Active safety control method and device for train

Abstract

The invention provides a train active safety control method and device, and relates to the technical field of train control. The method comprises the following steps: the sensing module preprocesses image information and radar point cloud information to obtain coordinate system fusion data; the edge calculation unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information; the network central control unit calls an API function based on the selected target bus driving identifier to determine a target communication system, and generates and sends a braking control instruction according to the target communication system and the security data information; and the brake control system performs brake control on an execution mechanism according to the brake control instruction. The device executes the method. According to the method and the device provided by the embodiment of the invention, the real-time performance and the effectiveness of active safety control of the train can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of train control, and in particular to a method and device for active safety control of a train. Background Art

[0002] With the continuous development of railway on-board control technology, train safety control methods have gradually adopted intelligent technology and fusion technology on the basis of existing artificial visual judgment, and developed into active safety control methods that support multiple transmission modes.

[0003] Train safety control is currently mainly achieved through the CTCS train control system, which receives movement authority (MA) and temporary speed limit information transmitted by the ground RBC (Radio Block Center), and the on-board equipment transmits train operation position report information to the RBC (Radio Block Center). The train driver takes relevant safety measures based on the CTCS information and visual operation status to ensure the safe operation of the train.

[0004] With the development of intelligent technology, active train safety control has been put on the agenda. It mainly changes the existing passive traffic safety management mode by using advanced information collection technology, network information transmission technology, intelligent information processing technology and modern control technology, realizes the active defense function before traffic accidents, adopts advanced computer technology and network transmission technology to realize information transmission between various data, realizes the active control function of each system on train safety, and effectively avoids the occurrence of traffic accidents.

[0005] Whether it is safety control based on the train control system or active safety based on intelligent technology, both provide guarantees for train safety from different angles, but there are also the following shortcomings:

[0006] 1. The safety assurance function is decentralized. The various detection systems and control systems transmit signals to each other to realize decentralized control and protection logic. Each controller takes time from receiving signals to signal processing, and the process is long. For example, in the process of emergency braking, the control needs to be transmitted between the CTCS system, hardware relays, input and output signal acquisition, central control unit, traction control unit, and brake control unit to complete the control.

[0007] 2. The interface design between systems is complex. Since different systems use different communication methods and network protocols and involve large amounts of information transmission, the network transmission function design is complex.

[0008] 3. The real-time performance of the system is low. Since data transmission goes through multiple links, some working conditions also involve conversion between different protocols and then sent to the control unit for execution. The transmission and processing links take a long time.

[0009] 4. Differences in communication methods lead to poor compatibility. Since it is necessary to consider the differences in communication methods when active safety equipment is connected to different systems, different communication standards need to be adapted. The hardware of systems with different functions is different and the compatibility is poor.

[0010] Therefore, the existing active safety control method adopts a decentralized functional layout, with a large amount of signal transmission between systems in the entire safety system, complex functional division of labor, and low real-time network transmission. When encountering emergency conditions or emergency stops, the real-time and effectiveness of the active safety system functions cannot be guaranteed. Summary of the invention

[0011] In view of the problems in the prior art, an embodiment of the present invention provides a train active safety control method and device, which can at least partially solve the problems in the prior art.

[0012] On the one hand, the present invention provides a train active safety control method, which is executed based on a train active safety control architecture; the train active safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein:

[0013] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0014] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0015] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0016] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0017] The brake control system performs brake control on the actuator according to the brake control instruction.

[0018] The perception module includes a machine vision acquisition unit, a signal comprehensive processing unit and a radar information acquisition unit; accordingly, the perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data, including:

[0019] The machine vision acquisition unit performs data acquisition to obtain the image information; the radar information acquisition unit performs data acquisition to obtain the radar point cloud information;

[0020] The image information and radar point cloud information are preprocessed by the signal comprehensive processing unit to obtain coordinate system fusion data.

[0021] Among them, the edge computing unit also determines security classification result information based on the security data information, and sends the security data information and the security classification result information to the network central control unit.

[0022] Wherein, the network central control unit includes a multi-bus selector and a network communication interface unit connected to each other; accordingly, the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, including:

[0023] Determine the target bus driver identifier through the multi-bus selector, and call the API function according to the target bus driver identifier;

[0024] The network communication interface unit traverses the standardized interface programs pre-stored locally, determines the target standardized interface program corresponding to the API function, and determines the target communication format corresponding to the target standardized interface program.

