AI-based unattended weighing-in system for factory exit and entry whole-process identification method

Abstract

The present application relates to the technical field of data analysis, and particularly relates to an AI-based unattended weighing-in and weighing-out factory full-process identification method, which comprises the following steps: acquiring the positions of pressure sensors in a weighbridge and the pressure amplitudes of the pressure sensors at each time, and acquiring the point cloud data of the weighbridge and the vehicle at each time, so as to acquire a plurality of support points at each time; acquiring the support force of each support point at each time in combination with the positions of the pressure sensors in the weighbridge and the pressure amplitudes of the pressure sensors at each time; acquiring the gravity center coordinates of the vehicle at the last time in combination with the positions of each support point and the support force at the last time and the pressure amplitudes of the pressure sensors; and acquiring and recording the real weight of the vehicle in combination with the position of the center of the weighbridge and the pressure amplitudes of all the pressure sensors at the last time. The present application acquires the real weight of the vehicle by analyzing the point cloud data and the pressure data in the process of the vehicle driving into the weighbridge.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to an AI-based method for identifying the entire process of an unattended weighing system for entering and leaving a factory. Background Technology

[0002] In factory logistics and trade management, reliable weighing of vehicles entering and leaving the factory is the foundation for material accounting, freight settlement and internal control. Therefore, it is necessary to accurately collect and record the true weight of vehicles when they enter and leave the factory. However, in modern unattended automatic weighing scenarios, the vehicle parking position is entirely controlled by the driver. When the vehicle deviates from its parking position or the load distribution is uneven (for example, when the vehicle is parked at the edge of the weighbridge), the vehicle's center of gravity will deviate from the mechanical center of the weighbridge, generating a significant tilting moment. This causes the weighbridge platform to tilt, putting the pressure sensors in a non-ideal stress state. In other words, the weighing result of the weighbridge cannot accurately represent the true weight of the vehicle. Summary of the Invention

[0003] This invention provides an AI-based unattended weighing system for full-process identification of vehicles entering and leaving the factory, in order to solve the existing problem that the weighing results of the weighbridge cannot accurately represent the true weight of the vehicle.

[0004] The AI-based unattended weighing system for full-process identification of inbound and outbound operations in this invention adopts the following technical solution: Includes the following steps: Acquire point cloud data of the weighbridge, point cloud data of vehicles at each time moment, the location of pressure sensors in the weighbridge, and the pressure amplitude of each pressure sensor at each time moment; Based on the point cloud data of the weighbridge and the point cloud data of the vehicle at each time moment, several support points are obtained at each time moment. Combined with the position of the pressure sensor in the weighbridge, the ideal distribution coefficient of each pressure sensor to each support point at each time moment is obtained. Combined with the pressure amplitude of the pressure sensor at each time moment, the support force evaluation model of each support point at each time moment is constructed. Based on the support force evaluation model of each support point at each time moment, the support force of each support point at each time moment is obtained. Based on the position of each support point at the last moment, obtain the coordinates of each support point at the last moment. Combine the supporting force of each support point at the last moment and the pressure amplitude of each pressure sensor at the last moment to construct a vehicle center of gravity coordinate fitting model. Based on the vehicle center of gravity coordinate fitting model, obtain the vehicle's center of gravity coordinates at the last moment. Combine the position of the weighbridge center to obtain the vehicle's eccentricity at the last moment. Combine the pressure amplitude of all pressure sensors at the last moment to obtain the weighbridge's tilt at the last moment. The vehicle's actual weight is obtained and recorded based on the tilt of the weighbridge at the last moment and the pressure amplitude of all pressure sensors at the last moment.

[0005] Preferably, the method for obtaining several support points at each time moment based on the point cloud data of the weighbridge and the point cloud data of the vehicle at each time moment, and combining the position of the pressure sensor in the weighbridge to obtain the ideal distribution coefficient of each pressure sensor to each support point at each time moment, includes the following specific methods: For the At the moment, the first The point cloud data that comes into contact with the vehicle's point cloud data in the point cloud data of the weighbridge at time point n is denoted as the i-th point. The support point cloud data at the current moment will be the first time. The region formed by adjacent support point cloud data at the nth time point is used as the nth time point. The support area at the current moment will be the first The center of the support region at time t is denoted as the th moment. Support points at any given moment; For the The first moment at the moment The pressure sensor and the first The first support point will be the first The first moment at the moment The pressure sensor and the first The reciprocal of the distance between the support points, compared to the previous one All pressure sensors at time n and the first The ratio obtained by summing the reciprocals of the distances between the support points is used as the th... At the moment of the first The pressure sensor is for the first Ideal coefficient for each support point.

