Visible light indoor positioning system and positioning method suitable for LED dust deposition scene

Abstract

The invention relates to the technical field of indoor positioning, in particular to a visible light indoor positioning system and method suitable for an LED dust deposition scene, and the system comprises a plurality of light source emitters, a photoelectric detector and a control unit. Wherein the plurality of light source emitters are all mounted on a ceiling, the photoelectric detector is arranged at a point to be measured, and the light intensity of a reference point vertically projected by a light source and a plurality of calibration points equidistantly distributed along the line are measured by taking the plane where the photoelectric detector is located as a benchmark; and the control unit intelligently screens the optimal calibration point, dynamically constructs and optimizes a visible light channel transmission model by using a double-point and single-point method and combining the dust deposition influence, and accurately obtains target position data. According to the invention, optimization is carried out especially for the dust deposition scene of the visible light emitter, and the influence of dust deposition on the light emitting characteristic of the visible light emitter can be accurately reflected through regular measurement and correction, so that high-precision indoor positioning can still be realized in a dust deposition environment.

Description

Technical Field

[0001] The present invention relates to the field of indoor positioning technology, and in particular to a visible light indoor positioning method suitable for LED dust accumulation scenes. Background Art

[0002] With the rapid development of technologies such as the Internet of Things, robots, and autonomous driving, the demand for indoor positioning services is increasing. Large public places such as hospitals, shopping malls, and garages, as well as modern application scenarios such as mobile machinery in factories, smart logistics warehouses, and smart homes, all require accurate indoor positioning technology to support their daily operations. However, although the Global Navigation Satellite System (GNSS) has achieved great success in outdoor positioning, it cannot provide accurate indoor positioning in indoor environments because satellite signals are easily blocked by building walls and severely attenuated.

[0003] In order to solve this problem, a variety of wireless communication technologies have been tried to be applied to indoor positioning in recent years, including WiFi, 5G, Bluetooth, ultra-wideband and visible light. Among them, visible light positioning (VLP) technology has attracted much attention due to its unique advantages. VLP technology has the advantages of wide spectrum, no electromagnetic interference, high positioning accuracy, low cost, low energy consumption and harmlessness to the human body. It is considered to be one of the most promising indoor positioning technologies.

[0004] At present, a variety of VLP methods have been proposed and applied in practical scenarios. These methods can be roughly divided into proximity method, received signal strength method (RSS), angle of arrival method (AOA), and time of arrival (TOA) or time difference of arrival (TDOA) method. Among them, the RSS method has become the most widely used VLP positioning method due to its low implementation cost and high positioning accuracy. In order to improve positioning accuracy, researchers have adopted a variety of methods, such as data preprocessing based on machine learning, building more accurate RSS or fingerprint library models, and trying to use different VLP system architectures such as LED arrays and detector arrays.

[0005] However, although the existing VLP indoor positioning schemes have made significant progress in improving positioning accuracy, they do not take into account the situation of LED dust accumulation. In scenes where dust is easily deposited and the concentration is high, such as factories, warehouses, and underground garages, dust is easily accumulated on the surface of LEDs, which will change the luminous characteristics of LEDs, resulting in differences between the LED channel model in the dust accumulation scene and the ideal channel model in the existing VLP positioning scheme. This difference may cause positioning errors, thereby affecting the accuracy and reliability of VLP technology. Therefore, in future research, it is necessary to pay attention to the impact of LED dust accumulation on VLP positioning accuracy and explore corresponding solutions to reduce or eliminate this impact. Summary of the invention

[0006] To this end, the technical problem to be solved by the present invention is to overcome the problem that the prior art does not consider the dust accumulation problem of the light source transmitter, resulting in large errors in the positioning results obtained by the constructed visible light channel transmission model.

[0007] In order to solve the above technical problems, the present invention provides a visible light indoor positioning system suitable for LED dust accumulation scenes, comprising: a plurality of light source emitters, a photodetector and a control unit;

[0008] Wherein, the multiple light source transmitters are all installed on the ceiling, the photodetector is located at the target point to be measured, and a reference point of any light source transmitter perpendicular to the receiving plane of the photodetector is recorded as a first reference point. The photodetector measures the received light intensity of the first reference point and the received light intensity of multiple first calibration points equidistantly spaced on a straight line extending from the first reference point on the receiving plane;

[0009] The control unit selects an optimal calibration point from multiple first calibration points in a dust-free scene, and constructs a visible light channel transmission model in a dust-free scene based on the first reference point and the optimal calibration point to obtain estimated three-dimensional coordinates of a target point to be measured.

[0010] In one embodiment of the present invention, the control unit is further configured to: after the visible light indoor positioning system runs for a specified period, only use the optimal calibration point to update the visible light channel transmission model, and calculate the estimated three-dimensional coordinates of the target point to be measured based on the updated visible light channel transmission model.

