Visible light imaging communication train positioning light source adaptive acquisition method

Abstract

The application discloses a visible light imaging communication train positioning light source adaptive acquisition method, first, the positional relationship between the LED positioning light source and the camera is analyzed, the critical condition of the camera azimuth angle is obtained by determining the minimum value and the maximum value of the camera azimuth angle; then, the inertial measurement unit (IMU) and the optical flow sensor are fused to construct an ESO model of the camera azimuth angle, the azimuth angle state and the external disturbance are estimated in real time; then, whether the obtained current camera azimuth angle meets the critical condition is judged; combining the Lyapunov theory and the adaptive backstepping method, an azimuth angle adaptive control strategy based on the train speed is designed; finally, the derivative of the virtual control quantity is approximately calculated by using the dynamic surface control (DSC) method, the response time of the camera azimuth angle control is optimized, and the positioning light source image is stably acquired. According to the application, the camera rotation amount is adaptively adjusted according to the curve radius of the line, the camera is ensured to stably acquire the positioning light source image, and the train positioning precision is improved.

Description

Technical Field

[0001] This invention belongs to the field of train positioning technology, specifically relating to an adaptive acquisition method for train positioning light source in visible light imaging communication. Background Technology

[0002] In recent years, Visible Light Communication (VLC) technology has been widely used in indoor positioning, underwater optical communication, medical applications, the Internet of Things (IoT), and intelligent transportation. The uniformly distributed LED lighting sources within subway tunnels provide a natural communication medium for train positioning using VLC, serving as the transmitter to transmit light signals containing the location information of the LED light sources. Visible light non-imaging communication technology, using photodetectors as receivers, is susceptible to interference from ambient tunnel light, resulting in lower positioning accuracy. In contrast, visible light imaging communication technology, employing complementary metal-oxide-semiconductor (CMOS) cameras as receivers, offers stronger resistance to background light interference, higher positioning accuracy, and is more suitable for train positioning in tunnel environments. When using visible light imaging communication for train positioning, a camera mounted on the top of the train's front receives light signals transmitted from LED light sources on the tunnel wall in real time. The onboard computer extracts the region of interest and feature information from the LED light source image to identify the LED light source's identity (ID) information and calculates the train's actual position based on the coordinates of the LED light sources. During train turns, due to the small radius of the curve on the subway line, the camera has difficulty in obtaining a stable image of the LED light source in front of the train, which may cause the positioning error to diverge rapidly or even interrupt the positioning.

[0003] Currently, there is a great deal of research on using VLC for positioning, such as:

[0004] (1) The modulation of the LED light source is used as the VLC signal transmitter to achieve positioning. It has achieved good positioning results in indoor positioning and low-speed vehicle positioning, but the influence of external interference light source on positioning performance is not considered.

[0005] (2) The visible light imaging communication positioning method using a camera as the receiver significantly reduces light source interference by taking advantage of the spatial separation of the camera, and the positioning accuracy reaches the centimeter level. However, when the camera's field of view is limited and the positioning light source is insufficient, the positioning accuracy will decrease significantly.

[0006] (3) Single light source positioning algorithm: When the LED light source image acquired by the camera cannot meet the positioning requirements, the actual position of the receiver can be calculated by a single light source, but its computational complexity is high and the positioning accuracy is low.

[0007] (4) Predicting subsequent light sources based on the current light source position in the image. This method can achieve high positioning accuracy when the spatial layout of the light sources remains stable. However, when the spatial layout of the light sources changes, the positional relationship of adjacent light sources in the image changes accordingly, and the prediction error will increase significantly.

[0008] (5) To address the issue of how to stably acquire target information in fields such as autonomous driving, mobile robots, and navigation, an Extended State Observer (ESO) can be used to dynamically adjust the camera angle to ensure that the target is always within the field of view, but the system response time is relatively slow.

[0009] Although existing technologies have verified the feasibility and effectiveness of using VLC for positioning using different methods, and provided a reference for how to adjust the camera rotation to stably acquire target information, they pay less attention to how to stably acquire positioning light sources for fast-moving objects on tracks with different curve radii in tunnel scenes, which makes it difficult to meet the train positioning requirements for vehicle-to-vehicle communication. Summary of the Invention

[0010] To address the problems existing in the above-mentioned background technology, the purpose of this invention is to provide an adaptive acquisition method for the positioning light source of a visible light imaging communication train, which adaptively acquires the positioning light source by adjusting the camera azimuth angle through an extended state observer (ESO).