[0006] Preferably, the specific method for constructing the support force evaluation model for each support point at each time point is as follows: In the formula, This represents the output of the support force assessment model for each support point at each time point; Indicates the number of moments; Indicates the number of pressure sensors; Indicates the first The number of support points at any given moment; Indicates the first At the moment of the first The pressure sensor is for the first Ideal coefficient for allocation of each support point; Indicates the first At the moment of the first The pressure amplitude of each pressure sensor; Indicates the first At the moment of the first The supporting force of each support point; Indicates the first At the moment of the first The supporting force of each support point.

[0007] Preferably, the specific method for obtaining the support force of each support point at each time point based on the support force evaluation model at each time point includes: By iterating through the support force evaluation model of each support point at each time point, when the output result of the support force evaluation model of each support point at each time point reaches the minimum, the support force of each support point at each time point traversed in the support force evaluation model of each support point at each time point is taken as the support force of each support point at each time point, thus obtaining the support force of each support point at each time point.

[0008] Preferably, the specific method for obtaining the coordinates of each support point at the last moment based on the position of each support point at the last moment is as follows: With the direction in which the vehicle enters the weighbridge as the positive half-axis of the longitudinal axis, and the direction in the weighbridge plane that is clockwise and perpendicular to the positive half-axis of the longitudinal axis as the positive half-axis of the transverse axis, a rectangular coordinate system is constructed with spatial distance as the unit of measurement and the center of the weighbridge as the origin, denoted as the weighbridge coordinate system. Based on the position of each support point in the weighbridge at the last moment, obtain the coordinates of each support point at the last moment.

[0009] Preferably, the specific method for constructing the vehicle center of gravity coordinate fitting model is as follows: In the formula, This represents the output of the vehicle's center of gravity coordinate fitting model. This indicates the number of support points at the last moment; Indicates the last moment of the next The supporting force of each support point; Indicates the last moment of the next The x-coordinate of each support point; Indicates the last moment of the next The ordinate of each support point; The x-coordinate representing the vehicle's center of gravity; The vertical coordinate representing the vehicle's center of gravity; Indicates the vehicle's center of gravity shifts to the last moment. The vector of each support point; This represents the modulo function.

[0010] Preferably, the specific method for obtaining the vehicle's center of gravity coordinates at the last moment based on the vehicle's center of gravity coordinate fitting model is as follows: By iterating through the horizontal and vertical coordinates of the vehicle's center of gravity in the vehicle center of gravity coordinate fitting model, when the output of the vehicle center of gravity coordinate fitting model reaches its minimum, the horizontal and vertical coordinates of the vehicle's center of gravity traversed in the vehicle center of gravity coordinate fitting model are used as the horizontal and vertical coordinates of the vehicle's center of gravity at the last moment, thus obtaining the vehicle's center of gravity coordinates at the last moment.

[0011] Preferably, the specific method for obtaining the vehicle's eccentricity at the last moment is as follows: In the weighbridge coordinate system, draw a straight line from the vehicle's center of gravity coordinate at the last moment to the center of the weighbridge. The line segment of the line within the weighbridge is denoted as the tilt axis. The ratio of the distance between the vehicle's center of gravity coordinate at the last moment and the center of the weighbridge to the length of the tilt axis is taken as the degree of vehicle eccentricity at the last moment.

[0012] Preferably, the specific method for obtaining the tilt degree of the weighbridge at the last moment is as follows: The sum of the pressure amplitudes of all pressure sensors at the last moment is multiplied by the vehicle's eccentricity at the last moment, and the product is normalized. The normalized result is used as the tilt degree of the weighbridge at the last moment.

