Railway foreign matter detection method and system based on cognitive waveform and semantic digital twinning

Abstract

This application relates to the field of rail transit technology and provides a railway foreign object detection method and system based on cognitive waveform and semantic digital twin. The method includes: adjusting the transmission waveform parameters of the train radar based on train operation status information; controlling the train radar to transmit cognitive waveform radar signals to the surrounding track scene according to the adjusted transmission waveform parameters and collecting radar echo information; calling a pre-constructed semantic digital twin scene model, matching the radar echo information with fixed structures in the model, calculating the train radar pose, and establishing a detection coordinate system; generating phase compensation parameters based on historical phase information; fusing the operation status information, candidate target data from the radar echo information, geometric constraint information of the detection coordinate system, and phase compensation parameters to obtain a target description result in the detection coordinate system; and performing foreign object identification based on the target description result and outputting the detection result. This effectively reduces the detection cost of foreign object detection.

Description

Technical Field

[0001] This application relates to the field of rail transit technology, specifically to a railway foreign object detection method and system based on cognitive waveform and semantic digital twin. Background Technology

[0002] Currently, the main method for detecting foreign objects is to deploy cameras, infrared sensors, and laser sensors along the track to detect foreign objects in key areas. While this method can detect foreign objects in specific sections, it typically requires large-scale deployment of detection equipment along the track, resulting in high construction costs, extensive maintenance, and susceptibility of the equipment to environmental factors along the track.

[0003] It is evident that foreign object detection solutions in related technologies suffer from high detection costs. Summary of the Invention

[0004] This application aims to address the problem of high detection costs in existing foreign object detection schemes in related technologies, and provides a railway foreign object detection method and system based on cognitive waveforms and semantic digital twins.

[0005] To solve the above problems, this application is implemented as follows:

[0006] In a first aspect, this application provides a railway foreign object detection method based on cognitive waveforms and semantic digital twins, applicable to trains equipped with train radar, the method comprising: The train's operating status information is obtained, and the transmission waveform parameters of the train radar are adjusted based on the operating status information to obtain the adjusted transmission waveform parameters. According to the adjusted transmission waveform parameters, the train radar is controlled to transmit cognitive waveform radar signals to the surrounding environment of the track and to acquire radar echo information of the surrounding environment of the track. The system calls a pre-built semantic digital twin scene model to match the radar echo information with the fixed structure in the semantic digital twin scene model, calculates the pose information of the train radar, and establishes a detection coordinate system based on the pose information. Historical phase time series are constructed based on the collected historical phase information. The historical phase time series are input into the vibration prediction model to predict the phase change trend, and the phase prediction results are obtained. Phase compensation parameters are generated based on the phase prediction results. The operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are fused to obtain the target description result in the detection coordinate system. Foreign object identification is performed based on the target description results, and the detection results are output.

[0007] Secondly, this application provides a railway foreign object detection system based on cognitive waveforms and semantic digital twins, applicable to trains equipped with train radar, the system comprising: An adjustment module is used to acquire the train's operating status information and adjust the transmission waveform parameters of the train radar based on the operating status information to obtain the adjusted transmission waveform parameters. The transmission and acquisition module is used to control the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track according to the adjusted transmission waveform parameters, and to acquire the radar echo information of the surrounding environment of the track. The calculation module is used to call a pre-built semantic digital twin scene model, match the radar echo information with the fixed structure in the semantic digital twin scene model, calculate the pose information of the train radar, and establish a detection coordinate system based on the pose information. The prediction generation module is used to construct a historical phase time series based on the collected historical phase information, input the historical phase time series into the vibration prediction model to predict the phase change trend, obtain the phase prediction result, and generate phase compensation parameters based on the phase prediction result. The fusion acquisition module is used to fuse the operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters to obtain the target description result in the detection coordinate system. The output module is used to identify foreign objects based on the target description results and output the detection results.

[0008] Thirdly, this application provides a terminal device including a processor and a memory, wherein the memory stores a program or instructions executable on the processor, and the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0009] Fourthly, this application provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0010] Fifthly, this application provides a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the steps of the method described in the first aspect.