[0011] The present invention also provides a visible light indoor positioning method applicable to LED dust accumulation scenes, which is applied to the above-mentioned visible light indoor positioning system applicable to LED dust accumulation scenes, and the method comprises the following steps:

[0012] S1: In a scene without dust accumulation, select the optimal calibration point from multiple first calibration points;

[0013] S2: In a dust accumulation scene, regularly measuring the received light intensity of the first reference point and the optimal calibration point, calculating a correction factor, and establishing a visible light channel transmission model according to the correction factor;

[0014] S3: Based on the visible light channel transmission model, the straight-line distance between the photodetector and the light source emitter is calculated to obtain the estimated three-dimensional coordinates of the target point to be measured.

[0015] In one embodiment of the present invention, the positioning method further includes:

[0016] S4: after the visible light indoor positioning system runs for a specified period, updating the correction factor only according to the received light intensity at different dust concentrations measured at the optimal calibration point to obtain an updated correction factor;

[0017] S5: updating the visible light channel transmission model in the dust accumulation scene according to the updated correction factor, and calculating the estimated three-dimensional coordinates of the target point to be measured according to the updated visible light channel transmission model.

[0018] In one embodiment of the present invention, in S1, the method steps of selecting the optimal calibration point from the plurality of first calibration points are as follows:

[0019] S11: In a dust-free scene, J first calibration points are taken at equal intervals on a straight line extending from the first reference point on the receiving plane, and the Lambertian order m in the dust-free scene is obtained according to the received light intensity of the J first calibration points. i :

[0020]

[0021] Wherein, d0 is the first reference point and the light source emitter LED i The distance, P r0 is the received light intensity measured at the first reference point, d j is the jth first calibration point and the light source emitter LED i distance, i represents the number of the light source emitter, i = 1, 2, ···, M, M is the total number of light source emitters; P rj is the received light intensity of the jth first calibration point, j = 1, 2, ···, J; P rj ,d j , P r0 , d0 follow the following relationship:

[0022]

[0023] S12: Calculate the Lambertian order m of each first calibration point according to the received light intensity of any first calibration point i,j :

[0024]

[0025] S13: Compare the Lambert order m of each first calibration point i,j and the Lambertian order m in the dust-free scene i The value of m i,j Closest to m i The first calibration point is recorded as Q point, that is, the optimal calibration point.

[0026] In one embodiment of the present invention, in S2, the method steps for establishing a visible light channel transmission model in a dust accumulation scene are as follows:

[0027] S21: Based on the optimal calibration point and the first reference point, respectively obtain the received light intensity in a dust-free scene and the received light intensity in a dust-accumulated scene to obtain the Lambertian order in the dust-accumulated scene Correction factor β:

[0028]

[0029] S22: According to the correction factor β, a correction factor α of the emission light intensity in the dust accumulation scene is obtained:

[0030]

[0031] S23: Based on the correction factor α and the correction factor β, a visible light channel transmission model in a dust accumulation scene is obtained, and its expression is:

[0032]

[0033] in, Indicates the received light intensity in dusty scenes, P ti represents the emission intensity in a dust-free scene, A is the response area of ​​the photodetector, d0 is the first reference point and the light source emitter LED i Distance; d Q Indicates the optimal calibration point to the light source emitter LED i The distance, P r0 represents the received light intensity measured at the first reference point in a dust-free scene, represents the received light intensity measured at the first reference point in the dust accumulation scene; P rQ It represents the received light intensity measured at the optimal calibration point in a dust-free scene. It represents the received light intensity measured at the optimal calibration point in the dust accumulation scene.

[0034] In one embodiment of the present invention, in S3, the method steps for obtaining the estimated three-dimensional coordinates of the target point to be measured are as follows:

[0035] S31: Based on the visible light channel transmission model in the dust accumulation scene, calculate the transmission distance from the photodetector to each light source emitter LED i The straight-line distance d i ;

[0036] S32: Each light source emitter LED i The coordinates are the center of the sphere, and the straight line distance d iThe spheres with a radius of intersect at one point, namely the target point to be measured, and the coordinates of the target point to be measured are calculated using the least squares method;

[0037] Among them, i represents the label of the light source emitter, i=1,2,···,M, and M is the total number of light source emitters.

[0038] In one embodiment of the present invention, the straight-line distance d i The calculation method is as follows:

[0039]

[0040] Where A is the response area of ​​the photodetector, m i represents the Lambert order in the scene without dust accumulation, and β represents the Lambert order in the scene with dust accumulation. The correction factor of the emission intensity in the dust scene is α, P ti represents the emission intensity, Indicates the received light intensity in dusty scenes, Δh i Indicates light source emitter LED i The vertical distance from the photodetector.

[0041] In one embodiment of the present invention, in S4, the method for obtaining the updated correction factor is as follows:

[0042] S41: Based on the optimal calibration point, obtain the received light intensity RSS under different dust concentrations i , construct the linear relationship between the received light intensity, correction factor and dust concentration:

[0043]

[0044] Among them, a1, a2, a3, b1, b2, and b3 are all fitting parameters, ρ is the dust concentration, α is the correction factor of the emission intensity in the dust scene, and β is the Lambertian order in the dust scene. The correction factor is , i represents the number of the light source emitter, i = 1, 2, ···, M, M is the total number of light source emitters;

[0045] S42: updating the correction factor according to the linear relationship to obtain an updated correction factor, that is, the correction factor α of the emission light intensity in the dust accumulation scene and the Lambertian order in the dust accumulation scene The correction factor β.