[0013] Preferably, the method for obtaining and recording the vehicle's true weight based on the tilt of the weighbridge at the last moment and the pressure amplitude of all pressure sensors at the last moment includes: A preset tilt correction angle is used. The product of the tilt degree of the weighbridge at the last moment and the tilt correction angle is used as the tilt angle of the weighbridge at the last moment. The sum of the pressure amplitudes of all pressure sensors at the last moment is divided by the cosine of the tilt angle of the weighbridge at the last moment. This ratio is taken as the vehicle's true weight, and the true weight of the vehicle entering and leaving the factory is recorded.

[0014] The beneficial effects of the technical solution of this invention are as follows: This application obtains the position of the pressure sensor in the weighbridge and its pressure amplitude at each time, and obtains point cloud data of the weighbridge and the vehicle at each time, thereby obtaining several support points at each time. During the process of the vehicle entering the weighbridge, the pressure amplitude collected by the pressure sensor in the weighbridge is transmitted from the vehicle's weight to the pressure sensor through the contact part between the vehicle's tires and the weighbridge. Therefore, by analyzing the spatial positional relationship between each pressure sensor and each support point, the pressure amplitude collected by the pressure sensor is ideally distributed, thereby establishing the correspondence between the amplitude of the pressure sensor and the support force of the support point, and obtaining the ideal distribution coefficient of each pressure sensor to each support point at each time. Furthermore, combined with the pressure amplitude of the pressure sensor at the corresponding time, a support force evaluation model of each support point at each time is constructed, thereby obtaining the support force of each support point at each time, providing a theoretical basis for subsequent analysis of the distribution characteristics of support points and the tilt of the weighbridge.

[0015] When the vehicle is in a balanced state, according to the principle of rigid body statics, the sum of the moments of each support point about the vehicle's center of gravity should be zero. Since the last moment is when the vehicle is stationary on the weighbridge, the coordinates of the vehicle's center of gravity at the last moment can be obtained by combining the position of each support point, the supporting force of each support point, and the pressure amplitude of each pressure sensor. The heavier the vehicle and the farther the distance between the vehicle's center of gravity and the weighbridge center, the more the weighbridge will tilt due to the vehicle's weight. Therefore, after obtaining the vehicle's center of gravity coordinates at the last moment, the degree of tilt of the weighbridge at the last moment can be obtained based on the relative position between the vehicle's center of gravity coordinates and the weighbridge center, as well as the pressure amplitude of all pressure sensors. This information is used to subsequently correct the vehicle weight collected by each pressure sensor on the weighbridge at the last moment, thereby obtaining and recording the vehicle's true weight and accurately calculating material flow. Attached Figure Description

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

[0017] Figure 1 This is a flowchart illustrating the steps of the AI-based unattended weighing system for identifying the entire process of entering and leaving the factory. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the AI-based unattended weighing system for full-process identification of factory entry and exit. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. In all division operations covered in this application, to prevent computer program crashes or invalid values ​​from being generated due to a denominator being zero, a preset minimum threshold can be set for the denominator term. When the absolute value of the denominator is less than the preset minimum threshold, it is adjusted to the preset minimum threshold with the same sign, thereby ensuring the robustness and feasibility of the algorithm under extreme conditions.

[0020] The following description, in conjunction with the accompanying drawings, details the specific solution of the AI-based unattended weighing system for identifying the entire process of entering and leaving the factory, provided by this invention.

[0021] Please see Figure 1 The diagram illustrates a flowchart of a method for identifying the entire process of an AI-based unattended weighing system for factory entry and exit, according to an embodiment of the present invention. The method includes the following steps: Step S001: Obtain the point cloud data of the weighbridge, the point cloud data of the vehicle at each time moment, the position of the pressure sensor in the weighbridge, and the pressure amplitude of each pressure sensor at each time moment.

[0022] It should be noted that, in order to accurately calculate material flow, vehicles need to be weighed when entering and leaving the factory. However, when a vehicle stops on the weighbridge, the weighbridge will tilt due to the weight of the vehicle, which will cause the weighbridge to fail to accurately collect the vehicle's weight. Therefore, this embodiment proposes an AI-based unattended weighing system for full-process identification of vehicle entry and exit. Specifically, it analyzes the pressure amplitude changes collected by various pressure sensors inside the weighbridge during the process of the vehicle entering the weighbridge, as well as the relative position relationship between the vehicle and the weighbridge, and analyzes the degree of tilt of the weighbridge when the vehicle finally stops on the weighbridge. This analysis is used to correct the weighbridge's collection of the vehicle's weight, thereby obtaining and recording the vehicle's true weight.