[0011] In a sixth aspect, this application provides a computer program product stored in a storage medium, which is executed by at least one processor to perform the steps of the method described in the first aspect.

[0012] Compared with the prior art, this application has the following beneficial effects: By employing a cognitive waveform adaptive design method, the transmitted waveform parameters of the train radar can be dynamically adjusted according to the train's operating speed. Combined with a vibration prediction compensation mechanism, the impact of high-speed motion and vehicle vibration on radar echo information is reduced, thereby improving the stability of foreign object detection and the accuracy of target localization in high-speed scenarios. Furthermore, by synergistically integrating cognitive waveform design, semantic scene modeling, and vibration prediction compensation, the system ensures detection accuracy, real-time performance, and environmental adaptability while reducing the need for additional hardware and complex manual maintenance. This simplifies system deployment and enhances the system's engineering practicality, economy, and promotional value.

[0013] In addition, by adopting the railway foreign object detection method based on cognitive waveform and semantic digital twin provided in this application, real-time detection and accurate positioning of foreign objects around the track can be achieved simply by using train radar installed on the train. Compared with foreign object detection schemes that deploy detection equipment on a large scale along the track, it can also effectively reduce the detection cost of foreign objects. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating a railway foreign object detection method based on cognitive waveforms and semantic digital twins provided in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a railway foreign object detection system based on cognitive waveform and semantic digital twin provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] The terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses. Additionally, the use of "and / or" in this application indicates at least one of the connected objects, such as A and / or B and / or C, representing seven possibilities: including A alone, B alone, C alone, and the presence of both A and B, both B and C, both A and C, and the presence of A, B, and C.

[0018] In this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0019] The railway foreign object detection method based on cognitive waveform and semantic digital twin provided in this application is described below.

[0020] See Figure 1 , Figure 1 This is a flowchart illustrating a railway foreign object detection method based on cognitive waveforms and semantic digital twins provided in an embodiment of this application. Figure 1 The railway foreign object detection method based on cognitive waveform and semantic digital twin shown can be executed by terminal devices such as mobile phones and computers, or it can be implemented by the train control system.

[0021] like Figure 1 As shown, the railway foreign object detection method based on cognitive waveform and semantic digital twin provided in this application is applied to trains equipped with train radar. The method includes the following steps: Step 101: Obtain the train's operating status information, and adjust the transmission waveform parameters of the train radar based on the operating status information to obtain the adjusted transmission waveform parameters.

[0022] In this embodiment, the train's operating speed, acceleration, attitude angular velocity, and position information can be obtained based on the train's control system.

[0023] In some embodiments, the above-mentioned running state information can be represented as a running state vector.

[0024] Specifically, a running state vector can be constructed. ,in, Indicates that the train is The running speed at any given moment, Indicates that the train is acceleration at any moment Indicates that the train is Attitude angular velocity at time t, Indicates that the train is Location information at any given time.

[0025] By adjusting the transmission waveform parameters of the train radar using operational status information, the impact on the detection results of the train radar under high-speed operating conditions can be reduced.

[0026] In some embodiments, the transmission waveform parameters of the train radar can be adjusted according to a preset speed mapping relationship and based on the operating speed.

[0027] In some embodiments, the operating status information includes train speed information and acceleration information, and the transmitted waveform parameters include frequency modulation slope and pulse repetition frequency; The adjustment of the train radar's transmission waveform parameters based on the operational status information includes: Based on the reference frequency modulation slope, the velocity information, and the acceleration information, the frequency modulation slope is adjusted to obtain the adjusted frequency modulation slope. Based on the reference pulse repetition frequency and the velocity information, the pulse repetition frequency is adjusted to obtain the adjusted pulse repetition frequency; The adjusted transmitted waveform parameters include the adjusted frequency modulation slope and the adjusted pulse repetition frequency.

[0028] In some embodiments, a set of transmit waveform parameters can be constructed. ,in, For frequency modulation slope, The pulse repetition frequency, These are the launch compensation parameters.