[0046] In one embodiment of the present invention, the positioning method also includes: when the light source transmitter is damaged or replaced with a homogeneous light source transmitter, using the linear relationship to calculate an updated correction factor, and updating the visible light channel transmission model through the updated correction factor, and based on the updated visible light channel transmission model, obtaining the estimated three-dimensional coordinates of the target point to be measured.

[0047] The above technical solution of the present invention has the following advantages compared with the prior art:

[0048] 1. The traditional Lambert model is modified by introducing correction factors α and β. The VLP channel transmission model in the dust accumulation scenario is constructed based on the modified Lambert model. This model can accurately reflect the impact of dust on RSS.

[0049] 2. Based on the channel model, a lightweight RSSVLP positioning solution is proposed in the case of dust accumulation. By measuring the RSS information of only two locations through the dual-point solution, the received signal strength and distance (RSS-d) curve in the dust accumulation scenario can be updated without re-establishing the complete RSS-d curve. This method not only significantly reduces the measurement and maintenance costs of the VLP system, but also effectively reduces the error caused by dust accumulation to the RSS VLP positioning solution, thereby improving the positioning accuracy.

[0050] 3. The linear relationship between the dust concentration at the optimal calibration point and the received signal strength and the correction factors α and β was explored in depth, and a more efficient single-point correction scheme was proposed on this basis. Under the premise of knowing the relationship curve between dust concentration and parameters, only the RSS of one position point needs to be measured to quickly construct a complete RSS-d relationship curve. This scheme further reduces the system measurement cost and makes RSSVLP positioning technology more portable and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0052] Figure 1 It is a structural schematic diagram of a visible light indoor positioning system applicable to LED dust accumulation scenes provided in the first embodiment of the present invention;

[0053] Figure 2 This is a flow chart of a visible light indoor positioning method applicable to LED dust accumulation scenarios provided in the second embodiment of the present invention;

[0054] Figure 3 This is a specific flow chart of a visible light indoor positioning method applicable to LED dust accumulation scenes provided in the second embodiment of the present invention;

[0055] Figure 4 is the Lambert order m provided in the second embodiment of the present invention i Calibration example graph;

[0056] Figure 5 (a) shows the relationship curve between the received light intensity and the dust concentration in the dust scene, (b) shows the relationship curve between the correction factor β and the dust concentration, and (c) shows the relationship curve between the correction factor α and the dust concentration;

[0057] Figure 6 (a) shows the block diagram of the transmitter implementation, and (b) shows the block diagram of the receiver implementation;

[0058] Figure 7 (a) shows the comparison between the RSS-d curve and the measured curve before and after the Lambert parameter calibration, and (b) shows the CDF diagram of the positioning error before and after the Lambert parameter calibration;

[0059] Figure 8 LED panels with different dust concentrations, where (a) indicates dust with a higher concentration and (b) indicates dust with a lower concentration;

[0060] Fig. 9 The RSS-d relationship curves under different dust concentrations, where (a) represents dust with a higher concentration and (b) represents dust with a lower concentration;

[0061] Fig.10 The CDF diagram of the positioning error, where (a) represents dust with a high concentration and (b) represents dust with a low concentration

[0062] Fig.11 It is a comparison diagram of RSS-d curves under different dust concentrations using the double-point scheme;

[0063] Fig.12 It is a comparison diagram of RSS-d curves under random dust accumulation concentration using single-point scheme;

[0064] Fig.13 Comparison of positioning results under Level 1 dust concentration, where (a) represents the positioning coordinate plane diagram, and (b) represents the positioning error box plot;

[0065] Fig.14 CDF diagram of positioning error of double-point scheme and single-point scheme under different dust concentrations;

[0066] Fig.15 The figure is a comparative bar chart of positioning errors under different dust concentrations. DETAILED DESCRIPTION

[0067] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0068] Embodiment 1

[0069] Reference Figure 1 As shown, the present invention provides a visible light indoor positioning system suitable for LED dust accumulation scenes, the system comprising: a plurality of light source emitters LED i (i=1,2,···,M, M is the total number of light source emitters), a photodetector (PD) and a control unit;

[0070] Among them, each light source emitter LED i They are installed vertically downward on the ceiling, and their three-dimensional coordinates are expressed as (x i ,y i ,z i The photoelectric detector (PD) is placed vertically on the ground, facing upward horizontally, and located at the target point to be measured. Its three-dimensional coordinates are expressed as (x p ,y p ,z p ). Each light source emitter LED i The vertical distance from PD is Δh i =z i -z p .