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] An adaptive acquisition method for a visible light imaging communication train positioning light source includes the following steps:

[0013] S1, Analyze the positional relationship between the LED positioning light source and the camera, and obtain the critical condition of the camera azimuth angle by determining the minimum and maximum values ​​of the camera azimuth angle;

[0014] S2. Construct the camera azimuth angle ESO model:

[0015]

[0016] Where x2=β is the drift angle caused by the external disturbance, representing the additional camera deviation caused by the disturbance; Angular velocity is the azimuth angle. Δ is the azimuth acceleration; β For the external disturbance in the drift angle, x5 = d(t) = f(x3,x4) + τ. For unmodeled dynamics, u is the control input, b1 is the model parameter, τ is the external disturbance, and h(t) is the derivative of the composite disturbance;

[0017] The camera azimuth ESO model is transformed into matrix equation form, ESO equations are constructed, and the azimuth state is estimated in real time based on the ESO equations.

[0018] S3. The camera azimuth angle is acquired in real time using an inertial measurement unit (IMU), the relative displacement of the LED light source image in the camera field of view is monitored using an optical flow sensor, the offset angle of the LED light source in the camera field of view is calculated, and the cumulative error of the camera azimuth angle acquired in real time by the IMU is corrected by the offset angle to obtain the current azimuth angle of the camera.

[0019] S4. Determine whether the current azimuth angle of the camera satisfies the critical condition obtained in step S1:

[0020] If the critical condition is met, it is further determined whether it is the desired azimuth angle; if so, the image of the positioning light source is directly captured and the positioning light source is determined; if not, the observation error between the current azimuth angle of the camera and the azimuth angle estimated by the ESO equation is calculated, and the disturbance information is estimated.

[0021] If the critical condition is not met, the critical error between the current azimuth angle of the camera and the critical value is calculated, and the disturbance information is estimated by combining the observation error between the current azimuth angle of the camera and the azimuth angle estimated by the ESO equation.

[0022] S5. Based on the estimated disturbance information, construct a Lyapunov function based on the adaptive backstepping method, and minimize the control error through the Lyapunov function;

[0023] Design the train speed gain function:

[0024]

[0025] Where K1 is the reference gain, μ is the adjustment coefficient for the relationship between speed and gain, and v is the real-time speed of the train;

[0026] The control gain is adaptively adjusted according to the train speed, and the disturbance is gradually compensated by the virtual control quantity to achieve adaptive adjustment of the camera azimuth angle, ensuring that the azimuth angle always meets the critical condition after adjustment. The derivative of the virtual control quantity is approximately calculated by the dynamic surface control (DSC) method to optimize the response time of the camera azimuth angle control and stably acquire the positioning light source image.

[0027] Furthermore, in step S1, the critical condition for obtaining the camera azimuth angle is obtained as follows: LED light sources are evenly distributed on the left tunnel wall along the train's running direction. The camera azimuth angle is the angle at which the camera deflects relative to both sides of the tunnel wall. The first LED light source in front of the train is used as the LED positioning light source. When the camera deflects from the right tunnel wall to the left, and the upper boundary of the camera's field of view can capture the first LED light source in front of the train and subsequent light sources, the camera azimuth angle reaches its minimum value. When the camera deflects towards the LED light source side, and the lower boundary of the camera's field of view can only capture the first LED light source in front of the train, the camera azimuth angle reaches its maximum value.

[0028] The minimum value of the camera azimuth angle ψ min The calculation formula is:

[0029]

[0030] Maximum value of camera azimuth angle ψ max The calculation formula is:

[0031]

[0032] Where F is the camera's field of view, L is the distance between adjacent LED light sources, and W is the vertical distance between the camera and the LED light source.

[0033] Furthermore, in step S2, the method for constructing the camera azimuth angle ESO model is as follows:

[0034] S21. Using the camera azimuth angle and drift angle as control parameters, establish a nonlinear state-space model of the camera azimuth angle:

[0035]

[0036] S22. Combine the external disturbances and unmodeled dynamics in the nonlinear state-space model of the camera azimuth angle into a composite disturbance, and define the state variable x5=d(t)=f(x3,x4)+τ, thereby obtaining the camera azimuth angle ESO model.