[0025] Wherein, the multi-bus selector includes a bus layer, a bus driver selection switch, a standardized interface driver module and a standardized interface connected in sequence; accordingly, determining the target bus driver identifier through the multi-bus selector and calling the API function according to the target bus driver identifier include:

[0026] Selecting a target bus driver from the bus layer through the bus driver selection switch;

[0027] Determine the target driver corresponding to the target bus driver identifier through the standardized interface driver module;

[0028] The target driver is loaded through the standardized interface to obtain API functions corresponding to the bus status APIs.

[0029] Wherein, the network central control unit further includes a network logic processing unit connected to the network communication interface unit; accordingly, generating and sending a braking control instruction according to the target communication format and the safety data information includes:

[0030] The network logic processing unit generates the braking control instruction according to the safety data information, determines a target vehicle-level bus according to the target communication format, and sends the braking control instruction to the braking control system through the target vehicle-level bus.

[0031] Wherein, the train active safety control method further includes:

[0032] The network logic processing unit determines a data receiving period for receiving the security data information and the security classification result information according to the security classification result information.

[0033] On the one hand, the present invention provides a train active safety control device, which is executed based on a train active safety control architecture; the train active safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein the train active safety control device includes:

[0034] An acquisition unit, used for controlling the perception module to pre-process the image information and the radar point cloud information to obtain coordinate system fusion data;

[0035] A detection unit, used to control the edge computing unit to perform target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0036] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0037] a determination unit, configured to control the network central control unit to call an API function based on the selected target bus driver identifier to determine a target communication format, and to generate and send a brake control instruction according to the target communication format and the safety data information;

[0038] The control unit is used to control the brake control system to perform brake control on the actuator according to the brake control instruction.

[0039] In another aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein:

[0040] The processor and the memory communicate with each other via the bus;

[0041] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the following method:

[0042] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0043] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0044] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0045] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0046] The brake control system performs brake control on the actuator according to the brake control instruction.

[0047] An embodiment of the present invention provides a non-transitory computer-readable storage medium, including:

[0048] The non-transitory computer-readable storage medium stores computer instructions, which cause the computer to execute the following method:

[0049] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0050] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0051] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0052] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0053] The brake control system performs brake control on the actuator according to the brake control instruction.

[0054] In the train active safety control method and device provided by the embodiment of the present invention, the perception module pre-processes the image information and the radar point cloud information to obtain the coordinate system fusion data; the edge computing unit performs target detection on the coordinate system fusion data according to the pre-trained target detection model to obtain the safety data information; wherein the target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm according to sample data; the network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends the braking control instruction according to the target communication format and the safety data information; the braking control system performs braking control on the actuator according to the braking control instruction, and selects the bus drive through the integrated layout of the multi-bus selector, which greatly simplifies the train active safety control architecture, and performs the data processing task before the target detection calculation in the signal integration processing unit, which greatly reduces the calculation amount of the edge computing unit, thereby improving the real-time and effectiveness of the train active safety control. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0056] Figure 1 It is a flow chart of a train active safety control method provided by an embodiment of the present invention.

[0057] Figure 2 It is a structural diagram of the active safety control architecture of a train provided by an embodiment of the present invention.

[0058] Figure 3 It is a structural schematic diagram of a multi-bus selector provided in an embodiment of the present invention.

[0059] Figure 4 It is a structural schematic diagram of a train active safety control device provided by an embodiment of the present invention.

[0060] Figure 5 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention. It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily.

[0062] Figure 1 FIG. 1 is a flow chart of a train active safety control method provided by an embodiment of the present invention. Figure 1 As shown, the train active safety control method provided by the embodiment of the present invention is executed based on the train active safety control architecture; the train active safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein:

[0063] Step S1: The perception module preprocesses the image information and radar point cloud information to obtain coordinate system fusion data.

[0064] Step S2: the edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0065] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm based on sample data.

[0066] Step S3: The network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information.

[0067] Step S4: the braking control system performs braking control on the actuator according to the braking control instruction.

[0068] In the above step S1, the device controls the perception module to pre-process the image information and the radar point cloud information to obtain coordinate system fusion data. The device can be a computer device that executes the corresponding steps.