[0071] Any light source emitter LED i A reference point perpendicular to the receiving plane of the photodetector PD is recorded as a first reference point, and the photodetector measures the received light intensity of the first reference point and the received light intensity of a plurality of first calibration points equally spaced on a straight line extending from the first reference point on the receiving plane;

[0072] The control unit selects an optimal calibration point from a plurality of first calibration points in a dust-free scene, and uses a double-point method to obtain coordinate data of a target point to be measured in a dust-accumulated scene, that is, constructs a visible light channel transmission model according to the first reference point and the optimal calibration point, and calculates estimated three-dimensional coordinates of the target point to be measured according to the visible light channel transmission model;

[0073] Optionally, as dust accumulates on the light source transmitter, in order to further improve computing efficiency, after the visible light indoor positioning system runs for a specified period, the single-point method is used to obtain the coordinate data of the target point to be measured, that is: only the optimal calibration point is used to update the visible light channel transmission model in the dust accumulation scene, and the estimated three-dimensional coordinates of the target point to be measured are calculated based on the updated visible light channel transmission model.

[0074] Embodiment 2

[0075] In the process of visible light transmission, the signal transmission channel between the light source transmitting end LED and the receiving end PD is usually considered to be a free space channel, which mainly includes two types of transmission links: line-of-sight (LOS) transmission and non-line-of-sight (NLOS) transmission. In indoor scenarios, because more than 95% of the optical signal received by the PD end comes from the LOS component, the NLOS component is usually negligible. ti Indicator LED i The emitted light power, PD from LED i Received optical power P ri It can be expressed as:

[0076] P ri =P ti H i (0)(1)

[0077] Among them, H i (0) Yes LED i The DC channel gain of the LOS link between the PD and the PD. According to Lambert's radiation law:

[0078]

[0079] Among them, m i is the Lambert order, A is the response area of ​​PD, d i Indicator LED i Straight-line distance from PD, θ i , ψ represents LED i The radiation angle, incident angle and field of view (FOV) of the PD are as follows: Figure 1 shown. and Represents optical filter gain and optical concentrator gain respectively. In this patent, PD does not use optical filters and lenses. Considering that the transmitting end plane and the receiving end plane are parallel, the radiation angle θ i With the incident angle Equal, according to the geometric relationship of the triangle In summary, formula (1) can be rewritten as:

[0080]

[0081] In the above visible light transmission system, it is usually assumed that the LED is an ideal Lambertian radiation source, and m i = 1. However, in reality, LEDs come in a variety of forms, and their radiation characteristics may not conform to the ideal Lambert model, i.e., m i It is not always 1, which causes a large deviation between the theoretical prediction value of positioning and the actual result.

[0082] Therefore, the present invention also provides a visible light indoor positioning method applicable to LED dust accumulation scenes, which is applied to the visible light indoor positioning system applicable to LED dust accumulation scenes in the first embodiment. Figure 2 to Figure 4 As shown, the method comprises the following steps:

[0083] S1: In a scene without dust accumulation, select the optimal calibration point from multiple first calibration points;

[0084] S2: In a dust accumulation scenario, regularly measuring the first reference point and the optimal calibration point ( Figure 4 The received light intensity of the light source (denoted as point Q in the figure) is calculated, a correction factor is calculated, and a visible light channel transmission model is established according to the correction factor;

[0085] S3: Based on the visible light channel transmission model, the straight-line distance between the photodetector and the light source emitter is calculated to obtain the estimated three-dimensional coordinates of the target point to be measured;

[0086] Optionally, as dust accumulates on the light source emitter, in order to further improve the calculation efficiency, the positioning method further includes the following steps:

[0087] S4: after the visible light indoor positioning system runs for a specified period, updating the correction factor only according to the received light intensity at different dust concentrations measured at the optimal calibration point to obtain an updated correction factor;

[0088] S5: updating the visible light channel transmission model in the dust accumulation scene according to the updated correction factor, and calculating the estimated three-dimensional coordinates of the target point to be measured according to the updated visible light channel transmission model.

[0089] Specifically, Figure 4 As shown, in step S1, the method steps for selecting the optimal calibration point from a plurality of first calibration points are as follows:

[0090] S11: In a dust-free scene, J first calibration points are taken at equal intervals on a straight line extending from the first reference point on the receiving plane, and the Lambertian order m in the dust-free scene is obtained according to the received light intensity of the J first calibration points. i :

[0091]

[0092] Wherein, d0 is the first reference point and the light source emitter LED i The distance, P r0 is the received light intensity measured at the first reference point, d j is the jth first calibration point and the light source emitter LED idistance, i represents the number of the light source emitter, i = 1, 2, ···, M, M is the total number of light source emitters; P rj is the received light intensity of the jth first calibration point, j = 1, 2, ···, J; further, P rj d j , P r0 , d0 follow the following relationship:

[0093]

[0094] S12: Calculate the Lambertian order m of each first calibration point according to the received light intensity of any first calibration point i,j :

[0095]

[0096] S13: Compare the Lambert order m of each first calibration point i,j and the Lambertian order m in the dust-free scene i The value of m i,j Closest to m i The first calibration point is recorded as the optimal calibration point, that is, the Q point.