[0037] Furthermore, in step S2, the method for constructing the ESO equation is as follows:

[0038] The camera azimuth angle ESO model is transformed into the following matrix equation form:

[0039]

[0040] Where x represents the camera azimuth angle. Here is the state matrix, and c1 and c2 are the attenuation coefficients of the drift angle; The input matrix is ​​the control matrix, y is the measurement output, C = [1 0 0 0] is the output matrix, and φ is the input matrix.x It is a composite disturbance of external disturbance and unmodeled dynamics;

[0041] Construct the ESO equations based on the matrix equation form described above:

[0042]

[0043] in, This is the estimated value of the camera azimuth angle. Let α be the ESO gain matrix. i >0 (i=1,2,3,4) is the ESO gain coefficient, and 0<ε<1 is the ESO convergence parameter.

[0044] Furthermore, in step S3, the method for the IMU to acquire the camera azimuth angle in real time is as follows:

[0045] The camera angular velocity measured by the IMU is ω = (ω x ,ω y ,ω z ), convert the angular velocity information into a quaternion q:

[0046]

[0047] in, Δt is the sampling time interval;

[0048] Using the quaternion conversion formula, q is converted into the camera azimuth angle ψ0:

[0049]

[0050] Where, q ω q x q y and q z Let each of the four components of the quaternion q be represented by a q. ω q is a scalar component, typically related to the rotation angle; x q y and q z All are vector components, representing the directions of the rotation axis (x, y, z).

[0051] Furthermore, in step S3, the offset angle Δψ of the LED light source in the camera's field of view is:

[0052] Δψ=kl;

[0053] Where l is the relative displacement of the LED light source image in the camera's field of view, l = (l x , l y ), l=(l x , l y ), l x and ly These represent the relative displacements of the LED light source image in the x and y directions within the camera's field of view, respectively, and k is the scaling factor of the optical flow sensor.

[0054] Then the current azimuth angle ψ of the camera is:

[0055] ψ = ψ0 + Δψ.

[0056] Compared with the shortcomings and deficiencies of existing technologies, the present invention has the following beneficial effects:

[0057] (1) This invention uses IMU and optical flow sensor data to determine the azimuth position of the camera, designs an azimuth control strategy to compensate for disturbance errors, and combines DCS to shorten the response time of azimuth control.

[0058] (2) The present invention adaptively adjusts the camera rotation amount according to the radius of the line curve, which can ensure that the camera can stably acquire the positioning light source image, thereby improving the positioning accuracy of the train. Attached Figure Description

[0059] Figure 1 This is a flowchart of an adaptive acquisition method for a visible light imaging communication train positioning light source provided in an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the structure for determining the critical conditions of the camera azimuth angle provided in the embodiment of the present invention. (a) shows the determination of the minimum value of the camera azimuth angle, and (b) shows the determination of the maximum value of the camera azimuth angle.

[0061] Figure 3 These are the experimental results of adaptive adjustment of camera azimuth angle provided in the embodiments of the present invention; in the figure, (a) represents the adjustment result when the curve radius is 500m; (b) represents the adjustment result when the curve radius is 400m; (c) represents the adjustment result when the curve radius is 300m; and (d) represents the adjustment result when the curve radius is 250m.

[0062] Figure 4 This refers to the success rate of LED light source acquisition under different acquisition methods provided in the embodiments of the present invention.

[0063] Figure 5 This refers to the success rate of LED light source acquisition at different speeds provided in the embodiments of the present invention;

[0064] Figure 6 These are the train positioning results at different speeds provided in the embodiments of the present invention;

[0065] Figure 7 These are train positioning results obtained using different light source acquisition methods, as provided in the embodiments of the present invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0067] I. During high-speed train operation, the system measures its own position and speed in real time, providing support for CBTC (Computer-Based Train Control) systems oriented towards vehicle-to-vehicle communication to achieve functions such as train interval control and overspeed protection. By establishing a visible light imaging communication link between the train and the LED light source, the onboard computer receives and decodes the light signals sent by the LED light source and autonomously calculates the train's position information. When the train passes through lines with different curve radii, this invention adaptively adjusts the camera rotation to stably acquire the positioning light source, improving train positioning accuracy. A flowchart of the visible light imaging communication train positioning light source adaptive acquisition method is shown below. Figure 1 As shown below, a detailed explanation will follow.

