Automatic guided vehicle positioning method, device, equipment and medium

Abstract

The application discloses an automatic guided vehicle positioning method, device, equipment and medium, and relates to the technical field of computers. The method comprises the following steps: acquiring a current characteristic value of an automatic guided vehicle, and screening a target signal fingerprint corresponding to the current characteristic value from each signal fingerprint of a fingerprint library constructed in advance by using a LoRa device; determining a historical coordinate in the target signal fingerprint as a reference position coordinate of the automatic guided vehicle; acquiring a reference optical flow vector of the automatic guided vehicle by using a preset optical flow device; and determining a current target position coordinate of the automatic guided vehicle based on the reference position coordinate and the reference optical flow vector. Through the above scheme, the reference position coordinate determined by using the LoRa device and the reference optical flow vector acquired by using the preset optical flow device are used, and then the current target position coordinate is determined based on the reference position coordinate and the reference optical flow vector, so that accurate positioning of the automatic guided vehicle can be realized more flexibly.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to methods, devices, equipment and media for positioning automated guided vehicles. Background Technology

[0002] Automated Guided Vehicles (AGVs) are commonly used industrial transport vehicles. During the operation of AGVs, the positioning and navigation system on the AGV is crucial due to the unmanned operation. Outdoors, satellite navigation and positioning technology can generally be used. However, the working environment of AGVs often involves workshops, warehouses, and other areas where satellite navigation and positioning systems may experience increased errors or even loss of lock. In such cases, it is necessary to obtain high-precision position information through other means.

[0003] Among existing positioning technologies, SLAM (Simultaneous Localization and Mapping) and millimeter-wave radar technologies are not ideal in dark environments or complex terrain. Traditional inertial navigation systems measure and save the motion state data of the vehicle during its movement using inertial sensors, but they have a serious problem: the system is divergent without the assistance of other sensors, and its error increases over time. Magnetic navigation uses magnetic materials laid on the ground for navigation and positioning, but later modifications will damage the original ground, which has limitations.

[0004] In summary, how to achieve more flexible and accurate positioning of automated guided vehicles is a problem that needs to be solved in this field. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide an automated guided vehicle (AGV) positioning method, apparatus, device, and medium, which can more flexibly achieve accurate positioning of AGVs. The specific solution is as follows:

[0006] In a first aspect, this application discloses an automated guided vehicle (AGV) positioning method, comprising:

[0007] The current feature value of the automated guided vehicle is obtained, and the target signal fingerprint corresponding to the current feature value is selected from the various signal fingerprints in the fingerprint library pre-built using LoRa devices.

[0008] The historical coordinates in the target signal fingerprint are determined as the reference position coordinates of the automated guided vehicle;

[0009] The reference optical flow vector of the automated guided vehicle is obtained using a preset optical flow device;

[0010] The current target position coordinates of the automated guided vehicle are determined based on the reference position coordinates and the reference optical flow vector.

[0011] Optionally, before obtaining the current feature value of the automated guided vehicle, the method further includes:

[0012] The LoRa device is used to obtain the signal strength, link quality, and time difference between transmitted and received signals for each channel at each location point along the preset driving route.

[0013] The coordinates of the location point, the signal strength, the link quality, and the time difference are stored in matrix form to obtain the corresponding signal fingerprint, and a fingerprint database of the driving route is constructed based on the signal fingerprint.

[0014] Optionally, before acquiring the signal strength, link quality, and time difference between transmitted and received signals for each channel at each location point along the preset driving route using the LoRa device, the method further includes:

[0015] The preset driving route is divided into several location points based on the starting point coordinates and the ending point coordinates of the preset driving route.

[0016] Optionally, the step of filtering the target signal fingerprint corresponding to the current feature value from various signal fingerprints in a fingerprint database pre-built using LoRa devices includes:

[0017] Obtain each signal fingerprint from a fingerprint database pre-built using LoRa devices, and obtain the weight value of each signal fingerprint using a preset probability model;

[0018] Based on the weight values ​​of each of the signal fingerprints, a particle filter algorithm is used to filter out the target signal fingerprint corresponding to the current feature value from each of the signal fingerprints.