[0029] Among them, frequency modulation slope Satisfy the following formula:

[0030] Pulse repetition frequency Satisfy the following formula:

[0031] in, As the reference frequency modulation slope, The reference pulse repetition frequency, , and This is the adjustment coefficient.

[0032] Step 102: According to the adjusted transmission waveform parameters, control the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track, and obtain radar echo information of the surrounding environment of the track.

[0033] In this embodiment, by transmitting cognitive waveform radar signals according to the adjusted transmission waveform parameters, the impact on the detection results of train radar under high-speed operating conditions can be reduced.

[0034] In some embodiments, controlling the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track includes: Based on the angle between the line-of-sight direction of the train radar and the direction of movement of the train, as well as the operating wavelength of the train radar, Doppler pre-compensation parameters are generated. The train radar signal to be transmitted is compensated based on the Doppler pre-compensation parameters to obtain a cognitive waveform radar signal. The train radar is controlled to transmit the compensated cognitive waveform radar signal to the surrounding environment of the track.

[0035] In this embodiment, by compensating the cognitive waveform radar signal with Doppler pre-compensation parameters, the train radar can be controlled to transmit a detection waveform that is adapted to the current operating state.

[0036] In some embodiments, the angle between the line-of-sight direction of the train radar and the direction of train movement can be used. and the operating wavelength of the train radar. A pre-compensation term is constructed, and Doppler pre-compensation parameters are generated based on the constructed pre-compensation term. In order to precode or pre-compensate the radar signal to be transmitted based on the Doppler pre-compensation parameters, and obtain the cognitive waveform radar signal.

[0037] Among them, the pre-compensation item The expression is:

[0038] In some embodiments, the acquired radar echo information can be preprocessed to obtain target candidate data.

[0039] Specifically, radar echo information can be processed by framing, buffering, noise suppression, outlier removal, and amplitude and phase information extraction to obtain target candidate data.

[0040] In some embodiments, the target candidate data can be understood as a set of candidate target points, and includes multiple candidate target points, wherein the candidate target points... It can be represented as:

[0041] in, For the first The position coordinates of the candidate target points For speed information, For amplitude information, This is phase information.

[0042] In some embodiments, phase information can be used as input data for subsequent vibration prediction and compensation, and target candidate data can be used as the basis data for subsequent foreign object identification and location calculation.

[0043] Step 103: Call the pre-built semantic digital twin scene model, match the radar echo information with the fixed structure in the semantic digital twin scene model, calculate the pose information of the train radar, and establish a detection coordinate system based on the pose information.

[0044] In some embodiments, a semantic digital twin scenario model can be pre-built through the following steps: Acquire low-speed inspection data or historical scene data as the raw data source for scene model construction; preprocess the acquired scene data, including filtering and noise reduction, data alignment, and invalid data removal. The preprocessed data is used to identify fixed structures along the railway line, including one or more of the following: catenary supports, tunnel walls, track boundaries, bridge structures, and sound barriers; and the identified fixed structures are semantically classified and labeled to form the output of the semantic recognition layer. Extract the spatial location and geometric features of the fixed structures, establish the relative geometric constraint relationships between the fixed structures, and form a geometric constraint layer; Based on semantic classification results and geometric constraints, a semantic digital twin scene model is constructed. This semantic digital twin scene model is used to output scene constraint information and localization reference information for subsequent detection processes.

[0045] In some embodiments, the step of invoking a pre-built semantic digital twin scene model to match the radar echo information with a fixed structure in the semantic digital twin scene model and calculating the pose information of the train radar includes: The radar echo information is matched with the fixed structure in the semantic digital twin scene model by calling a pre-built semantic digital twin scene model to obtain the target fixed structure that is successfully matched. The coordinates of the target fixed structure in the radar coordinate system and the coordinates of the target fixed structure in the semantic digital twin scene coordinate system are used to form a feature point pair; The coordinate rotation matrix and translation vector are solved by minimizing the geometric error of the feature point pairs; Based on the coordinate rotation matrix and the translation vector, the pose information of the train radar is determined; The pose information includes position information and attitude information.