[0069] In the above step S2, the device controls the edge computing unit to perform target detection on the coordinate system fusion data according to the pre-trained target detection model to obtain security data information;

[0070] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm based on sample data.

[0071] In the above step S3, the device controls the network central control unit to call the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a braking control instruction according to the target communication format and the safety data information.

[0072] In the above step S4, the device controls the brake control system to perform brake control on the actuator according to the brake control instruction.

[0073] like Figure 2 As shown, the train active safety control architecture is first described as follows:

[0074] 1. The perception module collects information about obstacles and the environment ahead of the train. The signal comprehensive processing unit completes the establishment of image information and radar point cloud information coordinate system information, and transmits the data information to the edge computing unit via Ethernet;

[0075] 2. The edge computing unit receives the coordinate system fusion data of the image information and radar point cloud information sent by the perception module, calibrates the obstacle position, size, coordinates, safety level information, and environmental limit information, and sends the data to the network central control unit via Ethernet;

[0076] 3. The network central control unit automatically matches the train-level communication protocol (vehicle-level bus), and generates brake control instructions in real time according to the real-time status of the train and safety information parameters, and sends them to the brake control unit for train control; at the same time, the relevant information is displayed on the human-computer interaction unit of the network system to provide operating instructions to the train crew;

[0077] 4. The braking control system initiates braking control instructions to the actuator and simultaneously feeds back the control effect to the network central control unit in real time to jointly realize the corresponding control function.

[0078] Through the comparison of the above control processes, the new active safety control method has generally reduced the data conversion links between the perception unit and the network central control unit, and the data conversion has achieved modularization and miniaturization, reducing the number of times the overall information is transmitted in the network during the function implementation process; reducing the performance requirements for the edge computing unit; the control function and human-computer interaction function are more centralized, reducing the overall time for information interaction between units; and saving hardware and software development costs.

[0079] The active train safety control architecture of the embodiment of the present invention has the following advantages:

[0080] The perception module adopts an integrated module unit. After the signal collection of each sensor component is completed, the preliminary signal processing is completed inside the perception module, and the respective physical coordinate systems are calibrated, which reduces the sensor installation space requirements and the high performance requirements for the edge computing unit, and shortens the model training cycle.

[0081] The adoption of autonomous support for multiple bus functions simplifies the data transmission method of different bus acquisition units after protocol conversion and then entering the network central control unit. Data in different protocol formats are uniformly processed in the network central control unit, reducing intermediate conversion and transmission nodes, improving data transmission and processing efficiency, and shortening safety alarm identification time.

[0082] The human-computer interaction unit adopts a functional fusion approach. By integrating the human-computer interaction protocol and information, it unifies the interface of the perception system human-computer interaction unit and the network system human-computer interaction unit, thereby realizing the centralized display of active safety information and vehicle control information.

[0083] The perception module includes a machine vision acquisition unit, a signal comprehensive processing unit and a radar information acquisition unit; accordingly, the perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data, including:

[0084] The machine vision acquisition unit performs data acquisition to obtain the image information; the radar information acquisition unit performs data acquisition to obtain the radar point cloud information;

[0085] The image information and radar point cloud information are preprocessed by the signal comprehensive processing unit to obtain coordinate system fusion data.

[0086] The machine vision acquisition unit adopts sensors with high safety level, protection level, vibration resistance and impact resistance, which can realize high-precision and high-frequency detection in the forward space; the radar information acquisition unit adopts a lidar sensor with long range, high precision, wide field of view and high reliability; the signal comprehensive processing unit adopts an industrial-grade embedded CPU board with high reliability and stability.

[0087] like Figure 2 As shown, the signal comprehensive processing unit acquires image information through the machine vision acquisition unit, and feeds back a first acquisition result, that is, whether the image information is successfully received, to the machine vision acquisition unit.

[0088] The signal integrated processing unit obtains the radar point cloud information through the radar information acquisition unit, and feeds back a second acquisition result to the radar information acquisition unit, that is, whether the radar point cloud information is successfully received.

[0089] The signal integration processing unit then sends the coordinate system fusion data to the edge computing unit.

[0090] The coordinate system fusion data can be input into the target detection model, and the output result of the target detection model can be used as safety data information, namely, obstacle position, size, coordinates and environmental limit information.

