Railway business line perimeter electronic protection method, system and terminal

Abstract

The application relates to the technical field of railway protection, and provides a railway operating line perimeter electronic protection method, a railway operating line perimeter electronic protection system and a terminal. The method comprises the following steps: a three-dimensional electronic protection wall is established for separating a construction area and a railway forbidden entry area; point cloud data and echo data of a target mechanical arm frame are collected through a laser radar and a millimeter wave radar respectively, fusion target three-dimensional coordinates are generated based on meteorological states and fusion weights of the two types of radars are determined; the current distance and the current orientation relationship of the target mechanical arm frame relative to the three-dimensional electronic protection wall are determined according to the fusion target three-dimensional coordinates, the predicted position of the target mechanical arm frame is predicted by combining the fusion target three-dimensional coordinates and motion information collected at continuous time points, and an early warning level and a three-dimensional intrusion early warning point are determined. The application can improve the reliability of three-dimensional positioning and intrusion limit early warning of the mechanical arm frame under the complex meteorological conditions and the adjacent operating line construction scene, and output a risk early warning before the target oversteps the boundary.

Description

Technical Field

[0001] This application relates to the field of railway protection technology, specifically to an electronic protection method, system and terminal for the perimeter of a railway operating line. Background Technology

[0002] During construction near operational railway lines, the booms of large construction machinery such as excavators, crawler cranes, and pump trucks have a large extension range and complex movement trajectories, making it easy for them to exceed the construction permit boundaries and intrude into railway safety clearances. Especially during continuous day and night operations, and in environments such as rain, snow, fog, dust, tree obstruction, and darkness, it is difficult for manual observation and ordinary video surveillance to continuously and accurately identify the spatial position of the machinery boom, and also difficult to determine the actual distance between the machinery boom and the railway safety boundary.

[0003] Existing construction work near operational railway lines typically employs physical barriers and is monitored on-site by dedicated safety personnel. However, manual monitoring is susceptible to factors such as personnel fatigue, obstructed visibility, and differences in experience, making it difficult to accurately judge the extension trend of the mechanical boom. Physical barriers primarily prevent construction personnel or equipment from entering designated areas and cannot predict the risk of the mechanical boom crossing boundaries based on its current position and movement. Therefore, existing safety measures usually only take action after the mechanical boom has approached or entered the railway safety clearance, representing reactive protection and failing to provide early prediction and warning of the risk of the mechanical boom extending beyond the clearance limits. Summary of the Invention

[0004] This application aims to address the problems in the prior art by providing a method, system, and terminal for electronic protection of railway operating line perimeter that can predict risks in advance.

[0005] To achieve the above objectives, the first aspect of this application provides an electronic protection method for the perimeter of a railway operating line, comprising:

[0006] The railway line centerline, safety distance threshold, catenary height threshold, and ground filtering threshold are obtained, and a three-dimensional electronic protective wall is established based on these thresholds to separate the construction area from the railway restricted area. Point cloud data of the target robotic arm is acquired by LiDAR, and the three-dimensional coordinates of the target robotic arm by LiDAR are determined based on the point cloud data; echo data of the target robotic arm is acquired by millimeter-wave radar, and the planar coordinates and motion information of the target robotic arm by millimeter-wave radar are determined based on the echo data. Obtain the weather conditions of the area where the target robotic arm is located, and determine the fusion weights of lidar and millimeter-wave radar based on the weather conditions; Based on the fusion weights of lidar and millimeter-wave radar, the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are fused, and the fused height coordinates are determined according to the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm. Based on the fused target's three-dimensional coordinates, determine the current distance from the target robotic arm to the three-dimensional electronic protective wall, as well as the current orientation of the target robotic arm relative to the three-dimensional electronic protective wall; Based on the fused target's three-dimensional coordinates and motion information corresponding to the continuous acquisition time, the three-dimensional motion speed and motion state of the target robotic arm are determined, and the predicted position of the target robotic arm at a preset future time is predicted based on the three-dimensional motion speed. Based on the current distance, current orientation, and the positional relationship between the predicted location and the three-dimensional electronic protective wall, the warning level and three-dimensional intrusion warning point corresponding to the target robotic arm are determined, and warning information corresponding to the warning level is generated.

[0007] A second aspect of this application provides an electronic perimeter protection system for railway operating lines, comprising: The protective wall establishment module is used to obtain the railway line centerline, safety distance threshold, catenary height threshold and ground filtering threshold, and based on the railway line centerline, the safety distance threshold, the catenary height threshold and the ground filtering threshold, to establish a three-dimensional electronic protective wall to separate the construction area from the railway restricted area; The target perception module is used to acquire point cloud data of the target robotic arm using lidar and determine the lidar target three-dimensional coordinates of the target robotic arm based on the point cloud data; and to acquire echo data of the target robotic arm using millimeter-wave radar and determine the millimeter-wave target planar coordinates and motion information of the target robotic arm based on the echo data. The fusion weight determination module is used to obtain the meteorological conditions of the area where the target robotic arm is located, and to determine the fusion weight of lidar and the fusion weight of millimeter-wave radar based on the meteorological conditions. The coordinate fusion module is used to fuse the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target based on the lidar fusion weight and the millimeter-wave radar fusion weight, and to determine the fused height coordinates based on the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm. The position relationship determination module is used to determine the current distance from the target robotic arm to the three-dimensional electronic protective wall, and the current orientation relationship of the target robotic arm relative to the three-dimensional electronic protective wall, based on the three-dimensional coordinates of the fused target. The trajectory prediction module is used to determine the three-dimensional motion speed and motion state of the target robotic arm based on the fused target three-dimensional coordinates and motion information corresponding to the continuous acquisition time, and to predict the predicted position of the target robotic arm at a preset future time based on the three-dimensional motion speed. The early warning generation module is used to determine the early warning level and three-dimensional intrusion early warning point corresponding to the target robotic arm based on the current distance, the current orientation relationship, and the positional relationship between the predicted position and the three-dimensional electronic protective wall, and to generate early warning information corresponding to the early warning level.

[0008] A third aspect of this application provides a terminal device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the above-mentioned electronic protection method for the perimeter of railway operating lines.

[0009] The fourth aspect of this application provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described electronic protection method for the perimeter of a railway operating line.

