A method and system for visible light positioning of fast moving carriers

Abstract

The application discloses a visible light positioning method for a fast-moving carrier, which is used for positioning of the fast-moving carrier with a photodiode as a signal receiver and comprises the following steps: calculating angular velocity and velocity in a Fourier time window according to state variables and inertial navigation data; calculating a position vector and a normal vector according to the state variables, the inertial navigation data and LED lamp positions; calculating a correction number of a received signal strength according to the angular velocity, the velocity, the position vector and the normal vector and in combination with a received signal strength obtained by a discrete Fourier transform; and optimizing the state variables according to the corrected received signal strength and the inertial navigation data, and outputting the position, the velocity and the attitude of the photodiode at a current time according to the optimized state variables. The application integrates the inertial navigation observation data and the state data in the window time of the discrete Fourier transform, restores the average RSS into the instantaneous RSS, and improves the navigation and positioning capability of the fast-moving carrier.

Description

Technical Field

[0001] This invention relates to the field of visible light positioning technology, and more particularly to the positioning of fast-moving vehicles using photodiodes as signal receivers, specifically a visible light positioning method and system for fast-moving vehicles. Background Technology

[0002] Visible light positioning technology is a novel high-precision indoor positioning technology that can provide centimeter-level or decimeter-level navigation and positioning services in indoor areas, and has advantages such as low cost, environmental friendliness, and low power consumption. In visible light positioning schemes, a common method is to measure the intensity of the light signal emitted by an LED using a photodiode (PD), and then calculate the distance using a Lambertian model to determine the PD's position. This method typically uses frequency division multiplexing to modulate different LEDs, allowing the PD to distinguish light signals from different LEDs. Visible light positioning systems using this method require signal extraction through discrete Fourier transform or similar methods. However, for rapidly moving objects, the light signal of each LED changes rapidly, and the discrete Fourier transform method, due to its reliance on time windows, struggles to accurately obtain instantaneous light intensity.

[0003] Numerous visible light positioning solutions have been designed for fast-moving vehicles. However, currently available literature and patents primarily focus on camera-based visible light positioning systems. For instance, the "High-Speed ​​Visible Light Positioning Method, System, and Medium Based on BN-CNN" invented by Fang Junbin et al. from Jinan University uses convolutional neural networks to address image blurring caused by high-speed vehicle movement. The "Visible Light Positioning Method and System Based on Enhanced Visual Target Tracking" invented by Liu Xiangyu et al. from Chongqing University of Posts and Telecommunications incorporates a moving target detection module to track LEDs in images during high-speed movement. However, there are no solutions for visible light positioning systems equipped with photodetectors specifically designed for fast-moving vehicles. While photodetector-based visible light positioning systems have been widely applied and discussed in academia and industry, the positioning effectiveness of such systems will be limited without correction and compensation of the observed values.

[0004] Therefore, proposing a visible light positioning technology for a fast-moving carrier equipped with a PD is of great significance in this field. Summary of the Invention

[0005] To improve the navigation and positioning capabilities of fast-moving vehicles, this invention provides a visible light positioning method for fast-moving vehicles. Since the received signal strength (RSS) required for visible light positioning in a computer system is obtained through Discrete Fourier Transform (DFT), and its value is the average of all RSS values ​​within the DFT time window, this method integrates the state data within the DFT window time, corrects the average RSS, and approximates the instantaneous RSS.

[0006] According to one aspect of the present invention, a visible light positioning method for a fast-moving vehicle is provided, for positioning a fast-moving vehicle using a PD as a signal receiver, comprising:

[0007] Calculate the angular velocity and velocity within the Fourier time window based on state variables and inertial measurement unit (IMU) data;

[0008] The position vector and normal vector are calculated based on the state variables, IMU data, and LED light positions.

[0009] The correction of RSS is calculated based on the angular velocity, velocity, position vector, and normal vector, combined with the RSS obtained by the discrete Fourier transform.

[0010] The state variables are jointly optimized based on the corrected RSS and IMU data, and the position, velocity and attitude of the PD at the current moment are output based on the optimized state variables.

[0011] As a further technical solution, the method also includes:

[0012] The state variables are constructed, including: the position and velocity of the volume downloaded in the indoor coordinate system, the attitude quaternion of the carrier from the coordinate system of the visible light receiver surface to the indoor coordinate system, and the accelerometer zero bias and gyroscope zero bias in the coordinate system of the visible light receiver surface.