[0019] Optionally, obtaining the reference optical flow vector of the automated guided vehicle using a preset optical flow device includes:

[0020] The previous frame image and the current frame image of the automated guided vehicle are obtained using a preset optical flow device.

[0021] Select a tracking block from the previous frame image, and determine the prediction block corresponding to the tracking block from the current frame image;

[0022] The reference optical flow vector is calculated using the tracking block and the prediction block.

[0023] Optionally, determining the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector includes:

[0024] The reference position coordinates and the reference optical flow vector are filtered using Kalman filtering to obtain the filtered reference position coordinates and the filtered reference optical flow vector.

[0025] The current target position coordinates of the automated guided vehicle are determined based on the filtered reference position coordinates and the filtered reference optical flow vector.

[0026] Optionally, the step of filtering the reference position coordinates and the reference optical flow vector using Kalman filtering includes:

[0027] The distillation factor is determined based on the positive definiteness principle of Kalman filtering;

[0028] A distillation matrix is ​​constructed using a distillation factor, and the distillation matrix is ​​used to filter the reference position coordinates and the reference optical flow vector.

[0029] Secondly, an automated guided vehicle positioning device includes:

[0030] The target fingerprint filtering module is used to obtain the current feature value of the automated guided vehicle and filter out the target signal fingerprint corresponding to the current feature value from each signal fingerprint in the fingerprint library pre-built using LoRa devices;

[0031] A reference position determination module is used to determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle;

[0032] The reference optical flow acquisition module is used to acquire the reference optical flow vector of the automated guided vehicle using a preset optical flow device;

[0033] The target position determination module is used to determine the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector.

[0034] Thirdly, this application discloses an electronic device, including:

[0035] Memory, used to store computer programs;

[0036] A processor is configured to execute the computer program to implement the steps of the aforementioned disclosed automated guided vehicle positioning method.

[0037] Fourthly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the aforementioned disclosed automatic guided vehicle positioning method.

[0038] The beneficial effects of this application are as follows: It obtains the current feature value of the automated guided vehicle (AGV) and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in a fingerprint database pre-built using LoRa devices; it determines the historical coordinates in the target signal fingerprint as the reference position coordinates of the AGV; it obtains the reference optical flow vector of the AGV using a preset optical flow device; and it determines the current target position coordinates of the AGV based on the reference position coordinates and the reference optical flow vector. Therefore, this application first uses LoRa devices to build a fingerprint database and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in the database, thereby determining the reference position coordinates of the AGV. Secondly, it uses the reference optical flow vector obtained by the preset optical flow device. Finally, it determines the current target position coordinates based on the reference position coordinates and the reference optical flow vector. Furthermore, it only requires the LoRa device and the preset optical flow device to be installed in the AGV, making subsequent modifications more convenient and unrestricted, and enabling more flexible and accurate positioning of the AGV. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application 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 embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0040] Figure 1 This is a flowchart of an automated guided vehicle positioning method disclosed in this application;

[0041] Figure 2 This is a schematic diagram of a specific tracking block and prediction block disclosed in this application;

[0042] Figure 3 This is a schematic diagram of a specific speed measurement model disclosed in this application;

[0043] Figure 4 This is a schematic diagram of the installation of a specific optical flow device disclosed in this application;

[0044] Figure 5 This is a flowchart of a specific automated guided vehicle positioning method disclosed in this application;

[0045] Figure 6 This is a flowchart of another specific automated guided vehicle positioning method disclosed in this application;

[0046] Figure 7 This is a schematic diagram of the structure of an automated guided vehicle positioning device disclosed in this application;

[0047] Figure 8This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0048] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] Among the existing positioning technologies, SLAM and millimeter-wave radar are not ideal in dark environments or complex terrain. Traditional inertial navigation systems measure and save the motion state data of the vehicle during its movement using inertial sensors, but they have a serious problem: the system diverges without the assistance of other sensors, and its error increases over time. Magnetic navigation uses magnetic materials laid on the ground for navigation and positioning, but later modifications will damage the original ground, which has limitations.

[0050] Therefore, this application provides an automated guided vehicle (AGV) positioning scheme that can more flexibly achieve accurate positioning of the AGV.