[0046] In some embodiments, the mapping relationship between the radar coordinate system and the semantic digital twin scene coordinate system can be expressed as:

[0047] in, This represents the target position vector in the radar coordinate system. Represents the target position vector in the semantic digital twin scene coordinate system. Represents the coordinate rotation matrix. This represents the coordinate translation vector.

[0048] In some embodiments, if the set of feature points of the fixed structure are respectively and The geometric constraint solution can be achieved by minimizing the error function as follows:

[0049] in, This represents the number of fixed structural feature points participating in the matching.

[0050] Step 104: Construct a historical phase time series based on the collected historical phase information, input the historical phase time series into the vibration prediction model to predict the phase change trend, obtain the phase prediction result, and generate phase compensation parameters based on the phase prediction result.

[0051] In this embodiment, by predicting the phase change trend and generating phase compensation parameters based on the phase prediction results, the stability of the detection results of the train radar can be improved.

[0052] In some embodiments, historical phase information can be extracted from the collected radar echo information, and a historical phase time series can be constructed based on the historical phase information. Then, the historical phase time series is input into the vibration prediction model to predict the phase change trend at future moments and obtain the phase prediction result.

[0053] In some embodiments, the vibration prediction model can be a prediction model built on an LSTM network, with the input being a phase sequence at multiple consecutive time points and the output being an estimate of the phase change at future time points.

[0054] In some embodiments, the historical phase time series can be set as follows:

[0055] Among them, the predicted phase prediction results It can be represented as:

[0056] In the formula, This represents a prediction function based on historical phase time series.

[0057] In some embodiments, phase compensation parameters can be generated based on the phase prediction results, and the current phase information can be adjusted accordingly. Feedforward correction is performed to obtain the compensated phase information. Phase information It can be represented as:

[0058] In some embodiments, the vibration compensation amount can be mapped to the coordinate correction amount. And can be based on coordinate correction amount For the target position before correction Make corrections to obtain the corrected target position. And the corrected target location It can be represented as:

[0059] The aforementioned target location can be understood as the position coordinates of candidate target points in the target candidate data.

[0060] This setting can reduce the impact of vehicle vibration and attitude changes on the test results.

[0061] Step 105: The running status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are fused to obtain the target description result in the detection coordinate system.

[0062] In this embodiment, by fusing operating status information, candidate target data, geometric constraint information, and phase compensation parameters, a reliable target description result can be formed, thereby improving the accuracy of foreign object detection results.

[0063] In some embodiments, fusing the operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters to obtain the target description result in the detection coordinate system includes: The running status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are uniformly spatiotemporally registered to obtain the spatiotemporal registration result. Based on the spatiotemporal registration results, the candidate target data is mapped to the detection coordinate system, and the candidate target data is subjected to position correction, data aggregation, and target reconstruction to obtain the corresponding target description results.

[0064] In some embodiments, the target description result can be represented as:

[0065] in, Indicates running status information. Represents candidate target data. Represents scene geometric constraint information. Indicates vibration compensation parameters, This represents the fusion mapping function.

[0066] In some embodiments, the state vector of a candidate target point after location correction, data aggregation, and target reconstruction can be further represented as:

[0067] in, The corrected position coordinates of the candidate target point. This is the corrected velocity.

[0068] Step 106: Based on the target description results, identify foreign objects and output the detection results.

[0069] In this embodiment, foreign object identification can be performed based on the target description results to identify the real foreign object target and output the detection results including the real foreign object target.

[0070] In some embodiments, the foreign object identification based on the target description result and the output of the detection result includes: Pre-defined track safety clearance zone; Based on the target description results, the target to be judged is determined to be located within the track safety clearance area; If the number of consecutive occurrences of the target to be determined is greater than a preset number, the target to be determined is identified as a real foreign object target, and a detection result including the real foreign object target is output.

[0071] In some embodiments, background structural echoes, random noise points, or targets in non-dangerous areas are removed to reduce the interference of non-real foreign objects on the detection results.

