A target optimization positioning method and system for railway platform canopy inspection

Abstract

The present invention discloses a target optimization positioning method and system for railway platform canopy inspection, comprising the following steps: obtaining the platform's longitude and latitude coordinate values to obtain the starting coordinates, end coordinates, and length of the platform's central axis; obtaining the inspector's GNSS position in real time to obtain the projection point of the positioning point on the central axis along the track; calculating the distance between the positioning point and different platform projection points, and taking the projection point with the absolute minimum value as the inspector's real-time positioning; calculating the distance between adjacent positioning information, judging the current inspector's status based on the difference between the acceleration sensor value and the distance of adjacent positioning, and returning real-time positioning information; converting the calculated positioning information into longitude and latitude coordinate values to drive a character model to move, and drive the character model to rotate in terms of direction and viewing angle. The present invention achieves low-cost positioning in indoor scenes, solves problems such as large positioning deviation and positioning information loss caused by canopy obstruction and electromagnetic interference, and improves real-time positioning accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of navigation and positioning, and in particular to a target optimization positioning method and system for railway platform canopy inspection. Background Art

[0002] Platform canopies are crucial to the operation, maintenance, and management of railway building equipment. Their safe operation is crucial to the safety of society and the lives and property of the people. To ensure safe railway operations, railway building management departments must conduct daily inspections of platform canopies. With the advent and development of digital railways, railway building operations and maintenance are gradually adopting electronic inspection systems instead of traditional manual inspections.

[0003] Real-time positioning technology is a key technology for the informatization of daily patrol inspections. Commonly used wireless positioning technologies include GNSS positioning (Global Navigation Satellite System), network positioning, UWB positioning technology, ZigBee positioning technology, Bluetooth positioning technology, etc.

[0004] GNSS positioning includes GPS, Beidou, etc. The positioning accuracy is greatly affected by occlusion. The advantage is that it does not rely on the network and does not require supporting hardware support.

[0005] Network positioning includes Wi-Fi positioning and base station positioning. It depends on the number of networks and base stations, and has low positioning accuracy. It is not suitable for scenarios where there is no Wi-Fi on the platform canopy and the operator's signal is weak.

[0006] UWB positioning technology has the dual advantages of real-time positioning and precise positioning. It requires the deployment of base stations and is easily affected by obstructions.

[0007] ZigBee positioning technology has high positioning accuracy, requires supporting base stations and receivers, and has poor anti-interference capabilities. It may cause unstable positioning in the railway high-voltage contact network environment.

[0008] Bluetooth positioning technology requires a large number of Bluetooth beacons as positioning terminals, which is costly and has high positioning accuracy.

[0009] Platform canopy inspections require visual tracking of inspection routes and real-time risk warnings, placing high demands on positioning accuracy. Platform canopies fully cover the entire platform, constituting an indoor positioning scenario. Surrounding the scene are numerous buildings and stopped trains, which block wireless signals. Furthermore, the railway's high-voltage overhead line interferes with wireless signal stability and transmission quality. The surrounding magnetic field is complex, and GNSS signals are weak, resulting in low positioning accuracy and significant latency.

[0010] The scene positioning space of the platform canopy is long and narrow, with a width of 450 meters or 550 meters along the track, but a smaller width perpendicular to the track. Daily inspections are usually carried out along the track, so the accuracy requirements for the along-track direction are high, while the vertical track direction needs to ensure that the positioning does not deviate from the platform range.

[0011] Platform canopy inspections are characterized by relatively fixed inspection personnel and the involvement of multiple platforms. Hardware layout within the yard poses significant safety hazards under the repeated effects of various dynamic loads such as train vibration and train wind. Therefore, considering safety and economic practicality requirements, base stations, beacons, smart handheld terminals and other hardware are not suitable for platform canopy positioning scenarios.

[0012] The problems and defects of the prior art are:

[0013] Given the indoor characteristics and positioning accuracy requirements of platform canopies, existing positioning technologies suffer from low network positioning accuracy, making them unsuitable for environments with no Wi-Fi and weak carrier signals. ZigBee positioning technology also suffers from poor interference resistance, making it unsuitable for railway high-voltage overhead lines. UWB and Bluetooth positioning technologies require the deployment of base stations and beacons, necessitating high hardware deployment requirements and costs, making them unsuitable for scenarios with multiple platforms and relatively fixed inspection personnel.