[0051] See Figure 1 As shown in the figure, this application discloses an automated guided vehicle (AGV) positioning method, including:

[0052] Step S11: Obtain the current feature value of the automated guided vehicle, and filter out the target signal fingerprint corresponding to the current feature value from the various signal fingerprints in the fingerprint library pre-built using LoRa devices.

[0053] LoRa (Long Range Radio) devices, MCU (Microcontroller Unit) processors, and pre-installed optical flow devices are installed in the automated guided vehicle (AGV) to enable subsequent navigation using optical flow-assisted LoRa combinations. The MCU processor is primarily used for controlling the various hardware components and processing the collected data using core algorithms. These core algorithms include sensor data acquisition, filtering and processing, fusion algorithm calculations, and machine learning algorithm inference.

[0054] In this embodiment, before obtaining the current feature value of the automated guided vehicle (AGV), the method further includes: using LoRa devices to obtain the signal strength, link quality, and time difference between the transmitted and received signals for each channel at each location point in the preset driving route; storing the coordinates of the location point, the signal strength, the link quality, and the time difference in matrix form to obtain the corresponding signal fingerprint; and constructing a fingerprint database for the driving route based on the signal fingerprint. LoRa transmitting and receiving devices are deployed within the AGV's operating area to prepare for constructing the fingerprint database.

[0055] First, the LoRa device carried by the AGV sends signals to multiple base stations simultaneously through multiple channels at location i, and records the first time of signal transmission. After receiving the signal, the base station immediately sends it back, and after receiving the feedback, the LoRa device immediately records the second time of signal reception for each channel.

[0056] Secondly, the time difference T between the first and second times at location point i. i Stored in matrix form; simultaneously, the signal strength M of each channel at position point i is recorded. i and link quality P i Stored in matrix form, the above matrix is ​​simultaneously stored as a signal fingerprint (x) in the corresponding form. i y i ,||M i ||,|T i ||,||P i ||), where, Represents the signal strength of channel n, ||T i ||=|T CH1 T CH2 ...T CHn |,T CHn This represents the time difference of channel n, (x i y i (x) represents the coordinates of position point i, that is, (x) i y i ) is a historical coordinate.

[0057] Finally, a fingerprint database of the driving route is constructed based on the signal fingerprints at each location point. Compared with traditional fingerprint databases, this database contains a larger amount of data and more comprehensive feature values ​​for each point, which can improve the success rate and accuracy of matching.

[0058] In this embodiment, before using LoRa devices to obtain the signal strength, link quality, and time difference between transmitted and received signals for each channel at each location point in the preset driving route, the method further includes: dividing the preset driving route into several location points based on the starting and ending coordinates of the preset driving route. The AGV operating area is divided into several straight driving routes (L1, L2, L3, ... L...). i The AGV moves at a constant speed along each straight line, recording the coordinates of the starting and ending points of the line. The line is then divided into several position points, each representing a step. Within each step, the LoRa data represents the signal fingerprint (x) of that grid point. i y i ,||M i ||,||T i ||,||P i ||).

[0059] It is understandable that the current feature value of the automated guided vehicle is obtained, which is the time difference, signal strength and link quality at the current position. Therefore, the current feature value is matched with the time difference, signal strength and link quality of each signal fingerprint in the fingerprint database to filter out the matching target signal fingerprint.

[0060] Step S12: Determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle.

[0061] The target signal fingerprint stores historical coordinates, time difference, signal strength, and link quality. Therefore, when the time difference, signal strength, and link quality in the target signal fingerprint match the current feature value, the historical coordinates in the target signal fingerprint can be determined as the reference position coordinates of the automated guided vehicle.

[0062] Step S13: Obtain the reference optical flow vector of the automated guided vehicle using a preset optical flow device.

[0063] In this embodiment, obtaining the reference optical flow vector of the automated guided vehicle using a preset optical flow device includes: obtaining the previous frame image and the current frame image of the automated guided vehicle at the current moment using the preset optical flow device; selecting a tracking block from the previous frame image and determining the prediction block corresponding to the tracking block from the current frame image; and calculating the reference optical flow vector using the tracking block and the prediction block.