[0010] Compared with the prior art, this application has the following beneficial effects: A three-dimensional electronic protective wall is established based on the railway line centerline and safety thresholds in both horizontal and vertical directions, effectively separating construction areas from railway restricted zones in three-dimensional space. By integrating the three-dimensional positioning capabilities of lidar with the all-weather detection and speed measurement capabilities of millimeter-wave radar, the continuity of the target robotic boom's coordinates and movement status can be improved under complex weather conditions. Simultaneously, by determining the target robotic boom's current distance, current orientation, and predicted position, it can not only determine whether the target robotic boom has intruded into the railway restricted zone but also provide early warnings before the target robotic boom crosses the boundary, outputting corresponding three-dimensional intrusion warning points. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a method provided in an embodiment of this application; Figure 2 This is a schematic diagram of a system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0013] To facilitate understanding of this application, the following description will be more comprehensive and detailed in conjunction with the accompanying drawings and preferred embodiments, but the scope of protection of this application is not limited to the following specific embodiments.

[0014] Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by those skilled in the art. The technical terms used herein are for the purpose of describing particular embodiments only and are not intended to limit the scope of this application.

[0015] Please see Figure 1 One embodiment of this application provides an electronic protection method for the perimeter of a railway operating line, comprising the following steps: S102. Obtain the railway line centerline, safety distance threshold, catenary height threshold, and ground filtering threshold, and based on the railway line centerline, safety distance threshold, catenary height threshold, and ground filtering threshold, establish a three-dimensional electronic protective wall to separate the construction area and the railway restricted area.

[0016] In this step, the centerline of the railway line can be determined based on on-site survey data, and a global rectangular coordinate system can be established in local sections where the centerline can be approximated as a straight line. The X-axis of the global rectangular coordinate system points horizontally towards the construction area, the Y-axis is parallel to the railway line direction, and the Z-axis is perpendicular to the ground and upwards. The equation of the railway line centerline in the XOY plane can be expressed as:

[0017] Where A, B, and C are the equation coefficients obtained by fitting the surveyed coordinates of the railway line centerline. For curved sections, the railway line centerline can be divided into multiple local segments, and the centerline equation of each local segment can be fitted separately, connecting the corresponding protection boundaries of adjacent local segments.

[0018] any target point in space Horizontal distance to the centerline of the railway line It can be represented as:

[0019] According to the safe distance threshold Determine the intervals from the centerline of the railway line on both sides. Two horizontal safety boundaries are used to define the two sides of the railway restricted area. The safety distance threshold can be determined based on the railway section type and on-site construction requirements. For example, 10m can be used for ordinary roadbed sections, 15m for electrified catenary sections, and 20m for bridge sections.

[0020] Contact wire height threshold The ground filtering threshold is used to determine the upper boundary of the height of the 3D electronic protective wall, and to filter out interfering targets such as weeds and gravel near the ground. In an embodiment where the ground filtering threshold is 0.5m, the discrete mesh vertex set of the 3D electronic protective wall can be represented as:

[0021] in, The coordinates of the discrete grid vertices are given. When other values ​​are used for the ground filtering threshold, the lower boundary of the height of set S is adopted using the corresponding ground filtering threshold. Connecting adjacent vertices in set S forms a three-dimensional mesh-like electronic protective wall. The grid cell size can be set to, for example, 0.5m × 0.5m. The area between the three-dimensional electronic protective walls on both sides of the railway line centerline is marked as the railway restricted area. The railway restricted area extends in the height direction from the ground filtering threshold to the contact wire height threshold. The areas on the sides of the two three-dimensional electronic protective walls away from the railway line centerline are marked as construction areas.

[0022] S104. Collect point cloud data of the target robotic arm using lidar, and determine the three-dimensional coordinates of the target robotic arm using lidar based on the point cloud data; collect echo data of the target robotic arm using millimeter-wave radar, and determine the millimeter-wave radar target plane coordinates and motion information of the target robotic arm based on the echo data.

[0023] In this step, the lidar emits laser pulses towards the target robotic arm and receives the laser echoes reflected from the target robotic arm. The distance from the lidar to the target robotic arm... It can be represented as:

[0024] in, At the speed of light, This refers to the round-trip time difference of a laser pulse from emission to reception. LiDAR obtains distance information through horizontal and vertical scanning. Horizontal angle and pitch angle The point cloud data is processed by ground filtering, outlier filtering, and target clustering. The end of the target robotic arm is then determined from the point cloud corresponding to the target robotic arm, and the end of the target robotic arm is used as the target reference position.

[0025] The three-dimensional coordinates of a lidar target can be represented as:

[0026] in, The three-dimensional coordinates of the lidar target. This refers to the installation height of the lidar, used to compensate for the height difference between the lidar origin and the ground zero point.

[0027] Millimeter-wave radar can be a frequency-modulated continuous wave (FMCV) millimeter-wave radar. Millimeter-wave radar transmits frequency-modulated electromagnetic waves and determines the target distance based on the frequency difference between the transmitted signal and the target echo signal.

[0028] in, The difference frequency between the transmitted signal and the target echo signal. This represents the frequency modulation slope of the millimeter-wave radar. The millimeter-wave radar determines the target's planar coordinates based on the target range and horizontal azimuth. .

[0029] Millimeter-wave radar can also determine the radial velocity of the target robotic arm relative to the millimeter-wave radar based on the Doppler frequency shift:

[0030] in, Radial velocity, For millimeter wave wavelength, This is due to the Doppler frequency shift. The motion information output by millimeter-wave radar can include radial velocity, direction of motion, target range, and target azimuth.

[0031] The same target reference position is defined for both the lidar target's 3D coordinates and the millimeter-wave radar target's planar coordinates. The target reference position is the end of the target robotic arm and remains unchanged during continuous tracking. The lidar determines the target's 3D coordinates based on the point cloud corresponding to the end of the target robotic arm. The millimeter-wave radar determines the corresponding millimeter-wave target planar coordinates based on the detection position where the planar position and direction of movement of the target robotic arm's end satisfy a preset association condition. Both types of radar use the target robotic arm's end as the object for coordinate calculation and fusion.

[0032] S106. Obtain the weather conditions of the area where the target robotic arm is located, and determine the fusion weights of the lidar and millimeter-wave radar based on the weather conditions.

[0033] In this step, the weather conditions of the area where the target robotic arm is located can be obtained through on-site weather sensors or weather data interfaces. Weather conditions can include sunny, rainy, snowy, foggy, and dusty conditions. Since laser signals are easily affected by scattering from water vapor and suspended particulate matter, while millimeter-wave signals experience relatively less attenuation in rain, snow, fog, and dusty environments, the proportion of the two types of radar in the planar coordinate fusion can be adjusted according to the weather conditions.