[0014] Although GNSS positioning meets the inspection and positioning requirements of platform canopies, the vertical track positioning accuracy is blocked by the canopy cover, and the positioning is offset out of the platform range and located on the track. The along-track positioning accuracy error is about 2-5 meters, and the positioning delay is large. Therefore, the GNSS positioning accuracy does not meet the requirements of platform canopy inspections. There are no mature algorithms and supporting systems on the market that can be used to improve GNSS positioning accuracy. It is necessary to develop corresponding algorithms to assist positioning for platform canopy inspection scenarios to improve positioning accuracy.

[0015] In summary, the significance of the present invention is:

[0016] Aiming at the characteristics of indoor positioning scenarios for railway platform canopy inspections, and based on the different positioning accuracy requirements in the along-track and perpendicular-track directions, in order to achieve low-cost GNSS-based positioning, the present invention proposes a positioning optimization algorithm based on the target scenario to solve the problems of GNSS positioning offsetting the platform in the perpendicular-track direction, large positioning accuracy error in the along-track direction, and GNSS signal loss due to shading by the canopy roof. This improves the positioning accuracy of inspection personnel and meets the positioning requirements of daily inspections of railway platform canopies.

[0017] This positioning method has higher robustness and positioning accuracy, does not require the deployment of a large number of additional hardware equipment, is easy to promote, and has great advantages and commercial prospects in platform canopy inspection and positioning application scenarios in complex environments. Summary of the Invention

[0018] In response to the above-mentioned problems, the present invention proposes a target optimization positioning method and system for railway platform canopy inspection. The specific scheme is as follows:

[0019] A target optimization positioning method for railway platform canopy inspection includes the following steps:

[0020] S1: Initialize the GNSS positioning, gyroscope sensor, and acceleration sensor data of the mobile device, obtain the longitude and latitude coordinates of the four corner points of each platform of the platform canopy, perform UTM plane rectangular coordinate conversion, and obtain the starting coordinates, end coordinates, and length of the platform along the track to the central axis;

[0021] S2: Obtain the GNSS positioning of the inspection personnel in real time, convert the obtained longitude and latitude coordinates into UTM plane rectangular coordinates, and then use vector dot product calculation to obtain the projection point of the positioning point on the along-track central axis;

[0022] S3: Calculate the distance between the positioning point and the projection points of different platforms, take the absolute minimum value to determine the platform where the current inspector is located, and use the projection point along the track to the central axis as the real-time positioning of the inspector, so that the inspector can move along the central axis of the platform without deviating from the platform range;

[0023] S4: Obtain data from the accelerometer, as well as the current and previous GNSS positioning information, calculate the distance between adjacent positioning information, and determine whether the current inspection personnel is stationary or in motion based on the difference between the accelerometer value and the distance between adjacent positioning information.

[0024] S5: When it is determined that the inspection personnel is stationary, the average value of the multiple GNSS positioning information obtained with delay is calculated and the positioning information is returned;

[0025] S6: When it is determined that the inspection personnel is in motion, the GNSS positioning information of the calibration point and its projection point on the central axis are obtained, the offset value between the inspection personnel in motion and the calibration point on the central axis is calculated, the sum of the inspection personnel's real-time GNSS positioning information and the offset value is used as the input value, the Kalman filter prediction model is introduced, the state value is updated in real time after multiple predictions, and the real-time positioning information is returned;

[0026] S7: Convert the calculated positioning information into longitude and latitude coordinate values to drive the movement of the character model, obtain data information from the gyroscope sensor, and drive the direction and perspective rotation of the character model.

[0027] Preferably, in step S2, the vector dot product operation refers to:

[0028] It is known that vector A=(A(lat),A(lng)) is the vector from the starting point of the central axis to the positioning point, and vector B=(B(lat),B(lng)) is the vector from the starting point to the end point along the central axis;

[0029] A(lat) and A(lng) are the components of vector A in the longitude and latitude directions, and B(lat) and B(lng) are the components of vector B in the longitude and latitude directions;

[0030] Through vector dot product operation , |A|cos(θ) is the projection length of A on vector B, and the projection point and projection length of the inspection personnel on the central axis are obtained. Through linear interpolation, the longitude and latitude coordinate values of the current positioning point are calculated.

[0031] Preferably, in step S3, the vertical distance between the positioning point and the projection points of different platforms is a positive value or a negative value. To determine the platform where the inspection personnel is located, the absolute minimum value of the vertical distance is taken.

[0032] Preferably, in step S4, determining whether the current inspection personnel is in a stationary state or a moving state includes the following methods:

[0033] S41: When the acceleration obtained by the acceleration sensor is less than the normal walking acceleration of an adult, the current state is static;

[0034] S42: When the difference between the positioning coordinate values of adjacent GNSS is less than the distance a normal adult moves per second, the current state is static;

[0035] S43: When the corresponding positioning distance is greater than the moving distance per second of a normal adult, and the acceleration obtained by the acceleration sensor is greater than the walking acceleration of a normal adult, the current state is motion.