[0072] In some embodiments, the track safety clearance zone can be defined as follows: The location coordinates of the target to be determined are: Then the intrusion limit determination function can be expressed as:

[0073] in, This indicates that the target to be determined is located within the track safety clearance area and is identified as a foreign object intruding into the clearance area; This indicates that the target to be determined is located outside the track safety clearance area.

[0074] In some embodiments, the number of consecutive occurrences of the target to be determined can be considered. and target confidence Establish foreign object detection criteria:

[0075] in, Indicates the first One of the targets to be determined was identified as a real foreign object target. For consecutive frames, This is the confidence threshold.

[0076] In some embodiments, for a real foreign object target, the location coordinates, distance information, orientation information, intrusion status and risk level of the real foreign object target can be calculated in the detection coordinate system, and corresponding detection results can be generated.

[0077] In some embodiments, the detection results may also include early warning information.

[0078] The detection results can be output to the vehicle-mounted display terminal, ground monitoring platform, or upper-level safety early warning system to achieve real-time detection, accurate positioning, and early warning of foreign objects around the track.

[0079] In some embodiments, the distance to the real foreign object target and direction They can be represented as:

[0080]

[0081] In the formula, The coordinates are the location coordinates of the actual foreign object target.

[0082] In some embodiments, risk level According to the infringement status ,distance and speed The comprehensive calculation is as follows:

[0083] in, , , These are the weighting coefficients. To prevent extremely small positive numbers with a denominator of zero.

[0084] In some embodiments, risk levels can be considered. Generate early warning information so that the detection results, including the early warning information, can be output.

[0085] The railway foreign object detection method based on cognitive waveform and semantic digital twin provided in this application, by adopting a technical solution that combines cognitive waveform design, semantic digital twin scene constraints and vibration prediction feedforward compensation, can achieve real-time detection and accurate positioning of foreign objects around the track under high-speed operation conditions.

[0086] By utilizing existing fixed structures along the railway line, such as overhead contact line supports and tunnel walls, as scene references, and establishing geometric constraints through a semantic digital twin scene model, there is no need to install additional reflective labels, corner reflectors, or large-scale fixed detection equipment. This saves on the installation and maintenance costs of external auxiliary facilities and improves the system's ability to be widely promoted and applied.

[0087] By adopting the cognitive waveform adaptive design method, the transmitted waveform parameters of the train radar can be dynamically adjusted according to the train's operating speed. Combined with the vibration prediction compensation mechanism, the impact of high-speed motion and vehicle vibration on radar echo information is reduced, thereby improving the stability of foreign object detection and the accuracy of target positioning in high-speed scenarios.

[0088] By performing pre-compensation during the signal transmission phase and introducing feedforward vibration prediction compensation during the detection process, compared with existing methods that rely solely on complex motion compensation and feedback correction at the receiving end, the processing link can be shortened, the computational latency reduced, and the real-time performance of foreign object detection and early warning output improved. This method is more suitable for online detection needs under high-speed train operation.

[0089] By combining semantic digital twin technology to model and constrain fixed structures in railway scenarios, not only can the stability of the detection coordinate system be improved, but the system's adaptability to changes in lighting, severe weather, and complex track environments can also be enhanced. Compared to detection schemes that rely solely on visual images, this application can maintain good detection performance and stability even at night, in rainy or foggy conditions, and in complex environments.

[0090] By synergistically integrating cognitive waveform design, semantic scene modeling, and vibration prediction compensation, the system ensures detection accuracy, real-time performance, and environmental adaptability while reducing the need for additional hardware facilities and complex manual maintenance. This simplifies system deployment and improves the system's engineering practicality, economy, and promotional value.

[0091] This application, by employing technologies such as cognitive waveform adaptive adjustment, semantic digital twin geometric constraints, and vibration prediction feedforward compensation, has achieved technological advancements in reducing external system dependence, improving detection accuracy, enhancing real-time performance, and improving environmental adaptability. It can better meet the application needs of foreign object detection and safety early warning in high-speed railways and other rail transit scenarios.