[0068] 1. Critical conditions for determining the camera azimuth angle

[0069] When the train is running on a straight section of track, the camera has a high success rate in acquiring images of the LED light sources ahead. However, at curves, the camera's field of view is limited due to the radius of the subway curve, making it more difficult to acquire images of the LED light sources. Therefore, the camera rotation is adaptively adjusted according to the curve radius to ensure that the acquired LED light sources meet the positioning requirements. Camera rotation refers to the camera's rotation angle in three-dimensional space, usually expressed as azimuth, pitch, and roll angles. Since the vertical height between the LED light source and the rail plane remains constant, and the train always runs along the rail, the impact of pitch and roll angles on the camera's acquisition of LED light source images is negligible. Therefore, adjusting the camera rotation mainly focuses on the azimuth angle. When adjusting the camera azimuth angle, two key factors need to be balanced: first, ensuring that there are always a sufficient number of LED light sources in the camera's field of view to prevent positioning interruptions; second, balancing the image quality of the LED light sources to prevent image distortion that could affect the train's positioning accuracy. Because the adjustment range of the camera azimuth angle is limited, an excessively large azimuth angle will result in an insufficient number of LED light sources in the field of view, while an excessively small azimuth angle will cause geometric distortion of the image. Therefore, determining the critical conditions for the camera azimuth angle is a prerequisite for reasonably adjusting the camera azimuth angle.

[0070] LED light sources are evenly distributed on the left side of the tunnel wall along the direction of train travel. The camera azimuth angle is the angle at which the camera is deflected relative to the two sides of the tunnel wall. The first LED light source in front of the train is used as the LED positioning light source.

[0071] (1) Determination of the minimum value of the camera azimuth angle

[0072] Reference Figure 2(a) When the camera deflects from the right tunnel wall to the left, and the upper boundary of the camera's field of view can capture the first LED light source in front of the train and subsequent light sources, the camera's azimuth angle reaches its minimum value ψ. min :

[0073]

[0074] (2) Determination of the maximum azimuth angle of the camera

[0075] Reference Figure 2 (b) When the camera deflects toward the LED light source, and the lower boundary of the camera's field of view can only capture the first LED light source in front of the train, the camera's azimuth angle reaches its maximum value ψ. max :

[0076]

[0077] Where F is the camera's field of view, L is the distance between adjacent LED light sources, and W is the vertical distance between the camera and the LED light source;

[0078] If the camera azimuth angle continues to increase, an effective positioning light source cannot be obtained.

[0079] 2. Camera azimuth information acquisition

[0080] Real-time acquisition of camera azimuth angle using an inertial measurement unit (IMU):

[0081] The camera angular velocity measured by the IMU is ω = (ω x ,ω y ,ω z ), convert the angular velocity information into a quaternion q:

[0082]

[0083] in, Δt is the sampling time interval;

[0084] Using the quaternion conversion formula, q is converted into the camera azimuth angle ψ0:

[0085]

[0086] Where, q ω q x q y and q z Let each of the four components of the quaternion q be represented by a q. ω q is a scalar component, typically related to the rotation angle; x q y and q z All are vector components, representing the directions of the rotation axis (x, y, z).

[0087] The relative displacement of the LED light source image within the camera's field of view is monitored using an optical flow sensor, and the offset angle Δψ of the LED light source within the camera's field of view is calculated.

[0088] Δψ=kl(5)

[0089] Where l is the relative displacement of the LED light source image in the camera's field of view, l = (l x , l y ), l=(l x , l y ), l x and l y , respectively, represent the relative displacements of the LED light source image in the x and y directions within the camera's field of view, and k is the scaling factor of the optical flow sensor.

[0090] The current azimuth angle ψ of the camera is obtained by correcting the cumulative error of the camera azimuth angle acquired in real time by the offset angle:

[0091] ψ=ψ0+Δψ(6).

[0092] 3. Camera azimuth angle disturbance estimation

[0093] Camera azimuth adjustment exhibits nonlinear characteristics and is susceptible to external disturbances and unmodeled dynamics such as azimuth angular velocity and angular acceleration. ESO is an intelligent observer that can estimate observation errors by estimating the azimuth state, thereby estimating disturbances caused by external interference and unmodeled dynamics.

[0094] A nonlinear state-space model of the camera azimuth angle is established using the camera azimuth angle and drift angle as control parameters:

[0095]

[0096] Where x2=β is the drift angle caused by the external disturbance, representing the additional camera deviation caused by the disturbance; Angular velocity is the azimuth angle. Δ is the azimuth acceleration; β External disturbances in the drift angle For unmodeled dynamics, u is the control input, b1 is the model parameter, and τ is the external disturbance.

[0097] To reduce estimation errors, the external disturbances and unmodeled dynamics in the nonlinear state-space model of the camera azimuth angle are combined into a composite disturbance, and the state variable x5 = d(t) = f(x3,x4) + τ is defined, thus obtaining the camera azimuth angle ESO model:

[0098]

[0099] Where h(t) is the derivative of the composite perturbation.

