Radar anomaly detection method and device, vehicle, storage medium and program product
By collecting and processing vehicle environmental data and radar signal data, and combining them with machine learning models to determine radar anomalies, the problem of low detection accuracy of ultrasonic radar in specific environments has been solved, achieving higher detection accuracy and safety.
Patent Information
- Application Number
- CN202511034394.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-25
- Publication Date
- 2025-11-14
AI Technical Summary
In existing technologies, ultrasonic radar has low detection accuracy in specific conditions such as open flat ground, no wind, and stable weather, resulting in a poor user experience and potentially causing safety hazards.
The system collects environmental data and radar signal data of the vehicle under different environmental factors. Combined with a pre-trained machine learning model, it determines radar anomalies by detecting whether the radar signal change value exceeds the threshold, including signal changes caused by environmental factors and obstacles. Data processing such as filtering, noise reduction and feature extraction are used to dynamically adjust radar operating parameters and improve safety through a three-level progressive alarm mechanism.
It improves the accuracy and reliability of radar detection, can intelligently identify environmental changes, reduce misoperation and safety risks, and enhance user satisfaction and system security.
Smart Images
Figure CN120949202A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of radar detection technology, and in particular to a radar anomaly detection method, device, vehicle, storage medium, and program product. Background Technology
[0002] Ultrasonic radar, as an important component of intelligent vehicle assistance systems, is widely used in parking assistance, obstacle detection, and other fields.
[0003] In related technologies, radar outlier detection can be applied directly to data anomaly detection using methods such as principal component analysis without the need for labeled data or training. For data with time-series correlations, outlier detection can be performed using deep learning-based methods after data labeling.
[0004] However, among related technologies, principal component analysis is sensitive to outliers, while deep learning methods rely on labeled data and time-series correlation data, which limits their application scenarios. In certain specific situations, such as open flat land, windless and stable weather, the detection accuracy is low, resulting in a poor user experience and even potential security risks, which urgently need to be improved. Summary of the Invention
[0005] This application provides a radar anomaly detection method, device, vehicle, storage medium, and program product to solve the problems in related technologies, such as limited application scenarios, low detection accuracy in certain specific situations, such as open flat ground, no wind, and stable weather, resulting in poor user experience and even potential safety hazards.
[0006] The first aspect of this application provides a method for detecting vehicle radar anomalies, comprising the following steps: collecting environmental data and radar signal data corresponding to the vehicle under different environmental factors; inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model to output a first radar signal change value caused by different environmental factors and / or a second radar signal change value caused by different obstacles; detecting whether the first radar signal change value is greater than a first change threshold, and / or whether the second radar signal change value is greater than a second change threshold; and determining that the vehicle radar is abnormal if the first radar signal change value is greater than the first change threshold and / or the second radar signal change value is greater than the second change threshold.
[0007] The above technical solution can collect environmental data and radar signal data corresponding to vehicles under different environmental factors, or use environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Then, if the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This can effectively correlate environmental data with radar data, improve the accuracy and reliability of ultrasonic radar, and intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems.
[0008] Optionally, in one embodiment of this application, before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: collecting training environment data and training radar signal data corresponding to the vehicle under different environmental factors; determining the structural information of the machine learning model based on the training environment data and the training radar signal data, or the training environment data, the training radar signal data, and pre-collected training obstacle data, so as to use the machine learning model to learn the first training radar signal change value of the vehicle radar caused by different environmental factors and / or the second training radar signal change value of the vehicle radar caused by different obstacles; and training the machine learning model based on the structural information using the training environment data and the training radar signal data, or the training environment data, the training radar signal data, and the pre-collected training obstacle data, until a preset training condition is met to obtain a trained machine learning model.
[0009] The above technical solution can utilize training environment data, training radar signal data, and pre-collected training obstacle data to determine the structural information of the machine learning model, and then train the machine learning model until certain training conditions are met, thereby obtaining the machine learning model. Multi-environment data coverage improves environmental adaptability, enhances model robustness, and thus improves obstacle detection accuracy and reliability.
[0010] Optionally, in one embodiment of this application, before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: acquiring the operating parameters of the vehicle radar; determining whether the operating parameters meet preset operating conditions; if the operating parameters do not meet the preset operating conditions, re-collecting the radar signal data until re-collected radar signal data is obtained when the operating parameters meet the preset operating conditions, and allowing the re-collected radar signal data to be input into the pre-trained machine learning model; if the operating parameters meet the preset operating conditions, allowing the radar signal data to be input into the pre-trained machine learning model.
[0011] The above technical solution can determine whether the vehicle radar's operating parameters meet certain operating conditions before the data is input into the pre-trained machine learning model. If the conditions are not met, the data can be re-acquired, thereby avoiding signal noise, distortion, or data loss. This improves the quality of input data while ensuring that the model can be trained and inferred based on reliable data, thus improving model performance, reducing maintenance costs, and enhancing user experience.