[0034] Under clear conditions and with high-quality point clouds, the lidar fusion weight can be increased. In heavy rain, fog, blizzards, or dust storms, the lidar fusion weight can be decreased, while the millimeter-wave radar fusion weight can be increased. Both the lidar and millimeter-wave radar fusion weights are non-negative values, and their sum is 1. For example, under clear conditions, the lidar fusion weight can be 0.9, and the millimeter-wave radar fusion weight can be 0.1; under heavy rain or fog conditions, the lidar fusion weight can be 0.4, and the millimeter-wave radar fusion weight can be 0.6.

[0035] S108. Based on the fusion weights of the lidar and the millimeter-wave radar, the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are fused, and the fused height coordinates are determined according to the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm.

[0036] In this step, after time alignment, coordinate system transformation, and target association are completed between the lidar target's three-dimensional coordinates and the millimeter-wave radar target's planar coordinates, coordinate fusion is performed on the same target reference position detected by both types of radars. The fused target three-dimensional coordinates are then used. It can be represented as:

[0037] in, The three-dimensional coordinates of the lidar target. For millimeter-wave radar target plane coordinates, For lidar fusion weights, For millimeter-wave radar fusion weights, and satisfy Millimeter-wave radar has low vertical resolution, therefore the fused height coordinates are derived from the height coordinates in the three-dimensional target coordinate system of the lidar. .

[0038] S110. Based on the fused target's three-dimensional coordinates, determine the current distance between the target robotic arm and the three-dimensional electronic protective wall, as well as the current orientation of the target robotic arm relative to the three-dimensional electronic protective wall.

[0039] In this step, the current distance from the target robotic arm to the 3D electronic protective wall can be determined based on the shortest spatial distance between the fused target's 3D coordinates and the boundary surface of the 3D electronic protective wall. For a discrete mesh 3D electronic protective wall, the mesh cell closest to the fused target's 3D coordinates can be found, and the shortest distance from the fused target's 3D coordinates to that mesh cell can be calculated.

[0040] Based on the position of the target's 3D coordinates in the railway line centerline equation, the pre-marked construction area direction, and the boundary position of the 3D electronic protective wall, it can be determined whether the target robotic boom is on one side of the construction area, on the 3D electronic protective wall, or on the side entering the railway restricted area. The current distance indicates the degree to which the target robotic boom is close to the 3D electronic protective wall; the current distance is a non-negative value. The current positional relationship is used to distinguish which side of the 3D electronic protective wall the target robotic boom is located on.

[0041] Specifically, when the height coordinate in the fused target's three-dimensional coordinates is between the ground filtering threshold and the contact wire height threshold, the target robotic boom is determined to be located in the construction area, on the three-dimensional electronic protective wall, or in a railway restricted area based on the relationship between the horizontal distance from the target robotic boom to the railway line centerline and the safety distance threshold. When the height coordinate is outside the aforementioned height range, the target robotic boom is determined to be outside the effective protection range of the three-dimensional electronic protective wall, and the current distance from the target robotic boom to the three-dimensional electronic protective wall is the shortest spatial distance from the fused target's three-dimensional coordinates to the three-dimensional electronic protective wall.

[0042] S112. Based on the fused target three-dimensional coordinates and motion information corresponding to the continuous acquisition time, determine the three-dimensional motion speed and motion state of the target robotic arm, and predict the predicted position of the target robotic arm at a preset future time based on the three-dimensional motion speed.

[0043] In this step, the three-dimensional motion direction and velocity of the target robotic arm are determined based on the fused three-dimensional coordinates of the target at multiple consecutive acquisition moments. The radial velocity output by the millimeter-wave radar is then used to correct the velocity determined based on the coordinate changes. The motion states can include a stationary state, a motion state close to the three-dimensional electronic defense wall, a motion state far from the three-dimensional electronic defense wall, and a motion state along the three-dimensional electronic defense wall.

[0044] Based on the current fused target 3D coordinates and 3D motion velocity of the target robotic arm, a first-order linear trajectory prediction method can be used to determine the predicted position of the target robotic arm at a preset future time.

[0045] in The current fusion target's three-dimensional coordinates. , and Let the target robotic arm move at its velocity along the three coordinate axes. To predict the time step, the predicted position is compared with a 3D electronic protective barrier to determine whether the target robotic arm will approach or enter the railway restricted area at a preset future time.

[0046] S114. Based on the current distance, current orientation, and the positional relationship between the predicted location and the three-dimensional electronic protective wall, determine the warning level and three-dimensional intrusion warning point corresponding to the target robotic arm, and generate warning information corresponding to the warning level.

[0047] In this step, when the target robotic boom is located on one side of the construction area, the proximity of the target robotic boom can be judged based on the current distance, and whether the predicted position will enter the railway restricted area can be used to determine if there is a risk of the target robotic boom crossing the boundary after continuing to move. If the target robotic boom is currently far away, but the predicted position will enter the railway restricted area, the warning level can be raised in advance.

[0048] In a specific example, a Level 1 warning can be issued when the target robotic boom is located on one side of the construction area and the current distance is greater than 3 meters but not more than 6 meters. A Level 2 warning can be issued when the target robotic boom is located on one side of the construction area and the current distance is greater than 0 meters but not more than 3 meters. An emergency intrusion can be issued when the current location indicates that the target robotic boom is located on a three-dimensional electronic protective wall or has entered a railway restricted area. The warning distance range and warning level can be set according to the safety management requirements of different railway sections.

[0049] When the target robotic arm has already entered the railway restricted area, the fused 3D coordinates of the target entering the restricted area or the target position closest to the 3D electronic protective wall can be determined as the 3D intrusion warning point. When the predicted position is about to enter the railway restricted area, the intersection of the predicted trajectory and the 3D electronic protective wall or the protective wall grid position corresponding to the predicted position can be determined as the 3D intrusion warning point.

[0050] The early warning information can include the sensor unit identifier, warning time, warning level, current distance, coordinates of the 3D intrusion warning point, motion status of the target robotic arm, radar point cloud snapshot, and on-site video. A level 1 warning can trigger a local light alert, a level 2 warning can trigger a local audible and visual alarm and send an alarm message to on-site security personnel, and an emergency intrusion can trigger a continuous audible and visual alarm and upload the 3D intrusion warning point to the digital twin platform for marking.

[0051] In this embodiment, a three-dimensional electronic protective wall is established based on the railway line centerline and safety thresholds in the horizontal and vertical directions, which can separate the construction area and the railway restricted area in three-dimensional space. By integrating the three-dimensional positioning capability of lidar and the all-weather detection and speed measurement capability of millimeter-wave radar, the continuity of the target robotic boom's coordinates and movement status under complex weather conditions can be improved. By simultaneously judging the target robotic boom's current distance, current orientation, and predicted position, it is possible not only to determine whether the target robotic boom has intruded into the railway restricted area, but also to provide early warning before the target robotic boom crosses the boundary, and output the corresponding three-dimensional intrusion warning point.

