Intelligent safety protection system, method and electronic device for rail transit engineering vehicle

Abstract

This invention relates to the field of rail transit safety protection technology, and provides an intelligent safety protection system, method, and electronic device for rail transit engineering vehicles. The system includes: a sensing module for real-time collection of the engineering vehicle's operating status data and environmental perception data; a positioning module for calculating the estimated location based on the engineering vehicle's initial positioning information and operating status data, and calibrating the estimated location according to environmental perception data and a pre-set electronic map to obtain real-time positioning information; and a control module for determining the current driving safety status based on the real-time positioning information and environmental perception data, and performing safety protection control on the engineering vehicle according to the current driving safety status. This invention achieves high-precision independent positioning and all-round safety protection based on intelligent perception by integrating operating status data, environmental perception data, and electronic maps for autonomous positioning calibration and safety control. It features low deployment cost, strong environmental adaptability, and supports flexible cross-line operations.

Description

Technical Field

[0001] This invention relates to the field of rail transit safety protection technology, and in particular to an intelligent safety protection system, method and electronic equipment for rail transit engineering vehicles. Background Technology

[0002] Rail transit engineering vehicles are mainly responsible for track maintenance, line repair, and emergency rescue, and are characterized by flexible operation and frequent cross-line operation. In order to ensure the safety of train operation, engineering vehicles must have the safety protection capabilities to prevent overshooting and signal failure, speeding, and collisions with obstacles.

[0003] Currently, safety protection solutions for engineering vehicles in the industry mainly include deploying Automatic Train Protection (ATP) systems, Train Operation Monitoring Devices (LKJ), or a fully driver-dependent manual driving mode. However, existing automatic protection solutions typically rely heavily on external track interlocking information and complex trackside and track surface equipment, resulting in high deployment costs and long construction periods. Furthermore, due to differences in signal systems across different lines, vehicles struggle to flexibly operate across lines. In addition, most existing systems focus on signal data-based protection, generally lacking the ability to actively detect track environment obstacles or foreign objects, failing to effectively meet collision avoidance safety requirements. Meanwhile, a purely manual driving mode is limited by the driver's condition, posing significant human-caused safety hazards. Summary of the Invention

[0004] This invention provides an intelligent safety protection system, method, and electronic device for rail transit engineering vehicles, which addresses the shortcomings of existing rail transit engineering vehicle safety protection schemes, such as high deployment costs due to reliance on external trackside equipment, difficulty in cross-line operation, and insufficient safety due to lack of obstacle detection capabilities.

[0005] This invention provides an intelligent safety protection system for rail transit engineering vehicles, comprising: The sensing module is used to collect real-time operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; The positioning module is used to perform positioning calculation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and to calibrate the estimated position according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. The control module is used to determine the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and to perform safety protection control on the engineering vehicle according to the current driving safety status.

[0006] According to the present invention, an intelligent safety protection system for rail transit engineering vehicles is provided, wherein the positioning module includes an initialization positioning submodule, and the initialization positioning submodule includes: The first positioning unit is used to determine the initial positioning information based on the mapping relationship between the satellite positioning signal and the preset electronic map when the satellite positioning signal is detected to be valid in the current environment. The second positioning unit is used to determine the initial positioning information based on the operating environment image and lidar point cloud data collected by the sensing module and the preset electronic map when the satellite positioning signal is detected to be invalid.

[0007] According to the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, the second positioning unit is specifically used for: Receive target information input by the user that corresponds to the current location of the engineering vehicle; Target recognition is performed on the operating environment image and the lidar point cloud data. If a reference target corresponding to the target information is identified, the first relative distance between the engineering vehicle and the reference target is calculated, and the initial positioning information is calculated based on the first relative distance and the absolute coordinates of the reference target in the preset electronic map. If the reference target is not identified, the user is prompted to input the second relative distance between the engineering vehicle and the reference target, and the initial positioning information is calculated based on the second relative distance and the absolute coordinates of the reference target.

[0008] According to the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, the positioning module further includes a positioning update submodule, which includes: The positioning calculation unit is used to perform positioning calculation based on the initial positioning information and the wheel speed information and inertial measurement information in the running status data to obtain the calculated position; The positioning calibration unit is used to identify route feature targets from the environmental perception data, match the route feature targets with the preset electronic map, and calibrate the estimated position according to the matching result to obtain the real-time positioning information.

[0009] According to the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, the positioning calibration unit is specifically used for: Identify the road feature targets located in front of the engineering vehicle from the environmental perception data; Detect whether the route feature target matches the positioning target expected based on the calculated position in the preset electronic map; If a match is found, the measured relative position of the route feature target relative to the engineering vehicle is obtained, and the estimated position is calibrated and updated based on the measured relative position and the absolute coordinates of the positioning target in the preset electronic map to obtain the real-time positioning information.

[0010] According to the present invention, an intelligent safety protection system for rail transit engineering vehicles is provided. The positioning calibration unit includes a track confirmation subunit. The track confirmation subunit is used to determine the running track of the engineering vehicle after passing the turnout when the engineering vehicle runs to the turnout area, and to lock the positioning trajectory of the engineering vehicle in the preset electronic map to the running track. The track confirmation subunit is specifically configured to perform at least one of the following operations: The operation task information of the engineering vehicle is obtained, and a preset driving route is generated in the preset electronic map based on the start position and end position in the operation task information. The turnout running direction is determined according to the preset driving route, and the running track is determined according to the turnout running direction. Based on the environmental perception data, the turnout direction information is identified, and the turnout direction information is mapped with the turnout topology in the preset electronic map to determine the running track; Identify the status of the signal associated with the turnout, determine the turnout opening direction based on the signal status, and determine the running track according to the turnout opening direction; Receive modification instructions from users regarding turnout direction prompts, and determine the operating track based on the modification instructions; Based on the operating status data, the vehicle body posture change characteristics are identified, and the operating track is determined according to the vehicle body posture change characteristics.

[0011] According to the present invention, an intelligent safety protection system for rail transit engineering vehicles is provided, wherein the control module includes a signal safety protection unit, the signal safety protection unit being used for: The system identifies the light status of the forward traffic signal from the environmental perception data and determines whether the forward traffic signal is a valid signal for the current route based on the real-time positioning information. When the signal ahead is a valid signal and the light color indicates a no-passing state, the signal ahead is determined to be a no-passing signal, and the braking distance of the engineering vehicle at the current operating speed is calculated in real time. If the distance between the engineering vehicle and the prohibition signal is less than or equal to the sum of the braking distance and the preset safety margin, a locomotive braking command is triggered to control the engineering vehicle to stop.

[0012] According to the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, the control module further includes an obstacle detection unit, which is used for: Based on the real-time positioning information and the operating status data, a three-dimensional track boundary space for the current driving section of the engineering vehicle is constructed in the three-dimensional space corresponding to the environmental perception data. Point cloud data is acquired from the environmental perception data, and the point cloud data is filtered based on the three-dimensional orbital boundary space to obtain point cloud data located within the three-dimensional orbital boundary space. Cluster analysis is performed on the point cloud data located within the three-dimensional track clearance space, and the presence of obstacles within the track clearance of the engineering vehicle is identified based on the analysis results. If the obstacle is detected, an alarm will be triggered or braking control will be applied based on the distance between the obstacle and the engineering vehicle and the current operating speed of the engineering vehicle.

[0013] According to the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, the control module further includes an overspeed safety protection unit, which is used for: The speed limit value of the current route of the engineering vehicle is obtained from the preset electronic map, and the current operating speed of the engineering vehicle is compared with the speed limit value of the route. If the current operating speed exceeds the line speed limit, an overspeed alarm will be output or normal braking control will be executed.

[0014] According to the present invention, an intelligent safety protection system for rail transit engineering vehicles further includes a human-machine interface terminal, which is deployed in the driver's cabs at both ends of the engineering vehicle. The human-machine interface terminal is used for: The system displays the real-time location information, the identification results in the environmental perception data, the operating status data, and the current driving safety status in real time. The system detects the current activation status of the driver's cab of the engineering vehicle, responds to configuration and control commands input by users in the active driver's cab, and blocks input operations from users in the inactive driver's cab.

