodiments of the present invention, the algorithm for obtaining the evaluation value by the estimation module includes:

[0096] Calculate the evaluation value S of the coherence between the first second operation scenario and the second second operation scenario:

[0097]

[0098] Among them, α is the weight coefficient of the spatial layout of the first second operating scene and the second second operating scene for the evaluation value of coherence, ranging from 0 to 1; H1 is the range of the spatial layout of the first second operating scene; H2 is the range of the spatial layout of the second second operating scene; W0 is the similarity of the operating parameters of the first second operating scene and the second second operating scene; W1 is the operating parameter of the first second operating scene; W2 is the operating parameter of the second second operating scene; β is the weight coefficient of the scene switching duration between the first second operating scene and the second second operating scene for the evaluation value of coherence, ranging from 0 to 1, α+β=1; T1 is the actual scene switching duration between the first second operating scene and the second second operating scene; T0 is the standard scene switching duration between the first second operating scene and the second second operating scene.

[0099] The working principle and beneficial effects of the above technical solution: the first second operation scene is the front second operation scene, and the second second operation scene is the back second operation scene. Based on the weight coefficient of the spatial layout of the first second operation scene and the second second operation scene for the evaluation value of coherence, the scene switching duration of the first second operation scene and the second second operation scene for the evaluation value of coherence and other parameters, the coherence is judged from the spatial layout level and the time level, which is convenient for accurately determining the evaluation value of the coherence of the first second operation scene and the second second operation scene, thereby improving the accuracy of judging the evaluation value and the preset threshold value, thereby facilitating the accurate determination of whether there is a situation of missing data.

[0100] According to some embodiments of the present invention, a detection module is further included, which is used to randomly select two second operation scenarios before the estimation module respectively estimates the state parameters of the second number of second operation scenarios, analyze the scene clustering results after clustering processing, and obtain an analysis value. When it is determined that the analysis value is less than a preset analysis threshold, it indicates that the scene clustering processing is unqualified and the scene clustering processing needs to be re-performed;

[0101] Calculate the analysis value D of the scene clustering result after clustering processing:

[0102]

[0103] Among them, H a is the number of sub-operation scenarios included in the a-th second operation scenario; Z is each sub-operation scenario in the a-th second operation scenario; c a is the average value of the distances between the sub-operation scenarios included in the a-th second operation scenario; H bis the number of sub-operation scenarios included in the b-th second operation scenario; V is each sub-operation scenario in the b-th second operation scenario; c b is the average value of the distances between the sub-operation scenarios included in the b-th second operation scenario; S(a, b) is the Euclidean distance between the cluster centers of the a-th second operation scenario and the b-th second operation scenario.

[0104] The working principle and beneficial effects of the above technical solution: Based on the detection module, before the estimation module estimates the state parameters of the second number of second operation scenes respectively, two second operation scenes are randomly selected, and the scene clustering results after clustering processing are analyzed to obtain an analysis value as an estimate for judging the classification effect. When it is determined that the analysis value is less than the preset analysis threshold, it means that the scene clustering processing is unqualified and the scene clustering processing needs to be re-performed, thereby facilitating the improvement of the accuracy of data classification integration and obtaining an accurate second operation scene. The distance between sub-operation scenes is the Euclidean distance. Based on comparing the sub-operation scenes in any two second operation scenes, the average value of the distance between the sub-operation scenes, the Euclidean distance between the cluster centers, etc., it is convenient to accurately calculate the analysis value of the scene clustering result after clustering processing, thereby improving the accuracy of judging the size of the analysis value and the preset analysis threshold, thereby improving the accuracy of judging whether the scene clustering processing is qualified.

[0105] like Figure 3 As shown, the second embodiment of the present invention proposes an automated monitoring method for an unmanned guided vehicle, comprising steps S1-S4:

[0106] S1. Sampling the operation data of each device in the unmanned guided vehicle to obtain an operation scenario set;

[0107] S2. Setting a time scale, dividing the operation scenario set according to the time scale, and obtaining a first number of first operation scenarios;

[0108] S3, performing scenario clustering processing on the first number of first operation scenarios to obtain a second number of second operation scenarios; wherein the second number is smaller than the first number;

[0109] S4. Estimating the state parameters of the second number of second operation scenarios respectively to obtain state parameter estimation results; and issuing an alarm prompt when it is determined that the state parameter estimation result indicates that an abnormal event has occurred.