[0052] In one embodiment, before fusing the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target, the point cloud data and echo data are timestamped and then the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are transformed to the same global coordinate system.

[0053] Specifically, lidar and millimeter-wave radar each add a timestamp to the collected data. The edge computing agent uses these timestamps to identify point cloud data and echo data within the same time synchronization range as data collected at the same time. For two types of radar with different sampling frequencies, the target positions detected by the two types of radar can be aligned to the same reference time based on data from adjacent collection times and the motion information output by the millimeter-wave radar.

[0054] The installation positions and orientations of the lidar and millimeter-wave radar in the global coordinate system are pre-calibrated. Based on the calibration results, the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are transformed to the same global coordinate system, so that the two types of coordinates have a consistent coordinate origin, coordinate axis direction, and length unit.

[0055] After timestamp alignment and coordinate system transformation, the spatial distance between the planar coordinates of the lidar target in three-dimensional coordinates and the planar coordinates of the millimeter-wave radar target is calculated. Simultaneously, based on the target position changes corresponding to multiple consecutive acquisition times, the planar motion directions of the lidar-detected target and the millimeter-wave radar-detected target are determined, and the angle between the two planar motion directions is also determined. Preset association conditions include: moving target association conditions and stationary target association conditions. The moving target association condition is that the angle between the planar motion directions of the two detected targets is not greater than a preset direction angle threshold. The stationary target association condition is that the position changes of the two detected targets within a preset association time window are not greater than a preset stationary determination threshold. When one detected target meets the stationary target association condition while the other does not, it is determined that the consistency of the motion directions of the two detected targets does not meet the preset association condition.

[0056] When the spatial distance is no greater than a preset association distance, and the two detected targets meet the association conditions for moving targets or stationary targets, the corresponding lidar-detected target and millimeter-wave radar-detected target are identified as the same target robotic arm, and the coordinates corresponding to that target robotic arm are fused. When multiple candidate detected targets exist, targets with smaller spatial distances and a higher degree of consistency in movement direction can be prioritized for association. Detected targets that do not meet the association conditions remain as different targets and are not subject to coordinate fusion.

[0057] This embodiment eliminates positional deviations caused by different acquisition times of the two types of radars by aligning timestamps, eliminates coordinate deviations caused by different sensor installation positions and orientations by using a unified coordinate system, and associates targets by using consistency in spatial distance and direction of motion. This can reduce the possibility of incorrectly fusing different reference positions on different machines or mechanical arms.

[0058] In one embodiment, determining the lidar fusion weight and the millimeter-wave radar fusion weight based on weather conditions includes: The degree of meteorological attenuation of the laser signal is determined based on the meteorological conditions. The number of effective points in the point cloud data and the continuity of the target position corresponding to the continuous acquisition time are obtained. The reliability of the lidar is determined based on the degree of meteorological attenuation, the number of effective points, and the continuity of the target position.

[0059] Specifically, a pre-established correspondence between weather conditions and the degree of meteorological attenuation of the laser signal can be established. Clear weather corresponds to a lower degree of meteorological attenuation, while rain, snow, fog, and dust storms correspond to different degrees of attenuation. When rainfall intensity, visibility, or dust concentration changes, the degree of meteorological attenuation of the laser signal can be adjusted accordingly. For example, with clear weather as the initial value, different attenuation values ​​can be added to the initial value for rain, snow, fog, or dust storms.

[0060] Valid points are points in the point cloud data that still belong to the target robotic arm after ground filtering, invalid distance filtering, and outlier filtering. The number of valid points indicates the completeness of the LiDAR's acquisition of the target robotic arm's outline. Target position continuity indicates whether the target position is continuous at adjacent acquisition times and whether the change in target position conforms to the previous movement trend of the target robotic arm. Within a preset continuity judgment time window, when the change in target position at adjacent acquisition times does not exceed a preset position change threshold, the corresponding target position is determined as a continuous position, and the target position continuity is determined based on the number of continuous positions. A LiDAR reliability correspondence table is pre-established. Each record in the LiDAR reliability correspondence table includes a meteorological attenuation degree interval, a valid point number interval, a target position continuity interval, and the corresponding LiDAR reliability. The reliability corresponding to the record in the correspondence table where the current meteorological attenuation degree, valid point number, and target position continuity all fall into the table is determined as the LiDAR reliability. The higher the meteorological attenuation degree, the lower the LiDAR reliability; the higher the valid point number and target position continuity, the higher the LiDAR reliability.

[0061] A millimeter-wave radar reliability mapping table is pre-established. Each record in the table includes the echo signal-to-noise ratio (SNR) range, the continuous tracking stability range, and the corresponding millimeter-wave radar reliability. The reliability of the record corresponding to which both the current echo SNR and continuous tracking stability fall is determined as the millimeter-wave radar reliability. Higher echo SNR and continuous tracking stability indicate higher millimeter-wave radar reliability.

[0062] The system acquires the echo signal-to-noise ratio (SNR) and continuous tracking stability of millimeter-wave radar echo data, and determines the reliability of the millimeter-wave radar based on these metrics. Continuous tracking stability can be determined based on the continuity of target markers, changes in target distance, azimuth, and radial velocity over continuous acquisition time, as well as target loss. Within a preset continuous tracking time window, if the target marker remains continuous and the changes in target distance, azimuth, and radial velocity do not exceed their respective preset thresholds, the corresponding target is identified as a continuously tracked target, and the continuous tracking stability is determined based on the number of continuously tracked targets.

[0063] The reliability scores of both the lidar and millimeter-wave radar are normalized to obtain the lidar fusion weight and the millimeter-wave radar fusion weight, and the sum of the two fusion weights is made equal to 1. Specifically, the lidar fusion weight is obtained by dividing the lidar reliability score by the sum of the lidar and millimeter-wave radar reliability scores; the millimeter-wave radar fusion weight is obtained by dividing the millimeter-wave radar reliability score by the sum of the lidar and millimeter-wave radar reliability scores.

[0064] This embodiment combines the number of effective points in the point cloud, the continuity of the target position, the echo signal-to-noise ratio, and the continuous tracking stability to determine the actual data quality of two types of radars. Under the same weather conditions, when the sensor data quality changes, the fusion weights can be adjusted accordingly to improve the continuity and reliability of the fused target's three-dimensional coordinates.