[0064] For example Figure 2The diagram illustrates a specific tracking block and prediction block. If the current time is T, then the previous time is T-1. The images at time T-1 and time T are acquired, i.e., the previous frame image and the current frame image are acquired. A tracking block of L×L pixels is randomly selected in the previous frame image, and the prediction block corresponding to the tracking block is determined in the current frame image. The determination process is as follows: There are several prediction blocks of L×L pixels in the current frame image. The sum of the absolute values ​​of the grayscale differences (Sum of Absolute Differences, i.e., SAD) between each corresponding pixel point of the tracking block and each prediction block is calculated. The prediction block with the smallest sum of absolute values ​​is the one corresponding to the tracking block.

[0065] It is understandable that if the tracking block is before the automated guided vehicle moves, then the prediction block is after the tracking block moves. That is, the reference optical flow vector u between the tracking block and the prediction block is calculated. Here, i and j represent the offsets of the tracking block at the current moment and the tracking block at the previous moment in the horizontal and vertical directions, respectively.

[0066] Step S14: Determine the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector.

[0067] The reference speed of the automated guided vehicle can be inferred by referring to the optical flow vector. The speed estimation model used is a planar model, with the camera mounted on the bottom of the vehicle and its optical axis perpendicular to the ground to achieve vertical imaging of the ground texture. Due to motion constraints, the vehicle's X and Z axes can be ignored. For example... Figure 3 The diagram shows a specific velocity measurement model. OXYZ is the camera coordinate system, oxyz is the imaging plane, and the camera is at the origin O. The distance f between point O and oxyz is the camera's focal length. Let there be a point P in space with coordinates (x, y, z) in OXYZ. Then there is a point P' in oxyz that is the projection of P. Assuming the coordinates of point P' are (x', y', f), the following relationship exists:

[0068] P′=f*P / Z

[0069] y′=f*y / z;

[0070] Assuming the vehicle is in relative motion with respect to P, and the terrain surface is flat during the motion, then:

[0071] V y =Z*v y / α y ;

[0072] In the formula, V y α represents the vehicle speed. y Focal length is expressed in horizontal pixels, v yZ represents the pixel velocity of the image, and Z represents the distance of the camera from the ground.

[0073] For example Figure 4 The diagram shows a specific installation schematic of an optical flow device. The AGV's movement direction is calculated by installing dual optical flow devices. Assume the rectangle represents the vehicle body, and the circles represent the two optical flow devices, installed on both sides of the AGV. For example, if a Cartesian coordinate system is selected with x parallel to the east direction, y parallel to the north direction, and h parallel to the sky direction as the navigation coordinate system, and without considering the installation error angle between the optical flow and the vehicle body, the reference speed of the automated guided vehicle output by the optical flow in the carrier coordinate system can be expressed as:

[0074]

[0075] The reference speed of the automated guided vehicle is determined based on the reference optical flow vector, so that the current target position coordinates of the automated guided vehicle can be determined based on the reference position coordinates and the reference speed of the automated guided vehicle.

[0076] The beneficial effects of this application are as follows: It obtains the current feature value of the automated guided vehicle (AGV) and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in a fingerprint database pre-built using LoRa devices; it determines the historical coordinates in the target signal fingerprint as the reference position coordinates of the AGV; it obtains the reference optical flow vector of the AGV using a preset optical flow device; and it determines the current target position coordinates of the AGV based on the reference position coordinates and the reference optical flow vector. Therefore, this application first uses LoRa devices to build a fingerprint database and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in the database, thereby determining the reference position coordinates of the AGV. Secondly, it uses the reference optical flow vector obtained by the preset optical flow device. Finally, it determines the current target position coordinates based on the reference position coordinates and the reference optical flow vector. Furthermore, it only requires the LoRa device and the preset optical flow device to be installed in the AGV, making subsequent modifications more convenient and unrestricted, and enabling more flexible and accurate positioning of the AGV.

[0077] See Figure 5 As shown in the figure, this application discloses a specific method for positioning an automated guided vehicle, including:

[0078] Step S21: Obtain the current feature value of the automated guided vehicle.