[0023] Specifically, the position of each pressure sensor in the weighbridge is obtained, and the point cloud data of the weighbridge is collected by a structured light camera. The time period from the moment the vehicle comes into contact with the weighbridge to the moment the vehicle stops on the weighbridge is recorded as the weighing period. The point cloud data of the vehicle at each moment in the weighing period and the pressure amplitude of each pressure sensor in the weighbridge at each moment are collected by the structured light camera. In this embodiment, 0.1 seconds is used as an example moment.

[0024] Step S002: Based on the point cloud data of the weighbridge and the point cloud data of the vehicle at each time moment, obtain several support points at each time moment. Combined with the position of the pressure sensor in the weighbridge, obtain the ideal distribution coefficient of each pressure sensor to each support point at each time moment. Combined with the pressure amplitude of the pressure sensor at each time moment, construct the support force evaluation model of each support point at each time moment. Based on the support force evaluation model of each support point at each time moment, obtain the support force of each support point at each time moment.

[0025] It should be noted that during the process of a vehicle entering the weighbridge, the pressure amplitude collected by the pressure sensors in the weighbridge transmits the vehicle's weight to the pressure sensors through the contact points between the vehicle's tires and the weighbridge. Furthermore, the pressure amplitude collected by each pressure sensor in the weighbridge is the external manifestation of the combined weight transmitted from multiple contact points. However, due to the different relative distances between different contact points and each pressure sensor, analyzing the vehicle's stress state solely based on the pressure sensor amplitudes will fail to distinguish the actual contribution of different support points to the weighbridge's stress, making it difficult to further analyze the intrinsic correlation between the vehicle's support state and the weighbridge's tilt. Therefore, this... The embodiment idealizes the distribution of pressure amplitude collected by each pressure sensor based on the spatial relationship between each pressure sensor and each support point, thereby establishing a correspondence between the pressure sensor amplitude and the support force of the support point, and obtaining the ideal distribution coefficient of each pressure sensor to each support point at each time. Furthermore, by obtaining the ideal distribution coefficient of each pressure sensor to each support point at each time and combining it with the pressure amplitude of the pressure sensor at the corresponding time, a support force evaluation model of each support point at each time is constructed, thereby obtaining the support force of each support point at each time, providing a theoretical basis for subsequent analysis of the distribution characteristics of support points and the tilt of the weighbridge.

[0026] Preferably, in a specific embodiment of the present invention, for the first At the moment, the first The point cloud data that comes into contact with the vehicle's point cloud data in the point cloud data of the weighbridge at time point n is denoted as the i-th point. The support point cloud data at the current moment will be the first time. The region formed by adjacent support point cloud data at the nth time point is used as the nth time point. The support area at the current moment will be the first The center of the support region at time t is denoted as the th moment. Support points at any given moment; Furthermore, regarding the first The first moment at the moment The pressure sensor and the first The first support point will be the first The first moment at the moment The pressure sensor and the first The reciprocal of the distance between the support points, compared to the previous one All pressure sensors at time n and the first The ratio obtained by summing the reciprocals of the distances between the support points is used as the th... At the moment of the first The pressure sensor is for the first Ideal coefficient for each support point.

[0027] It should be noted that the vehicle transmits its weight to the pressure sensors through each support point. The ideal distribution coefficient of each pressure sensor to each support point at each time represents the proportion of the force supporting the vehicle that each support point distributes to each pressure sensor at each time. The greater the distance between the support point and the pressure sensor, the more difficult it is for the support point to distribute the force supporting the vehicle to each pressure sensor. Therefore, this is the basis for obtaining the proportion of the force supporting the vehicle that each support point distributes to each pressure sensor at each time.