[0036] Preferably, in step S5, the average value is calculated by directly summing the number of GNSS positioning information obtained and then taking the average value, or by assigning weights according to the order in which the GNSS positioning information is obtained and then calculating the weighted average value.

[0037] Preferably, in step S6, the Kalman filter algorithm inputs the measurement value, calculates the Kalman gain through the predicted state evaluation value and covariance matrix, and then updates the state evaluation value and covariance matrix according to the measurement value, and inputs the sum of the current GNSS positioning and offset value as the measurement value, and performs prediction updates by obtaining uninterrupted positioning information input in real time.

[0038] Preferably, in step S6, the offset value is the offset value between the current GNSS positioning of the inspection personnel and the current calibration point on the central axis. There is one or more calibration points, and the number of corresponding offset values is consistent with the number of calibration points.

[0039] Further preferably, the calibration point is a reference point whose positioning information is unique and easily identifiable.

[0040] The second purpose of the present invention is:

[0041] A target optimization positioning system for railway platform canopy inspection, which applies a target optimization positioning method for railway platform canopy inspection, includes the following modules:

[0042] Data acquisition and storage module, used to collect and store GNSS positioning data, acceleration sensor data, gyroscope sensor information and other related information;

[0043] Platform data processing module, used to realize UTM conversion of platform coordinates and calculation of platform centerline coordinates;

[0044] The positioning data processing module calculates and projects the real-time GNSS positioning data of the inspection personnel onto the platform's track-aligned central axis to obtain the coordinate value of the projection point.

[0045] Platform positioning and identification module, used to calculate the distance between the inspection personnel and the projection points of different platforms and identify the current platform;

[0046] Inspection status judgment module, used to calculate the acceleration sensor value and the distance difference between adjacent positioning, and identify whether the current inspection personnel is in a stationary state or in motion;

[0047] Inspection positioning optimization module, which is used to adopt different algorithms according to the different status of the inspection personnel to return optimized positioning information;

[0048] The model drive control module is used to drive the character model and adjust the viewing angle according to the optimized positioning information.

[0049] Preferably, the target optimization positioning system for railway platform canopy inspection is mounted on a mobile terminal with GNSS positioning, a gyroscope sensor and an acceleration sensor.

[0050] The beneficial effects of the present invention are:

[0051] The present invention obtains the projection point of the positioning point along the track to the central axis through vector dot product operation, and determines the positioning algorithm of the platform where it is located, which solves the problem that GNSS positioning cannot be used for indoor positioning scenarios with full coverage of platform canopies, and ensures that the positioning of inspection personnel does not deviate from the platform range.

[0052] By calculating the difference between the acceleration sensor value and the distance to the corresponding positioning point, the real-time status of the inspection personnel is judged, and the along-track positioning calibration is achieved through the average value method and Kalman filter principle, which solves the problem of large delays in GNSS positioning caused by interference from the platform's magnetic field. At the same time, it reduces the risk of large positioning deviations of inspection personnel, effectively improves the accuracy of inspection personnel's positioning, and ensures the inspection efficiency and real-time performance.

[0053] Compared with the prior art, the advantages of the present invention further include:

[0054] Through targeted optimization of the along-track and perpendicular-track positioning algorithms, the problem of low GNSS positioning accuracy and its failure to meet the positioning requirements for railway platform canopy inspections was resolved. This improved the real-time positioning accuracy of inspectors, thereby enhancing the visual tracking level of inspection trajectories and the real-time early warning capabilities of inspections. This is beneficial to improving the structural safety of platform canopies, ensuring the operational safety of railways and the personal safety of passengers.

[0055] The technical solution of the present invention is based on the scene positioning requirements of railway platform canopy inspections in complex surrounding environments. It meets the existing railway building construction, operation and maintenance management and operation modes, and does not require additional hardware equipment. It has good economic applicability and is easy to promote. It has great advantages and commercial prospects.

[0056] The technical solution of the present invention is conducive to improving the work efficiency of platform canopy inspectors, improving the ability to detect defects during inspections and provide real-time warnings, which is conducive to reducing the overall risk of platform canopies and improving the efficiency and level of operation and maintenance management.