[0079] Step S22: Obtain each signal fingerprint in the fingerprint library pre-built using LoRa devices, and obtain the weight value of each signal fingerprint using a preset probability model; based on the weight value of each signal fingerprint, use a particle filter algorithm to filter out the target signal fingerprint corresponding to the current feature value from each signal fingerprint.

[0080] An improved particle filter algorithm is used to filter out the target signal fingerprint corresponding to the current feature value. Particle filtering is a non-parametric filter based on the Monte Carlo method. It replaces the integral operation with the sample mean and represents the posterior distribution of the state with a finite number of particles, thereby obtaining the minimum variance estimate of the state. It can effectively handle nonlinear and non-Gaussian systems, and therefore has been widely used in various fields.

[0081] Particle filtering comprises four processes: particle initialization, importance sampling, resampling, and state estimate output. In this embodiment, the system states 's' represent position, signal strength, and time difference of arrival, as shown below:

[0082] s = (x, y, γ, β);

[0083] In the formula, x and y represent position coordinates, γ represents signal strength, and β represents the time difference. In the particle filtering process, each particle represents an assumption about the current state and carries a weight ω; a larger weight indicates that the particle is closer to the true state. The weight of the i-th particle is determined by a probabilistic model of the measured values. The calculation is obtained, where P represents the probability model and t represents the number of samplings. Since P(z) t+1 -z t |s t s t+1 Since this unknown quantity cannot be accurately obtained in practical applications, it is set as Gaussian noise in this invention. Therefore, the measurement equation is as shown in the following formula:

[0084]

[0085] In the formula, This indicates that the i-th particle obtains the LoRa fingerprint corresponding to state s from the fingerprint database during the t-th sampling, z t Let t represent the LoRa fingerprint collected in the t-th iteration, and σ be the standard deviation of the LoRa offset.

[0086] Once all particles have weights, particles with low weights are filtered out. These particles are usually equated to being far from the actual state. Through resampling, the particles are concentrated in the region near the high-weight particles, causing the particle swarm to converge. In other words, the one with the highest degree of convergence is the target signal fingerprint corresponding to the current feature value.

[0087] Step S23: Determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle.

[0088] Step S24: Obtain the reference optical flow vector of the automated guided vehicle using a preset optical flow device.

[0089] Step S25: Determine the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector.

[0090] Therefore, this application uses an improved particle filter algorithm to select the most matching target signal fingerprint from various signal fingerprints in the fingerprint database, which means that the reference position coordinates of the automated guided vehicle are determined more accurately. In this way, the reliability of the current target position coordinates determined based on the reference position coordinates and the reference optical flow vector can be improved.

[0091] See Figure 6 As shown in the embodiment, this application discloses another specific method for positioning an automated guided vehicle, including:

[0092] Step S31: Obtain the current feature value of the automated guided vehicle, and filter out the target signal fingerprint corresponding to the current feature value from the various signal fingerprints in the fingerprint library pre-built using LoRa devices.

[0093] Step S32: Determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle.

[0094] Step S33: Obtain the reference optical flow vector of the automated guided vehicle using a preset optical flow device.

[0095] Step S34: Use Kalman filtering to filter the reference position coordinates and the reference optical flow vector to obtain filtered reference position coordinates and filtered reference optical flow vector; determine the current target position coordinates of the automated guided vehicle based on the filtered reference position coordinates and the filtered reference optical flow vector.

[0096] In this embodiment, the step of using Kalman filtering to filter the reference position coordinates and the reference optical flow vector includes: determining a distillation factor according to the positive definite principle of Kalman filtering; constructing a distillation matrix using the distillation factor; and using the distillation matrix to filter the reference position coordinates and the reference optical flow vector.

[0097] In practical applications, it is difficult to ensure that the LoRa coordinate system axis is aligned with the optical flow axis, resulting in a certain installation angle error. Typically, the optical flow is fixed on the AGV, with a small installation angle that hardly changes. Let's denote the LoRa coordinate system as the b-system and the navigation coordinate system as the n-system. Let b and n represent the transition matrix. Then, the actual form of the optical flow estimate's velocity and position in the navigation coordinate system is:

[0098]

[0099]

[0100] In the formula, θ represents the pitch installation error angle. Indicates the heading installation error angle. This indicates the output velocity of the optical flow in the navigation coordinate system. This represents the position coordinates of the optical flow in the navigation coordinate system. This represents the output velocity of optical flow in the carrier coordinate system.