[0100] To facilitate subsequent analysis, the camera azimuth angle ESO model is transformed into the following matrix equation form:

[0101]

[0102] Where x represents the camera azimuth angle. Here is the state matrix, and c1 and c2 are the attenuation coefficients of the drift angle; The input matrix is ​​the control matrix, y is the measurement output, C = [1 0 0 0] is the output matrix, and φ is the input matrix. x It is a composite disturbance of external disturbance and unmodeled dynamics.

[0103] Construct the ESO equations in matrix form:

[0104]

[0105] in, This is the estimated value of the camera azimuth angle. Let α be the ESO gain matrix. i >0 (i=1,2,3,4) is the ESO gain coefficient, and 0<ε<1 is the ESO convergence parameter.

[0106] The azimuth state is estimated in real time based on the ESO equation.

[0107] When the current azimuth angle measured by the IMU and optical flow sensor does not meet the critical condition, the azimuth angle is estimated based on the current azimuth angle of the camera and the ESO equation, and the observation error e is calculated. o And calculate the critical error e in combination with the critical conditions. c At this time, the total azimuth error e of the camera is:

[0108] e = e o +e c (11)

[0109] in:

[0110]

[0111] If the current azimuth angle of the camera meets the critical condition, then the total error e = e o .

[0112] By adjusting the parameters of the observer gain matrix H, e is amplified and fed back into the ESO, thus updating the ESO equations:

[0113]

[0114] in, These represent the state variables estimated by ESO.

[0115] The updated camera azimuth status and disturbance information are as follows:

[0116]

[0117] 4. Camera azimuth control strategy design

[0118] To compensate for disturbances caused by external interference and unmodeled dynamics, an adaptive azimuth control strategy is designed to reduce azimuth adjustment deviations during high-speed train operation. The adaptive backstepping method, a recursive design approach based on Lyapunov theory, allows for online parameter adjustment to achieve disturbance compensation. The Dynamic Surface Control (DSC) method simplifies higher-order differential calculations by introducing a first-order filter, reducing computational complexity and shortening control response time.

[0119] (1) To compensate for the camera drift angle β, the azimuth control error e1 is defined using the adaptive backstepping method as:

[0120] e1=ψ-ψ d +β=x1-ψ d +x2(16)

[0121] Where, ψ d The desired azimuth angle;

[0122] To minimize e1, the Lyapunov function V1 is constructed as follows:

[0123]

[0124] Differentiating the Lyapunov function V1, we get:

[0125]

[0126] On straight sections of the road, acquiring the location light source is relatively easy for the camera, requiring no significant adjustment of the azimuth angle. Furthermore, the train travels at higher speeds, and excessive azimuth angle adjustments could blur the light source image. However, on curves, acquiring the location light source is more difficult, and the train travels at lower speeds, allowing for a larger adjustment range. Therefore, a speed gain function K(v) is designed to adaptively adjust the control gain based on the train speed.

[0127] Train speed gain function K(v):

[0128]

[0129] Where K is the reference gain, μ is the adjustment coefficient for the relationship between speed and gain, and v is the real-time speed of the train.

[0130] In order to To make e1 converge, a virtual control variable q1 is introduced:

[0131]

[0132] Where K1(v) is the control gain, Perturbation estimation for drift angle:

[0133]

[0134] Where, α β and σ β To design positive constants.

[0135] Dynamic surface control (DSC) is employed, and a filter is introduced to approximate the derivative of the virtual control quantity:

[0136]

[0137] Where γ1 is the filter time constant and q1′ is the filtered output of q1.

[0138] As shown in equation (7), the control parameter x3 can directly drive x1 and x2. Let the filtering error E1 = q1 - q1′ be used to compensate for the unmodeled dynamic angular velocity. The resulting disturbance is defined as follows: The new azimuth control error e2 is:

[0139] e2=x3-q1′=x3-q1+E1(23)

[0140] Substituting equations (20) and (23) into (18), we get:

[0141]

[0142] To stabilize e2, the Lyapunov function V2 is chosen as:

[0143]

[0144] Differentiating equation (25), we get:

[0145]

[0146] For V2, select virtual control variable q2:

[0147]

[0148] Where K2(v) is the control gain, and by adjusting K2(v), we can ensure... Make e1 and e1 converge.

[0149]

[0150] Where γ2 is the filter time constant and q2′ is the filtered output of q2.