[0028] It should be further explained that the vehicle's tires are equivalent to support points. Under normal circumstances, the support force provided by the vehicle's tires is stable, that is, the amplitude of the support force at each support point is stable at each moment. Therefore, this embodiment constructs a support force evaluation model for each support point at each moment by using the pressure amplitude change of the pressure sensor during the vehicle's entry into the weighbridge and the ideal coefficient of each pressure sensor's allocation to each support point at each moment, in order to obtain the support force of each support point at each moment.

[0029] Preferably, in a specific embodiment of the present invention, for the first For each support point, based on the pressure amplitude of each pressure sensor at each time moment, and combined with the ideal distribution coefficient of each pressure sensor for each support point at each time moment, a support force evaluation model for each support point at each time moment is constructed. The specific expression is as follows: In the formula, This represents the output of the support force assessment model for each support point at each time point; Indicates the number of moments; Indicates the number of pressure sensors; Indicates the first The number of support points at any given moment; Indicates the first At the moment of the first The pressure sensor is for the first Ideal coefficient for allocation of each support point; Indicates the first At the moment of the first The pressure amplitude of each pressure sensor; Indicates the first At the moment of the first The supporting force of each support point; Indicates the first At the moment of the first The supporting force of each support point (support points with the same index at different times represent the support points corresponding to the same tire of the vehicle).

[0030] It should be noted that, since the ideal distribution coefficient of each pressure sensor to each support point at each time moment represents the proportional coefficient by which each support point distributes the force supporting the vehicle to each pressure sensor at each time moment, therefore the... At the moment of the first The pressure sensor is for the first The ideal coefficient of the distribution of the first support point, and the first support point At the moment of the first The product of the supporting forces at each support point represents the product of the forces at the th support point. At the moment of the first The pressure amplitude of the pressure sensor is determined by the first... At the moment of the first The portion transmitted by the first support point, therefore the first At the moment of the first The pressure amplitude of the pressure sensor is determined by the first... The sum of the portions transmitted by each support point at the i-th moment equals the sum of the portions transmitted at the j-th moment. At the moment of the first The pressure amplitude of each pressure sensor, i.e. and Since they are equal, we can construct a support force assessment model for each support point at each time point. Furthermore, by traversing the support force assessment model for each support point at each time point, we can obtain the support force of each support point at each time point.

[0031] Preferably, in a specific embodiment of the present invention, by traversing the support force evaluation model of each support point at each time, the support force of each support point at each time is taken as the support force of each support point at each time when the output result of the support force evaluation model of each support point at each time reaches the minimum, thus obtaining the support force of each support point at each time.

[0032] Thus, the supporting force of each support point at each moment is obtained.

[0033] Step S003: Based on the position of each support point at the last moment, obtain the coordinates of each support point at the last moment. Combine the supporting force of each support point at the last moment and the pressure amplitude of each pressure sensor to construct a vehicle center of gravity coordinate fitting model. Based on the vehicle center of gravity coordinate fitting model, obtain the vehicle's center of gravity coordinates at the last moment. Combine the position of the weighbridge center to obtain the vehicle's eccentricity at the last moment. Combine the pressure amplitudes of all pressure sensors at the last moment to obtain the weighbridge's tilt at the last moment.

[0034] It should be noted that the last moment is when the vehicle is stationary on the weighbridge. At this time, the vehicle is in a balanced state. According to the principle of rigid body statics, the sum of the moments of each support point about the vehicle's center of gravity should be zero. Therefore, after obtaining the support force of each support point at each moment through step S002, a vehicle center of gravity coordinate fitting model can be constructed by combining the position of each support point at the last moment, the support force of each support point at the last moment, and the pressure amplitude of each pressure sensor. Based on the vehicle center of gravity coordinate fitting model, the vehicle's center of gravity coordinates at the last moment can be obtained. When the vehicle is heavier and the distance between the vehicle's center of gravity coordinates and the center of the weighbridge is farther, the weighbridge will be more affected by the vehicle's weight and tilt. Therefore, after obtaining the vehicle's center of gravity coordinates at the last moment, the degree of tilt of the weighbridge at the last moment can be obtained based on the relative position between the vehicle's center of gravity coordinates and the center of the weighbridge at the last moment, as well as the pressure amplitude of all pressure sensors at the last moment. This is used to subsequently correct the vehicle weight collected by each pressure sensor in the weighbridge at the last moment, thereby obtaining the vehicle's true weight.