[0013] As a further technical solution, the angular velocity and velocity within the Fourier time window are calculated based on the state variables and IMU data, including:

[0014] The direction cosine matrix is ​​obtained by transforming the attitude quaternion in the state variables. Based on the direction cosine matrix, combined with the gyroscope observations and the gyroscope zero bias in the coordinate system of the visible light receiver surface, the angular velocity within the Fourier time window is obtained.

[0015] Based on the velocity at the previous moment, combined with the direction cosine matrix, IMU sampling interval, accelerometer observations, and accelerometer zero bias in the coordinate system of the visible light receiver surface, the velocity within the Fourier time window is obtained.

[0016] As a further technical solution, the position vector and normal vector are calculated based on the state variables, IMU data, and LED light positions, including:

[0017] The position of the volume in the indoor coordinate system and the attitude quaternion of the carrier transformed from the coordinate system of the visible light receiver surface to the indoor coordinate system are directly used in the state variables at time k to calculate the position vector and normal vector.

[0018] As a further technical solution, the position vector and normal vector are calculated based on the state variables, IMU data, and LED light positions, including:

[0019] The position of the PD is calculated based on the speed, and the position vector from the PD to the LED is calculated based on the PD position and the LED position.

[0020] The attitude quaternion is derived from the angular velocity, and the normal vector is calculated from the attitude quaternion.

[0021] As a further technical solution, calculating the RSS correction also includes:

[0022]

[0023] Where, m l Let be the Lambertian coefficient of the l-th LED light. and n u These are the plane normal vectors of the l-th LED and the PD, respectively, pointing upwards; D is the distance vector from PD to the l-th LED. l It is its model; P l RSS represents the value obtained by the discrete Fourier transform, and T represents the length of the Fourier time window.

[0024] As a further technical solution, the state variables are jointly optimized based on the corrected RSS and IMU data, including:

[0025] The RSS obtained from the Discrete Fourier Transform is corrected based on the corrected RSS to obtain the corrected RSS observation.

[0026] Based on the corrected RSS and IMU data, an objective function to be optimized is constructed to perform state estimation and achieve navigation and positioning.

[0027] According to one aspect of the present invention, a visible light positioning system for a fast-moving vehicle is provided for positioning a fast-moving vehicle with a PD as a signal receiver, comprising: a plurality of LEDs, a PD, an IMU, and a processor for receiving PD signals; the PD and IMU are placed on the fast-moving vehicle, and the processor is used to execute the steps of the visible light positioning method for the fast-moving vehicle according to the signals received by the PD and the IMU data.

[0028] According to one aspect of the present invention, a visible light positioning device for a fast-moving vehicle is provided, comprising a memory and a processor, the memory storing program instructions executed by the processor, the processor invoking the program instructions to execute the visible light positioning method for a fast-moving vehicle.

[0029] According to one aspect of the present invention, a non-transitory computer-readable storage medium is provided, the non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the visible light positioning method for a fast-moving carrier.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] This invention corrects the RSS (Real-Side Scale) using IMU (Infrastructure Utilization) data and state variables, thereby improving the navigation and positioning capabilities of fast-moving vehicles. This invention incorporates a portion of the IMU's output to assist visible light positioning in correcting observations, thus obtaining more accurate instantaneous RSS observations. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A time relationship diagram provided for an embodiment of the present invention.

[0034] Figure 2 This is a schematic flowchart of a visible light positioning method for a rapidly moving carrier provided in an embodiment of the present invention.

[0035] Figure 3 This is a schematic diagram of a visible light positioning system for a fast-moving carrier provided in an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention 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, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined to form new technical solutions. Such combinations are not bound by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0037] This invention provides a visible light positioning method for a fast-moving vehicle. First, the angular velocity and velocity within a Fourier time window are calculated based on state variables and IMU data. Next, the position vector and normal vector are calculated based on the state variables, IMU data, and LED position. Then, the RSS correction is calculated based on the angular velocity, velocity, position vector, and normal vector, combined with the received signal strength obtained from the Discrete Fourier Transform. Finally, the state variables are jointly optimized based on the corrected RSS and IMU data, and the angular velocity and velocity are calculated again at the next state time. Simultaneously, the position, velocity, and attitude of the PD at the current time are output based on the optimized state variables.