[0092] By adopting the railway foreign object detection method based on cognitive waveform and semantic digital twin provided in this application, real-time detection and accurate positioning of foreign objects around the track can be achieved simply by using train radar installed on the train. Compared with foreign object detection schemes that deploy detection equipment on a large scale along the track, it can also effectively reduce the detection cost of foreign objects.

[0093] See Figure 2 , Figure 2 This is a schematic diagram of the structure of a railway foreign object detection system based on cognitive waveforms and semantic digital twins, provided in one embodiment of this application. Figure 2 As shown, system 200 can be applied to trains equipped with train radar, and system 200 includes: The adjustment module 201 is used to acquire the train's operating status information and adjust the transmission waveform parameters of the train radar based on the operating status information to obtain the adjusted transmission waveform parameters. The transmission acquisition module 202 is used to control the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track according to the adjusted transmission waveform parameters, and to acquire the radar echo information of the surrounding environment of the track. The calculation module 203 is used to call a pre-built semantic digital twin scene model, match the radar echo information with the fixed structure in the semantic digital twin scene model, calculate the pose information of the train radar, and establish a detection coordinate system based on the pose information. The prediction generation module 204 is used to construct a historical phase time series based on the collected historical phase information, input the historical phase time series into the vibration prediction model to predict the phase change trend, obtain the phase prediction result, and generate phase compensation parameters based on the phase prediction result. The fusion acquisition module 205 is used to fuse the operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters to obtain the target description result in the detection coordinate system. The output module 206 is used to identify foreign objects based on the target description results and output the detection results.

[0094] Optionally, the solution module 203 is specifically used for: The radar echo information is matched with the fixed structure in the semantic digital twin scene model by calling a pre-built semantic digital twin scene model to obtain the target fixed structure that is successfully matched. The coordinates of the target fixed structure in the radar coordinate system and the coordinates of the target fixed structure in the semantic digital twin scene coordinate system are used to form a feature point pair; The coordinate rotation matrix and translation vector are solved by minimizing the geometric error of the feature point pairs; Based on the coordinate rotation matrix and the translation vector, the pose information of the train radar is determined; The pose information includes position information and attitude information.

[0095] Optionally, the fusion acquisition module 205 is specifically used for: The running status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are uniformly spatiotemporally registered to obtain the spatiotemporal registration result. Based on the spatiotemporal registration results, the candidate target data is mapped to the detection coordinate system, and the candidate target data is subjected to position correction, data aggregation, and target reconstruction to obtain the corresponding target description results.

[0096] Optionally, the output module 206 is specifically used for: Pre-defined track safety clearance zone; Based on the target description results, the target to be judged is determined to be located within the track safety clearance area; If the number of consecutive occurrences of the target to be determined is greater than a preset number, the target to be determined is identified as a real foreign object target, and a detection result including the real foreign object target is output.

[0097] Optionally, the operating status information includes the train's speed and acceleration information, and the transmitted waveform parameters include the frequency modulation slope and the pulse repetition frequency; The adjustment acquisition module 201 is specifically used for: Based on the reference frequency modulation slope, the velocity information, and the acceleration information, the frequency modulation slope is adjusted to obtain the adjusted frequency modulation slope. Based on the reference pulse repetition frequency and the velocity information, the pulse repetition frequency is adjusted to obtain the adjusted pulse repetition frequency; The adjusted transmitted waveform parameters include the adjusted frequency modulation slope and the adjusted pulse repetition frequency.

[0098] Optionally, the transmission acquisition module 202 is specifically used for: Based on the angle between the line-of-sight direction of the train radar and the direction of movement of the train, as well as the operating wavelength of the train radar, Doppler pre-compensation parameters are generated. The train radar signal to be transmitted is compensated based on the Doppler pre-compensation parameters to obtain a cognitive waveform radar signal. The train radar is controlled to transmit the compensated cognitive waveform radar signal to the surrounding environment of the track.

[0099] The railway foreign object detection system based on cognitive waveform and semantic digital twin provided in this application can achieve the goals outlined in this application. Figure 1 The various processes in the method embodiments, and the ways to achieve the same beneficial effects, will not be repeated here to avoid repetition.