[0035] Preferably, in a specific embodiment of the present invention, the direction in which the vehicle enters the weighbridge is taken as the positive half-axis of the longitudinal axis, the direction in the weighbridge plane that is clockwise and perpendicular to the positive half-axis of the longitudinal axis is taken as the positive half-axis of the transverse axis, the spatial distance is taken as the unit of measurement, and the center of the weighbridge is taken as the origin, and a rectangular coordinate system is constructed and denoted as the weighbridge coordinate system. Furthermore, based on the position of each support point in the weighbridge at the last moment, the coordinates of each support point at the last moment are obtained.

[0036] It should be noted that since the last moment is when the vehicle is stationary on the weighbridge, the vehicle is in a balanced state. According to the principle of rigid body statics, the sum of the moments of each support point about the vehicle's center of gravity should be zero. Therefore, this can be used as the basis for constructing a vehicle center of gravity coordinate fitting model.

[0037] Preferably, in a specific embodiment of the present invention, a vehicle center of gravity coordinate fitting model is constructed based on the coordinates of each support point at the last moment and the supporting force of each support point at the last moment. The specific expression of this model is as follows: In the formula, This represents the output of the vehicle's center of gravity coordinate fitting model. This indicates the number of support points at the last moment; Indicates the last moment of the next The supporting force of each support point; Indicates the last moment of the next The x-coordinate of each support point; Indicates the last moment of the next The ordinate of each support point; The x-coordinate representing the vehicle's center of gravity; The vertical coordinate representing the vehicle's center of gravity; Indicates the vehicle's center of gravity shifts to the last moment. The vector of each support point; This represents the modulo function.

[0038] It should be noted that since the last moment is when the vehicle is stationary on the weighbridge, it is in a balanced state. According to the principle of rigid body statics, the sum of the moments of each support point about the vehicle's center of gravity should be zero. Therefore, at the last moment, the moments of each support point... The closer the magnitude of the sum is to 0, the better. and The more a vehicle's center of gravity characteristics it possesses, the more likely it can be traversed. and Obtain the coordinates of the vehicle's center of gravity at the last moment.

[0039] Preferably, in a specific embodiment of the present invention, by traversing the abscissa and ordinate of the vehicle's center of gravity in the vehicle center of gravity coordinate fitting model, when the output result of the vehicle center of gravity coordinate fitting model reaches the minimum, the abscissa and ordinate of the vehicle's center of gravity traversed in the vehicle center of gravity coordinate fitting model are used as the abscissa and ordinate of the vehicle's center of gravity at the last moment, so as to obtain the vehicle's center of gravity coordinates at the last moment.

[0040] It should be noted that the greater the distance between the vehicle's center of gravity and the weighbridge center, the more the weighbridge will tilt due to the vehicle's weight. Therefore, this embodiment analyzes the relative positional relationship between the vehicle's center of gravity and the weighbridge center at the last moment to obtain the degree of vehicle eccentricity at the last moment, which is used to quantify the degree of weighbridge tilt in subsequent steps.

[0041] Preferably, in a specific embodiment of the present invention, a straight line is drawn between the vehicle's center of gravity coordinates at the last moment and the center of the weighbridge in the weighbridge coordinate system. The line segment of the straight line located in the weighbridge is denoted as the tilt axis. The ratio obtained by dividing the distance between the vehicle's center of gravity coordinates at the last moment and the center of the weighbridge by the length of the tilt axis is used as the degree of eccentricity of the vehicle at the last moment.

[0042] It should be noted that, since the heavier the vehicle on the weighbridge, the more the weighbridge will tilt due to the vehicle's weight, this embodiment, after obtaining the degree of vehicle eccentricity at the last moment, further combines the pressure amplitude of all pressure sensors at the last moment to obtain the degree of tilt of the weighbridge at the last moment, which is used to quantify the degree of tilt of the weighbridge.

[0043] Preferably, in a specific embodiment of the present invention, the sum of the pressure amplitudes of all pressure sensors at the last moment is multiplied by the vehicle's eccentricity at the last moment and normalized (the product can be normalized using a maximum-minimum normalization function), and the normalized result is used as the tilt degree of the weighbridge at the last moment.