[0065] In one embodiment, the initial three-dimensional motion velocity of the target robotic arm is determined based on the fused three-dimensional coordinates of the target at the continuous acquisition time.

[0066] Specifically, the initial three-dimensional motion velocity is determined based on the time interval between adjacent acquisition moments and the positional changes of the fused target's three-dimensional coordinates along the X, Y, and Z axes. Alternatively, the positional change trend can be determined based on the fused target's three-dimensional coordinates from multiple consecutive acquisition moments to reduce the impact of coordinate fluctuations at a single acquisition moment on the velocity calculation results.

[0067] Based on the installation location and target plane coordinates of the millimeter-wave radar, the radial direction of the target robotic arm relative to the millimeter-wave radar is determined. This radial direction is the horizontal direction pointing from the installation location of the millimeter-wave radar to the target plane coordinates. The planar motion velocity in the initial three-dimensional motion velocity is decomposed into a radial velocity component along the radial direction and a tangential velocity component perpendicular to the radial direction.

[0068] After unifying the positive and negative directions of the radial velocity from the millimeter-wave radar with the initial radial velocity component, the initial radial velocity component is corrected using the radial velocity from the motion information. In one specific implementation, the radial velocity directly measured by the millimeter-wave radar through Doppler frequency shift can be used as the corrected radial velocity component to reduce the radial velocity error caused by continuous coordinate difference.

[0069] The corrected radial velocity component, tangential velocity component, and height velocity component from the initial three-dimensional motion velocity are synthesized to obtain the three-dimensional motion velocity of the target robotic arm. Based on the magnitude and direction of the three-dimensional motion velocity and its orientation relative to the three-dimensional electronic protective wall, the motion state of the target robotic arm—whether it is stationary, close to the three-dimensional electronic protective wall, far from the three-dimensional electronic protective wall, or moving along the three-dimensional electronic protective wall—is determined, and the predicted position is calculated using the corrected three-dimensional motion velocity.

[0070] This embodiment uses the radial velocity directly measured by millimeter-wave radar to correct the radial velocity component obtained from continuous coordinate changes, while retaining the tangential velocity component and the height velocity component provided by the fused three-dimensional coordinates of the target. This can reduce the impact of coordinate fluctuations on radial velocity calculation and improve the accuracy of judging the motion state of the target robotic arm and predicting its future position.

[0071] In one embodiment, predicting the predicted position of the target robotic arm at a preset future time based on the three-dimensional motion velocity includes: The system obtains the predicted duration between the current acquisition time and a preset future time, and sets multiple prediction sampling times within the predicted duration according to a preset time step. The last prediction sampling time is the preset future time.

[0072] Specifically, the current acquisition time is used as the prediction start time, and multiple prediction sampling times are set sequentially according to a preset time step. The preset time step is less than the prediction duration and can be determined based on the sampling period of the sensing unit, so as to form multiple prediction sampling times between the current acquisition time and the preset future time. When the prediction duration is not an integer multiple of the preset time step, the last prediction sampling time can be directly set as the preset future time, so that the predicted motion trajectory covers the complete prediction duration. For example, the prediction duration of the preset future time relative to the current acquisition time can be 0.5s.

[0073] Within a relatively short prediction time, the three-dimensional motion velocity determined based on continuously acquired data can be used as the motion velocity at each prediction sampling moment. The velocity components of the three-dimensional motion velocity in the X, Y, and Z axis directions are obtained, and the time difference between each prediction sampling moment and the current acquisition moment is determined. The velocity components in each of the three coordinate axes are multiplied by their corresponding time differences to obtain the predicted displacement in the X, Y, and Z axis directions at each prediction sampling moment, thus yielding the three-dimensional predicted displacement corresponding to each prediction sampling moment.

[0074] Each predicted 3D displacement is superimposed onto the fused target 3D coordinates at the current acquisition time to obtain the predicted 3D coordinates corresponding to each predicted sampling time. Each predicted 3D coordinate can be calculated according to the aforementioned formula (9). Wherein, when calculating for different predicted sampling times, the prediction time step in formula (9) is replaced with the time difference between the corresponding predicted sampling time and the current acquisition time.

[0075] The predicted 3D coordinates corresponding to a preset future time are used as the predicted position of the target robotic arm, and the predicted 3D coordinates are arranged in chronological order. By connecting temporally adjacent predicted 3D coordinates, the predicted motion trajectory extending from the current fused target 3D coordinates to the predicted position can be obtained. The predicted motion trajectory is used not only to indicate the position that the target robotic arm may reach at the preset future time, but also to determine whether the target robotic arm has passed through a 3D electronic protective barrier before reaching that position.

[0076] This embodiment expands a single predicted position into a continuous predicted motion trajectory by setting multiple predicted sampling times between the current acquisition time and a preset future time. Therefore, even if the target robotic arm has already crossed the three-dimensional electronic protective wall at the preset future time, the position where the target robotic arm first reaches the three-dimensional electronic protective wall can be determined based on the predicted motion trajectory, providing a basis for determining the intrusion time and the location of the three-dimensional intrusion warning point.

[0077] In one embodiment, the warning level and three-dimensional intrusion warning point are determined based on the current distance, current orientation, and the positional relationship between the predicted location and the three-dimensional electronic protective wall. Specifically, this includes: Given that the current orientation indicates that the target robotic arm is located on the three-dimensional electronic protective wall, the warning level of the target robotic arm is determined to be the highest level of emergency intrusion, and the current fused target three-dimensional coordinates are determined as the three-dimensional intrusion warning point.

[0078] If the current orientation indicates that the target robotic boom has entered the railway restricted area, the current distance represents the shortest distance from the fused target's 3D coordinates to the 3D electronic protective barrier. Since the current orientation indicates that the target robotic boom is located on one side of the railway restricted area, this current distance is taken as the current intrusion depth of the target robotic boom. The warning level is determined to be the highest risk level, emergency intrusion, and the fused target's 3D coordinates are determined as the 3D intrusion warning point. In this case, the fused target's 3D coordinates at the current acquisition time are determined as the 3D intrusion warning point, so as to display the current location of the target robotic boom intruding into the railway restricted area in the digital twin platform.

[0079] Given that the target robotic arm is located within the construction area based on the current orientation, the trajectory segments in the predicted motion trajectory are checked sequentially to determine whether they intersect with the 3D electronic protective wall. For the 3D electronic protective wall represented by a discrete mesh, it can be determined whether each trajectory segment passes through the corresponding protective wall mesh cell.