[0015] This invention also provides an intelligent safety protection method for rail transit engineering vehicles, comprising: Real-time collection of operational status data of engineering vehicles and environmental perception data in the direction of operation of engineering vehicles; Based on the initial positioning information of the engineering vehicle and the operating status data, the estimated position of the engineering vehicle is calculated, and the estimated position is calibrated according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. Based on the real-time positioning information and the environmental perception data, the current driving safety status of the engineering vehicle is determined, and safety protection control is performed on the engineering vehicle according to the current driving safety status.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the intelligent safety protection method for rail transit engineering vehicles as described above.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent safety protection method for rail transit engineering vehicles as described above.

[0018] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the intelligent safety protection method for rail transit engineering vehicles as described above.

[0019] The intelligent safety protection system, method, and electronic equipment for rail transit engineering vehicles provided by this invention collect real-time operating status data and environmental perception data of the engineering vehicle through a sensing module. Combined with positioning calculations based on operating status data and a calibration mechanism based on environmental perception data and a pre-set electronic map in the positioning module, it eliminates the strong dependence on external trackside equipment and can achieve high-precision autonomous positioning without relying on external signal input. Furthermore, the control module comprehensively utilizes real-time positioning information and environmental perception data to perform dual safety status determination. It can not only prevent overstepping, overshooting, and speeding based on location information, but also actively identify track obstacles or abnormal conditions using environmental perception data. Thus, it constructs an independent, complete, and proactively environmental detection-enabled vehicle safety protection system. While significantly reducing the complexity and cost of system deployment, it effectively solves the safety protection problem of engineering vehicles flexibly crossing lines between different track types. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this invention or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1This is one of the structural schematic diagrams of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention; Figure 2 This is the second structural schematic diagram of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention; Figure 3 This is a schematic diagram of the workflow of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention; Figure 4 This is a schematic diagram of positioning initialization based on intelligent perception provided by the present invention; Figure 5 This is a schematic diagram of the positioning update and persistence based on intelligent perception provided by the present invention; Figure 6 This is a schematic diagram of positioning calibration update based on multi-frame tracking fusion detection provided by the present invention; Figure 7 This is a schematic diagram of turnout identification and updating based on intelligent sensing provided by the present invention; Figure 8 This is a schematic diagram of the signal recognition process based on intelligent sensing provided by the present invention; Figure 9 This is a schematic diagram of the obstacle detection process based on intelligent perception provided by the present invention; Figure 10 This is a schematic diagram of the interaction between the human-computer interaction terminal and the computing control unit provided by the present invention; Figure 11 This is a flowchart illustrating the intelligent safety protection method for rail transit engineering vehicles provided by the present invention; Figure 12 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0023] As the foundational vehicles of rail transit operating groups, rail transit engineering vehicles undertake crucial tasks such as track maintenance, line repair, and train breakdown rescue. Compared to conventionally operating trains, rail transit engineering vehicles are characterized by their operational flexibility, frequent operations, and the need to frequently cross different lines. To ensure safe operation, engineering vehicles must meet stringent safety protection requirements during operation, specifically including preventing accidental entry into tracks with closed signals and preventing accidental exit from closed signals (referred to as "2-accident protection"), preventing exceeding track speed limits or temporary operating speed limits (referred to as "1-overspeed protection"), and preventing collisions with personnel, vehicles, and objects on and beside the track (referred to as "1-collision protection"). Therefore, constructing a comprehensive safety protection system is a necessary measure to ensure the safe operation of engineering vehicles.

[0024] Currently, the main implementation plans for the safety protection of rail transit engineering vehicles in the industry include the following: The first approach involves deploying an Automatic Train Protection (ATP) system on the engineering vehicles and integrating it with track interlocking information. While this approach provides protection against two types of accidents and one type of overrun, it has significant limitations in practical application. First, this approach is highly dependent on external track interlocking systems and requires the installation of complex trackside and track surface equipment, resulting in extremely high implementation costs, long deployment cycles, and high maintenance costs. Second, due to differences in interlocking signals across different lines, engineering vehicles equipped with specific ATP systems often cannot operate across lines, limiting their scope of use. Furthermore, this approach primarily focuses on signal protection and cannot effectively detect and protect against obstacles or abnormal collisions on the track.

[0025] The second option is to deploy train operation monitoring and recording devices (LKJ) on rail transit engineering vehicles. However, this option also suffers from problems such as heavy reliance on external information, lack of collision avoidance capabilities, lack of support for cross-line operation, and the need to deploy costly trackside and ground equipment for assistance.

[0026] In response, this invention provides an intelligent safety protection system for rail transit engineering vehicles. By integrating operational status data, environmental perception data, and pre-set electronic maps for autonomous positioning calibration and safety control, the system achieves high-precision independent positioning and all-round active safety protection for engineering vehicles without the support of external trackside equipment. The system has low deployment cost, does not rely on external interlocking information, has strong environmental adaptability, and supports flexible cross-line operations, thereby overcoming the above-mentioned defects.

[0027] It should be noted that all actions involving the acquisition of signals, information, or data in this invention are carried out in compliance with the relevant data protection laws and policies of the country where the invention is located, and with authorization from the owner of the relevant device.

[0028] Figure 1This is one of the structural schematic diagrams of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, such as... Figure 1 As shown, the system includes: The sensing module 110 is used to collect real-time operating status data of the engineering vehicle and environmental sensing data in the direction of the engineering vehicle's operation. The positioning module 120 is used to perform positioning calculation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and to calibrate the estimated position according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. The control module 130 is used to determine the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and to perform safety protection control on the engineering vehicle according to the current driving safety status.

[0029] It should be noted that this invention provides an intelligent safety protection system for rail transit engineering vehicles (hereinafter referred to as the EV-IPS system). This system aims to address the technical problems existing in rail transit engineering vehicles (such as basic vehicles used for subway operation and maintenance, line repair, and electric train breakdown rescue) during operation, including reliance on external interlocking systems, inability to operate across lines, lack of obstacle detection, and high maintenance costs of trackside equipment. This system, through intelligent sensing and computing equipment deployed on the vehicle, achieves independent high-precision positioning and safety protection without relying on trackside equipment, specifically fulfilling the core safety requirements of "2 omissions, 1 overspeed, and 1 prevention" (i.e., prevention of omissions, prevention of overspeeding, and prevention of collisions).

[0030] Specifically, the system mainly includes a sensing module, a positioning module, and a control module. The sensing module is primarily used to collect real-time operational status data of the engineering vehicle and environmental perception data along the vehicle's direction of travel.

[0031] Here, operational status data refers to data reflecting the mechanical motion characteristics of the engineering vehicle itself, mainly including but not limited to real-time operating speed (i.e., wheel speed), acceleration, and driving direction. In specific applications, the sensing module may include vehicle speed sensors, inertial measurement units (IMUs), etc. These sensor units can be connected to the on-board control host (i.e., control module) of the engineering vehicle through the vehicle-to-everything (V2X) interface to transmit the collected operational status data such as speed, acceleration, and driving direction to the control module.

[0032] Environmental perception data refers to data reflecting the external working environment of the engineering vehicle, especially the track conditions ahead in the direction of travel. To acquire this data, the perception module can also include various sensors deployed at the front of the engineering vehicle (and, if necessary, the rear, for dual-end operation protection), such as high-definition industrial cameras, LiDAR, or millimeter-wave radar. These sensors can collect real-time visual image data and point cloud data from the front and rear of the vehicle to identify key elements on the track, such as signal lights (including mainline red / green / yellow signals, section blue / white signals, etc.), track alignment, turnout direction, and obstacles within the track clearance (such as personnel, vehicles, and foreign objects).

[0033] Through the sensing module, the system can autonomously acquire the key environmental information required for vehicle operation without relying on external ground equipment (such as transponders and track circuits), providing the raw data foundation for subsequent positioning and control.

[0034] The positioning module is used to calculate the location of the engineering vehicle based on its initial positioning information and operating status data. After obtaining the calculated location of the engineering vehicle, the module further calibrates the calculated location based on environmental perception data and a pre-set electronic map to obtain more accurate real-time positioning information.