[0044] At this point, we have obtained the tilt level of the weighbridge at the last moment.

[0045] Step S004: Based on the tilt of the weighbridge at the last moment and the pressure amplitude of all pressure sensors at the last moment, obtain and record the actual weight of the vehicle.

[0046] It should be noted that after obtaining the tilt degree of the weighbridge at the last moment through step S003, the vehicle weight collected by each pressure sensor in the weighbridge at the last moment can be corrected based on the tilt degree of the weighbridge at the last moment, thereby obtaining the true weight of the vehicle and recording the true weight of the vehicle when entering and leaving the factory, so as to accurately calculate the material flow.

[0047] Preferably, in a specific embodiment of the present invention, a tilt correction angle is preset. The specific value of the tilt correction angle can be set by the user according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, the tilt correction angle is 5° as an example. The product of the tilt degree of the weighbridge at the last moment and the tilt correction angle is used as the tilt angle of the weighbridge at the last moment. Furthermore, the sum of the pressure amplitudes of all pressure sensors at the last moment is divided by the cosine of the tilt angle of the weighbridge at the last moment, and the ratio is taken as the true weight of the vehicle. The true weight of the vehicle entering and leaving the factory is also recorded.

[0048] This concludes the embodiment.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An AI-based unattended weighing system for full-process identification of inbound and outbound operations, characterized in that, The method includes the following steps: Acquire point cloud data of the weighbridge, point cloud data of vehicles at each time moment, the location of pressure sensors in the weighbridge, and the pressure amplitude of each pressure sensor at each time moment; Based on the point cloud data of the weighbridge and the point cloud data of the vehicle at each time moment, several support points are obtained at each time moment. Combined with the position of the pressure sensor in the weighbridge, the ideal distribution coefficient of each pressure sensor to each support point at each time moment is obtained. Combined with the pressure amplitude of the pressure sensor at each time moment, the support force evaluation model of each support point at each time moment is constructed. Based on the support force evaluation model of each support point at each time moment, the support force of each support point at each time moment is obtained. Based on the position of each support point at the last moment, obtain the coordinates of each support point at the last moment. Combine the supporting force of each support point at the last moment and the pressure amplitude of each pressure sensor at the last moment to construct a vehicle center of gravity coordinate fitting model. Based on the vehicle center of gravity coordinate fitting model, obtain the vehicle's center of gravity coordinates at the last moment. Combine the position of the weighbridge center to obtain the vehicle's eccentricity at the last moment. Combine the pressure amplitude of all pressure sensors at the last moment to obtain the weighbridge's tilt at the last moment. The vehicle's actual weight is obtained and recorded based on the tilt of the weighbridge at the last moment and the pressure amplitude of all pressure sensors at the last moment.

2. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The method for obtaining several support points at each time moment based on the point cloud data of the weighbridge and the point cloud data of the vehicle at each time moment, and combining the position of the pressure sensor in the weighbridge to obtain the ideal distribution coefficient of each pressure sensor to each support point at each time moment, includes the following specific methods: For the At the moment, the first The point cloud data that comes into contact with the vehicle's point cloud data in the point cloud data of the weighbridge at time point n is denoted as the i-th point. The support point cloud data at the current moment will be the first time. The region formed by adjacent support point cloud data at the nth time point is used as the nth time point. The support area at the current moment will be the first The center of the support region at time t is denoted as the th moment. Support points at any given moment; For the The first moment at the moment The pressure sensor and the first The first support point will be the first The first moment at the moment The pressure sensor and the first The reciprocal of the distance between the support points, compared to the previous one All pressure sensors at time n and the first The ratio obtained by summing the reciprocals of the distances between the support points is used as the th... At the moment of the first The pressure sensor is for the first Ideal coefficient for each support point.

3. The AI-based unattended weighing system for full-process identification of factory entry and exit as described in claim 1, characterized in that, The specific method for constructing the support force assessment model for each support point at each time point is as follows: In the formula, This represents the output of the support force assessment model for each support point at each time point; Indicates the number of moments; Indicates the number of pressure sensors; Indicates the first The number of support points at any given moment; Indicates the first At the moment of the first The pressure sensor is for the first Ideal coefficient for allocation of each support point; Indicates the first At the moment of the first The pressure amplitude of each pressure sensor; Indicates the first At the moment of the first The supporting force of each support point; Indicates the first At the moment of the first The supporting force of each support point.