[0080] When the predicted trajectory intersects with the 3D electronic protective wall, the first intersection point appearing in chronological order is determined as the predicted intrusion point. The warning level is then set as the highest-risk emergency intrusion limit, and the first intersection point indicates the location where the target robotic arm will first reach the 3D electronic protective wall if it continues to move according to its current motion state.

[0081] The predicted trajectory segment containing the predicted intrusion point is determined, and the predicted sampling times corresponding to both ends of this trajectory segment are obtained. Based on the position of the predicted intrusion point within the predicted trajectory segment, time interpolation is performed between the two predicted sampling times to obtain the predicted time when the target robotic arm arrives at the predicted intrusion point. The time difference between the predicted time and the current acquisition time is determined as the remaining time. The shorter the remaining time, the closer the target robotic arm is to the three-dimensional electronic defense wall.

[0082] When the predicted trajectory intersects with the 3D electronic protective wall, the predicted intrusion point corresponding to the first intersection position is determined as the 3D intrusion warning point. Therefore, the 3D intrusion warning point indicates the position where the target robotic arm is expected to first reach the 3D electronic protective wall, and can be used to mark potential intrusion locations in advance within a digital twin platform.

[0083] If the predicted trajectory does not intersect with the 3D electronic protective wall, the current distance is compared with a preset distance threshold. When the current distance meets the corresponding distance warning condition, the distance warning level is determined based on the current distance, and no 3D intrusion warning point is generated. When the current distance does not meet any distance warning condition, the warning level is determined to be a no-intrusion-risk level, and no 3D intrusion warning point is generated.

[0084] In one embodiment, multiple sensing units with overlapping monitoring areas are arranged along the centerline of the railway line, and each sensing unit is equipped with an edge computing agent. Adjacent sensing units can exchange target sensing data through an industrial local area network, thereby enabling collaborative monitoring of target robotic arms that span adjacent monitoring areas.

[0085] Specifically, each sensing unit may include lidar, millimeter-wave radar, and corresponding edge computing agents. Multiple sensing units are spaced apart along the centerline of the railway line, with partial overlap in the monitoring areas of adjacent sensing units to avoid blind spots between adjacent monitoring areas. When the target robotic arm simultaneously enters the monitoring area of ​​an adjacent sensing unit, the adjacent edge computing agents exchange target point cloud data and target coordinate data.

[0086] The exchanged data includes the acquisition time, sensor unit pose, and target identifier. The acquisition time is used for time alignment of data acquired by adjacent sensor units. The sensor unit pose includes the sensor unit's installation position and orientation in the global coordinate system, used for coordinate transformation. The target identifier is used to distinguish different detection targets continuously tracked by each sensor unit. Since adjacent sensor units may assign different internal target identifiers to the same target, the target identifier can serve as auxiliary information for target association, rather than as the sole basis for determining whether two detection targets belong to the same target robotic arm.

[0087] Based on the poses of adjacent sensing units, the target point cloud data and target coordinate data acquired by the adjacent sensing units are transformed to the same global coordinate system, and time alignment is performed according to the acquisition time. For data with differences in sampling times between adjacent sensing units, the target coordinates can be aligned to the same reference time by combining the corresponding target's motion speed and direction.

[0088] After completing time alignment and coordinate system transformation, based on the spatial continuity, motion direction consistency, and motion speed consistency between targets acquired by adjacent sensing units, it is determined whether the targets detected by adjacent sensing units belong to the same target robotic arm.

[0089] Spatial continuity is used to determine whether the target coordinates and target point clouds acquired by adjacent sensing units are close to each other within the overlapping monitoring area, and whether the robotic arm contours represented by the two sets of target point clouds can be continuously connected. Motion direction consistency is used to determine whether the movement directions of the two detected targets are the same or similar during continuous acquisition. Motion speed consistency is used to determine whether the magnitudes of the motion speeds of the two detected targets and the speed changes along each coordinate axis are matched.

[0090] When two detection targets are spatially continuous and their direction and speed of movement meet the corresponding consistency conditions, the two detection targets are identified as the same target robotic arm. If either condition is not met, the two detection targets can be kept as different targets to avoid incorrectly splicing different robotic arms operating simultaneously within overlapping monitoring areas.

[0091] When adjacent sensing units detect targets belonging to the same robotic arm, spatial registration is performed on the target point cloud data acquired by the adjacent sensing units. For point clouds repeatedly acquired within overlapping monitoring areas, overlapping points can be merged and outliers filtered. The data is then stitched together according to the positions of each point cloud in the global coordinate system to obtain the spatial contour of the target robotic arm spanning the monitoring areas of adjacent sensing units. This spatial contour represents the overall three-dimensional shape of the target robotic arm within the monitoring range of multiple sensing units.

[0092] Determine the orientation and distance of each target point in the spatial contour of the target robotic boom relative to the 3D electronic protective wall. When a target point exists within a railway restricted area, select the target point with the greatest intrusion depth within the railway restricted area as the current risk point. When no target point exists within a railway restricted area, but a target point exists within the 3D electronic protective wall, select the target point within the 3D electronic protective wall as the current risk point. When all target points are located within the construction area, select the target point with the smallest distance from the 3D electronic protective wall as the current risk point. Determine the 3D coordinates of the current risk point. The current risk point can be located at the end of the robotic boom.

[0093] Based on the current 3D coordinates of the risk point, the current distance and orientation determined by the fused target's 3D coordinates are corrected. Specifically, the shortest distance from the current risk point to the 3D electronic protective wall is used as the corrected current distance, and the corrected current orientation is determined based on whether the current risk point is located on one side of the construction area, on the 3D electronic protective wall, or on the side of the railway restricted area. Thus, the current distance and orientation are corrected from the judgment result corresponding to the target reference position to the judgment result corresponding to the highest risk position in the overall spatial contour of the robotic boom.

[0094] The corrected current distance and current orientation are used to determine the warning level of the target robotic boom. When the corrected current orientation indicates that the current risk point is still within the construction area, the fused target's 3D coordinates are replaced with the current risk point's 3D coordinates, and the predicted position and trajectory are re-determined based on the target robotic boom's motion state. The warning level is then determined based on the corrected current distance and the re-determined predicted trajectory. When the corrected current orientation indicates that the current risk point has reached the 3D electronic protective wall or entered the railway restricted area, the current risk point's 3D coordinates are determined as the 3D intrusion warning point, and the warning level is set as emergency intrusion.

[0095] This embodiment obtains the spatial contour of a target robotic boom spanning adjacent monitoring areas by exchanging and stitching target point cloud data between adjacent edge computing agents. Using the target point closest to the 3D electronic protective wall within the spatial contour as the current risk point avoids the problem of failing to detect when the robotic boom's end is approaching or entering a railway restricted area, which is common when judging solely based on the robotic boom's center position or a single target reference position. It also reduces monitoring blind spots between adjacent sensing units.