[0038] The visible light positioning system for a fast-moving carrier used in this embodiment of the invention includes:

[0039] The first LED light, the second LED light, ..., the Lth LED light, PD receiver, IMU, drone;

[0040] The first LED light, the second LED light, ..., the Lth LED light are installed on the indoor ceiling;

[0041] like Figure 3 As shown, the PD receiver includes a PD and a microcontroller for analog-to-digital conversion (ADC). The PD receiver and IMU are placed on the UAV, and the PD receiver and IMU are wired to the computing unit on the UAV. The computing unit is wirelessly connected to the server and is used to transmit data to the server for navigation and positioning calculations, while receiving motion control provided by the server.

[0042] Figure 2 This is a flowchart of a visible light positioning method for a rapidly moving carrier, provided as an embodiment of the present invention.

[0043] In this embodiment, the length T of the Fourier time window is 1s, the time interval between state variables is 1s, the navigation system uses graph optimization as the state estimation method, and the IMU data is processed by integration using the predictive sub-model.

[0044] In this embodiment, the coordinate system is defined as follows: subscript v k Indicates t k The visible light coordinate system (v system) at any given time, with the subscript u representing the indoor local coordinate system (u system).

[0045] In this embodiment, the state variables are configured as follows:

[0046]

[0047] p u v u These represent the position and velocity of the carrier in the U-system, respectively. For the attitude quaternion of the carrier from the v system to the u system, b a and b g For accelerometer zero bias and gyroscope zero bias in the v-frame.

[0048] This embodiment calculates the state variables for the next time step using the state variables from the previous time step, serving as the initial values ​​for subsequent optimization. This process only requires a rough estimate, and the method is not unique; one optional recursive method is given below:

[0049]

[0050] in, Indicates t k to t k+1 IMU pre-integration between time points, g u It is the local gravitational acceleration.

[0051] The recursive calculation method for the expected score is as follows:

[0052]

[0053] in They represent from t k Time to t i+1 Pre-integration of velocity, acceleration, and angular velocity at time t. They represent from t k Time to t i Pre-integration of velocity, acceleration, and angular velocity at time t i and t i+1 Located at visible light observation time t k and t k+1 The time interval with IMU observations between. and t i The accelerometer and gyroscope data at each moment need to be converted to the v-frame, where δt is the time difference between two adjacent IMU observations. and These represent the zero bias of the angular velocity meter and the gyroscope, respectively. R(·) is a function that converts the quaternion into a direction cosine matrix. It is quaternion multiplication.

[0054] Specifically, the method includes the following steps:

[0055] S1. Calculate the angular velocity and velocity within the Fourier time window based on the state variables and IMU data. The angular velocity can be obtained directly from the gyroscope observations after adding a zero-bias correction.

[0056]

[0057] The velocity can be obtained stepwise recursively from the accelerometer observations after zero bias correction:

[0058]

[0059] in The direction cosine matrix can be obtained through attitude quaternion transformation, where δt is the IMU sampling interval (e.g., ...). Figure 1 As shown), and These are the gyroscope and accelerometer observations at time i, respectively.

[0060] S2. Calculate the position vector and normal vector based on the state variables, IMU data, and LED positions.

[0061] Alternatively, in S2, p in the state variables at time k can be used directly. u and Calculate the position vector and normal vector. Position vector The normal vector is obtained by subtracting the PD coordinates from the LED coordinates. Obtained through calculation.

[0062] Alternatively, to improve temporal resolution, the PD position can be calculated from the velocity in S2, and the attitude quaternion can be calculated from the angular velocity.

[0063]

[0064] And calculate the position vector from PD to LED. Based on attitude quaternions Calculate the normal vector.

[0065] In this embodiment, the PD position is calculated recursively based on velocity, and the attitude quaternion is calculated based on angular velocity. Position vector The plane normal vector of the l-th LED is obtained by subtracting the PD coordinates from the LED coordinates. The plane normal vector n of PD u It can be achieved through attitude quaternions calculate:

[0066]

[0067] S3. The formula for calculating the RSS motion correction based on velocity, angular velocity, position vector, and normal vector is as follows:

[0068]

[0069] The relationship between the various times is as follows: Figure 1 As shown, m l Let m be the Lambertian coefficient of the l-th LED. l This information can be obtained through the manufacturer's technical documentation or through independent calibration. and n u These are the plane normal vectors of the l-th LED and the PD, respectively, pointing upwards; D is the distance vector from the l-th LED to the PD. l It is its model; P l RSS represents the value obtained by the discrete Fourier transform, and T represents the length of the Fourier time window.

[0070] S4. Optimize the state variables jointly based on the corrected RSS and IMU data.