[0151] Similarly, control parameter x4 can directly drive x3, and the filtering error E2 = q2 - q2′ is used to compensate for the unmodeled dynamic angular acceleration. The disturbance caused is defined as follows: The new azimuth control error e3 is:

[0152] e3=x4-q2′=x4-q2+E2(29)

[0153] Substituting equations (27) and (29) into (26), we get:

[0154]

[0155] To ensure stability of e3, the Lyapunov function is chosen:

[0156]

[0157] Differentiating equation (31), we get:

[0158]

[0159] By minimizing e1, e2, and e3, the disturbances caused by drift angle, angular velocity, and angular acceleration are compensated, ensuring that the camera azimuth control error is consistent and bounded. The final azimuth control strategy is as follows:

[0160]

[0161] Where K3(v) is the control gain.

[0162] 5. Acquire the image of the positioning light source.

[0163] Determine whether the current azimuth angle of the camera meets the critical condition obtained in step S1:

[0164] If the critical condition is met, it is further determined whether it is the desired azimuth angle; if so, the image of the positioning light source is directly captured and the positioning light source is determined; if not, the observation error between the current azimuth angle of the camera and the azimuth angle estimated by the ESO equation is calculated, and the disturbance information is estimated.

[0165] If the critical condition is not met, calculate the critical error between the current azimuth angle of the camera and the critical value, and estimate the disturbance information together with the observation error;

[0166] Based on the estimated disturbance information, a camera azimuth angle control strategy is used to compensate for the disturbance and adaptively adjust the camera azimuth angle. Specifically, the control gain is adaptively adjusted according to the train speed, and the disturbance is gradually compensated using virtual control quantities to achieve adaptive adjustment of the camera azimuth angle, ensuring that the adjusted azimuth angle always meets the critical conditions. The derivative of the virtual control quantity is approximately calculated using the Dynamic Surface Control (DSC) method to optimize the response time of the camera azimuth angle control and stably acquire the positioning light source image.

[0167] Due to the specific spatial layout of LED light sources in the tunnel and the high precision requirements of the CBTC system for vehicle-to-vehicle communication for train positioning, a dual-LED light source positioning algorithm with high reliability and low computational complexity is adopted to solve the train position information and obtain the actual position of the train.

[0168] II. Simulation Experiments and Result Analysis

[0169] Based on the line data and equipment information of a certain subway line 1, a train positioning experimental scenario was constructed to simulate the train running in the subway tunnel. The performance of ESO adaptive adjustment of camera azimuth angle was verified under different curve radii. The influence of light source acquisition method and train running speed on the success rate of positioning light source acquisition was analyzed. The positioning results of the train at different running speeds and when the camera azimuth angle was adjusted by different methods were compared and analyzed.

[0170] 1. Simulation Experiment Design

[0171] The curve radius of subway lines is closely related to train speed. When the train speed is below 80 km / h, the minimum curve radius should generally not be less than 300 m, and not less than 250 m under difficult conditions. If the train speed exceeds 80 km / h, the minimum curve radius should generally not be less than 500 m, and not less than 400 m under difficult conditions. Curve radii of 500 m, 400 m, 300 m, and 250 m were selected. Using the distance between two LED light sources as a positioning unit, a 20 m × 2.0 m × 1.5 m experimental platform was built. The vertical and horizontal distances of the CMOS camera from the LED light sources were 1.5 m and 2.0 m, respectively.

[0172] The camera's initial position is set as the origin, the LED light source side as the X-axis, the experimental trolley's running direction as the Y-axis, and the direction perpendicular to the ground as the Z-axis. Twenty test points are set at 0.5m intervals along the experimental trolley's running direction, from (-2.0, 0, -1.5) to (-2.0, 10, -1.5), and ten tests are performed for each point. When the experimental trolley runs at 0 km / h and 20 km / h, the actual images of the LED light source captured by the camera are recorded sequentially. Based on the image shift at different train speeds, motion-blurred images of the LED light source at train speeds of 40, 60, 80, and 100 km / h are simulated using MATLAB software. The simulation parameters are shown in Table 1.

[0173] Table 1 Experimental parameters

[0174]

[0175] 2. Analysis of Experimental Results

[0176] (1) Performance analysis of adaptive adjustment of camera azimuth angle

[0177] To verify the performance of the adaptive adjustment of the camera azimuth angle, the proposed method was used to adjust the camera azimuth angle as the curve radius changed. During t = 0–100s, the train ran on a straight section; at t = 100s, the curve radius changed. The adjusted camera azimuth angle ψ and the expected value ψ were compared when the curve radii were 500m, 400m, 300m, and 250m. d The maximum deviations were 0.26°, 0.28°, 0.31°, and 0.34°, respectively. As the radius of the line curve decreased, the error in azimuth adjustment tended to increase, but no significant overshoot or lag was observed throughout the adjustment process. Figure 3 As shown.