[0110] The beneficial effects of the above technical solution are as follows: based on sampling the operating data of each device in the unmanned guided vehicle, an operating scenario set is constructed, the operating scenario set is segmented and clustered, and integrated processing of the operating scenarios is realized; based on judging the operating scenarios of the unmanned guided vehicle, the amount of data is reduced, the monitoring efficiency is improved, the state parameters of the unmanned guided vehicle are accurately determined, and comprehensive monitoring of the unmanned guided vehicle is achieved based on the state parameters of the unmanned guided vehicle, thereby improving the accuracy of monitoring the unmanned guided vehicle and ensuring the safe and reliable operation of the unmanned guided vehicle.

[0111] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An unmanned guided vehicle state parameter estimation system, characterized in that: include: A sampling module is used to sample the operating data of each device in the unmanned guided vehicle to obtain an operating scenario set; A division module, used to set a time scale, divide the operation scenario set according to the time scale, and obtain a first number of first operation scenarios; A clustering module, configured to perform scene clustering processing on a first number of first operation scenes to obtain a second number of second operation scenes; wherein the second number is smaller than the first number; an estimation module, configured to respectively estimate the state parameters of the second number of second operation scenarios to obtain state parameter estimation results; and to issue an alarm prompt when it is determined that the state parameter estimation result indicates that an abnormal event has occurred; It also includes a detection module, which is used to randomly select two second operation scenes before the estimation module estimates the state parameters of the second number of second operation scenes respectively, analyze the scene clustering results after clustering processing, and obtain an analysis value. When it is determined that the analysis value is less than a preset analysis threshold, it indicates that the scene clustering processing is unqualified and the scene clustering processing needs to be re-performed; Calculate the analysis value D of the scene clustering result after clustering processing: Among them, H a is the number of sub-operation scenarios included in the a-th second operation scenario; Z is each sub-operation scenario in the a-th second operation scenario; c a is the average value of the distances between the sub-operation scenarios included in the a-th second operation scenario; H b is the number of sub-operation scenarios included in the b-th second operation scenario; V is each sub-operation scenario in the b-th second operation scenario; c b is the average value of the distances between the sub-operation scenarios included in the b-th second operation scenario; S(a, b) is the Euclidean distance between the cluster centers of the a-th second operation scenario and the b-th second operation scenario.

2. The state parameter estimation system of the unmanned guided vehicle according to claim 1, characterized in that: The sampling module comprises: Establish a module for obtaining device information of each device in the unmanned guided vehicle, arranging the devices from large to small according to the importance of each device included in the device information, and then establishing a queue for each device; The data acquisition module is used to sample the operating data of each device in the unmanned guided vehicle based on the queue to obtain an operating scenario set.

3. The state parameter estimation system of the unmanned guided vehicle according to claim 2, characterized in that: The sampling module further includes: The preprocessing module is used to preprocess the sampled data, remove outlier data and abnormal data in the sampled data, and obtain valid data; The generation module is used to generate a set of running scenarios based on valid data.

4. The state parameter estimation system of the unmanned guided vehicle according to claim 3, characterized in that: The generating module comprises: Aggregation module for: Based on the functional characteristics of the unmanned guided vehicle components included in the valid data, the basic functional categories of the unmanned guided vehicle components are classified and aggregated to form the basic functional categories of the unmanned guided vehicle components as the functional layer; the basic functional categories include architecture, assets and configuration roles; Based on the interaction features of the unmanned guided vehicle components included in the valid data, the basic association categories of the unmanned guided vehicle components are classified and aggregated to form the basic association categories of the unmanned guided vehicle components as the association layer; the basic association categories include the configuration role interaction relationship definition and the role behavior influence model of each component; Based on the data features of the unmanned guided vehicle constituent elements included in the valid data, the data are classified and aggregated to form basic logical categories of the unmanned guided vehicle constituent elements as a logical layer; the basic logical categories include the interaction logic between the architecture, assets and configuration roles; The combination module is used to combine based on the functional layer, the association layer and the logic layer to obtain a set of operating scenarios.