[0071] S41. Correct the RSS observations based on the correction calculated according to equation (11):

[0072] P l corr =P l +ΔP l (12)

[0073] Calculate the normal vector and displacement vector based on the state variables, and then calculate the predicted RSS observation value:

[0074]

[0075] Where m l The Lambertian coefficient of the LED light, A R For the effective area of ​​PD, P Tl The luminous power of the LED lamp can be obtained by consulting the product manual. The difference between the predicted RSS observation value and the corrected actual observation value is used to obtain the visible light residual, and the Jacobian matrix of the visible light residual is calculated.

[0076] Specifically, the expression for calculating the visible light residual is as follows:

[0077]

[0078] Where X = [x0, x1, ..., x n ] is a vector consisting of all state variables within the sliding window. It is t k The RSS observation value of the l-th LED at time l.

[0079] S42. The pre-integration residual calculated based on the state variables at two adjacent time points and the IMU pre-integration is:

[0080]

[0081] in t represents the definition of equations (3)-(5) k to t k+1 IMU pre-integration between time steps. The Jacobian matrix of the pre-integrated residuals can be easily obtained by differentiation.

[0082] S43. Select the visible light residual and the pre-integration residual within the sliding window range, construct the cost function by the weighted sum of squares of the visible light residual and the pre-integration residual, and optimize the state variables based on the Jacobian matrix of the visible light residual and the Jacobian matrix of the pre-integration residual.

[0083] The expression for the cost function is:

[0084]

[0085] The sliding window has a length of n+1 and contains n+1 state variables, X = [x0, x1, ..., x n ], n pre-integral residuals and n+1 visible light residuals can be formed according to equations (15) and (14), respectively. Σ l Let X and Y represent the covariance matrices of the pre-integration residual and the visible light residual, respectively, estimated by the error levels of the IMU and the visible light. Expression (16) needs to be optimized using a least-squares optimization algorithm to find the state variable X that minimizes the cost function.

[0086] In real-time navigation applications, the last state value in the sliding window is usually output because it is the most up-to-date.

[0087] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides a visible light positioning system for a fast-moving vehicle, comprising: a plurality of LED lights, a PD, an IMU, and a processor for receiving PD signals, wherein the IMU includes an accelerometer and a gyroscope; the PD and IMU are placed on the fast-moving vehicle, and the processor is used to execute the steps of the visible light positioning method for the fast-moving vehicle according to the signals received by the PD and the IMU data.

[0088] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides a visible light positioning device for a fast-moving vehicle, including a memory and a processor. The memory stores program instructions that are executed by the processor, and the processor calls the program instructions to execute the visible light positioning method for a fast-moving vehicle.

[0089] In embodiments of the present invention, the memory can be non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it can be volatile memory, such as random-access memory (RAM). Memory is any other medium capable of carrying or storing desired program code having an instruction or data structure form and accessible by a computer, but is not limited thereto. The memory in embodiments of the present invention can also be a circuit or any other device capable of implementing a storage function for storing program instructions and / or data.

[0090] In this embodiment of the invention, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in this embodiment of the invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in this embodiment of the invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0091] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides a non-transitory computer-readable storage medium storing computer instructions that cause the computer to execute the visible light positioning method for a fast-moving carrier as follows:

[0092] Calculate the angular velocity and velocity within the Fourier time window based on the state variables and IMU data;

[0093] The position vector and normal vector are calculated based on the state variables, IMU data, and LED light positions.

[0094] The correction of RSS is calculated based on the angular velocity, velocity, position vector, and normal vector, combined with the RSS obtained by the discrete Fourier transform.

[0095] The state variables are jointly optimized based on the corrected RSS and IMU data, and the position, velocity and attitude of the PD at the current moment are output based on the optimized state variables.

[0096] In summary, this invention relates to the field of visible light positioning technology, and particularly to the positioning of fast-moving vehicles using a PD (Digital Photon Transform) as a signal receiver. Visible light positioning systems employing this method typically require signal extraction via Discrete Fourier Transform (DFT) or similar methods. However, for fast-moving objects, the light signal of each LED changes rapidly, making it difficult for the DFT method to accurately obtain the instantaneous RSS (Real-Solution Time). Therefore, to accurately obtain the instantaneous RSS, this invention integrates the IMU (Integrated Memory Unit) observation data and the state data within the window time of the DFT, restoring the average RSS to the instantaneous RSS, thereby improving the navigation and positioning capabilities of fast-moving vehicles.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A visible light positioning method for a fast-moving carrier, characterized in that, The method for positioning a fast-moving carrier that uses a photodiode as a signal receiver includes: Calculate the angular velocity and velocity within the Fourier time window based on the state variables and inertial navigation data; The position vector and normal vector are calculated based on the state variables, inertial navigation data, and the position of the LED lights. The correction of the received signal strength is calculated based on the angular velocity, velocity, position vector, and normal vector, combined with the received signal strength obtained by discrete Fourier transform. The state variables are jointly optimized based on the corrected received signal strength and inertial navigation data, and the position, velocity and attitude of the photodiode at the current moment are output based on the optimized state variables.