[0035] It is understandable that, when acquiring the real-time positioning information of the engineering vehicle, this embodiment of the invention adopts a fusion positioning strategy of calculation and calibration to adapt to the complex operating environment of rail transit, including open-air, tunnel, and elevated scenarios.

[0036] Specifically, the positioning and estimation process is conducted based on the known initial positioning information of the engineering vehicle (i.e., the starting position coordinates or odometer markers when the vehicle departed). The positioning module uses the operational status data (such as wheel speed pulses or acceleration integrals) provided by the sensing module to perform dead reckoning. For example, the travel distance is calculated by multiplying the real-time speed by the time interval, thus continuously accumulating from the initial position to obtain a dynamically updated estimated position. However, due to factors such as wheel slippage and sensor drift, relying solely on estimation will lead to increasingly large accumulated errors, thus requiring calibration.

[0037] During the calibration process of the estimated position, the positioning module uses feature information from the environmental perception data (such as the identified signal number, signal position, and switch position) to match it with a pre-stored electronic map within the system. This pre-stored electronic map contains the absolute coordinates or mileage information of key facilities on the line (such as signals, switches, and platforms). When the perception module identifies a specific signal, the positioning module uses the signal's accurate position on the map as a reference to correct the current estimated position. For example, it forces the estimated position back to the signal's accurate coordinates, thereby eliminating accumulated errors and obtaining high-precision real-time positioning information.

[0038] This positioning method does not rely on continuous GPS (Global Positioning System) or RTK (Real-time kinematic) signals, solving the positioning problem in underground tunnels. It also does not require the installation of expensive RFID (Radio Frequency Identification) tags or transponders, and can achieve high-precision positioning of the entire road network (including tunnels) solely based on vehicle-side sensing.

[0039] The control module is the decision-making and execution center of the system. It is used to determine the current driving safety status of the engineering vehicle based on real-time positioning information and environmental perception data, and to perform safety protection control on the engineering vehicle according to the current driving safety status. Specifically, the control module determines the current driving safety status by comprehensively analyzing where the vehicle is (provided by real-time positioning information) and what is in front of the vehicle (provided by environmental perception data).

[0040] Here, the current driving safety status can include normal driving status, speeding risk status, risk of overshooting / overshooting status, and collision risk status. Normal driving status means the vehicle is within the permitted range, there are no obstacles ahead, and the speed does not exceed the speed limit. Speeding risk status means the vehicle's current speed exceeds the route speed limit or temporary work-related speed limit at that location (based on electronic map query). Risk of overshooting / overshooting status means the vehicle is approaching a signal indicating no passage (such as a red or blue light) and shows no sign of slowing down, posing a risk of overshooting the signal. Collision risk status means environmental perception data indicates the presence of an obstacle within the track clearance, and the distance is less than the safe braking distance.

[0041] Based on the determined driving safety status, the control module will implement safety protection controls for the engineering vehicle. This control can be tiered. For example, in the early stages of risk (such as slight speeding or the detection of obstacles at a distance), an audible and visual alarm will be issued through the onboard HMI (Human Machine Interface) terminal to prompt the driver to intervene manually. When the risk escalates (such as about to run a red light or the obstacle is at a critical distance), the control module will send commands directly to the braking system of the engineering vehicle through the locomotive-vehicle interconnection interface (such as cutting off traction or triggering the locomotive brakes) to force the vehicle to stop, thereby achieving active safety protection of 2 risk levels, 1 speeding violation, and 1 prevention.

[0042] The system provided in this invention combines vehicle-side sensing, fusion positioning, and intelligent control to achieve autonomous intelligent protection for engineering vehicles without requiring modifications to trackside signaling facilities. This significantly reduces deployment costs and improves the system's flexibility and cross-line operation capabilities.

[0043] Based on the above embodiments, Figure 2 This is the second structural schematic diagram of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, as shown below. Figure 2 As shown, the system (EV-IPS) mainly consists of five core hardware components: the vehicle control host, environmental perception sensors, driver's cab HMI terminal, vehicle-to-vehicle interconnection interface, and positioning equipment. These hardware components form a complete data interaction closed loop through physical connection.

[0044] The onboard control unit, serving as the core computing unit of the system, corresponds to the computing components of the control and positioning modules. Deployed inside the engineering vehicle, it is responsible for collecting all sensor data, executing fusion positioning algorithms, image recognition algorithms, and safety logic judgments, and issuing control commands to the vehicle through the vehicle-to-vehicle interface.

[0045] Environmental perception sensors are part of the perception module. To accommodate the bidirectional operation of engineering vehicles, environmental perception sensors are deployed at both ends (I and II) of the vehicle (i.e., the front and rear). These sensors typically include LiDAR, industrial cameras, etc., and are used to collect environmental data along the vehicle's direction of travel.

[0046] The driver's cab HMI terminals are deployed in both the I-end and II-end cabs, serving as the human-machine interface. They display real-time positioning, speed, forward signal status, obstacle distance, and other information, and receive configuration commands from the driver. It should be understood that the system typically only responds to HMI commands from the currently active terminal (i.e., the driver's actual operating terminal) to avoid accidental operation.

[0047] The locomotive interconnection interface is connected to the engineering vehicle's own control system via cable. On the one hand, it reads wheel speed and locomotive operating conditions (such as traction / braking status and steering handle position) as auxiliary data for positioning calculation; on the other hand, it enables locomotive control, that is, directly applying braking to the vehicle in emergency situations.

[0048] The positioning device is typically an RTK positioning antenna, used to provide absolute position reference in open-air scenarios to assist the system in initial positioning. It should be understood that this device is optional for open-air track applications, allowing EV-IPS to achieve locomotive positioning at any time and any location via RTK equipment. For complex tracks and underground sections where RTK coverage is ineffective or poor, this positioning device is unnecessary. For mixed open-air and underground tracks, EV-IPS automatically determines the availability of the RTK equipment, automatically calculates and uses the data, requiring no manual intervention.

[0049] Figure 3 This is a schematic diagram of the workflow of the intelligent safety protection system for rail transit engineering vehicles provided by the present invention, as shown below. Figure 3As shown, this process demonstrates a closed-loop process from system startup to the execution of security protection, which mainly includes the following steps: Step S1, System Initialization and Configuration Once the operation begins, the system is first initialized and configured. This step is crucial, as it includes determining the initial location (i.e., initial positioning information). Subsequently, the driver configures the task via the system's HMI, inputting relevant parameters for the operation (such as destination, operation type, etc.).

[0050] Step S2, Multi-dimensional Intelligent Sensing and Processing During the operation of the engineering vehicle, the system enters a parallel processing state, simultaneously performing signal recognition based on intelligent perception, line feature recognition for positioning and updating, obstacle detection, and health monitoring.

[0051] Step S3, Driving Safety Decision-Making and Execution The onboard control unit aggregates the aforementioned perception data to make EV-IPS driving safety decisions. The system will assess in real time whether there is a risk of running a red light (running into or out of a lane), speeding, or collision. If a risk is detected, the system will execute two actions in parallel: one is to issue an audible and visual warning to the driver via the EV-IPS onboard HMI; the other, if the risk level is high, will directly trigger the locomotive braking system via the EV-IPS linkage locomotive interface safety control.

[0052] Step S4, Task Loop Once a task is completed or a risk is eliminated, the system is ready, and EV-IPS receives new tasks and enters the next work cycle.

[0053] The embodiments of this invention provide a brief overview of the overall architecture and workflow of the system. The specific structure and working principle of each functional module of the system will be described in detail below.

[0054] Based on any of the above embodiments, the positioning module 120 includes an initialization positioning submodule 121, which includes: The first positioning unit 1211 is used to determine the initial positioning information based on the mapping relationship between the satellite positioning signal and the preset electronic map when the satellite positioning signal is detected to be valid in the current environment. The second positioning unit 1212 is used to determine the initial positioning information based on the operating environment image and lidar point cloud data collected by the sensing module and the preset electronic map when the satellite positioning signal is detected to be invalid.