[0091] The edge computing unit uses a server-type edge computing device to undertake model training and centralized data processing; the communication module is integrated into the embedded edge computing host and communicates in real time through the backplane. The perception module and the edge computing unit are connected via Ethernet, and the edge computing unit and the network central control unit are connected via Ethernet.

[0092] The edge computing unit carries the target detection model, completes image recognition and radar detection of intrusions on the train track, establishes a digital information model of the line, automatically identifies intrusions that may endanger driving safety, and sends warning information with different safety levels to the network central control unit. The target detection model, that is, the automatic perception algorithm, consists of a target detection algorithm based on deep learning, a lidar point cloud data detection algorithm, and a data fusion algorithm for the two.

[0093] Among them, the target detection algorithm based on deep learning uses intelligent sensing equipment through cameras, lidar and other sensor components to collect foreign objects and limit information in the vehicle's running direction in real time, locate and identify targets in the image; the lidar point cloud data detection algorithm establishes high-precision distance information of the target, 3D geometric coordinate information of the surrounding environment, reflection intensity and other attribute information, and performs three-dimensional reconstruction of the surrounding environment, and forms limit warnings and obstacle hazard level identification based on point cloud data; the fusion algorithm realizes the fusion of the two coordinate systems and data of the image and radar point cloud, and further matches the obstacle information.

[0094] The edge computing unit also determines security classification result information based on the security data information, and sends the security data information and the security classification result information to the network central control unit. The security classification result information can be determined based on the above-mentioned obstacle position, size, coordinates and environmental limit information. Taking the classification result as two levels as an example, the security classification result information can be determined as high risk and low risk.

[0095] like Figure 3 As shown, current mainstream train network communication drivers are provided, including TCN bus driver, TRDP bus driver, TSN bus driver, CAN bus driver, etc. Each bus driver is an independent functional module, which can realize the sending and receiving of corresponding types of train bus data. One or more bus drivers can be selected by bus driver selection switch to generate bus data interface. The multi-bus selector provided by the present invention, on the basis of protocol converter, carries out the selection of bus driver and the standardized interface design of driver, so that different communication formats can be seamlessly switched in the development platform.

[0096] The train active safety control architecture is an autonomous application software development platform that supports visual programming. It uses block diagram elements and other visual forms to display different functional units, input and output interfaces, connection relationships between functional units, and data and control signal flow information, making the control logic clearer and more intuitive, and conducive to collaborative programming and software management, greatly facilitating the software integration of large systems. Based on this platform, after receiving the intrusion category information, size information, location information and security classification pre-alarm information in real time, it can trigger the corresponding level of warning and protection measures, give vehicle driving suggestions and emergency operation instructions, and provide support for drivers and passengers to make quick judgments in dealing with emergencies. The centralized deployment of safety functions is realized, so that the safety control function reduces the interactive confirmation links between different systems. At the same time, through the support of multiple bus technologies such as Ethernet, MVB (one of the vehicle-level buses), serial communication, CAN communication, etc., multiple communication methods are realized to adapt to different application environments and better meet the functional requirements of safety control in different application scenarios such as high-speed EMUs, urban rail, and locomotives.

[0097] The network central control unit includes a multi-bus selector and a network communication interface unit connected to each other; accordingly, the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, including:

[0098] The target bus driver identifier is determined by the multi-bus selector, and the API function is called according to the target bus driver identifier; the API function corresponds to the bus status API respectively, such as Figure 3 As shown, the bus status API includes a bus configuration API, a bus start API and a bus stop API.

[0099] The network communication interface unit traverses the standardized interface programs pre-stored locally, determines the target standardized interface program corresponding to the API function, and determines the target communication format corresponding to the target standardized interface program. That is, multiple standardized interface programs are pre-stored in the network communication interface unit.

[0100] The multi-bus selector includes a bus layer, a bus driver selection switch, a standardized interface driver module and a standardized interface connected in sequence; accordingly, determining the target bus driver identifier through the multi-bus selector and calling the API function according to the target bus driver identifier includes:

[0101] Selecting a target bus driver from the bus layer through the bus driver selection switch;

[0102] Determine the target driver corresponding to the target bus driver identifier through the standardized interface driver module;

[0103] The target driver is loaded through the standardized interface to obtain API functions corresponding to the bus status APIs.