4. The AI-based unattended weighing system for full-process identification of factory entry and exit as described in claim 1, characterized in that, The specific method for obtaining the support force of each support point at each time point based on the support force assessment model at each time point is as follows: By iterating through the support force evaluation model of each support point at each time point, when the output result of the support force evaluation model of each support point at each time point reaches the minimum, the support force of each support point at each time point traversed in the support force evaluation model of each support point at each time point is taken as the support force of each support point at each time point, thus obtaining the support force of each support point at each time point.

5. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The specific method for obtaining the coordinates of each support point at the last moment based on the position of each support point at the last moment is as follows: With the direction in which the vehicle enters the weighbridge as the positive half-axis of the longitudinal axis, and the direction in the weighbridge plane that is clockwise and perpendicular to the positive half-axis of the longitudinal axis as the positive half-axis of the transverse axis, a rectangular coordinate system is constructed with spatial distance as the unit of measurement and the center of the weighbridge as the origin, denoted as the weighbridge coordinate system. Based on the position of each support point in the weighbridge at the last moment, obtain the coordinates of each support point at the last moment.

6. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The specific method for constructing the vehicle center of gravity coordinate fitting model is as follows: In the formula, This represents the output of the vehicle's center of gravity coordinate fitting model. This indicates the number of support points at the last moment; Indicates the last moment of the next The supporting force of each support point; Indicates the last moment of the next The x-coordinate of each support point; Indicates the last moment of the next The ordinate of each support point; The x-coordinate representing the vehicle's center of gravity; The vertical coordinate representing the vehicle's center of gravity; Indicates the vehicle's center of gravity shifts to the last moment. The vector of each support point; This represents the modulo function.

7. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The specific method for obtaining the vehicle's center of gravity coordinates at the last moment based on the vehicle's center of gravity coordinate fitting model is as follows: By iterating through the horizontal and vertical coordinates of the vehicle's center of gravity in the vehicle center of gravity coordinate fitting model, when the output of the vehicle center of gravity coordinate fitting model reaches its minimum, the horizontal and vertical coordinates of the vehicle's center of gravity traversed in the vehicle center of gravity coordinate fitting model are used as the horizontal and vertical coordinates of the vehicle's center of gravity at the last moment, thus obtaining the vehicle's center of gravity coordinates at the last moment.

8. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The specific method for obtaining the vehicle's eccentricity at the last moment is as follows: In the weighbridge coordinate system, draw a straight line from the vehicle's center of gravity coordinate at the last moment to the center of the weighbridge. The line segment of the line within the weighbridge is denoted as the tilt axis. The ratio of the distance between the vehicle's center of gravity coordinate at the last moment and the center of the weighbridge to the length of the tilt axis is taken as the degree of vehicle eccentricity at the last moment.

9. The AI-based unattended weighing system for full-process identification of inbound and outbound operations according to claim 1, characterized in that, The specific method for obtaining the tilt degree of the weighbridge at the last moment is as follows: The sum of the pressure amplitudes of all pressure sensors at the last moment is multiplied by the vehicle's eccentricity at the last moment, and the product is normalized. The normalized result is used as the tilt degree of the weighbridge at the last moment.

10. The AI-based unattended weighing system for full-process identification of factory entry and exit as described in claim 1, characterized in that, The method for obtaining and recording the vehicle's true weight based on the tilt of the weighbridge at the last moment and the pressure amplitude of all pressure sensors at the last moment includes the following specific methods: A preset tilt correction angle is used. The product of the tilt degree of the weighbridge at the last moment and the tilt correction angle is used as the tilt angle of the weighbridge at the last moment. The sum of the pressure amplitudes of all pressure sensors at the last moment is divided by the cosine of the tilt angle of the weighbridge at the last moment. This ratio is taken as the vehicle's true weight, and the true weight of the vehicle entering and leaving the factory is recorded.

Images