[0097] Furthermore, in S2, the method steps for establishing a visible light channel transmission model in a dust accumulation scene are as follows:

[0098] S21: Based on the Q point and the first reference point, respectively obtain the received light intensity in a dust-free scene and the received light intensity in a dust-accumulated scene to obtain the Lambertian order in the dust-accumulated scene Correction factor β:

[0099]

[0100] Among them, d Q Indicates the optimal calibration point to the light source emitter LED i The distance, P r0 represents the received light intensity measured at the first reference point in a dust-free scene, represents the received light intensity measured at the first reference point in the dust accumulation scene; P rQ It represents the received light intensity measured at the optimal calibration point in a dust-free scene. It represents the received light intensity measured at the optimal calibration point in the dust accumulation scene.

[0101] In formula (7) and when no dust accumulation is predicted, m i Based on the Lambertian order in the dust accumulation scene It can be expressed as:

[0102]

[0103] Lambert order based on dust accumulation scene The received light intensity at point Q and the distance d Q The relationship is:

[0104]

[0105] Comparing equation (9) with the received light intensity at point Q in the dust-free scene, we can further obtain LED i The expression of the emitted light intensity under dust accumulation is:

[0106]

[0107] S22: According to the correction factor β, a correction factor α of the emission light intensity in the dust accumulation scene is obtained:

[0108]

[0109] Simplify formula (10) to:

[0110]

[0111] S23: Based on the correction factor α and the correction factor β, a visible light channel transmission model in a dust accumulation scene is obtained, and its expression is:

[0112]

[0113] in, Indicates the received light intensity in dusty scenes, P ti represents the emission light intensity in a dust-free scene, and A is the response area of ​​the photodetector.

[0114] According to the above technical solution, the receiving end regularly switches the LED i Measure the received signal strength at the reference point and Q point directly below and The correction factors α and β are calculated to establish the VLP channel model in the dust accumulation scenario. This eliminates the need to repeatedly measure the RSS information of the complete J calibration points, greatly reducing the implementation complexity and measurement cost of the system.

[0115] In this embodiment, in S3, the method steps for obtaining the three-dimensional estimated position of the target point to be measured based on the visible light channel transmission model are as follows:

[0116] S31: When the receiving end PD enters the test point, first extract the LED of each light source transmitter i Light intensity information Based on the visible light channel transmission model in the dust accumulation scene represented by the above formula (13), the transmission distance from the photodetector to each light source emitter LED is calculated. i The straight-line distance d i :

[0117]

[0118] Where A is the response area of ​​the photodetector, m i represents the Lambert order in the scene without dust accumulation, and β represents the Lambert order in the scene with dust accumulation. The correction factor of the emission intensity in the dust scene is α, P ti represents the emission intensity, Indicates the received light intensity in dusty scenes, Δh i Indicates light source emitter LED i a vertical distance from the photodetector;

[0119] S32: Each light source emitter LED i The coordinates are the center of the sphere, and the straight line distance d i The spheres with a radius of intersect at one point, which is the target point to be measured (x p ,y p ,z p ). According to the principle of three-sided positioning, the coordinates of the target point to be measured can be calculated by the following equations:

[0120]

[0121] Since the actual straight-line distance d i The estimated value of is usually inaccurate, which increases the difficulty of solving equation (11). Therefore, the RSSVLP algorithm usually uses the least squares method to estimate the target point coordinates X R '=(x p ',y p ',z p ') T ,have:

[0122] X' R =(T T T) -1 T T D(16)

[0123] in,

[0124]

[0125] (x i ,y i ,z i ) is the light source emitter LED iThe three-dimensional coordinates, d i Indicates light source emitter LED i The straight-line distance from the photodetector, i=1, 2,..., M, where M is the total number of light source emitters.

[0126] Further, in S4, after the visible light indoor positioning system runs for a specified period, the correction factor is updated only according to the received light intensity at different dust concentrations measured at the optimal calibration point to obtain an updated correction factor, and the method steps are as follows:

[0127] S41: Based on the optimal calibration point, obtain the received light intensity RSS under different dust concentrations ρ i It is found that the correction factors α and β have an approximately linear correlation with the dust concentration ρ, and the received signal strength at this point also decreases approximately linearly with the increase of dust concentration, such as Figure 5 Therefore, the received light intensity RSS is constructed i , the linear relationship between the correction factors α, β and the dust concentration ρ:

[0128]

[0129] Among them, a1, a2, a3, b1, b2, and b3 are all fitting parameters, ρ is the dust concentration, α is the correction factor of the emission intensity in the dust scene, and β is the Lambertian order in the dust scene. Correction factor for

[0130] S42: updating the correction factor according to the linear relationship to obtain an updated correction factor, that is, the correction factor α of the emission light intensity in the dust accumulation scene and the Lambertian order in the dust accumulation scene The correction factor β.