[0057] The technical solution of the present invention improves the positioning accuracy of indoor positioning scenarios without adding hardware equipment. It can be extended to narrow and long indoor positioning scenarios with different positioning accuracy requirements in two directions. The technical solution has strong versatility and great promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that these drawings are designed for illustrative purposes only and are not intended to limit the scope of the present invention. Furthermore, unless otherwise specified, these drawings are intended only to conceptually illustrate the structures described herein and are not necessarily drawn to scale.

[0059] Figure 1 This is a flow chart of a target optimization positioning method for railway platform canopy inspection provided by Example 1 of the present invention;

[0060] Figure 2 This is a flow chart of the central axis positioning algorithm mentioned in steps S1 to S3 provided in Example 1 of the present invention;

[0061] Figure 3 This is a logic flow chart for determining the current state of step S4 provided in Example 1 of the present invention;

[0062] Figure 4 This is a flow chart of the average value algorithm and Kalman filter algorithm in steps S5 to S7 provided in Example 1 of the present invention;

[0063] Figure 5 This is a schematic diagram of obtaining the projection point of the positioning point on the along-track central axis through vector dot product calculation provided by Example 1 of the present invention;

[0064] Figure 6 This is a schematic diagram of the system modules provided in Example 2 of the present invention.

[0065] In the picture:

[0066] 1-Data acquisition and storage module; 2-Platform data processing module; 3-Positioning data processing module; 4-Platform positioning identification module; 5-Inspection status judgment module; 6-Inspection positioning optimization module; 7-Model driven control module. DETAILED DESCRIPTION

[0067] First of all, it should be noted that the specific structure, features and advantages of the present invention will be described in detail below by way of example. However, all descriptions are for illustration only and should not be understood as limiting the present invention. In addition, any single technical feature described or implied in the embodiments mentioned herein, or any single technical feature shown or implied in the drawings, can still be combined or deleted between these technical features to obtain more other embodiments of the present invention that may not be directly mentioned herein. In addition, in order to simplify the drawings, the same or similar technical features may be marked in only one place in the same drawing.

[0068] In the present invention, unless otherwise clearly stipulated and limited, the terms "install", "set", "connect", "fix", "screw" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integrated connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. Unless otherwise clearly defined, ordinary technicians in this field can understand the specific meanings of the above terms in the present invention according to the specific circumstances.

[0069] The following is combined with Figure 1 -Attached Figure 6 The present invention will be described in detail. Example 1

[0070] A target optimization positioning method for railway platform canopy inspection includes the following steps:

[0071] S1: Initialize the GNSS positioning, gyroscope sensor, and acceleration sensor data of the mobile device, obtain the longitude and latitude coordinates of the four corner points of each platform of the platform canopy, perform UTM plane rectangular coordinate conversion, and obtain the starting coordinates, end coordinates, and length of the platform along the track to the central axis;

[0072] S2: Obtain the GNSS positioning of the inspection personnel in real time, convert the obtained longitude and latitude coordinates into UTM plane rectangular coordinates, and then use vector dot product calculation to obtain the projection point of the positioning point on the along-track central axis;

[0073] S3: Calculate the distance between the positioning point and the projection points of different platforms, take the absolute minimum value to determine the platform where the current inspector is located, and use the projection point along the track to the central axis as the real-time positioning of the inspector, so that the inspector can move along the central axis of the platform without deviating from the platform range;

[0074] This step is used to solve the problem of excessive GNSS positioning deviation and out-of-station range;

[0075] S4: Obtain data from the accelerometer, as well as the current and previous GNSS positioning information, calculate the distance between adjacent positioning information, and determine whether the current inspection personnel is stationary or in motion based on the difference between the accelerometer value and the distance between adjacent positioning information.

[0076] S5: When it is determined that the inspection personnel is stationary, the average value of the multiple GNSS positioning information obtained with delay is calculated and the positioning information is returned;

[0077] This step is used to solve the problem of large delays in GNSS positioning caused by interference;

[0078] S6: When it is determined that the inspection personnel is in motion, the GNSS positioning information of the calibration point and its projection point on the central axis are obtained, the offset value between the inspection personnel in motion and the calibration point on the central axis is calculated, the sum of the inspection personnel's real-time GNSS positioning information and the offset value is used as the input value, the Kalman filter prediction model is introduced, the state value is updated in real time after multiple predictions, and the real-time positioning information is returned;

[0079] This step is used to reduce the risk of large deviations in the inspection personnel's positioning along the track;

[0080] S7: Convert the calculated positioning information into longitude and latitude coordinate values to drive the movement of the character model, obtain data information from the gyroscope sensor, and drive the direction and perspective rotation of the character model.