5. The state parameter estimation system of the unmanned guided vehicle according to claim 4, characterized in that: The combined module comprises: The first determination module is used to determine the leaf nodes of the operation element tree corresponding to the function layer, the association layer and the logic layer and the elements corresponding to the leaf nodes respectively; The second determination module is used to determine all elements in the path where the element corresponding to each leaf node is located and combine them to obtain an operation scenario set.

6. The state parameter estimation system of the unmanned guided vehicle according to claim 1, characterized in that: The clustering module comprises: An extraction module, configured to perform feature extraction on the first number of first operation scenarios respectively, and determine a feature vector for each operation scenario; Compute module for: Calculating similarities between all first running scenarios based on a similarity measurement method; Based on the K-means clustering algorithm, clustering is performed according to the similarities between all the first operation scenarios to obtain a second number of second operation scenarios.

7. The state parameter estimation system of the automated guided vehicle according to claim 1, characterized in that: The estimation module is also used to evaluate the coherence between two adjacent second operating scenarios, obtain an evaluation value, and determine whether it is less than a preset threshold; when it is determined that the evaluation value is less than the preset threshold, it indicates that there is missing data and a prompt message is issued.

8. The state parameter estimation system of the automated guided vehicle according to claim 7, characterized in that: The algorithm for obtaining the evaluation value by the estimation module includes: Calculate the evaluation value S of the coherence between the first second operation scenario and the second second operation scenario: Among them, α is the weight coefficient of the spatial layout of the first second operating scene and the second second operating scene for the evaluation value of coherence; H1 is the range of the spatial layout of the first second operating scene; H2 is the range of the spatial layout of the second second operating scene; W0 is the similarity of the operating parameters of the first second operating scene and the second second operating scene; W1 is the operating parameter of the first second operating scene; W2 is the operating parameter of the second second operating scene; β is the weight coefficient of the scene switching duration between the first second operating scene and the second second operating scene for the evaluation value of coherence, α+β=1; T1 is the actual scene switching duration between the first second operating scene and the second second operating scene; T0 is the standard scene switching duration between the first second operating scene and the second second operating scene.

9. An automated monitoring method for an unmanned guided vehicle, characterized in that: include: Sampling the operating data of each device in the unmanned guided vehicle to obtain an operating scenario set; Setting a time scale, dividing the operation scenario set according to the time scale, and obtaining a first number of first operation scenarios; Performing scenario clustering processing on the first number of first operating scenarios to obtain a second number of second operating scenarios; wherein the second number is smaller than the first number; Performing state parameter estimation on the second number of second operation scenarios respectively to obtain state parameter estimation results; and issuing an alarm prompt when it is determined that the state parameter estimation result indicates that an abnormal event has occurred; The method further includes: before respectively estimating the state parameters of the second number of second operation scenarios, randomly selecting two second operation scenarios, analyzing the scene clustering results after the clustering process to obtain an analysis value, and when it is determined that the analysis value is less than a preset analysis threshold, indicating that the scene clustering process is unqualified and needs to be re-performed; Calculate the analysis value D of the scene clustering result after clustering processing: Among them, H a is the number of sub-operation scenarios included in the a-th second operation scenario; Z is each sub-operation scenario in the a-th second operation scenario; c a is the average value of the distances between the sub-operation scenarios included in the a-th second operation scenario; H b is the number of sub-operation scenarios included in the b-th second operation scenario; V is each sub-operation scenario in the b-th second operation scenario; c b is the average value of the distances between the sub-operation scenarios included in the b-th second operation scenario; S(a, b) is the Euclidean distance between the cluster centers of the a-th second operation scenario and the b-th second operation scenario.

Images