[0131] In this way, the receiving end PD measures the received light intensity at point Q. After that, the correction factors α and β can be updated only according to formula (17); this means that the channel model in the dust accumulation scene no longer needs to measure the RSS information of the vertical reference point (i.e., the aforementioned “first reference point”). Only the Q point can be used for measurement. Compared with the aforementioned two-point measurement method of the vertical reference point and the Q point, the workload of data collection is further reduced, achieving a lighter-weight positioning.

[0132] In addition, the positioning method also includes: when the light source transmitter is damaged or replaced with a homogeneous light source transmitter, the previously constructed linear relationship is used to directly adopt a single-point method (only using the Q point) to measure and calculate the updated correction factor without starting all over again.

[0133] In order to verify the effectiveness of the technical solution of the present invention, a 3m×3m×3m indoor positioning hardware experimental platform was built. Figure 6 The four LEDs at the transmitting end are installed on a horizontal plane at a height of 3m, with a power of 25W each, and coordinates of (2.6, 2.6, 2.65), (2.6, 0.4, 2.65), (0.4, 2.6, 2.65), and (0.4, 0.4, 2.65). Figure 6 (a) summarizes the implementation of the transmitter. The microcontroller outputs a square wave signal, which is amplified by a linear power amplifier (OPA541) to drive the LED to send a light signal and provide normal indoor lighting. Each LED sends a square wave signal of different frequencies to work together. When setting the frequency, it is necessary to avoid the superposition of the harmonic components of each LED signal, which will affect the extraction of the received signal strength by the receiving end, and at the same time, avoid the damage to the human eye caused by slow flickering of the light. The signal frequencies set in the experiment are 22.5kHz, 32kHz, 62.5kHz and 72kHz respectively.

[0134] The receiving end photodiode PD (Thorlabs FDS1010) is facing upward horizontally, 80cm from the ground. Figure 6 As shown in (b), the PD receives the mixed light signal from the LED and converts it into a current signal. The current signal is converted into a voltage signal through the IV conversion (LT1818IS8) module, and then enters the microcontroller through the amplifier circuit. The microcontroller collects the signal at a sampling frequency of 512kHz (number of sampling points = 1024) and performs a fast Fourier transform FFT to extract the signal amplitude at each fundamental frequency as the received signal strength value of the corresponding LED. In the experiment, the 3m×3m positioning plane was divided into small squares of 20cm×20cm. Every 80cm interval in the horizontal and vertical directions is used as a positioning point, with a total of 36 points to be tested.

[0135] If the LED is a non-ideal Lambertian light source, the Lambertian order m needs to be i Calibration is performed to achieve more accurate RSS positioning. To this end, in the initial dust-free condition, the LED i Eleven equidistant points are selected on the diagonal line as calibration points. First, i To reduce the measurement error, the median average filtering method was used to remove the maximum and minimum values ​​of the 20 measurement results and take the average value as the received light intensity at each point.

[0136] from Figure 7 (a) It can be seen that the RSS-d curve corresponding to the calibrated Lambert model is closer to the measured value than that without calibration. Figure 7(b) shows the CDF graph of the positioning error. When the Lambertian parameters are not calibrated, 80% of the positioning errors are within 12.5 cm after calibration, and the maximum error can reach more than 20 cm. After calibration, 80% of the positioning errors are within 5 cm, and the maximum positioning error is within 7.5 cm. The experimental results show that calibrating the Lambertian parameters greatly improves the positioning accuracy and reduces the positioning error.

[0137] In order to verify the influence of dust on the RSSVLP positioning scheme proposed in this invention, soil mixture and single component silicon dioxide SiO2 were used to simulate different indoor dusts. For each type of dust, the positioning errors of four LED dust concentrations were verified, namely Level 1 (0.0062g / dm2), Level 2 (0.0124g / dm2), Level 3 (0.0248g / dm2) and Level 4 (0.0496g / dm2). Figure 8 shown.

[0138] Depend on Fig. 9 It can be seen that with the accumulation of dust, the received light intensity at the PD end weakens, and the fitted RSS-d curve decreases gradually, deviating more and more from the initial fitting curve when there is no dust. Fig.10 The impact of the deviation of this channel model on positioning accuracy is clearly demonstrated. When there is no dust, 80% of the positioning errors are less than 5cm, and the maximum positioning error does not exceed 8cm. In the lightest dust accumulation scenario (Level 1), regardless of whether the dust concentration is high or low, 80% of the positioning error lines are close to 15cm, and the maximum positioning error exceeds 20cm. When the dust accumulation is heavy (Level 4), the minimum error exceeds 5cm, and the maximum can reach 40cm.

[0139] from Fig.11 It can be seen that the existing RSSVLP scheme does not consider the influence of dust. The channel model established when there is no dust at the beginning will deviate more and more from the actual measurement value as dust accumulates. Fig.11 The RSS information of the reference point and the Q point is measured periodically, and then the correction factors α and β are calculated to establish the channel model under the current dust accumulation concentration. Compared with measuring J+1 points to rebuild a complete channel model, the dual-point method of this patent greatly reduces the measurement workload and maintenance cost of the system. And Fig.11 It clearly shows that the RSS-d curve established by the dual-point method of this patent has a high degree of match with the actual measurement value, proving the accuracy of the channel model proposed in this patent.