[0101] When using LoRa output and optical flow output as measurements, optical flow-assisted LoRa integrated navigation must consider the installation error angle θ. Otherwise, it will cause some errors.

[0102] The state equations use nine state parameters, and the specific state variables (representing the state information contained in the system) are as follows:

[0103] X(t)=[Δα 3×1 Δv 3×1 Δp 3×1 ] T ;

[0104] In the formula, Δα 3×1 Indicates the installation angle error, Δv 3×1 Indicates the velocity error, δp 3×1 The position error is represented by the following formulas for obtaining the installation angle error, velocity error, and position error:

[0105] Δα 3×1 =[α x α y α h ] T ;

[0106] Δv 3×1 =[Δv x Δv y Δv h ] T ;

[0107] δp 3×1 =[δx δ y δh] T ;

[0108] In the formula, α x Indicates the X-axis mounting angle error, α y Indicates the Y-axis mounting angle error, α h The height axis mounting angle error is represented by Δv. x The value Δv represents the X-axis velocity error. y The Y-axis velocity error, Δv h δx represents the height axis velocity error, and δx represents the X-axis position error. yδh represents the position error along the Y-axis, and δh represents the position error along the height axis.

[0109] The combined state equations are:

[0110] X(t) = F(t)X(t) + W(t);

[0111] In the formula, F(t) represents the state transition matrix, X(t) represents the system state error, and W(t) represents the system process noise.

[0112] The difference between the position and velocity calculated by LoRa and the position and velocity derived by optical flow is used as the measurement value Z, and the following formula applies:

[0113]

[0114] In the formula, This indicates the speed of the LoRa system within the navigation coordinate system. This indicates the position of the LoRa system within the navigation coordinate system.

[0115] The measurement equation is the value obtained through sensors, etc., and can be expressed as:

[0116] Z(t) = H(t)x(t) + V(t);

[0117] In the formula, H(t) represents the nonlinear observation equation for position and velocity, and V(t) represents white noise.

[0118] As can be seen from the above, the optical flow-assisted LoRa integrated navigation discrete system can be represented as:

[0119]

[0120] Among them, F k Let X represent the state transition matrix at time k. k H represents the state vector at time k. k Let Z represent the observation equation. k W represents the measurement vector at time k. k and V k This represents Gaussian white noise that is uncorrelated.

[0121] In addition to being affected by its own nonlinearity, the LoRa and optical flow combined navigation system is also affected by measurement noise. The strict requirements of the standard Kalman filter on the system model increase the computational complexity and make accurate positioning difficult under the influence of a large amount of measurement noise. Therefore, it is necessary to improve the standard Kalman filter. This embodiment derives the solution method of the distillation factor based on the positive definite principle of Kalman filtering to determine the distillation factor and then perform filtering. The specific derivation process is as follows:

[0122] According to the definition of residual sequences, we know that:

[0123]

[0124] In the formula, ε k H represents the residual sequence at time K. k Let X represent the observation equation. k This represents the state vector at time k. v represents the state value at time k derived from time k-1. k β represents white noise at time k. k Let k represent the diagonal matrix at time k. w represents the state value at time k-1. k Let f represent the system noise at time k, where f k It can be expressed by the following formula:

[0125]

[0126] In the formula, F k / k-1 Let represent the state transition matrix at time k derived from the state transition matrix at time k-1.

[0127] Then the residual matrix at time k+j can be expressed as:

[0128]

[0129] According to the positive definiteness principle:

[0130]

[0131] Substituting the two equations, we can see that:

[0132]

[0133] In the formula, β i Let f be the diagonal matrix at time i. i with f k Similarly, K i H represents the Kalman gain matrix at time i. i Describe the observation equation at time i. Let K represent the state covariance matrix. k Let K represent the Kalman gain matrix at time k.