[0012] Optionally, in one embodiment of this application, the method further includes: determining the anomaly level of the vehicle radar based on the first radar signal change value and / or the second radar signal change value; when the anomaly level is Level 1, sending a voice broadcast to the user to remind the user that the vehicle radar has malfunctioned; when the anomaly level is Level 2, issuing an audible and visual alarm to attract the user's attention; and when the anomaly level is Level 3, issuing an audible and visual alarm while simultaneously issuing a vehicle body vibration alarm to ensure the user's timely perception.
[0013] The above technical solution can classify the abnormality level of the vehicle radar into three levels: Level 1 is voice broadcast, Level 2 is sound and light alarm, and Level 3 is sound, light and vehicle vibration. Through the three-level progressive alarm mechanism, multiple safety protection barriers are built to achieve precise defense and improve vehicle safety and user experience.
[0014] Optionally, in one embodiment of this application, the collection of environmental data and radar signal data corresponding to the vehicle under different environmental factors includes: performing at least one of filtering, denoising, and feature extraction on the environmental data and the radar signal data respectively to obtain corresponding processed environmental data and processed radar signal data.
[0015] The above technical solutions can perform at least one of the following operations during the data acquisition stage: filtering, denoising, and feature extraction. This enables noise suppression, signal purification, data correction, and outlier removal, thereby fundamentally improving data quality, increasing data transmission efficiency, and saving time.
[0016] A second aspect of this application provides a vehicle radar anomaly detection device, comprising: a first acquisition module for acquiring environmental data and radar signal data corresponding to the vehicle under different environmental factors; an output module for inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-acquired obstacle data, into a pre-trained machine learning model to output a first radar signal change value caused by different environmental factors and / or a second radar signal change value caused by different obstacles; a detection module for detecting whether the first radar signal change value is greater than a first change threshold, and / or whether the second radar signal change value is greater than a second change threshold; and a determination module for determining that the vehicle radar is abnormal when the first radar signal change value is greater than the first change threshold, and / or the second radar signal change value is greater than the second change threshold.
[0017] The above technical solution can collect environmental data and radar signal data corresponding to vehicles under different environmental factors, or use environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Then, if the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This can effectively correlate environmental data with radar data, improve the accuracy and reliability of ultrasonic radar, and intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems.
[0018] Optionally, in one embodiment of this application, it further includes: a second acquisition module, configured to acquire training environment data and training radar signal data corresponding to the vehicle under different environmental factors before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data and pre-acquired obstacle data, into the pre-trained machine learning model; a first determination module, configured to determine the structural information of the machine learning model based on the training environment data and the training radar signal data, or the training environment data, the training radar signal data and pre-acquired training obstacle data, so as to use the machine learning model to learn the first training radar signal change value of the vehicle radar caused by different environmental factors and / or the second training radar signal change value of the vehicle radar caused by different obstacles; and a training module, configured to train the machine learning model based on the structural information, using the training environment data and the training radar signal data, or the training environment data, the training radar signal data and the pre-acquired training obstacle data, until a preset training condition is met, so as to obtain the trained machine learning model.
[0019] The above technical solution can utilize training environment data, training radar signal data, and pre-collected training obstacle data to determine the structural information of the machine learning model, and then train the machine learning model until certain training conditions are met, thereby obtaining the machine learning model. Multi-environment data coverage improves environmental adaptability, enhances model robustness, and thus improves obstacle detection accuracy and reliability.
[0020] Optionally, in one embodiment of this application, it further includes: an acquisition module, configured to acquire the operating parameters of the vehicle radar before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model; a judgment module, configured to judge whether the operating parameters meet preset operating conditions; a first input module, configured to re-collect the radar signal data when the operating parameters do not meet the preset operating conditions, until re-collected radar signal data is obtained when the operating parameters meet the preset operating conditions, and allow the re-collected radar signal data to be input into the pre-trained machine learning model; and a second input module, configured to allow the radar signal data to be input into the pre-trained machine learning model when the operating parameters meet the preset operating conditions.
[0021] The above technical solution can determine whether the vehicle radar's operating parameters meet certain operating conditions before the data is input into the pre-trained machine learning model. If the conditions are not met, the data can be re-acquired, thereby avoiding signal noise, distortion, or data loss. This improves the quality of input data while ensuring that the model can be trained and inferred based on reliable data, thus improving model performance, reducing maintenance costs, and enhancing user experience.
[0022] Optionally, in one embodiment of this application, it further includes: a second determining module, configured to determine the anomaly level of the vehicle radar based on the first radar signal change value and / or the second radar signal change value; a first alarm module, configured to send a voice broadcast to the user when the anomaly level is Level 1, to remind the user that the vehicle radar has malfunctioned; a second alarm module, configured to issue an audible and visual alarm when the anomaly level is Level 2, to attract the user's attention; and a third alarm module, configured to issue an audible and visual alarm and a vehicle body vibration alarm simultaneously when the anomaly level is Level 3, to ensure the user's timely perception.
[0023] The above technical solution can classify the abnormality level of the vehicle radar into three levels: Level 1 is voice broadcast, Level 2 is sound and light alarm, and Level 3 is sound, light and vehicle vibration. Through the three-level progressive alarm mechanism, multiple safety protection barriers are built to achieve precise defense and improve vehicle safety and user experience.