[0081] The innovation of the target optimization positioning method for railway platform canopy inspection provided by an embodiment of the present invention lies in: without adding hardware equipment, the present invention solves the problems of low GNSS positioning accuracy and failure to meet the positioning scenarios of railway platform canopy inspection through target optimization of along-track and perpendicular-track positioning algorithms, thereby improving the positioning accuracy of indoor positioning scenarios. The technical solution has strong versatility and great promotion value.

[0082] Furthermore, in the embodiment, it can also be considered that in step S2, the vector dot product operation refers to:

[0083] It is known that vector A=(A(lat),A(lng)) is the vector from the starting point of the central axis to the positioning point, and vector B=(B(lat),B(lng)) is the vector from the starting point to the end point along the central axis;

[0084] A(lat) and A(lng) are the components of vector A in the longitude and latitude directions, and B(lat) and B(lng) are the components of vector B in the longitude and latitude directions;

[0085] Through vector dot product operation , |A|cos(θ) is the projection length of A on vector B, and the projection point and projection length of the inspection personnel on the central axis are obtained. Through linear interpolation, the longitude and latitude coordinate values of the current positioning point are calculated.

[0086] Furthermore, in the embodiment, it can also be considered that in step S3, the vertical distance between the positioning point and the projection points of different platforms is positive or negative. In order to determine the platform where the inspection personnel is located, the absolute minimum value of the vertical distance is taken.

[0087] In this embodiment, the vertical distance between the positioning point and the projection points of different platforms can be positive or negative. Therefore, to determine the platform where the inspector is located, the absolute minimum value is used. Furthermore, the inspector's displayed location can only appear on the platform and will not appear in other locations such as the station building or on the track.

[0088] Furthermore, in the embodiment, it can also be considered that in step S4, determining whether the current inspection personnel is in a stationary state or a moving state includes the following methods:

[0089] S41: When the acceleration obtained by the acceleration sensor is less than the normal walking acceleration of an adult, the current state is static;

[0090] S42: When the difference between the positioning coordinate values of adjacent GNSS is less than the distance a normal adult moves per second, the current state is static;

[0091] S43: When the corresponding positioning distance is greater than the moving distance per second of a normal adult, and the acceleration obtained by the acceleration sensor is greater than the walking acceleration of a normal adult, the current state is motion.

[0092] In this embodiment, the walking acceleration of a normal adult is 1-1.5 m / s, and the distance a normal adult moves per second is 1-1.5 meters.

[0093] Furthermore, in the embodiment, it can also be considered that in step S5, the average value is calculated by directly summing the number of GNSS positioning information obtained and then taking the average value, or by assigning weights according to the order in which the GNSS positioning information is obtained and then calculating the weighted average value.

[0094] Furthermore, it can also be considered in the embodiment that in step S6, the Kalman filter algorithm inputs the measurement value, calculates the Kalman gain through the predicted state evaluation value and covariance matrix, and then updates the state evaluation value and covariance matrix according to the measurement value, and inputs the sum of the current GNSS positioning and offset value as the measurement value, and performs prediction updates by obtaining uninterrupted positioning information input in real time.

[0095] Furthermore, in the embodiment, it can also be considered that in step S6, the offset value is the offset value between the current GNSS positioning of the inspection personnel and the current calibration point on the central axis. The calibration point can be one or more, and the number of corresponding offset values is consistent with the number of calibration points.

[0096] Furthermore, in the embodiments, it can also be considered that the calibration point is a reference point whose positioning information is unique and easily identifiable.

[0097] In this embodiment, the calibration points are reference points such as platform corners, awning poles, and station buildings, which have unique positioning information and are easily identifiable.

[0098] Furthermore, in the embodiment, it can also be considered that in step S1, the mobile device should have the functions of GNSS positioning, gyroscope sensor and acceleration sensor, including mobile phones, tablet computers, smart terminals, handheld terminals, etc. The mobile device system can be IOS, Android, Hongmeng, etc.

[0099] Furthermore, in the embodiment, it is also considered that in step S1, the platform canopy can be a canopy with platform columns or a canopy without platform columns at stations such as high-speed railways, conventional railways, heavy-haul railways, and intercity railways. The platform canopy can be a concrete canopy, a steel structure canopy, a composite structure canopy, etc.

[0100] Furthermore, in the embodiment, it can also be considered that in step S1, the longitude and latitude coordinate values of the platform corner points can be obtained through map data such as Baidu and Amap, or through surveying and mapping.

[0101] Furthermore, in the embodiment, it can be considered that in step S1, the length of the platform centerline is generally 450 meters, 550 meters, etc. The width of the platform is generally 8 meters, 10.5 meters, 12 meters, etc.