[0055] It should be noted that in the actual operation of rail transit engineering vehicles, the vehicle may start at an open mainline, or it may start in a tunnel or garage without satellite signal coverage. To ensure that the system can obtain accurate initial coordinates at any starting location, the positioning module includes an initialization positioning submodule. Figure 4 This is a schematic diagram of the positioning initialization based on intelligent perception provided by the present invention, as shown below. Figure 4 As shown, this submodule employs two different initialization strategies based on the differences in the external signal environment.

[0056] Specifically, the initialization positioning submodule includes a first positioning unit and a second positioning unit. The first positioning unit is used to handle scenarios involving outdoor operation or good satellite signal coverage. Engineering vehicles are typically equipped with vehicle positioning devices, such as RTK positioning devices.

[0057] When the engineering vehicle starts, the system first checks the validity of satellite positioning signals (such as GPS, BeiDou, and other GNSS signals) in the current environment. This can be achieved by judging the strength or quality of the satellite signals. If the satellite positioning signal is detected as valid (e.g., good RTK signal coverage), the first positioning unit can directly obtain high-precision latitude and longitude coordinates. Subsequently, based on the mapping relationship between the satellite positioning signal and the system's internal pre-set electronic map (i.e., the conversion between latitude and longitude and the track mileage marker / electronic map coordinate system), the initial positioning information of the engineering vehicle is directly determined. This method requires no manual intervention and can achieve rapid and accurate automatic initialization in open-air stations or mainline areas.

[0058] The second positioning unit is used to handle complex scenarios where satellite positioning signals are invalid (including no signal or poor signal quality that cannot meet positioning accuracy requirements) in non-open-air locations (such as tunnels, underground platforms, or parking garages). In this scenario, relying solely on satellite positioning will lead to initialization failure or position drift.

[0059] Therefore, the second positioning unit adopts a fusion approach combining visual and lidar perception with human-computer interaction. It determines the initial positioning information based on the operating environment images collected by the perception module, lidar point cloud data, and a pre-set electronic map. The specific implementation process is as follows: Receive target information input by the user that corresponds to the current location of the engineering vehicle; Target recognition is performed on the operating environment image and the lidar point cloud data. If a reference target corresponding to the target information is identified, the first relative distance between the engineering vehicle and the reference target is calculated, and the initial positioning information is calculated based on the first relative distance and the absolute coordinates of the reference target in the preset electronic map. If the reference target is not identified, the user is prompted to input the second relative distance between the engineering vehicle and the reference target, and the initial positioning information is calculated based on the second relative distance and the absolute coordinates of the reference target.

[0060] Specifically, when the system detects an invalid satellite signal, it will prompt the user to perform auxiliary initialization on the HMI (Hybrid Management Interface) in the driver's cab. The second positioning unit first receives the target information corresponding to the current position of the engineering vehicle, input by the user. Here, target information usually refers to prominent features along the track that the driver sees visually. In actual operation, the driver stops the vehicle near a specific signal, platform end, or parking point marker. At this time, the driver selects or inputs the current track name, direction of travel, and the nearest reference target ahead on the HMI interface; for example, the driver selects "Signal D198 ahead".

[0061] The second positioning unit invokes a sensing module (such as an industrial camera) to perform target recognition on images of the operating environment and LiDAR point cloud data. The system will attempt to search for a reference target (such as the D198 signal machine) that matches the target information input by the user in the sensing results.

[0062] If a reference target is identified, the system automatically calculates the initial relative distance between the construction vehicle (i.e., the position of the vehicle's front-end sensing device) and the reference target using a fusion of vision and LiDAR ranging algorithms, such as geometric ranging based on monocular vision or point cloud ranging fused with LiDAR. For example, if the target is identified as 15 meters from a traffic signal. Simultaneously, the system retrieves the precise absolute coordinates of the reference target (e.g., traffic signal D198) from a pre-set electronic map. Finally, by subtracting (or adding, depending on the direction) the measured initial relative distance from the absolute coordinates of the reference target, the system can reverse-engineer the current absolute position of the construction vehicle, thus obtaining the initial positioning information. This method utilizes intelligent sensing technology, solving the problem of no satellite signal and avoiding errors in distance estimation by the driver.

[0063] In certain extreme cases, such as when visibility is extremely low, or when the vehicle is parked too far or too close to the traffic signal, exceeding the perception range, the system determines that the reference target has not been identified. To ensure the system can still start, the second positioning unit will execute the following logic: First, the system displays a prompt on the HMI, asking the user to manually input the second relative distance between the vehicle and the reference target. For example, the driver can park the vehicle approximately 5 meters from the traffic signal (or align the front of the vehicle with a landmark), and then confirm or input 5 meters or 0 meters on the screen. Subsequently, the second positioning unit calculates the initial positioning information based on the second relative distance input by the user and the absolute coordinates of the reference target in the preset electronic map. This is a forced alignment mechanism. Although relying on manual estimation may introduce a small amount of error, this error is acceptable during the initialization phase and can be gradually corrected through subsequent positioning calibration functions during driving.

[0064] In this embodiment of the invention, the combination of the first positioning unit and the second positioning unit ensures that the engineering vehicle can reliably complete system initialization whether it is in the open or underground, whether it is automatically sensed or manually assisted, thus solving the technical defects of existing solutions that rely heavily on external interlocking information and cannot be initialized in an independent environment.

[0065] Based on any of the above embodiments, the positioning module 120 further includes a positioning update submodule 122, the positioning update submodule 122 including: The positioning calculation unit 1221 is used to perform positioning calculation based on the initial positioning information and the wheel speed information and inertial measurement information in the running status data to obtain the calculated position; The positioning calibration unit 1222 is used to identify line feature targets from the environmental perception data, match the line feature targets with the preset electronic map, and calibrate the estimated position according to the matching result to obtain the real-time positioning information.

[0066] It should be noted that, considering the characteristics of long rail transit lines and significant environmental changes (such as passing through tunnels, curves, and switch areas), the positioning module also includes a positioning update submodule to eliminate accumulated errors caused by long-term operation. This submodule mainly consists of a positioning calculation unit and a positioning calibration unit, which work together.

[0067] Specifically, the positioning and estimation unit is responsible for providing continuous, high-frequency basic positioning data. Based on the initial positioning information obtained in the aforementioned embodiments (as the starting point for calculation), it performs dead reckoning using the operational status data collected in real time by the sensing module.

[0068] Specifically, operational status data can include wheel speed information (pulse count or velocity values ​​from vehicle wheel axle sensors) and inertial measurement information (acceleration and angular velocity from the IMU). The positioning estimation unit uses the wheel speed information to calculate the longitudinal distance traveled by the vehicle (mileage accumulation), while simultaneously monitoring changes in the vehicle's attitude (such as changes in heading angle during cornering) using inertial measurement information. Through integration calculations, the vehicle's coordinates on a pre-set electronic map are updated in real time to obtain the estimated position. It should be understood that estimated positioning has the advantages of good continuity and is not affected by external environmental obstructions, but it is limited by wheel spin / slippage and sensor drift, and its error will diverge over time, thus requiring the introduction of a calibration mechanism.

[0069] The positioning calibration unit is used to calibrate the estimated position using environmental sensing data during operation. Figure 5 This is a schematic diagram of the positioning update and persistence based on intelligent perception provided by the present invention, as shown below. Figure 5 As shown, this intuitively demonstrates the alternating logic of the positioning modes during the operation of the engineering vehicle. During the speed-mileage accumulation phase, in the section between two traffic signals, where there are no significant feature targets, the system primarily relies on the positioning estimation unit to accumulate speed and mileage based on wheel speed and inertial information, calculating the vehicle's position in real time. During the position calibration phase, when the engineering vehicle approaches a traffic signal beside the line, the system's intelligent sensing function is activated, automatically sensing and measuring the relative distance between the vehicle and the traffic signal. At this point, the system uses the absolute position of the traffic signal to correct the calculated position. Through this cyclical mechanism of calculation during operation and calibration upon encountering a target, the system can effectively eliminate mileage accumulation errors and ensure positioning accuracy over long-term operation.