[0096] like Figure 2 As shown, this application also provides an electronic perimeter protection system 300 for railway operating lines, comprising: The protective wall establishment module 302 is used to obtain the railway line centerline, safety distance threshold, catenary height threshold and ground filtering threshold, and based on the railway line centerline, safety distance threshold, catenary height threshold and ground filtering threshold, to establish a three-dimensional electronic protective wall to separate the construction area from the railway restricted area.

[0097] The target perception module 304 is used to acquire point cloud data of the target robotic arm through lidar and determine the lidar target three-dimensional coordinates of the target robotic arm based on the point cloud data; it is also used to acquire echo data of the target robotic arm through millimeter-wave radar and determine the millimeter-wave radar target planar coordinates and motion information of the target robotic arm based on the echo data.

[0098] The fusion weight determination module 306 is used to acquire the meteorological conditions of the area where the target robotic arm is located, and to determine the fusion weights of the lidar and the millimeter-wave radar based on the meteorological conditions. The fusion weight determination module 306 can also determine the reliability of the lidar and the millimeter-wave radar based on the meteorological attenuation of the laser signal, the number of valid points in the point cloud data, the continuity of the target position, the echo signal-to-noise ratio, and the continuous tracking stability, and then normalize the two reliability values.

[0099] The coordinate fusion module 308 is used to fuse the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target based on the lidar fusion weight and the millimeter-wave radar fusion weight, and to determine the fused height coordinates based on the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm.

[0100] The positional relationship determination module 310 is used to determine the current distance from the target robotic boom to the three-dimensional electronic protective wall, and the current orientation of the target robotic boom relative to the three-dimensional electronic protective wall, based on the fused target's three-dimensional coordinates. The current orientational relationship is used to characterize whether the target robotic boom is located on one side of the construction area, on the three-dimensional electronic protective wall, or on the side entering the railway restricted area.

[0101] The trajectory prediction module 312 is used to determine the motion state of the target robotic arm based on the fused target 3D coordinates and motion information corresponding to the continuous acquisition time, and predict the predicted position of the target robotic arm at a preset future time based on the motion state. The trajectory prediction module 312 can also generate a predicted motion trajectory according to a preset time step to determine whether the target robotic arm has passed through the 3D electronic protective wall before reaching the preset future time.

[0102] The early warning generation module 314 is used to determine the early warning level and three-dimensional intrusion early warning point of the target robotic arm based on the current distance, current orientation, and the positional relationship between the predicted position and the three-dimensional electronic protective wall, and to generate early warning information corresponding to the early warning level. The early warning information may include the sensing unit identifier, early warning time, early warning level, current distance, coordinates of the three-dimensional intrusion early warning point, motion status of the target robotic arm, radar point cloud snapshot, and on-site video.

[0103] Each module in the above-mentioned device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0104] like Figure 3As shown, this application also provides a terminal device 400, including a processor 401 and a memory 402. The memory 402 stores a program or instructions that can run on the processor 401. When the program or instructions are executed by the processor 401, they implement the various steps of the above method embodiments and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0105] This application also provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the various processes of the above-described method embodiments and achieve the same technical effects. To avoid repetition, these will not be described again here.

[0106] The processor can be the processor in the aforementioned system. Readable storage media include computer-readable storage media, such as computer read-only memory, random access memory, magnetic disks, or optical disks.

[0107] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0108] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as a read-only memory). The device includes a number of instructions in a ROM (random access memory), RAM (magnetic disk), or optical disk to cause a terminal to execute the methods described in the various embodiments of this application.

[0109] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for electronic protection of the perimeter of a railway operating line, characterized in that, include: The railway line centerline, safety distance threshold, catenary height threshold, and ground filtering threshold are obtained, and a three-dimensional electronic protective wall is established based on the railway line centerline, the safety distance threshold, the catenary height threshold, and the ground filtering threshold to separate the construction area from the railway restricted area. Point cloud data of the target robotic arm is acquired by LiDAR, and the three-dimensional coordinates of the target robotic arm by LiDAR are determined based on the point cloud data; echo data of the target robotic arm is acquired by millimeter-wave radar, and the planar coordinates and motion information of the target robotic arm by millimeter-wave radar are determined based on the echo data. The meteorological conditions of the area where the target robotic arm is located are obtained, and the fusion weights of lidar and millimeter-wave radar are determined based on the meteorological conditions. Based on the fusion weights of the lidar and the millimeter-wave radar, the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are fused, and the fused height coordinates are determined according to the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm. Based on the fused target's three-dimensional coordinates, determine the current distance from the target robotic arm to the three-dimensional electronic protective wall, as well as the current orientation of the target robotic arm relative to the three-dimensional electronic protective wall; Based on the fused target's three-dimensional coordinates and motion information corresponding to the continuous acquisition time, the three-dimensional motion speed and motion state of the target robotic arm are determined, and the predicted position of the target robotic arm at a preset future time is predicted based on the three-dimensional motion speed. Based on the current distance, the current orientation relationship, and the positional relationship between the predicted position and the three-dimensional electronic protective wall, the warning level and three-dimensional intrusion warning point corresponding to the target robotic arm are determined, and warning information corresponding to the warning level is generated.

2. The electronic protection method for the perimeter of a railway operating line according to claim 1, characterized in that, Before fusing the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target, the method further includes: The point cloud data and the echo data are timestamped and aligned, and the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target are transformed to the same global coordinate system. Based on the spatial distance between the planar coordinates of the laser radar target and the planar coordinates of the millimeter-wave radar target in the three-dimensional coordinate system, and the consistency of the motion direction corresponding to the continuous acquisition time, the laser radar detection target and the millimeter-wave radar detection target are associated. When the spatial distance meets the preset association distance and the consistency of the motion direction meets the preset association condition, the corresponding laser radar detection target and millimeter-wave radar detection target are identified as the same target robotic arm, and the coordinates corresponding to the same target robotic arm are fused.

3. The electronic protection method for the perimeter of a railway operating line according to claim 2, characterized in that, The step of determining the lidar fusion weight and millimeter-wave radar fusion weight based on the meteorological conditions includes: The degree of meteorological attenuation of the laser signal is determined based on the meteorological conditions. The number of valid points in the point cloud data and the continuity of the target position corresponding to the continuous acquisition time are obtained, and the reliability of the lidar is determined based on the meteorological attenuation degree, the number of valid points, and the continuity of the target position. The echo signal-to-noise ratio and continuous tracking stability of the echo data are obtained, and the reliability of the millimeter-wave radar is determined based on the echo signal-to-noise ratio and the continuous tracking stability. The reliability of the lidar and the reliability of the millimeter-wave radar are normalized to obtain the lidar fusion weight and the millimeter-wave radar fusion weight.