[0100] like Figure 3 As shown, this application also provides a terminal device, including a processor 301 and a memory 302. The memory 302 stores a program or instructions that can run on the processor 301. When the program or instructions are executed by the processor 301, they implement the various steps of the above-described embodiment of the railway foreign object detection method based on cognitive waveform and semantic digital twin, and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0101] It should be noted that the terminal device in this application can be a terminal or other devices besides a terminal. For example, the terminal device can be a mobile phone, tablet computer, laptop computer, etc., and this application does not make any specific limitation.

[0102] This application also provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the various processes of the above-described embodiments of the railway foreign object detection method based on cognitive waveforms and semantic digital twins, and achieve the same technical effect. To avoid repetition, these will not be described again here.

[0103] The processor is the processor in the terminal device described in the above embodiments. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (Read-Only Memory). Only memory (ROM), random access memory (RAM), magnetic disks or optical disks, etc.

[0104] This application also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described embodiments of the railway foreign object detection method based on cognitive waveforms and semantic digital twins, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0105] It should be understood that the chip mentioned in this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0106] This application provides a computer program product stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-described embodiment of the railway foreign object detection method based on cognitive waveforms and semantic digital twins, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

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

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

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

Claims

1. A railway foreign object detection method based on cognitive waveform and semantic digital twinning, characterized in that, Applied to trains equipped with train radar, the method includes: The train's operating status information is acquired, and the transmission waveform parameters of the train radar are adjusted based on the operating status information to obtain the adjusted transmission waveform parameters. According to the adjusted transmission waveform parameters, the train radar is controlled to transmit cognitive waveform radar signals to the surrounding environment of the track and to acquire radar echo information of the surrounding environment of the track. The system calls a pre-built semantic digital twin scene model to match the radar echo information with the fixed structure in the semantic digital twin scene model, calculates the pose information of the train radar, and establishes a detection coordinate system based on the pose information. Historical phase time series are constructed based on the collected historical phase information. The historical phase time series are input into the vibration prediction model to predict the phase change trend, and the phase prediction results are obtained. Phase compensation parameters are generated based on the phase prediction results. The operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are fused to obtain the target description result in the detection coordinate system. Foreign object identification is performed based on the target description results, and the detection results are output. The operating status information includes the train's speed and acceleration information, and the transmitted waveform parameters include the frequency modulation slope and pulse repetition frequency; The adjustment of the train radar's transmission waveform parameters based on the operational status information includes: Based on the reference frequency modulation slope, the velocity information, and the acceleration information, the frequency modulation slope is adjusted to obtain the adjusted frequency modulation slope. Based on the reference pulse repetition frequency and the speed information, the pulse repetition frequency is adjusted to obtain the adjusted pulse repetition frequency; The adjusted transmitted waveform parameters include the adjusted frequency modulation slope and the adjusted pulse repetition frequency; The control of the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track includes: Based on the angle between the line-of-sight direction of the train radar and the direction of movement of the train, as well as the operating wavelength of the train radar, Doppler pre-compensation parameters are generated. The train radar signal to be transmitted is compensated based on the Doppler pre-compensation parameters to obtain a cognitive waveform radar signal. The train radar is controlled to transmit the compensated cognitive waveform radar signal to the surrounding environment of the track.

2. The method of claim 1, wherein, The step of calling a pre-built semantic digital twin scene model, matching the radar echo information with fixed structures in the semantic digital twin scene model, and calculating the pose information of the train radar includes: The radar echo information is matched with the fixed structure in the semantic digital twin scene model by calling a pre-built semantic digital twin scene model to obtain the target fixed structure that is successfully matched. The coordinates of the target fixed structure in the radar coordinate system and the coordinates of the target fixed structure in the semantic digital twin scene coordinate system are used to form a feature point pair; The coordinate rotation matrix and translation vector are solved by minimizing the geometric error of the feature point pairs; Based on the coordinate rotation matrix and the translation vector, the pose information of the train radar is determined; The pose information includes position information and attitude information.