[0134] To ensure that the matrix is ​​positive definite, the following equation must hold:

[0135]

[0136] According to the Kalman gain formula, the above equation can be transformed into:

[0137]

[0138] Since the gain matrix is ​​a non-zero matrix, the above equation can be simplified to:

[0139]

[0140] Expanding the covariance, we get:

[0141]

[0142] In the formula, λ k Represents the distillation matrix. Indicates the weight. Indicates the distillation point. R represents the measurement value at time k calculated from time k-1. k Let represent a nonnegative definite matrix.

[0143] Define N k M k for:

[0144] N k =V k -R k ;

[0145]

[0146] Use N k M k Represented as λ k M k =N k Taking the product of both sides, we obtain the distillation factor τ. k :

[0147]

[0148] Then, the distillation factor is used to compose the distillation matrix:

[0149] λ k =diag(λ 1，1 , λ 2，2 ,…λ n，n );

[0150]

[0151] λ i，i =τ k κ i ;

[0152] In the formula, κ i p represents the coefficient. i，i p represents the probability coefficient at position i. j，j Let λ represent the probability coefficient at position j. i，iThis represents the distillation factor in the i-th row and i-th column of the distillation matrix.

[0153] Therefore, this invention constructs a distillation matrix from the distillation factor to perform targeted filtering on each state variable. It directly calculates the prediction covariance matrix using the distillation matrix, eliminating the need for the steps in traditional algorithms that first calculate the covariance matrix of the state equation and then recalculate the covariance matrix of the prediction equation. This effectively reduces the computational load and is suitable for embedded devices.

[0154] See Figure 7 As shown in the figure, this application discloses an automated guided vehicle positioning device, including:

[0155] The target fingerprint screening module 11 is used to obtain the current feature value of the automated guided vehicle and screen out the target signal fingerprint corresponding to the current feature value from each signal fingerprint in the fingerprint library pre-built using LoRa devices;

[0156] The reference position determination module 12 is used to determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle;

[0157] The reference optical flow acquisition module 13 is used to acquire the reference optical flow vector of the automated guided vehicle using a preset optical flow device;

[0158] The target position determination module 14 is used to determine the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector.

[0159] The beneficial effects of this application are as follows: It obtains the current feature value of the automated guided vehicle (AGV) and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in a fingerprint database pre-built using LoRa devices; it determines the historical coordinates in the target signal fingerprint as the reference position coordinates of the AGV; it obtains the reference optical flow vector of the AGV using a preset optical flow device; and it determines the current target position coordinates of the AGV based on the reference position coordinates and the reference optical flow vector. Therefore, this application first uses LoRa devices to build a fingerprint database and filters out the target signal fingerprint corresponding to the current feature value from various signal fingerprints in the database, thereby determining the reference position coordinates of the AGV. Secondly, it uses the reference optical flow vector obtained by the preset optical flow device. Finally, it determines the current target position coordinates based on the reference position coordinates and the reference optical flow vector. Furthermore, it only requires the LoRa device and the preset optical flow device to be installed in the AGV, making subsequent modifications more convenient and unrestricted, and enabling more flexible and accurate positioning of the AGV.

[0160] Furthermore, embodiments of this application also provide an electronic device. Figure 8This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0161] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Specifically, it may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the automated guided vehicle positioning method performed by the electronic device disclosed in any of the foregoing embodiments.

[0162] In this embodiment, the power supply 23 is used to provide operating voltage for various hardware devices on the electronic device; the communication interface 24 can create a data transmission channel between the electronic device and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0163] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0164] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon include operating system 221, computer program 222 and data 223, etc., and the storage method can be temporary storage or permanent storage.

[0165] The operating system 221 manages and controls the various hardware devices and computer programs 222 on the electronic device to enable the processor 21 to perform calculations and processing on the massive amounts of data 223 in the memory 22. The operating system can be Windows, Unix, Linux, etc. The computer program 222, in addition to including a computer program capable of performing the automated guided vehicle positioning method executed by the electronic device as disclosed in any of the foregoing embodiments, may further include computer programs capable of performing other specific tasks. The data 223 may include data received by the electronic device from external devices, as well as data collected by its own input / output interface 25.

[0166] Furthermore, embodiments of this application also disclose a computer-readable storage medium storing a computer program. When the computer program is loaded and executed by a processor, it implements the method steps performed during the positioning process of the automated guided vehicle as disclosed in any of the foregoing embodiments.