[0104] The network central control unit also includes a network logic processing unit connected to the network communication interface unit; accordingly, generating and sending a braking control instruction according to the target communication format and the safety data information includes:

[0105] The network logic processing unit generates the braking control instruction according to the safety data information, determines a target vehicle-level bus according to the target communication format, and sends the braking control instruction to the braking control system through the target vehicle-level bus.

[0106] The train active safety control method also includes:

[0107] The network logic processing unit determines a data receiving period for receiving the security data information and the security classification result information according to the security classification result information.

[0108] Combination Figure 2 The instructions are as follows:

[0109] The network communication interface unit receives the API function sent by the multi-bus selector in advance, determines the target communication format according to the API function, and feeds back the target communication format to the multi-bus selector.

[0110] When the network communication interface unit receives security data information and security classification result information, it sends the security classification result information to the network logic processing unit. If the security classification result information is high risk, the data receiving period is determined to be the first period value; if the security classification result information is low risk, the data receiving period is determined to be the second period value. The first period value is less than the second period value. For example, the first period value can be selected as 10 seconds, and the second period value can be selected as 1 minute.

[0111] The data receiving cycle is fed back to the edge computing unit so that the edge computing unit can send security data information and security classification result information to the network communication interface unit according to the data receiving cycle.

[0112] The train active safety control method also includes:

[0113] The network logic processing unit generates a driver operation instruction according to the safety data information, and directly sends the driver operation instruction to the network system human-computer interaction unit through the internal bus, so that the network system human-computer interaction unit displays the driver operation instruction and provides the driver with an operation reference, thereby realizing active safety control of the train.

[0114] like Figure 2As shown, the network communication interface unit sends a brake control instruction to the brake control unit.

[0115] Figure 2 The brake control unit in the embodiment may be specifically an MCU, the execution control module may be specifically a switch solenoid valve and a relay, etc., and the actuator may be specifically a brake pad. Accordingly, the MCU sends a brake control instruction to the switch solenoid valve and the relay, etc., and the switch solenoid valve and the relay perform brake control on the brake pad and feed back the brake control state of the brake pad to the MCU.

[0116] The train active safety control method provided in the embodiment of the present invention realizes the automatic selection and centralized processing of different communication formats through the support of multi-bus functions, and realizes the unification of the communication mode between the perception system and the train-level control system; it saves various communication conversion units or communication conversion boards from the hardware, and also improves the data transmission efficiency from the software, realizes the functional integration of perception warning and train network control, reduces the interaction links between systems, improves control accuracy, and compresses control time.

[0117] The active safety control method of the embodiment of the present invention has the following characteristics:

[0118] Adopting an independent development platform, each bus function is independent of each other, and each function can be used separately or simultaneously, which can meet the needs of different application scenarios, and the external interface is standardized; it is convenient for the expansion of new buses, and the overall architecture of the platform remains unchanged, and modularization is achieved in functional expansion, which greatly improves the efficiency of software development. At the same time, due to the decoupling between the interfaces, the development cycle of new functions is shortened, the system development cost is reduced, and the reliability of system implementation is improved, breaking through the current technical barriers that hinder the development of active safety control of trains, and has good development prospects.

[0119] In the train active safety control method provided by an embodiment of the present invention, the perception module pre-processes image information and radar point cloud information to obtain coordinate system fusion data; the edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain safety data information; wherein the target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm according to sample data; the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a braking control instruction according to the target communication format and the safety data information; the braking control system performs braking control on the actuator according to the braking control instruction, and selects the bus drive through a multi-bus selector integrated layout, which greatly simplifies the train active safety control architecture, and performs data processing tasks before target detection calculation in the signal integration processing unit, which greatly reduces the calculation amount of the edge computing unit, thereby improving the real-time and effectiveness of the train active safety control.

[0120] Furthermore, the perception module includes a machine vision acquisition unit, a signal comprehensive processing unit and a radar information acquisition unit; accordingly, the perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data, including:

[0121] The image information is obtained by collecting data through the machine vision collection unit; the radar point cloud information is obtained by collecting data through the radar information collection unit; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0122] The image information and the radar point cloud information are preprocessed by the signal comprehensive processing unit to obtain coordinate system fusion data. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.

[0123] Furthermore, the edge computing unit further determines security classification result information according to the security data information, and sends the security data information and the security classification result information to the network central control unit. The above embodiments can be referred to for explanation and will not be described in detail.