[0102] Furthermore, in the embodiment, it can also be considered that in step S2, the GNSS positioning includes Beidou, GPS, GLONASS, Galileo, etc. Example 2

[0103] A target optimization positioning system for railway platform canopy inspections, which applies a target optimization positioning method for railway platform canopy inspections, includes the following modules:

[0104] Data acquisition and storage module 1, used to collect and store GNSS positioning data, acceleration sensor data, gyroscope sensor information and other related information;

[0105] Platform data processing module 2, used to realize the UTM conversion of platform coordinates and the calculation of platform centerline coordinates;

[0106] Positioning data processing module 3 projects the real-time GNSS positioning data of the patrol personnel onto the platform's track-aligned central axis after calculation to obtain the coordinate value of the projection point;

[0107] Platform positioning and identification module 4, used to calculate the distance between the inspection personnel and the projection points of different platforms and identify the current platform;

[0108] Inspection status judgment module 5, used to calculate the acceleration sensor value and the distance difference between adjacent positioning, and identify whether the current inspection personnel is in a stationary state or a moving state;

[0109] Inspection positioning optimization module 6, used to adopt different algorithms according to different states of inspection personnel to return optimized positioning information;

[0110] The model driving control module 7 is used to drive the character model and adjust the viewing angle according to the optimized positioning information.

[0111] Furthermore, in an embodiment, it can also be considered that the target optimization positioning system for the railway platform canopy inspection is mounted on a mobile terminal having GNSS positioning, a gyroscope sensor and an acceleration sensor.

[0112] Application Example 1:

[0113] To further illustrate the present invention, a high-speed railway passenger station cross-line steel structure canopy is taken as an example, and the present invention is described in detail in conjunction with the accompanying drawings:

[0114] Step S1: Initialize the GNSS positioning, gyroscope sensor, and acceleration sensor data information of an Android phone, obtain the longitude and latitude coordinates of the four corner points of each platform of a platform canopy, perform UTM plane rectangular coordinate conversion, and obtain the starting coordinates, end coordinates, and central axis length of the platform along the track.

[0115] In step S1, a high-speed railway passenger station cross-line steel structure canopy includes two platforms, each of which is 450 meters long and 12 meters wide.

[0116] In step S1, data information initialization is to obtain the GNSS positioning information of the current patrol personnel, obtain the horizontal rotation angle information and pitch angle information of the gyroscope sensor, determine the current holding state of the mobile phone, obtain the acceleration of the acceleration sensor, and determine the motion state of the current patrol personnel.

[0117] In step S1, the topography around a platform canopy is obtained through surveying and as-built drawings, and the longitude and latitude coordinates of a total of 8 corner points of the two platforms are extracted. The starting and end points of the two central axes of the two platforms are obtained after calculation. The calculated length of the central axis is 450 meters, which is consistent with the actual length.

[0118] Step S2: Get the GPS location of the patrol personnel in real time, convert the longitude and latitude coordinates obtained into UTM plane rectangular coordinates, and then get the projection point of the positioning point on the central axis along the track through vector dot product operation, such as Figure 5 shown.

[0119] In step S2, vector A = (A(lat), A(lng)) is the vector from the bottom midpoint of the central axis to the anchor point. A(lat) = the longitude coordinate value of the anchor point - the longitude coordinate value of the bottom midpoint, and A(lng) = the latitude coordinate value of the anchor point - the latitude coordinate value of the bottom midpoint.

[0120] Vector B = (B(lat), B(lng)) is the vector along the track from the bottom midpoint to the top midpoint of the central axis. B(lat) = the longitude coordinate of the bottom midpoint - the longitude coordinate of the top midpoint, and B(lng) = the latitude coordinate of the bottom midpoint - the latitude coordinate of the top midpoint.

[0121] Vector dot product operation . , |A| is the length of vector A, |B| is the length of vector B, so |A|cos(θ) is the length of the projection of vector A on vector B. ,The projection length of the current inspection personnel on the central axis can be obtained, and the longitude and latitude coordinates of the projection point can be calculated through linear interpolation.

[0122] Step S3: Calculate the distance between the current GNSS positioning point of the inspector and the projection points of different platforms, take the absolute minimum value to determine the platform where the current inspector is located, and use the projection point along the track to the central axis as the real-time positioning of the inspector, so that the inspector can move on the central axis of the platform without deviating from the platform range.