[0178] (2) The impact of camera azimuth angle adjustment on the acquisition of positioning light source

[0179] Using both fixed and adaptive azimuth angle methods, a positioning light source was acquired under different curve radii. The influence of the camera azimuth angle on acquiring the positioning light source was analyzed. Experimental results are as follows: Figure 4 As shown, both methods have a success rate of nearly 99.06% on straight road sections (i.e., curve radii of ∞). When the curve radii are 500m, 400m, 300m, and 250m, the success rates of the fixed azimuth angle are 79.98%, 71.89%, 62.40%, and 57.01%, respectively, while the success rates of the adaptive azimuth angle adjustment are 99.87%, 99.01%, 95.89%, and 93.95%, respectively.

[0180] (3) The impact of train speed on obtaining positioning light source

[0181] When the train travels at different speeds, the proposed method is used to adaptively adjust the camera azimuth angle to acquire the positioning light source, and the influence of train speed on the acquisition of the positioning light source is analyzed. When the train travels at speeds of 20 km / h, 40 km / h, 60 km / h, 80 km / h, and 100 km / h through a line with a curve radius of 250 m, the success rates of acquiring the positioning light source are 92.29%, 91.84%, 91.63%, 91.48%, and 91.39%, respectively. Figure 5 As shown, the faster the train travels, the lower the success rate of the camera acquiring the LED light source. However, when the train is traveling at different speeds, the success rate of acquiring the positioning light source is higher than 91%.

[0182] (4) Analysis of train positioning results at different operating speeds

[0183] To analyze the impact of train speed on positioning results, the proposed method was used to verify the train positioning accuracy at different speeds. When the train speeds were 20 km / h, 40 km / h, 60 km / h, 80 km / h, and 100 km / h, the maximum positioning errors were 20.87 cm, 22.35 cm, 24.97 cm, 27.84 cm, and 30.04 cm, respectively, and the average positioning errors were 10.59 cm, 11.45 cm, 12.41 cm, 13.69 cm, and 15.18 cm, respectively. Figure 6 As shown in the figure. Experimental results show that the proposed method can achieve good positioning results at different train operating speeds, meeting the requirements of IEEE 1474.1-2025 standard for train positioning accuracy.

[0184] (5) Comparison of train positioning results from different methods

[0185] To verify the positioning performance of the proposed method under different curve radii, the train positioning results were compared and analyzed when the train was running at 100 km / h, using no ESO, using traditional ESO, and using the proposed method. The maximum positioning errors without ESO and with traditional ESO under different curve radii were 25.78, 29.56, 32.28, 35.15, 39.66 cm and 24.99, 27.56, 30.28, 33.15, 38.26 cm, respectively, while the maximum positioning errors of the proposed method were 23.36, 25.69, 27.86, 29.69, and 30.04 cm. Compared with the methods without ESO and traditional ESO, the positioning accuracy of the proposed method is improved by 24.25% and 21.48% respectively when the curve radius is smallest. Figure 7 As shown.