[0124] Further, the network central control unit includes a multi-bus selector and a network communication interface unit connected to each other; accordingly, the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, including:

[0125] The target bus driver identifier is determined by the multi-bus selector, and the API function is called according to the target bus driver identifier; the above-mentioned embodiment can be referred to for description and will not be described in detail.

[0126] The network communication interface unit traverses the standardized interface programs pre-stored locally, determines the target standardized interface program corresponding to the API function, and determines the target communication format corresponding to the target standardized interface program.

[0127] Further, the multi-bus selector includes a bus layer, a bus driver selection switch, a standardized interface driver module and a standardized interface connected in sequence; accordingly, determining the target bus driver identifier through the multi-bus selector and calling the API function according to the target bus driver identifier includes:

[0128] The target bus driver is selected from the bus layer through the bus driver selection switch; the description can be made with reference to the above embodiment and will not be repeated here.

[0129] The target driver corresponding to the target bus driver identifier is determined by the standardized interface driver module; the description may refer to the above embodiment and will not be repeated here.

[0130] The target driver is loaded through the standardized interface to obtain API functions corresponding to the bus status APIs. The above embodiments can be referred to for explanation and will not be described in detail.

[0131] Further, the network central control unit also includes a network logic processing unit connected to the network communication interface unit; accordingly, generating and sending a braking control instruction according to the target communication format and the safety data information includes:

[0132] The network logic processing unit generates the brake control instruction according to the safety data information, determines the target vehicle level bus according to the target communication format, and sends the brake control instruction to the brake control system through the target vehicle level bus.

[0133] Furthermore, the train active safety control method further includes:

[0134] The network logic processing unit determines the data receiving period for receiving the security data information and the security classification result information according to the security classification result information.

[0135] Figure 4 FIG. 1 is a schematic diagram of the structure of a train active safety control device provided by an embodiment of the present invention. Figure 4As shown, an active train safety control device provided by an embodiment of the present invention is executed based on an active train safety control architecture; the active train safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein the active train safety control device includes an acquisition unit 401, a detection unit 402, a determination unit 403 and a control unit 404, wherein:

[0136] The acquisition unit 401 is used to control the perception module to pre-process the image information and the radar point cloud information to obtain the coordinate system fusion data; the detection unit 402 is used to control the edge computing unit to perform target detection on the coordinate system fusion data according to the pre-trained target detection model to obtain the safety data information; wherein, the target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm according to sample data; the determination unit 403 is used to control the network central control unit to call the API function based on the selected target bus driver identifier to determine the target communication format, and generate and send the braking control instruction according to the target communication format and the safety data information; the control unit 404 is used to control the braking control system to perform braking control on the actuator according to the braking control instruction.

[0137] Specifically, the acquisition unit 401 in the device is used to control the perception module to pre-process the image information and radar point cloud information to obtain coordinate system fusion data; the detection unit 402 is used to control the edge computing unit to perform target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain safety data information; wherein, the target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm based on sample data; the determination unit 403 is used to control the network central control unit to call an API function based on the selected target bus driver identifier to determine the target communication format, and generate and send a braking control instruction according to the target communication format and the safety data information; the control unit 404 is used to control the braking control system to perform braking control on the actuator according to the braking control instruction.

[0138] In the train active safety control device provided by an embodiment of the present invention, the perception module pre-processes image information and radar point cloud information to obtain coordinate system fusion data; the edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain safety data information; wherein the target detection model is obtained by training a fusion algorithm including a target detection algorithm and a lidar point cloud data detection algorithm according to sample data; the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a braking control instruction according to the target communication format and the safety data information; the braking control system performs braking control on the actuator according to the braking control instruction, and selects the bus drive through a multi-bus selector integrated layout, which greatly simplifies the train active safety control architecture, and performs data processing tasks before target detection calculation in the signal integration processing unit, greatly reducing the calculation amount of the edge computing unit, thereby improving the real-time and effectiveness of the train active safety control.

[0139] The embodiments of the active train safety control device provided by the embodiments of the present invention can be specifically used to execute the processing flow of the above-mentioned method embodiments. Its functions are not repeated here, and reference can be made to the detailed description of the above-mentioned method embodiments.