[0123] In step S3: the actual location of the inspection personnel should be on the platform, but the current GNSS positioning may be on the platform due to awning cover, signal interference, etc., or it may appear on the track, station building, etc. By calculating the distance between the current GNSS positioning point and the projection point of the central axis of Platform 1 and Platform 2, taking the absolute minimum value of the distance, it is determined that the current inspection personnel is located at Platform 1, and the longitude and latitude coordinate values of the projection point of Platform 1 are returned to drive the movement of the character model on the central axis.

[0124] Step S4: Obtain data information from the acceleration sensor, as well as the current GNSS positioning information and the previous GNSS positioning information, calculate the distance between adjacent positioning information, and determine whether the current inspection personnel is in a stationary state or a moving state based on the acceleration sensor value and the distance difference between adjacent positioning information.

[0125] In step S4, the acceleration sensor is initialized first, and the data information returned by the acceleration sensor is monitored in real time. If the acceleration sensor does not return data information, it is determined that the current state is static. If the acceleration sensor returns data information, the acceleration at this time is less than 0.8m / s 2 , it is determined that the current state is static.

[0126] In step S4, the returned GNSS positioning information is monitored in real time. The current GNSS positioning information is compared with the previous positioning information to calculate the difference. If the difference is 0, the current GNSS positioning information is invalid. If the difference between the two positioning information is less than 0.5 meters, the current state is determined to be stationary.

[0127] In step S4, when the difference between the two positioning information is not less than 0.5 meters and the acceleration at this time is not less than 0.8m / s 2 , it is judged that the current state is motion.

[0128] Step S5: If the inspector is determined to be stationary, multiple GNSS positioning information acquired with delays is aggregated and a weighted average is calculated based on the order in which the information was returned. The weighted average is given to the later-acquired positioning information, which is closer to the true information. If a large amount of positioning information is aggregated, a weighted average of 3 seconds or 3 seconds of collected positioning information can also be used.

[0129] In step S5, the returned positioning information is converted into UTM coordinates, and the vector dot product is performed according to step S2, and finally the longitude and latitude coordinate values of the projection point on the platform centerline are returned to drive the movement of the character model.

[0130] Step S6: When it is determined that the inspection personnel is in motion, a certain awning column is taken as the current positioning calibration point, the GNSS positioning information of the awning column is obtained, and its projection point information on the central axis is obtained according to step S2. The distance difference between the inspection personnel in motion and the projection point of the awning column calibration point on the central axis is calculated as the offset value.

[0131] In step S6, the Kalman filter algorithm takes the measurement as input, calculates the Kalman gain using the predicted state estimate and covariance matrix, and then updates the state estimate and covariance matrix based on the measurement. Therefore, the sum of the current GNSS position and the offset value serves as the measurement input, and the prediction is updated by continuously acquiring positioning information in real time.

[0132] In step S6, the Kalman filter prediction model is introduced, and the coordinate value of the inspection personnel's projection point on the central axis plus the offset value are used as input variables. Iterative prediction is performed at a frequency of 100 times per second. After the iteration of the Kalman filter prediction, it can be close to the real state. The 100th result after the iteration is returned, and the returned longitude and latitude coordinate values drive the movement of the character model.

[0133] In step S6, the positioning calibration can be repeated multiple times, and multiple reference points with unique and easily identifiable positioning information, such as awning columns, station building corners, and platform corners, can be selected as calibration points to obtain GNSS positioning information.

[0134] Step S7: Convert the positioning information returned in the static or moving state into longitude and latitude coordinate values to drive the movement of the character model. Use the horizontal rotation angle information and pitch angle information obtained by the gyroscope sensor to drive the movement of the character model's perspective.

[0135] The above embodiments describe the present invention in detail, but the contents described are only preferred embodiments of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A target optimization positioning method for railway platform canopy inspection, characterized in that: The following steps are involved: S1: Initialize the GNSS positioning, gyroscope sensor, and acceleration sensor data of the mobile device, obtain the longitude and latitude coordinates of the four corner points of each platform of the platform canopy, perform UTM plane rectangular coordinate conversion, and obtain the starting coordinates, end coordinates, and length of the platform along the track to the central axis; S2: Obtain the GNSS positioning of the inspection personnel in real time, convert the obtained longitude and latitude coordinates into UTM plane rectangular coordinates, and then use vector dot product calculation to obtain the projection point of the positioning point on the along-track central axis; S3: Calculate the distance between the positioning point and the projection points of different platforms, take the absolute minimum value to determine the platform where the current inspector is located, and use the projection point along the track to the central axis as the real-time positioning of the inspector, so that the inspector can move along the central axis of the platform without deviating from the platform range; S4: Obtain data from the accelerometer, as well as the current and previous GNSS positioning information, calculate the distance between adjacent positioning information, and determine whether the current inspection personnel is stationary or in motion based on the difference between the accelerometer value and the distance between adjacent positioning information. S5: When it is determined that the inspection personnel is stationary, the average value of the multiple GNSS positioning information obtained with delay is calculated and the positioning information is returned; S6: When it is determined that the inspection personnel is in motion, the GNSS positioning information of the calibration point and its projection point on the central axis are obtained, the offset value between the inspection personnel in motion and the calibration point on the central axis is calculated, the sum of the inspection personnel's real-time GNSS positioning information and the offset value is used as the input value, the Kalman filter prediction model is introduced, the state value is updated in real time after multiple predictions, and the real-time positioning information is returned; The offset value is the offset between the inspector's current GNSS positioning and the current calibration point on the central axis. There can be one or more calibration points, and the number of corresponding offset values is the same as the number of calibration points. S7: Convert the calculated positioning information into longitude and latitude coordinate values to drive the movement of the character model, obtain data information from the gyroscope sensor, and drive the direction and perspective rotation of the character model.

2. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized in that: In step S2, the vector dot product operation refers to: It is known that vector A=(A(lat),A(lng)) is the vector from the starting point of the central axis to the positioning point, and vector B=(B(lat),B(lng)) is the vector from the starting point to the end point along the central axis; A(lat) and A(lng) are the components of vector A in the longitude and latitude directions, and B(lat) and B(lng) are the components of vector B in the longitude and latitude directions; Through the vector dot product operation A・B=A(lat)・B(lat) + A(lng)・B(lng), |A|cos(θ) is the projection length of A on vector B. The projection point and projection length of the inspector on the central axis can be obtained. Through linear interpolation, the longitude and latitude coordinates of the current positioning point are calculated.

3. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized by: In step S3, the vertical distance between the positioning point and the projection points of different platforms is a positive value or a negative value. To determine the platform where the inspection personnel is located, the absolute minimum value of the vertical distance is taken.

4. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized in that: In step S4, it is determined whether the current patrol officer is in a stationary state or a moving state, including the following methods: S41: When the acceleration obtained by the acceleration sensor is less than the normal walking acceleration of an adult, the current state is static; S42: When the difference between the positioning coordinate values of adjacent GNSS is less than the distance a normal adult moves per second, the current state is static; S43: When the corresponding positioning distance is greater than the moving distance per second of a normal adult, and the acceleration obtained by the acceleration sensor is greater than the walking acceleration of a normal adult, the current state is motion.

5. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized in that: In step S5, the average value is calculated by directly summing the number of GNSS positioning information obtained and then taking the average value, or by assigning weights according to the order in which the GNSS positioning information is obtained and then calculating the weighted average value.

6. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized by: In step S6, the Kalman filter algorithm inputs the measurement value, calculates the Kalman gain through the predicted state evaluation value and covariance matrix, and then updates the state evaluation value and covariance matrix based on the measurement value. The sum of the current GNSS positioning and offset value is input as the measurement value, and the prediction update is performed by obtaining uninterrupted positioning information input in real time.

7. The target optimization positioning method for railway platform canopy inspection according to claim 1 is characterized by: Calibration points are reference points with unique positioning information that are easily identifiable.

8. A target optimization positioning system for railway platform canopy inspection, applying the target optimization positioning method for railway platform canopy inspection according to any one of claims 1 to 7, characterized in that: Includes the following modules: A data acquisition and storage module (1) is used to collect and store GNSS positioning data, acceleration sensor data, gyroscope sensor information and other related information; Platform data processing module (2), used to realize UTM conversion of platform coordinates and calculation of platform centerline coordinates; The positioning data processing module (3) projects the real-time GNSS positioning data of the patrol personnel onto the platform along the track axis after calculation to obtain the coordinate value of the projection point; Platform positioning and identification module (4), used to calculate the distance between the inspection personnel and the projection points of different platforms and identify the current platform; The inspection status judgment module (5) is used to calculate the acceleration sensor value and the distance difference between adjacent positionings to identify whether the current inspection personnel is in a stationary state or a moving state; Inspection positioning optimization module (6), used to adopt different algorithms according to different states of inspection personnel to return optimized positioning information; The model driving control module (7) is used to drive the character model and adjust the viewing angle according to the optimized positioning information.

9. The target optimization positioning system for railway platform canopy inspection according to claim 8, characterized in that: The target optimization positioning system for railway platform canopy inspection is mounted on a mobile terminal with GNSS positioning, a gyroscope sensor and an acceleration sensor.

Images