4. The electronic protection method for the perimeter of a railway operating line according to claim 1, characterized in that, The step of determining the three-dimensional motion speed and motion state of the target robotic arm based on the fused target three-dimensional coordinates and motion information corresponding to the continuous acquisition time includes: Based on the fused target's three-dimensional coordinates corresponding to the continuous acquisition time, the initial three-dimensional motion velocity of the target robotic arm is determined; Based on the installation location of the millimeter-wave radar and the target plane coordinates of the millimeter-wave radar, the radial direction of the target robotic arm relative to the millimeter-wave radar is determined; The planar motion velocity in the initial three-dimensional motion velocity is decomposed into a radial velocity component along the radial direction and a tangential velocity component perpendicular to the radial direction; The radial velocity component is corrected based on the radial velocity in the motion information, and the corrected radial velocity component, the tangential velocity component, and the height direction velocity component in the initial three-dimensional motion velocity are combined to obtain the three-dimensional motion velocity of the target robotic arm. The motion state of the target robotic arm is determined based on the three-dimensional motion velocity.

5. The electronic protection method for the perimeter of a railway operating line according to claim 4, characterized in that, Predicting the target robotic arm's position at a preset future time based on the three-dimensional motion velocity; including: Obtain the predicted duration between the current acquisition time and the preset future time, and set multiple predicted sampling times within the predicted duration according to a preset time step, wherein the last predicted sampling time is the preset future time; The velocity components of the three-dimensional motion velocity in the three coordinate axes are multiplied by the time difference between each predicted sampling time and the current acquisition time to obtain the three-dimensional predicted displacement corresponding to each predicted sampling time. The predicted three-dimensional displacements are superimposed onto the fused target three-dimensional coordinates to obtain the predicted three-dimensional coordinates corresponding to each predicted sampling time. The predicted three-dimensional coordinates corresponding to the preset future time are used as the predicted position, and the predicted three-dimensional coordinates are arranged in chronological order to obtain the predicted motion trajectory extending from the fused target three-dimensional coordinates to the predicted position.

6. The electronic protection method for the perimeter of a railway operating line according to claim 5, characterized in that, The step of determining the early warning level and the three-dimensional intrusion early warning point based on the current distance, the current orientation relationship, and the positional relationship between the predicted location and the three-dimensional electronic protective wall includes: If the current orientation relationship indicates that the target robotic arm is located on the three-dimensional electronic protective wall, the warning level is determined to be the highest level of emergency intrusion, and the fused target three-dimensional coordinates are determined as the three-dimensional intrusion warning point. If the current orientation relationship indicates that the target robotic boom has entered the railway restricted area, the current distance is taken as the current intrusion depth of the target robotic boom, the warning level is determined as the emergency intrusion limit with the highest risk level, and the fused target three-dimensional coordinates are determined as the three-dimensional intrusion warning point. If the current orientation relationship indicates that the target robotic arm is located in the construction area, determine whether the predicted motion trajectory intersects with the three-dimensional electronic protective wall; wherein, When the predicted motion trajectory intersects with the three-dimensional electronic protective wall, the first intersection point of the predicted motion trajectory and the three-dimensional electronic protective wall is determined as the predicted intrusion point; based on the times corresponding to the two ends of the predicted trajectory segment where the predicted intrusion point is located, and the position of the predicted intrusion point in the predicted trajectory segment, the predicted time when the target robotic arm arrives at the predicted intrusion point is determined, the warning level is determined as the emergency intrusion limit with the highest risk level, and the predicted intrusion point is determined as the three-dimensional intrusion warning point; If the predicted motion trajectory does not intersect with the three-dimensional electronic protective wall, the current distance is compared with a preset distance threshold; if the current distance reaches the corresponding distance warning condition, the corresponding distance warning level is determined as the warning level, and the three-dimensional intrusion warning point is not generated; if the current distance does not reach any distance warning condition, the warning level is determined as a no-intrusion risk level, and the three-dimensional intrusion warning point is not generated.

7. An electronic protection system for the perimeter of a railway operating line, characterized in that, include: The protective wall establishment module is used to obtain the railway line centerline, safety distance threshold, catenary height threshold and ground filtering threshold, and based on the railway line centerline, the safety distance threshold, the catenary height threshold and the ground filtering threshold, to establish a three-dimensional electronic protective wall to separate the construction area from the railway restricted area; The target perception module is used to acquire point cloud data of the target robotic arm using lidar and determine the lidar target three-dimensional coordinates of the target robotic arm based on the point cloud data; and to acquire echo data of the target robotic arm using millimeter-wave radar and determine the millimeter-wave target planar coordinates and motion information of the target robotic arm based on the echo data. The fusion weight determination module is used to obtain the meteorological conditions of the area where the target robotic arm is located, and to determine the fusion weight of lidar and the fusion weight of millimeter-wave radar based on the meteorological conditions. The coordinate fusion module is used to fuse the planar coordinates in the three-dimensional coordinates of the lidar target and the planar coordinates of the millimeter-wave radar target based on the lidar fusion weight and the millimeter-wave radar fusion weight, and to determine the fused height coordinates based on the height coordinates in the three-dimensional coordinates of the lidar target, so as to obtain the fused three-dimensional coordinates of the target robotic arm. The position relationship determination module is used to determine the current distance from the target robotic arm to the three-dimensional electronic protective wall, and the current orientation relationship of the target robotic arm relative to the three-dimensional electronic protective wall, based on the three-dimensional coordinates of the fused target. The trajectory prediction module is used to determine the three-dimensional motion speed and motion state of the target robotic arm based on the fused target three-dimensional coordinates and motion information corresponding to the continuous acquisition time, and to predict the predicted position of the target robotic arm at a preset future time based on the three-dimensional motion speed. The early warning generation module is used to determine the early warning level and three-dimensional intrusion early warning point corresponding to the target robotic arm based on the current distance, the current orientation relationship, and the positional relationship between the predicted position and the three-dimensional electronic protective wall, and to generate early warning information corresponding to the early warning level.

8. A terminal device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, implements the method according to any one of claims 1 to 6.

9. A readable storage medium, characterized in that, It stores a computer program thereon, which, when executed by a processor, implements the method described in any one of claims 1 to 6.

Images