[0186] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A visible light imaging communication train positioning light source adaptive acquisition method, characterized in that, Comprise the following steps: S1, analyze the positional relationship between the LED positioning light source and the camera, and obtain the critical condition of the camera azimuth angle by determining the minimum and maximum values of the camera azimuth angle; The critical condition of the camera azimuth angle is obtained by: uniformly arranging LED light sources on the left tunnel wall along the running direction of the train, the camera azimuth angle is the angle of the camera relative to the deflection of the two sides of the tunnel wall, the first LED light source in front of the train is taken as the LED positioning light source, when the camera deflects from the right tunnel wall to the left, the upper boundary of the camera field of view can capture the first LED light source and the subsequent light source in front of the train, the camera azimuth angle reaches the minimum value; when the camera deflects to the LED light source side, the lower boundary of the camera field of view can only obtain the first LED light source in front of the train, the camera azimuth angle reaches the maximum value; S2, construct the camera azimuth angle ESO model: where x2= β is the drift angle due to external disturbance, and x3= δ is the extra deviation of the camera due to disturbance; is the azimuth angle acceleration; Δ is the azimuth angle acceleration; Δ β is the external disturbance in the drift angle, x5= d(t) = f(x3, x4) + τ, is the unmodeled dynamics, u is the control input, b1 is a model parameter, τ is the external disturbance, and h(t) is the derivative of the compound disturbance; Convert the camera azimuth angle ESO model into a matrix equation form, construct an ESO equation, and estimate the azimuth angle state in real time according to the ESO equation; The construction method of the ESO equation is: Convert the camera azimuth angle ESO model into the following matrix equation form: where x is the camera azimuth state, is the state matrix, c1 and c2 are the decay coefficients of the drift angle; is the control input matrix, y is the measurement output, C = [1 0 0 0] is the output matrix, φ x is the composite disturbance of external disturbance and unmodeled dynamics; Construct the ESO equation according to the matrix equation form: wherein, is a camera azimuth state estimate, is an ESO gain matrix, a i are ESO gain coefficients, and 0 < ε < 1 is an ESO convergence parameter. S3, use the inertial measurement unit to collect the camera azimuth angle in real time, use the optical flow sensor to monitor the relative displacement of the LED light source image in the camera field of view, calculate the offset angle of the LED light source in the camera field of view, correct the cumulative error of the camera azimuth angle collected by the IMU in real time through the offset angle, and obtain the current camera azimuth angle; S4, judge whether the current camera azimuth angle meets the critical condition obtained in step S1: If the critical condition is met, further judge whether it is the expected azimuth angle; if yes, directly capture the positioning light source image and determine the positioning light source; if not, calculate the observation error between the current camera azimuth angle and the estimated azimuth angle of the ESO equation, and estimate the disturbance information; If the critical condition is not met, calculate the critical error between the current camera azimuth angle and the critical value, and estimate the disturbance information together with the observation error between the current camera azimuth angle and the estimated azimuth angle of the ESO equation; S5, according to the estimated disturbance information, construct a Lyapunov function based on adaptive backstepping method, and minimize the control error through the Lyapunov function; Design a train speed gain function: Wherein, K is the reference gain, μ is the adjustment coefficient of the relationship between speed and gain, v is the real-time speed of the train; According to the train speed, the control gain is adjusted adaptively, the disturbance is compensated gradually by using the virtual control quantity, the camera azimuth angle is adjusted adaptively, and it is ensured that the adjusted azimuth angle always meets the critical condition; the derivative of the virtual control quantity is calculated approximately by using the dynamic surface control method, the response time of the camera azimuth angle control is optimized, and the positioning light source image is stably obtained.

2. The method of claim 1, wherein the method further comprises: determining the location of the train based on the received image data. In step S1, the minimum value ψ of the camera azimuth angle min The calculation formula is: Maximum value of camera azimuth angle ψ max The calculation formula is: Wherein, F is the camera field of view angle, L is the distance between adjacent LED light sources, and W is the vertical distance between the camera and the LED light source.

3. The adaptive acquisition method for visible light imaging communication train positioning light source as described in claim 1, characterized in that, In step S2, the construction method of the camera azimuth angle ESO model is: S21, take the camera azimuth angle and the drift angle as control parameters, and establish a nonlinear state space model of the camera azimuth angle: S22, the external disturbance and unmodeled dynamics in the camera azimuth nonlinear state space model are combined into a composite disturbance, and a state variable x5=d(t)=f(x3,x4)+τ is defined, so as to obtain the camera azimuth ESO model.

4. The adaptive acquisition method for the positioning light source of a visible light imaging communication train as described in claim 1, characterized in that, In step S3, the method for the inertial measurement unit to collect the camera azimuth in real time is: The camera angular velocity measured by the inertial measurement unit is ω = (ω x , ω y , ω z ), and the angular velocity information is converted into a quaternion q: wherein At is the sampling time interval; The q is converted into the camera azimuth ψ0 by using a quaternion conversion formula: where q ω , q x , q y and q z represent four components of the quaternion q, respectively, q ω is a scalar component, related to the rotation angle; q x , q y and q z are vector components, representing the direction of the rotation axis (x, y, z).

5. The method of claim 4, wherein the method further comprises: In step S3, the offset angle Δψ of the LED light source in the camera field of view is: Δψ=kl; wherein l is the relative displacement of the LED light source image in the camera field of view, l = (l x , y ) l x and l y are the relative displacement of the LED light source image in the x direction and y direction of the camera field of view, respectively, and k is the scale factor of the optical flow sensor; Then, the current camera azimuth ψ is: ψ=ψ0+Δψ.

Images