[0070] Furthermore, the specific workflow of the positioning calibration unit is as follows: Identify the road feature targets located in front of the engineering vehicle from the environmental perception data; Detect whether the route feature target matches the positioning target expected based on the calculated position in the preset electronic map; If a match is found, the measured relative position of the route feature target relative to the engineering vehicle is obtained, and the estimated position is calibrated and updated based on the measured relative position and the absolute coordinates of the positioning target in the preset electronic map to obtain the real-time positioning information.

[0071] Specifically, the positioning calibration unit processes image or point cloud data collected by the perception module in real time, identifying line feature targets located in front of the engineering vehicle from the environmental perception data. Here, line feature targets refer to facilities with fixed positions along the line, obvious characteristics, and already recorded on the map. The most typical example is a signal, but it can also be a specific kilometer marker, transponder location marker, etc. The perception algorithm extracts the appearance features of these targets, such as type, color, and shape, and further extracts their spatial location attributes, including the target's left-right orientation relative to the current line (e.g., located on the left or right side of the track) and its morphological height features (e.g., whether it is a high-pole signal or a low-pole signal). It should be understood that the positioning calibration unit does not simply perform image recognition when identifying line feature targets, but rather uses a fusion detection method combining image recognition and LiDAR. The system does not blindly use any identified object for calibration, but requires logical verification. The positioning calibration unit checks whether the identified line feature targets match the expected positioning targets in the pre-set electronic map based on the calculated location. This matching verification includes not only checking the target type, but also verifying the target's left-right orientation and height / low-pole attributes. For example, if the estimated location indicates that the vehicle is about to reach the D200 traffic signal, and the preset electronic map records the D200 traffic signal as being on the left and having a high post, then if the visual and lidar fusion perception happens to identify a traffic signal ahead whose shape matches the characteristics of D200, and the identification result confirms that the traffic signal is located on the left side of the driving direction and has a high post shape, then it is considered a match. This indicates that the identified line feature target is the expected equipment. Conversely, if the traffic signal identified by the visual and lidar fusion perception is close in distance but located on the right side of the driving direction (possibly an adjacent line traffic signal), or if the identification result is a low post traffic signal (not matching the high post recorded on the map), or if the estimated location indicates that there is a tunnel wall ahead, but the visual and lidar fusion perception still identifies a traffic signal, then it is considered a mismatch. This indicates that the identified line feature target is not the expected equipment. In this case, no vehicle position calibration is performed, and if calibration fails continuously, a manual confirmation is prompted.

[0072] If a match is successful, it indicates that the currently identified device is indeed the one marked on the map (i.e., the expected device). At this point, the positioning calibration unit performs the following operations: First, using visual fusion lidar ranging, it accurately calculates the distance and orientation of the feature target on the route relative to the front of the engineering vehicle; for example, the target is 20.5 meters in front of the vehicle. Then, it retrieves the precise absolute coordinates of the positioning target (such as the D200 signal light) from a pre-set electronic map. Finally, using the target's absolute coordinates and the measured relative position, it calculates the current precise absolute coordinates of the engineering vehicle and replaces the current estimated position with these coordinates. That is, the estimated position is calibrated and updated to obtain real-time positioning information.

[0073] To further improve the accuracy and anti-interference capability of calibration, this embodiment of the invention also employs multi-frame tracking fusion detection technology. Figure 6 This is a schematic diagram of positioning calibration update based on multi-frame tracking fusion detection provided by the present invention, as shown below. Figure 6 As shown, to avoid position jumps caused by momentary noise or misidentification from the sensors, the system employs a rigorous calibration triggering logic. Specifically, based on the vehicle's current estimated position, the system calculates and generates a virtual detection window in real time from a pre-set electronic map (e.g., ...). Figure 6 The yellow area in the middle represents, for example, a range of ±15 meters before and after the target. Calibration is only initiated when the target identified by the sensing system falls within this expected window.

[0074] As a vehicle approaches a target (such as a traffic signal), the perception system continuously tracks and detects the target. This means the system doesn't draw conclusions based solely on a single frame of visual or LiDAR detection, but rather requires consistent and stable identification of the target across multiple consecutive frames of combined visual and LiDAR data. Figure 6 As shown, the system will only perform the operation of updating and calibrating the locomotive position when the distance between the vehicle's current position and the target is less than the preset calibration threshold (i.e., the vehicle enters the optimal observation zone) and the continuous tracking condition is met.

[0075] Through the above mechanism, the system can continuously correct its position by utilizing features such as traffic lights along the route, even in underground tunnels or areas with signal blockage where there are no satellite signals, thus eliminating mileage accumulation errors and achieving high-precision autonomous positioning in all weather conditions and along all road sections.

[0076] Based on any of the above embodiments, the positioning calibration unit 1222 includes a track confirmation subunit, which is used to determine the running track of the engineering vehicle after passing the turnout when the engineering vehicle runs to the turnout area, and lock the positioning trajectory of the engineering vehicle in the preset electronic map to the running track. The track confirmation subunit is specifically configured to perform at least one of the following operations: The operation task information of the engineering vehicle is obtained, and a preset driving route is generated in the preset electronic map based on the start position and end position in the operation task information. The turnout running direction is determined according to the preset driving route, and the running track is determined according to the turnout running direction. Based on the environmental perception data, the turnout direction information is identified, and the turnout direction information is mapped with the turnout topology in the preset electronic map to determine the running track; Identify the status of the signal associated with the turnout, determine the turnout opening direction based on the signal status, and determine the running track according to the turnout opening direction; Receive modification instructions from users regarding turnout direction prompts, and determine the operating track based on the modification instructions; Based on the operating status data, the vehicle body posture change characteristics are identified, and the operating track is determined according to the vehicle body posture change characteristics.

[0077] Specifically, Figure 7 This is a schematic diagram of turnout identification and updating based on intelligent sensing provided by the present invention, as shown below. Figure 7 As shown, when an engineering vehicle approaches an area with a separating turnout, simple mileage calculation cannot determine whether the vehicle is going straight or turning sideways. This can easily lead to the vehicle jumping onto the wrong parallel track on the electronic map. For example, the vehicle may actually be heading to destination B, but the map may show it on the track leading to destination A. To address this, the track confirmation subunit of this system uses a combination of technologies to determine the operating track of the engineering vehicle after passing the turnout and locks the vehicle's positioning trajectory on the pre-set electronic map onto that determined operating track.

[0078] To ensure accuracy, the stock confirmation subunit supports multiple recognition strategies, which can be used individually or in combination. The specific implementation is as follows: Method 1: The EV-IPS system is equipped with an HMI terminal in the train cab. Before the train departs, the driver can configure the task information for this operation (including the starting position, the destination position, etc.). Based on the above information, the system can generate a preset driving route in a pre-set electronic map. Based on the route, the running direction of the turnouts along the line can be automatically generated, realizing the initial line turnout configuration and positioning calculation update.

[0079] Method Two: Because multiple reachable paths exist in some areas of the line, the automatically generated path and turnout direction based on the start and end points of the task cannot be guaranteed to be completely accurate. Therefore, during train operation, the EV-IPS system performs real-time environmental awareness to update the direction of the leading turnout. First, signal status can be used for linkage identification. In typical mainline operations, the opening direction of a turnout is usually linked to the status of the signal protecting that route. The track confirmation subunit can determine the turnout opening direction by identifying the signal status associated with the turnout. Specifically, the system pre-stores the linkage relationship between signals and turnouts in the electronic map. For example, signal D1 protects a lateral route, and signal D2 protects a straight route. When the sensing module detects that signal D1 displays a green light (open) while D2 displays a red light (closed), the system infers that the turnout must have been opened to the lateral direction based on the linkage relationship. Therefore, the system directly determines that the running track is a lateral track.