2. The visible light positioning method for a fast-moving carrier according to claim 1, characterized in that, The method further includes: The state variables are constructed, including: the position and velocity of the volume downloaded in the indoor coordinate system, the attitude quaternion of the carrier from the coordinate system of the visible light receiver surface to the indoor coordinate system, and the accelerometer zero bias and gyroscope zero bias in the coordinate system of the visible light receiver surface.

3. The visible light positioning method for a fast-moving carrier according to claim 2, characterized in that, Calculate the angular velocity and velocity within the Fourier time window based on the state variables and inertial navigation data, including: The direction cosine matrix is ​​obtained by transforming the attitude quaternion in the state variables. Based on the direction cosine matrix, combined with the gyroscope observations and the gyroscope zero bias in the coordinate system of the visible light receiver surface, the angular velocity within the Fourier time window is obtained. Based on the velocity at the previous moment, combined with the direction cosine matrix, inertial navigation sampling interval, accelerometer observations, and accelerometer zero bias in the coordinate system of the visible light receiver surface, the velocity within the Fourier time window is obtained.

4. The visible light positioning method for a fast-moving carrier according to claim 2, characterized in that, The position vector and normal vector are calculated based on the state variables, inertial navigation data, and LED light positions, including: Use directly k The position of the volume in the indoor coordinate system and the attitude quaternion of the carrier transformed from the coordinate system of the visible light receiver surface to the indoor coordinate system are calculated in the time state variables. The position vector and normal vector are calculated.

5. The visible light positioning method for a fast-moving carrier according to claim 2, characterized in that, The position vector and normal vector are calculated based on the state variables, inertial navigation data, and LED light positions, including: The position of the photodiode is estimated based on the speed, and the position vector from the photodiode to the LED is calculated based on the position of the photodiode and the position of the LED. The attitude quaternion is derived from the angular velocity, and the normal vector is calculated from the attitude quaternion.

6. The visible light positioning method for a fast-moving carrier according to claim 1, characterized in that, The calculation of the correction for the received signal strength also includes: ， in, For the first The Lambertian coefficient of an LED light bulb. and They are the first The normal vector of the LED light and photodiode is pointing upwards; It is a photodiode up to the first The distance vector of each LED light. It is its model; This represents the received signal strength obtained from the Discrete Fourier Transform. T This indicates the length of the Fourier time window. Indicates time i, Let k represent time. Indicates the IMU sampling interval. Represents the angular velocity at time i. The value represents the velocity at time i, and the subscript u indicates the indoor local coordinate system.

7. The visible light positioning method for a fast-moving carrier according to claim 1, characterized in that, The state variables are jointly optimized based on the corrected received signal strength and inertial navigation data, including: The received signal strength obtained by the Discrete Fourier Transform is corrected based on the corrected received signal strength to obtain the corrected received signal strength observation value. Based on the corrected received signal strength and inertial measurement data, an objective function to be optimized is constructed, thereby performing state estimation and achieving navigation and positioning.

8. A visible light positioning system for a fast-moving carrier, characterized in that, The system for positioning a fast-moving vehicle using a photodiode as a signal receiver includes: a plurality of LEDs, a photodiode, an inertial measurement unit, and a processor for receiving signals from the photodiode; the photodiode and the inertial measurement unit are placed on the fast-moving vehicle, and the processor is used to execute the steps of the visible light positioning method for a fast-moving vehicle according to any one of claims 1-7 based on the signals received by the photodiode and the inertial navigation data.

9. A visible light positioning device for a fast-moving carrier, characterized in that, It includes a memory and a processor, the memory storing program instructions that are executed by the processor, the processor invoking the program instructions to perform the visible light positioning method for a fast-moving carrier as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions that cause the computer to perform the visible light positioning method for a fast-moving carrier as described in any one of claims 1 to 7.

Images