[0140] Figure 5 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Figure 5 As shown, the electronic device includes: a processor (processor) 501, a memory (memory) 502 and a bus 503;

[0141] The processor 501 and the memory 502 communicate with each other via a bus 503;

[0142] The processor 501 is used to call the program instructions in the memory 502 to execute the methods provided by the above method embodiments, for example, including:

[0143] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0144] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0145] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0146] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0147] The brake control system performs brake control on the actuator according to the brake control instruction.

[0148] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can perform the methods provided by the above method embodiments, for example, including:

[0149] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0150] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0151] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0152] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0153] The brake control system performs brake control on the actuator according to the brake control instruction.

[0154] This embodiment provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program enables the computer to execute the methods provided by the above method embodiments, for example, including:

[0155] The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data;

[0156] The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information;

[0157] The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data;

[0158] The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information;

[0159] The brake control system performs brake control on the actuator according to the brake control instruction.

[0160] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0162] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0164] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0165] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A train active safety control method, characterized in that: The train active safety control method is executed based on the train active safety control architecture; the train active safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein: The perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data; The edge computing unit performs target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information; The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data; The network central control unit calls the API function based on the selected target bus driver identifier to determine the target communication format, and generates and sends a brake control instruction according to the target communication format and the safety data information; The brake control system performs brake control on the actuator according to the brake control instruction.

2. The train active safety control method according to claim 1, characterized in that: The perception module includes a machine vision acquisition unit, a signal comprehensive processing unit and a radar information acquisition unit; accordingly, the perception module pre-processes the image information and the radar point cloud information to obtain coordinate system fusion data, including: The machine vision acquisition unit performs data acquisition to obtain the image information; the radar information acquisition unit performs data acquisition to obtain the radar point cloud information; The image information and radar point cloud information are preprocessed by the signal comprehensive processing unit to obtain coordinate system fusion data.

3. The train active safety control method according to claim 1, characterized in that: The edge computing unit also determines security classification result information based on the security data information, and sends the security data information and the security classification result information to the network central control unit.

4. The train active safety control method according to claim 3, characterized in that: The network central control unit includes a multi-bus selector and a network communication interface unit connected to each other; accordingly, the network central control unit calls an API function based on the selected target bus driver identifier to determine the target communication format, including: Determine the target bus driver identifier through the multi-bus selector, and call the API function according to the target bus driver identifier; The network communication interface unit traverses the standardized interface programs pre-stored locally, determines the target standardized interface program corresponding to the API function, and determines the target communication format corresponding to the target standardized interface program.

5. The train active safety control method according to claim 4, characterized in that: The multi-bus selector includes a bus layer, a bus driver selection switch, a standardized interface driver module and a standardized interface connected in sequence; accordingly, determining the target bus driver identifier through the multi-bus selector and calling the API function according to the target bus driver identifier includes: Selecting a target bus driver from the bus layer through the bus driver selection switch; Determine the target driver corresponding to the target bus driver identifier through the standardized interface driver module; The target driver is loaded through the standardized interface to obtain API functions corresponding to the bus status APIs.

6. The train active safety control method according to claim 4, characterized in that: The network central control unit also includes a network logic processing unit connected to the network communication interface unit; accordingly, generating and sending a braking control instruction according to the target communication format and the safety data information includes: The network logic processing unit generates the braking control instruction according to the safety data information, determines a target vehicle-level bus according to the target communication format, and sends the braking control instruction to the braking control system through the target vehicle-level bus.

7. The train active safety control method according to claim 6, characterized in that: The train active safety control method also includes: The network logic processing unit determines a data receiving period for receiving the security data information and the security classification result information according to the security classification result information.

8. A train active safety control device, characterized in that: The train active safety control device is executed based on the train active safety control architecture; the train active safety control architecture includes a perception module, an edge computing unit, a network central control unit and a braking control system connected in sequence, wherein the train active safety control device includes: An acquisition unit, used for controlling the perception module to pre-process the image information and the radar point cloud information to obtain coordinate system fusion data; A detection unit, used to control the edge computing unit to perform target detection on the coordinate system fusion data according to a pre-trained target detection model to obtain security data information; The target detection model is obtained by training a fusion algorithm including a target detection algorithm and a laser radar point cloud data detection algorithm according to sample data; a determination unit, configured to control the network central control unit to call an API function based on the selected target bus driver identifier to determine a target communication format, and to generate and send a brake control instruction according to the target communication format and the safety data information; The control unit is used to control the brake control system to perform brake control on the actuator according to the brake control instruction.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images