3. The method of claim 1, wherein, The process of fusing the operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters to obtain the target description result in the detection coordinate system includes: The running status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are uniformly spatiotemporally registered to obtain the spatiotemporal registration result. Based on the spatiotemporal registration results, the candidate target data is mapped to the detection coordinate system, and the candidate target data is subjected to position correction, data aggregation, and target reconstruction to obtain the corresponding target description results.

4. The method according to any one of claims 1 to 3, characterized in that, The foreign object identification based on the target description result and the output of the detection result include: Pre-defined track safety clearance zone; Based on the target description results, the target to be judged is determined to be located within the track safety clearance area; If the number of consecutive occurrences of the target to be determined is greater than a preset number, the target to be determined is identified as a real foreign object target, and a detection result including the real foreign object target is output.

5. A railway foreign object detection system based on cognitive waveform and semantic digital twinning, characterized in that, The system, applicable to trains equipped with train radar, includes: An adjustment module is used to acquire the train's operating status information and adjust the transmission waveform parameters of the train radar based on the operating status information to obtain the adjusted transmission waveform parameters. The transmission and acquisition module is used to control the train radar to transmit cognitive waveform radar signals to the surrounding environment of the track according to the adjusted transmission waveform parameters, and to acquire the radar echo information of the surrounding environment of the track. The calculation module is used to call a pre-built semantic digital twin scene model, match the radar echo information with the fixed structure in the semantic digital twin scene model, calculate the pose information of the train radar, and establish a detection coordinate system based on the pose information. The prediction generation module is used to construct a historical phase time series based on the collected historical phase information, input the historical phase time series into the vibration prediction model to predict the phase change trend, obtain the phase prediction result, and generate phase compensation parameters based on the phase prediction result. The fusion acquisition module is used to fuse the operating status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters to obtain the target description result in the detection coordinate system. The output module is used to identify foreign objects based on the target description results and output the detection results; The operating status information includes the train's speed and acceleration information, and the transmitted waveform parameters include the frequency modulation slope and pulse repetition frequency; The acquisition and adjustment module is specifically used for: adjusting the frequency modulation slope based on the reference frequency modulation slope, the velocity information, and the acceleration information to obtain the adjusted frequency modulation slope; and adjusting the pulse repetition frequency based on the reference pulse repetition frequency and the velocity information to obtain the adjusted pulse repetition frequency; wherein the adjusted transmitted waveform parameters include the adjusted frequency modulation slope and the adjusted pulse repetition frequency; The transmission acquisition module is specifically used for: generating Doppler pre-compensation parameters based on the angle between the line-of-sight direction of the train radar and the direction of movement of the train, and the operating wavelength of the train radar; compensating the radar signal to be transmitted by the train radar based on the Doppler pre-compensation parameters to obtain a cognitive waveform radar signal; and controlling the train radar to transmit the compensated cognitive waveform radar signal to the surrounding scene of the track.

6. The system of claim 5, wherein, The solution module is specifically used for: The radar echo information is matched with the fixed structure in the semantic digital twin scene model by calling a pre-built semantic digital twin scene model to obtain the target fixed structure that is successfully matched. The coordinates of the target fixed structure in the radar coordinate system and the coordinates of the target fixed structure in the semantic digital twin scene coordinate system are used to form a feature point pair; The coordinate rotation matrix and translation vector are solved by minimizing the geometric error of the feature point pairs; Based on the coordinate rotation matrix and the translation vector, the pose information of the train radar is determined; The pose information includes position information and attitude information.

7. The system of claim 5, wherein, The fusion acquisition module is specifically used for: The running status information, the candidate target data obtained after preprocessing the radar echo information, the geometric constraint information of the detection coordinate system, and the phase compensation parameters are uniformly spatiotemporally registered to obtain the spatiotemporal registration result. Based on the spatiotemporal registration results, the candidate target data is mapped to the detection coordinate system, and the candidate target data is subjected to position correction, data aggregation, and target reconstruction to obtain the corresponding target description results.

8. A terminal device, comprising: It includes a processor and a memory, wherein the memory stores a program or instructions executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method as described in any one of claims 1 to 4.

Images