[0167] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0168] The above provides a detailed description of the automatic guided vehicle positioning method, apparatus, equipment, and medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An automated guided vehicle positioning method, characterized by, include: The current feature value of the automated guided vehicle is obtained, and the target signal fingerprint corresponding to the current feature value is selected from the various signal fingerprints in the fingerprint library pre-built using LoRa devices. The historical coordinates in the target signal fingerprint are determined as the reference position coordinates of the automated guided vehicle; The reference optical flow vector of the automated guided vehicle is obtained using a preset optical flow device; The current target position coordinates of the automated guided vehicle are determined based on the reference position coordinates and the reference optical flow vector; Determining the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector includes: The distillation factor is determined based on the positive definiteness principle of Kalman filtering; A distillation matrix is ​​constructed using a distillation factor, and the reference position coordinates and the reference optical flow vector are filtered using the distillation matrix to obtain filtered reference position coordinates and filtered reference optical flow vector; the current target position coordinates of the automated guided vehicle are determined based on the filtered reference position coordinates and the filtered reference optical flow vector.

2. The automated guided vehicle positioning method of claim 1, wherein, Before obtaining the current feature value of the automated guided vehicle, the process also includes: The LoRa device is used to obtain the signal strength, link quality, and time difference between transmitted and received signals for each channel at each location point along the preset driving route. The coordinates of the location point, the signal strength, the link quality, and the time difference are stored in matrix form to obtain the corresponding signal fingerprint, and a fingerprint database of the driving route is constructed based on the signal fingerprint.

3. The automated guided vehicle positioning method of claim 2, wherein, Before acquiring the signal strength, link quality, and time difference between transmitted and received signals for each channel at each location point along the preset driving route using LoRa devices, the process also includes: The preset driving route is divided into several location points based on the starting point coordinates and the ending point coordinates of the preset driving route.

4. The automated guided vehicle positioning method of claim 1, wherein, The step of filtering the target signal fingerprint corresponding to the current feature value from various signal fingerprints in a fingerprint database pre-built using LoRa devices includes: Obtain each signal fingerprint from a fingerprint database pre-built using LoRa devices, and obtain the weight value of each signal fingerprint using a preset probability model; Based on the weight values ​​of each of the signal fingerprints, a particle filter algorithm is used to filter out the target signal fingerprint corresponding to the current feature value from each of the signal fingerprints.

5. The automated guided vehicle positioning method of claim 1, wherein, The step of obtaining the reference optical flow vector of the automated guided vehicle using a preset optical flow device includes: The previous frame image and the current frame image of the automated guided vehicle are obtained using a preset optical flow device. Select a tracking block from the previous frame image, and determine the prediction block corresponding to the tracking block from the current frame image; The reference optical flow vector is calculated using the tracking block and the prediction block.

6. An automated guided vehicle positioning apparatus, characterized by, include: The target fingerprint filtering module is used to obtain the current feature value of the automated guided vehicle and filter out the target signal fingerprint corresponding to the current feature value from each signal fingerprint in the fingerprint library pre-built using LoRa devices; A reference position determination module is used to determine the historical coordinates in the target signal fingerprint as the reference position coordinates of the automated guided vehicle; The reference optical flow acquisition module is used to acquire the reference optical flow vector of the automated guided vehicle using a preset optical flow device; The target position determination module is used to determine the current target position coordinates of the automated guided vehicle based on the reference position coordinates and the reference optical flow vector; The automated guided vehicle positioning device is specifically used for: According to the positive definite principle of Kalman filtering, the distillation factor is determined; a distillation matrix is ​​constructed using the distillation factor, and the reference position coordinates and the reference optical flow vector are filtered using the distillation matrix to obtain the filtered reference position coordinates and the filtered reference optical flow vector; the current target position coordinates of the automated guided vehicle are determined based on the filtered reference position coordinates and the filtered reference optical flow vector.

7. An electronic device, comprising: include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the automated guided vehicle positioning method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, Used to store a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the automated guided vehicle positioning method as described in any one of claims 1 to 5.

Images