[0140] Fig.12 This is a verification of the single-point solution proposed in this patent. After the VLP system has been running for a certain period of time, Figure 5 The relationship curve between dust accumulation concentration and correction factor. At this time, only the RSS of point Q is measured in a random dust accumulation scene. According to formula (17) and Figure 5 The relationship curve of the channel model is established. Fig.12 It can be seen that the channel model established by the single-point method can still well reflect the influence of dust and maintain a high degree of fit with the measured RSS curve. At the same time, compared with the double-point method, the measurement workload is further reduced.

[0141] In order to observe the performance of the positioning solution more intuitively, Fig.13 (a) First, taking the lightest dust accumulation scene as an example (Level 1), the actual positioning effects of the RSSVLP scheme (two-point method) proposed in this patent and the existing typical RSSVLP scheme at each test point are compared. It can be seen that the positioning results of the scheme proposed in this patent at each test point (marked with '*' in the figure) are relatively closer to the actual position (marked with '☆' in the figure), and in most cases they are almost identical. Fig.13 (b) gives the statistical results of the positioning errors of the two schemes. It can be seen that in the case of slight dust accumulation, the median positioning error of the existing typical RSSVLP positioning scheme exceeds 10cm, and the maximum positioning error exceeds 20cm, which has a great impact on the VLP scheme whose error is usually on the order of 10cm. The median positioning error of the RSSVLP scheme proposed in this patent is close to 4cm, the minimum positioning error is less than 2cm, and the maximum positioning error does not exceed 10cm, which significantly improves the positioning accuracy.

[0142] Fig.14 The CDF values ​​of the positioning errors of the RSSVLP schemes in different dust accumulation scenarios were compared, including the dual-point and single-point schemes proposed in this patent. As dust accumulates, the maximum positioning errors of each VLP scheme increase. Because the single-point scheme is only based on Q-point RSS fitting, the amount of information is small and the error is slightly higher than that of the dual-point scheme. However, compared with the existing typical RSSVLP schemes, the single-point and dual-point schemes proposed in this patent have achieved higher positioning accuracy in various scenarios, with the maximum error always controlled within 10 cm, and 80% of the positioning errors not exceeding 7 cm.

[0143] Fig.15 The average positioning errors obtained by each scheme were further compared, and compared with the RSSVLP positioning error in the dust-free case. It can be seen that the average error of the dual-point scheme proposed in this patent does not exceed 5cm, and the single-point scheme does not exceed 6cm, and the effect is stable in various dust accumulation scenarios, which is very close to the positioning accuracy in the dust-free case. Compared with the existing typical RSSVLP scheme, the positioning accuracy is improved by more than 50%, and the higher the dust concentration, the more obvious the advantage. This shows that the scheme proposed in this patent effectively suppresses the influence of dust on the VLP positioning accuracy.

[0144] Embodiment 3

[0145] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the visible light indoor positioning method for LED dust accumulation scenarios described in Example 2 is implemented.

[0146] Embodiment 4

[0147] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, instructions for implementing the visible light indoor positioning method applicable to LED dust accumulation scenarios described in Embodiment 2 are implemented.

[0148] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0149] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0150] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0151] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0152] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A visible light indoor positioning system suitable for LED dust accumulation scenes, characterized in that: include: a plurality of light source emitters, a photodetector and a control unit; Wherein, the multiple light source transmitters are all installed on the ceiling, the photodetector is located at the target point to be measured, and a reference point of any light source transmitter perpendicular to the receiving plane of the photodetector is recorded as a first reference point. The photodetector measures the received light intensity of the first reference point and the received light intensity of multiple first calibration points equidistantly spaced on a straight line extending from the first reference point on the receiving plane; The control unit selects an optimal calibration point from multiple first calibration points in a dust-free scene, and in a dust-accumulated scene, constructs a visible light channel transmission model based on the first reference point and the optimal calibration point to obtain the estimated three-dimensional coordinates of the target point to be measured.

2. The visible light indoor positioning system suitable for LED dust accumulation scenes according to claim 1, characterized in that: The control unit is further configured to: after the visible light indoor positioning system runs for a specified period, update the visible light channel transmission model using only the optimal calibration point, and calculate the estimated three-dimensional coordinates of the target point to be measured based on the updated visible light channel transmission model.

3. A visible light indoor positioning method applicable to LED dust accumulation scenes, applied to the visible light indoor positioning system applicable to LED dust accumulation scenes as claimed in claim 1 or claim 2, characterized in that: The following steps are involved: S1: In a scene without dust accumulation, select the optimal calibration point from multiple first calibration points; S2: In a dust accumulation scene, regularly measure the received light intensity of the first reference point and the optimal calibration point, calculate a correction factor, and establish a visible light channel transmission model according to the correction factor; S3: Based on the visible light channel transmission model, the straight-line distance between the photodetector and the light source emitter is calculated to obtain the estimated three-dimensional coordinates of the target point to be measured.

4. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 3 is characterized in that: The positioning method further includes: S4: after the visible light indoor positioning system runs for a specified period, updating the correction factor only according to the received light intensity at different dust concentrations measured at the optimal calibration point to obtain an updated correction factor; S5: updating the visible light channel transmission model in the dust accumulation scene according to the updated correction factor, and calculating the estimated three-dimensional coordinates of the target point to be measured according to the updated visible light channel transmission model.

5. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 3, characterized in that: In S1, the method steps of selecting the optimal calibration point from the multiple first calibration points are as follows: S11: In a dust-free scene, J first calibration points are taken at equal intervals on a straight line extending from the first reference point on the receiving plane, and the Lambertian order m in the dust-free scene is obtained according to the received light intensity of the J first calibration points. i : Wherein, d0 is the first reference point and the light source emitter LED i The distance, P r0 is the received light intensity measured at the first reference point, d j is the jth first calibration point and the light source emitter LED i distance, i represents the number of the light source emitter, i = 1, 2, ···, M, M is the total number of light source emitters; P rj is the received light intensity of the jth first calibration point, j = 1, 2, ···, J; P rj ,d j , P r0 , d0 follow the following relationship: S12: Calculate the Lambertian order m of each first calibration point according to the received light intensity of any first calibration point i,j : S13: Compare the Lambert order m of each first calibration point i,j and the Lambertian order m in the dust-free scene i The value of m i,j Closest to m i The first calibration point is recorded as the optimal calibration point.

6. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 3, characterized in that: In S2, the method steps for establishing a visible light channel transmission model in a dust accumulation scene are as follows: S21: Based on the optimal calibration point and the first reference point, respectively obtain the received light intensity in a dust-free scene and the received light intensity in a dust-accumulated scene to obtain the Lambertian order in the dust-accumulated scene Correction factor β: S22: According to the correction factor β, a correction factor α of the emission light intensity in the dust accumulation scene is obtained: S23: Based on the correction factor α and the correction factor β, a visible light channel transmission model in a dust accumulation scene is obtained, and its expression is: in, Indicates the received light intensity in dusty scenes, P ti represents the emission intensity in a dust-free scene, A is the response area of ​​the photodetector, d0 is the first reference point and the light source emitter LED i Distance; d Q Indicates the optimal calibration point to the light source emitter LED i The distance, P r0 represents the received light intensity measured at the first reference point in a dust-free scene, represents the received light intensity measured at the first reference point in the dust accumulation scene; P rQ It represents the received light intensity measured at the optimal calibration point in a dust-free scene. It represents the received light intensity measured at the optimal calibration point in the dust accumulation scene.

7. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 3, characterized in that: In S3, the method steps for obtaining the estimated three-dimensional coordinates of the target point to be measured are as follows: S31: Based on the visible light channel transmission model in the dust accumulation scene, calculate the transmission distance from the photodetector to each light source emitter LED i The straight-line distance d i ; S32: Each light source emitter LED i The coordinates are the center of the sphere, and the straight line distance d i The spheres with a radius of intersect at one point, namely the target point to be measured, and the estimated three-dimensional coordinates of the target point to be measured are calculated using the least squares method; Among them, i represents the label of the light source emitter, i=1,2,···,M, and M is the total number of light source emitters.

8. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 7, characterized in that: The straight-line distance d i The calculation method is as follows: Where A is the response area of ​​the photodetector, m i represents the Lambert order in the scene without dust accumulation, and β represents the Lambert order in the scene with dust accumulation. The correction factor of the emission intensity in the dust scene is α, P ti represents the emission intensity, Indicates the received light intensity in dusty scenes, Δh i Indicates light source emitter LED i The vertical distance from the photodetector.

9. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 4, characterized in that: In S4, the method for obtaining the updated correction factor is as follows: S41: Based on the optimal calibration point, obtain the received light intensity RSS under different dust concentrations i , construct the linear relationship between the received light intensity, correction factor and dust concentration: Among them, a1, a2, a3, b1, b2, and b3 are all fitting parameters, ρ is the dust concentration, α is the correction factor of the emission intensity in the dust scene, and β is the Lambertian order in the dust scene. The correction factor is , i represents the number of the light source emitter, i = 1, 2, ···, M, M is the total number of light source emitters; S42: updating the correction factor according to the linear relationship to obtain an updated correction factor, that is, the correction factor α of the emission light intensity in the dust accumulation scene and the Lambertian order in the dust accumulation scene The correction factor β.

10. The visible light indoor positioning method applicable to LED dust accumulation scenes according to claim 9, characterized in that: The positioning method also includes: when the light source transmitter is damaged or replaced with a homogeneous light source transmitter, using the linear relationship to calculate an updated correction factor, and updating the visible light channel transmission model by the updated correction factor, and based on the updated visible light channel transmission model, obtaining the estimated three-dimensional coordinates of the target point to be measured.

Images