[0080] Secondly, the system is linked to the turnout direction identification. The track confirmation subunit uses a sensing module (such as a high-resolution camera) to collect image data of the turnout ahead, enabling intelligent identification of the turnout's direction. Specifically, the system uses image processing algorithms to identify the contact status between the turnout switch rail and the stock rail. For example, if the switch rail is in contact on the left, it means the turnout is open to the left; if it's in contact on the right, it means the turnout is open to the straight or right. The system maps and matches the identified physical turnout direction information with the turnout topology in a pre-set electronic map. If the map topology shows that the turnout connects to track A on the left and track B on the right, and the sensing result indicates left-opening, the system determines that the engineering vehicle is about to enter track A and pre-locks the navigation path.

[0081] Furthermore, the EV-IPS system reserves basic manual warning and input functions. After the system completes the identification and update of the turnout ahead, it provides distance and direction warnings through the onboard HMI. If the driver verifies the direction incorrectly, the direction can be modified directly on the HMI terminal. Manual input has the highest priority at any time and can be maintained to ensure the correct operating route and accurate positioning calculation.

[0082] Furthermore, regarding the positioning maintenance and updating in the turnout area, in addition to the aforementioned intelligent detection before the turnout passes, intelligent pose detection is also supported for updating the positioning while passing the turnout. The EV-IPS system's detection sensors, equipped with IMUs and other devices, continuously monitor and determine the pose of the train as it passes through the turnout area. Through intelligent analysis and calculation, the cumulative pose change trend is obtained, enabling accurate positioning updates after the train passes the turnout. This solution can serve as a supplementary measure for EV-IPS when various positioning updates before the turnout fail.

[0083] By combining the above-mentioned methods, the embodiments of the present invention effectively solve the problem of positioning drift of engineering vehicles in complex turnout areas. Even when running across lines to other lines, as long as the system has map data, it can autonomously determine the driving path through intelligent perception, without relying on expensive ground transponder equipment, thus achieving low-cost and highly reliable turnout positioning and locking.

[0084] Based on any of the above embodiments, the control module 130 includes a signal security protection unit 131, which is used for: The system identifies the light status of the forward traffic signal from the environmental perception data and determines whether the forward traffic signal is a valid signal for the current route based on the real-time positioning information. When the signal ahead is a valid signal and the light color indicates a no-passing state, the signal ahead is determined to be a no-passing signal, and the braking distance of the engineering vehicle at the current operating speed is calculated in real time. If the distance between the engineering vehicle and the prohibition signal is less than or equal to the sum of the braking distance and the preset safety margin, a locomotive braking command is triggered to control the engineering vehicle to stop.

[0085] It should be noted that the embodiments of the present invention mainly illustrate how the system can achieve the function of preventing overrun / overrun, that is, how the engineering vehicle can autonomously identify red or blue lights and implement parking control without external ATP support.

[0086] Specifically, Figure 8 This is a schematic diagram of the signal recognition process based on intelligent sensing provided by the present invention, as shown below. Figure 8 As shown, the signal safety protection unit achieves a complete closed loop from signal recognition to safe braking through deep fusion of perception and positioning.

[0087] In complex rail transit environments, multiple signals may appear in front of the engineering vehicle, such as adjacent signal signals and signals facing away from it. The first task of the signal safety protection unit is to accurately identify the signal that has a binding effect on it.

[0088] Specifically, the system utilizes onboard sensing sensors to intelligently perceive the track ahead and identify the light status of the signals ahead (such as red, green, and yellow signals within the mainline operating range, and blue and white signals within stations and depots). The sensing algorithm not only extracts the color but also extracts features such as the signal's position coordinates, height, and type.

[0089] To prevent accidental stops due to misreading adjacent signal lights, the signal safety protection unit performs logical checks. It identifies whether the signal ahead is a valid signal for the current route based on real-time positioning information. Specifically, the system uses high-precision real-time positioning information to search for the nearest next signal (i.e., the expected target) on the current lane ahead of the vehicle in a pre-set electronic map. The system compares the perceived signal's location / attributes with the expected signal's location / attributes on the map. Only when the perceived signal matches the expected location and attributes on the map is it marked as a valid signal. Otherwise, if it is an adjacent signal or an invalid light source, it is rejected.

[0090] To further improve reliability, the system performs intelligent sensing signal tracking and identification confirmation. Through stable tracking of multiple consecutive frames of images, it ensures the stability of signal status judgment and prevents misjudgments caused by flickering or momentary obstruction.

[0091] After locking onto a valid traffic signal, the system enters the safety decision-making phase. When the system determines that the traffic signal ahead is valid and its light status is "no passage" (e.g., red light on the main line, blue light in the depot), it identifies the signal as a prohibitory signal and triggers the protection logic. At this time, the HMI terminal will simultaneously display the identification result and issue an audible and visual warning.

[0092] In addition, the signal safety protection unit calculates the braking distance of the construction vehicle at its current speed in real time based on the vehicle's physical characteristics. Simultaneously, the system monitors the distance (remaining distance) between the construction vehicle's current position and the traffic signal displaying a red or blue light (i.e., a stop signal). When the distance between the construction vehicle and the stop signal ahead is less than or equal to the sum of the braking distance and the safety margin, a risk of overshooting / overshooting is identified. Here, the preset safety margin refers to a buffer zone (e.g., 20 meters) reserved by the system to prevent increased braking distance due to slippery road conditions or braking system delays, ensuring that the vehicle can come to a stable stop before the signal in any extreme situation and never overshoots it.

[0093] Once the above conditions are met, the signal safety protection unit immediately triggers a locomotive braking command to stop the engineering vehicle. This command is sent to the vehicle's braking system via the locomotive interconnection interface, cutting off traction and applying maximum braking force until the vehicle comes to a complete stop, thereby preventing overshooting and undershooting.

[0094] Based on any of the above embodiments, the control module 130 further includes an obstacle detection unit 132, the obstacle detection unit 132 being used for: Based on the real-time positioning information and the operating status data, a three-dimensional track boundary space for the current driving section of the engineering vehicle is constructed in the three-dimensional space corresponding to the environmental perception data. Point cloud data is acquired from the environmental perception data, and the point cloud data is filtered based on the three-dimensional orbital boundary space to obtain point cloud data located within the three-dimensional orbital boundary space. Cluster analysis is performed on the point cloud data located within the three-dimensional track clearance space, and the presence of obstacles within the track clearance of the engineering vehicle is identified based on the analysis results. If the obstacle is detected, an alarm will be triggered or braking control will be applied based on the distance between the obstacle and the engineering vehicle and the current operating speed of the engineering vehicle.

[0095] It should be noted that the embodiments of the present invention mainly illustrate how the system uses intelligent sensing technology to solve the problem that traditional rail transit can only detect train occupancy but cannot detect obstacles, and achieve the function of 1 prevention (i.e., anti-collision).

[0096] Specifically, Figure 9This is a schematic diagram of the obstacle detection process based on intelligent perception provided by the present invention, as shown below. Figure 9 As shown, the obstacle detection unit employs a multi-sensor fusion strategy to perform 3D reconstruction and real-time monitoring of the operating environment ahead of the track. During obstacle detection, although tunnel walls, the ground, and overhead contact line supports are objects, they are outside the safety clearance and should not trigger alarms. Therefore, the obstacle detection unit first needs to define a monitoring range.

[0097] Based on real-time positioning information and operational status data, the obstacle detection unit constructs a three-dimensional track clearance space for the current travel segment of the engineering vehicle within the three-dimensional space corresponding to the environmental perception data. Here, the three-dimensional track clearance space refers to a three-dimensional virtual space formed by adding necessary safety margins to the train's external dimensions. Only objects intruding into this space are considered a threat to driving safety. This step is equivalent to extracting a region of interest from a complex point cloud.

[0098] Specifically, the obstacle detection unit acquires real-time video image data and LiDAR point cloud data based on the locomotive's position and speed information. It then uses intelligent algorithms to identify the track alignment and maps the identified track alignment to the LiDAR coordinate system. Based on the current position and direction of travel, it queries the track centerline within a certain range ahead. Using the track centerline as a reference center, it constructs boundary slices from the real-time point cloud, extracting obstacle point clouds within the boundary. Subsequently, it merges and logically judges the distribution and quantity of obstacle point clouds in each slice, and calculates the location of the nearest obstacle. After detecting an obstacle, the system fuses and analyzes the obstacle location with the engineering vehicle's travel path to ensure that warnings and protection are only provided for targets that affect vehicle safety.

[0099] When an obstacle posing a threat is detected, the obstacle detection unit calculates the distance between the obstacle and the engineering vehicle in real time (based on point cloud ranging) as well as the current speed of the engineering vehicle. If the obstacle is far away, or if a non-emergency target is detected, the system triggers an alarm. At this time, the onboard HMI terminal displays an image of the obstacle, its distance, and its type, and issues an audible alarm to alert the driver to take precautions or manually reduce speed. When the obstacle distance is less than the safe braking distance at the current speed (i.e., there is a risk of imminent collision), the system no longer waits for the driver's reaction and directly applies braking control through the locomotive interface, forcing the vehicle to decelerate or stop, thereby achieving the active safety objective of first-strike protection.

[0100] Based on any of the above embodiments, the control module 130 further includes an overspeed safety protection unit 133, the overspeed safety protection unit 133 being used for: The speed limit value of the current route of the engineering vehicle is obtained from the preset electronic map, and the current operating speed of the engineering vehicle is compared with the speed limit value of the route. If the current operating speed exceeds the line speed limit, an overspeed alarm will be output or normal braking control will be executed.

[0101] It should be noted that the embodiments of the present invention mainly illustrate how the system can monitor vehicle speed at all times and prevent speeding by combining electronic maps with real-time speed measurement.

[0102] Specifically, in rail transit operations, different sections (such as curves, switch areas, and tunnel entrances) have strict speed limits, and manual driving is highly susceptible to speeding due to negligence. The overspeed safety protection unit solves this problem through digital means.

[0103] This unit first uses real-time positioning information provided by the positioning module to obtain the speed limit value of the current route of the engineering vehicle from a pre-set electronic map. It's understandable that the system's pre-set electronic map not only stores the geometry of the route but also stores the speed limit information for the entire route in the form of attribute data. This information includes fixed speed limits (such as permanent speed limits determined by curve radii) and temporary speed limits for work operations (such as temporary low-speed requirements in construction areas). When the vehicle reaches a certain mileage segment, the system automatically retrieves the maximum permissible speed for that segment.

[0104] The system monitors the current operating speed of the engineering vehicle in real time, collected by the sensing module, and compares it with the speed limit of the route. For example, if the current speed limit for the road section is 40 km / h, and the vehicle's actual speed reaches 42 km / h, the system determines that it is speeding.

[0105] To balance driving experience and safety, overspeed protection typically employs a tiered strategy to avoid frequent sudden braking. If the current operating speed exceeds the route's speed limit, the overspeed safety protection unit takes appropriate action based on the degree of overspeed: If the overspeed is minor (e.g., exceeding the speed limit but not reaching the danger threshold, such as 1-2 km / h overspeed), the system outputs an overspeed alarm. At this time, the onboard HMI terminal displays the current speed in a prominent color (e.g., flashing red) and sounds an alarm to alert the driver. If the overspeed is significant, or if the driver fails to take action after the alarm, causing the speed to continue increasing, the system will implement service braking control. Here, service braking is a relatively gentle braking method designed to smoothly reduce the vehicle speed back below the speed limit, rather than bringing the vehicle to a complete stop. This design ensures safety while maintaining driving continuity, avoiding severe impacts on onboard equipment and personnel.

[0106] Based on any of the above embodiments, the system further includes a human-machine interface terminal 140, which is deployed in both driver's cabs of the engineering vehicle. The human-machine interface terminal 140 is used for: The system displays the real-time location information, the identification results in the environmental perception data, the operating status data, and the current driving safety status in real time. The system detects the current activation status of the driver's cab of the engineering vehicle, responds to configuration and control commands input by users in the active driver's cab, and blocks input operations from users in the inactive driver's cab.

[0107] It should be noted that rail transit engineering vehicles typically adopt a dual-cab design (i.e., there is a driver's cab at both the front and rear of the vehicle). In order to adapt to this two-way operation mode, this system deploys human-machine interaction terminals with identical hardware in both driver's cabs at both ends of the engineering vehicle.

[0108] Specifically, Figure 10 This is a schematic diagram of the interaction between the human-computer interaction terminal and the computing control unit provided by the present invention, as shown below. Figure 10 As shown, a closed-loop information flow is formed between the human-machine interface terminal (HMI) and the system's core computing and control unit (i.e., the control module and positioning module). The computing and control unit can push processed key data to the HMI in real time. The HMI is used to display real-time positioning information, recognition results from environmental perception data, operating status data, and the current driving safety status.

[0109] Specifically, the HMI screen displays location information, perception results, status data, and safety status in an intuitive graphical interface. Location information includes the current line name, the specific kilometer marker, and the next station name. Perception results may include the signal number and color (e.g., red / green light icon), the distance and type icon of obstacles ahead, and the direction indication of switches. Status data may include the vehicle's current digital speed display and the speed limit indicator for the current section. Safety status is indicated by background color or pop-up windows to display the current risk level.

[0110] Since both cabs are equipped with screens, simultaneous operation by both drivers could lead to system logic chaos. Therefore, this invention employs an intelligent access control mechanism. The system detects the current activation status of the cab by reading the key signal, the status of the occupancy relay, or the position of the master control handle. At any given time, the vehicle can only have one master control unit (activated) and one slave control unit (inactive). For example, when the driver inserts the key and activates cab I, cab I is determined to be activated, while cab II is automatically determined to be inactive.

[0111] The human-machine interface terminal responds to configuration and control commands input by the user in the active driver's cab. This means that the driver on the main control unit can perform system configuration management (such as setting running tasks and selecting destinations) and result confirmation and processing (such as confirming alarm information and clearing alarms). On the HMI screen of the inactive terminal, although the image is displayed synchronously, all touch buttons and physical buttons are locked or grayed out, and system parameters cannot be changed.

[0112] The intelligent safety protection method for rail transit engineering vehicles provided by the present invention is described below. The intelligent safety protection method for rail transit engineering vehicles described below can be referred to in correspondence with the intelligent safety protection system for rail transit engineering vehicles described above.

[0113] Based on any of the above embodiments Figure 11 This is a flowchart illustrating the intelligent safety protection method for rail transit engineering vehicles provided by the present invention, as shown below. Figure 11 As shown, this method can be applied to the intelligent safety protection system for rail transit engineering vehicles described in any of the above embodiments. The method includes the following steps: Step S10: Real-time collection of the operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; Step S20: Based on the initial positioning information of the engineering vehicle and the operating status data, the positioning is estimated to obtain the estimated position of the engineering vehicle, and the estimated position is calibrated according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. Step S30: Based on the real-time positioning information and the environmental perception data, determine the current driving safety status of the engineering vehicle, and perform safety protection control on the engineering vehicle according to the current driving safety status.

[0114] The method provided in this invention, by collecting real-time operating status data and environmental perception data of engineering vehicles, and combining positioning calculation based on operating status data with a calibration mechanism based on environmental perception data and pre-set electronic maps, eliminates the strong dependence on external trackside equipment and enables high-precision autonomous positioning without relying on external signal input. Furthermore, by comprehensively utilizing real-time positioning information and environmental perception data for dual safety status determination, it can not only prevent overstepping, overshooting, and speeding based on location information, but also actively identify track obstacles or abnormal situations using environmental perception data. This constructs an independent, complete, and proactively environmental detection-enabled vehicle safety protection system, significantly reducing system deployment complexity and cost while effectively solving the safety protection problem of engineering vehicles flexibly crossing lines between different track types.

[0115] It should be noted that other embodiments of the intelligent safety protection method for rail transit engineering vehicles provided by the present invention can refer to the various embodiments of the intelligent safety protection system for rail transit engineering vehicles described above, and will not be repeated here.

[0116] Figure 12 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 12 As shown, the electronic device may include a processor 1210, a communication interface 1220, a memory 1230, and a communication bus 1240, wherein the processor 1210, the communication interface 1220, and the memory 1230 communicate with each other through the communication bus 1240. The processor 1210 can call logical instructions in the memory 1230 to execute an intelligent safety protection method for rail transit engineering vehicles. This method includes: real-time acquisition of the operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; performing positioning estimation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and calibrating the estimated position according to the environmental perception data and a preset electronic map to obtain the real-time positioning information of the engineering vehicle; determining the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and performing safety protection control on the engineering vehicle according to the current driving safety status.

[0117] Furthermore, the logical instructions in the aforementioned memory 1230 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to related technologies, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the intelligent safety protection method for rail transit engineering vehicles provided by the above methods. The method includes: real-time acquisition of the operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; performing positioning estimation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and calibrating the estimated position according to the environmental perception data and a preset electronic map to obtain the real-time positioning information of the engineering vehicle; determining the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and performing safety protection control on the engineering vehicle according to the current driving safety status.

[0119] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the intelligent safety protection method for rail transit engineering vehicles provided by the above methods. The method includes: real-time acquisition of the operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; performing positioning estimation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and calibrating the estimated position according to the environmental perception data and a preset electronic map to obtain the real-time positioning information of the engineering vehicle; determining the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and performing safety protection control on the engineering vehicle according to the current driving safety status.

[0120] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0121] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent safety protection system for rail transit engineering vehicles, characterized in that, include: The sensing module is used to collect real-time operating status data of the engineering vehicle and environmental perception data in the direction of the engineering vehicle's operation; The positioning module is used to perform positioning calculation based on the initial positioning information of the engineering vehicle and the operating status data to obtain the estimated position of the engineering vehicle, and to calibrate the estimated position according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. The control module is used to determine the current driving safety status of the engineering vehicle based on the real-time positioning information and the environmental perception data, and to perform safety protection control on the engineering vehicle according to the current driving safety status.

2. The intelligent safety protection system for rail transit engineering vehicles according to claim 1, characterized in that, The positioning module includes an initialization positioning submodule, which includes: The first positioning unit is used to determine the initial positioning information based on the mapping relationship between the satellite positioning signal and the preset electronic map when the satellite positioning signal is detected to be valid in the current environment. The second positioning unit is used to determine the initial positioning information based on the operating environment image and lidar point cloud data collected by the sensing module and the preset electronic map when the satellite positioning signal is detected to be invalid.

3. The intelligent safety protection system for rail transit engineering vehicles according to claim 2, characterized in that, The second positioning unit is specifically used for: Receive target information input by the user that corresponds to the current location of the engineering vehicle; Target recognition is performed on the operating environment image and the lidar point cloud data. If a reference target corresponding to the target information is identified, the first relative distance between the engineering vehicle and the reference target is calculated, and the initial positioning information is calculated based on the first relative distance and the absolute coordinates of the reference target in the preset electronic map. If the reference target is not identified, the user is prompted to input the second relative distance between the engineering vehicle and the reference target, and the initial positioning information is calculated based on the second relative distance and the absolute coordinates of the reference target.

4. The intelligent safety protection system for rail transit engineering vehicles according to claim 1, characterized in that, The positioning module further includes a positioning update submodule, which includes: The positioning calculation unit is used to perform positioning calculation based on the initial positioning information and the wheel speed information and inertial measurement information in the running status data to obtain the calculated position; The positioning calibration unit is used to identify route feature targets from the environmental perception data, match the route feature targets with the preset electronic map, and calibrate the estimated position according to the matching result to obtain the real-time positioning information.

5. The intelligent safety protection system for rail transit engineering vehicles according to claim 4, characterized in that, The positioning calibration unit is specifically used for: Identify the road feature targets located in front of the engineering vehicle from the environmental perception data; Detect whether the route feature target matches the positioning target expected based on the calculated position in the preset electronic map; If a match is found, the measured relative position of the route feature target relative to the engineering vehicle is obtained, and the estimated position is calibrated and updated based on the measured relative position and the absolute coordinates of the positioning target in the preset electronic map to obtain the real-time positioning information.

6. The intelligent safety protection system for rail transit engineering vehicles according to claim 4, characterized in that, The positioning calibration unit includes a track confirmation subunit, which is used to determine the running track of the engineering vehicle after passing the turnout when the engineering vehicle runs to the turnout area, and lock the positioning trajectory of the engineering vehicle in the preset electronic map to the running track. The track confirmation subunit is specifically configured to perform at least one of the following operations: The operation task information of the engineering vehicle is obtained, and a preset driving route is generated in the preset electronic map based on the start position and end position in the operation task information. The turnout running direction is determined according to the preset driving route, and the running track is determined according to the turnout running direction. Based on the environmental perception data, the turnout direction information is identified, and the turnout direction information is mapped with the turnout topology in the preset electronic map to determine the running track; Identify the status of the signal associated with the turnout, determine the turnout opening direction based on the signal status, and determine the running track according to the turnout opening direction; Receive modification instructions from users regarding turnout direction prompts, and determine the operating track based on the modification instructions; Based on the operating status data, the vehicle body posture change characteristics are identified, and the operating track is determined according to the vehicle body posture change characteristics.

7. The intelligent safety protection system for rail transit engineering vehicles according to claim 1, characterized in that, The control module includes a signal security protection unit, which is used for: The system identifies the light status of the forward traffic signal from the environmental perception data and determines whether the forward traffic signal is a valid signal for the current route based on the real-time positioning information. When the signal ahead is a valid signal and the light color indicates a no-passing state, the signal ahead is determined to be a no-passing signal, and the braking distance of the engineering vehicle at the current operating speed is calculated in real time. If the distance between the engineering vehicle and the prohibition signal is less than or equal to the sum of the braking distance and the preset safety margin, a locomotive braking command is triggered to control the engineering vehicle to stop.

8. The intelligent safety protection system for rail transit engineering vehicles according to claim 1, characterized in that, The control module further includes an obstacle detection unit, which is used for: Based on the real-time positioning information and the operating status data, a three-dimensional track boundary space for the current driving section of the engineering vehicle is constructed in the three-dimensional space corresponding to the environmental perception data. Point cloud data is acquired from the environmental perception data, and the point cloud data is filtered based on the three-dimensional orbital boundary space to obtain point cloud data located within the three-dimensional orbital boundary space. Cluster analysis is performed on the point cloud data located within the three-dimensional track clearance space, and the presence of obstacles within the track clearance of the engineering vehicle is identified based on the analysis results. If the obstacle is detected, an alarm will be triggered or braking control will be applied based on the distance between the obstacle and the engineering vehicle and the current operating speed of the engineering vehicle.

9. The intelligent safety protection system for rail transit engineering vehicles according to claim 1, characterized in that, The control module further includes an overspeed safety protection unit, which is used for: Obtain the speed limit value of the current route of the engineering vehicle from the preset electronic map, and compare the current operating speed of the engineering vehicle with the speed limit value. If the current operating speed exceeds the line speed limit, an overspeed alarm will be output or normal braking control will be executed.

10. The intelligent safety protection system for rail transit engineering vehicles according to any one of claims 1 to 9, characterized in that, The system also includes human-machine interface terminals, which are deployed in the driver's cabs at both ends of the engineering vehicle. These human-machine interface terminals are used for: The system displays the real-time location information, the identification results in the environmental perception data, the operating status data, and the current driving safety status in real time. The system detects the current activation status of the driver's cab of the engineering vehicle, responds to configuration and control commands input by users in the active driver's cab, and blocks input operations from users in the inactive driver's cab.

11. A method for intelligent safety protection of rail transit engineering vehicles, characterized in that, include: Real-time collection of operational status data of engineering vehicles and environmental perception data in the direction of operation of engineering vehicles; Based on the initial positioning information of the engineering vehicle and the operating status data, the estimated position of the engineering vehicle is calculated, and the estimated position is calibrated according to the environmental perception data and the preset electronic map to obtain the real-time positioning information of the engineering vehicle. Based on the real-time positioning information and the environmental perception data, the current driving safety status of the engineering vehicle is determined, and safety protection control is performed on the engineering vehicle according to the current driving safety status.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the intelligent safety protection method for rail transit engineering vehicles as described in claim 11.

Images