[0024] Optionally, in one embodiment of this application, the first acquisition module includes: a processing unit, configured to perform at least one of filtering, denoising, and feature extraction on the environmental data and the radar signal data respectively, to obtain corresponding processed environmental data and processed radar signal data.
[0025] The above technical solutions can perform at least one of the following operations during the data acquisition stage: filtering, denoising, and feature extraction. This enables noise suppression, signal purification, data correction, and outlier removal, thereby fundamentally improving data quality, increasing data transmission efficiency, and saving time.
[0026] A third aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle radar anomaly detection method as described in the above embodiments.
[0027] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle radar anomaly detection method described above.
[0028] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the vehicle radar anomaly detection method described above.
[0029] This application embodiment can collect environmental data and radar signal data of the vehicle under different environmental factors, or use environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Then, when the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This effectively correlates environmental data with radar data, improving the accuracy and reliability of ultrasonic radar. It can intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems. Therefore, it solves the problems in related technologies, such as limited application scenarios, low detection accuracy in certain specific situations, such as open flat ground, windless and stable weather, leading to poor user experience and even potential safety hazards.
[0030] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0031] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:
[0032] Figure 1 This is a flowchart of a vehicle radar anomaly detection method according to an embodiment of this application;
[0033] Figure 2 This is a block diagram of a vehicle radar anomaly detection device provided according to an embodiment of this application;
[0034] Figure 3 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application.
[0035] Figure label:
[0036] 10-Vehicle radar anomaly detection device; 100-First acquisition module, 200-Output module, 300-Detection module, 400-Judgment module; 301-Memory, 302-Processor, 303-Communication interface. Detailed Implementation
[0037] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0038] The radar anomaly detection method, apparatus, vehicle, storage medium, and program product of this application are described below with reference to the accompanying drawings. Addressing the limitations of application scenarios mentioned in the background art, such as low detection accuracy in certain specific situations (e.g., open flat ground, windless conditions, and stable weather), leading to a poor user experience and potentially causing safety hazards, this application provides a vehicle radar anomaly detection method. This method collects environmental data and radar signal data corresponding to different environmental factors of the vehicle, or utilizes environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain a first radar signal change value caused by different environmental factors and / or a second radar signal change value caused by different obstacles. Then, if the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This method effectively correlates environmental data with radar data, improving the accuracy and reliability of ultrasonic radar. It can intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems. This solves the problem that the application scenarios of related technologies are limited, and that the detection accuracy is low in certain specific situations, such as open flat ground, windless and stable weather, resulting in a poor user experience and even potential safety hazards.
[0039] Specifically, Figure 1 This is a flowchart of a vehicle radar anomaly detection method provided according to an embodiment of this application.
[0040] like Figure 1 As shown, the vehicle radar anomaly detection method includes the following steps:
[0041] In step S101, environmental data and radar signal data of the vehicle under different environmental factors are collected.
[0042] It is understood that the data collected in the embodiments of this application may include, but is not limited to, environmental data and radar signal data. Environmental data may include, but is not limited to, data on various environmental factors such as ambient temperature, ambient humidity, wind speed, and ambient air pressure; this application does not impose specific limitations.
[0043] Furthermore, embodiments of this application may collect ambient temperature data via a temperature sensor, ambient humidity data via a humidity sensor, wind speed data via a wind speed sensor, and ambient air pressure data via a pressure sensor, etc., without any specific limitations; or may collect radar signal data via ultrasonic radar, etc., without any specific limitations.
[0044] In addition, it should be noted that the installation positions of temperature sensors, humidity sensors, wind speed sensors, air pressure sensors, ultrasonic radars, etc., can be set by those skilled in the art according to the actual situation, and this application does not impose specific restrictions.
[0045] As one possible approach, embodiments of this application can collect environmental data and radar signal data of the vehicle under different environmental factors.
[0046] For example, in the embodiments of this application, the vehicle is equipped with a variety of environmental sensors, such as temperature sensors, humidity sensors, wind speed sensors, and air pressure sensors, to collect environmental data on radar normal operation and radar abnormal alarm under open, sunny, windless, and suitable temperature conditions; and to collect radar signal data on radar normal operation and radar abnormal alarm under open, sunny, windless, and suitable temperature conditions.
[0047] In addition, embodiments of this application can also collect obstacle data and radar signal data when the radar encounters an obstacle and alarms normally under clear, windless, and suitable temperature conditions.
[0048] Optionally, in one embodiment of this application, collecting environmental data and radar signal data of the vehicle under different environmental factors includes: performing at least one of filtering, denoising, and feature extraction on the environmental data and radar signal data respectively to obtain the corresponding processed environmental data and processed radar signal data.
[0049] It is understood that the embodiments of this application can use digital filters such as low-pass filters, high-pass filters or band-pass filters to filter and remove noise at a specific frequency; noise can be removed by wavelet transform, Kalman filtering or machine learning methods; feature extraction is to extract statistical features, time-frequency features, etc. from the data, and this application does not impose specific limitations.
[0050] In some embodiments, the present application can filter, denoise, and extract features from environmental data to obtain corresponding processed environmental data.
[0051] In some embodiments, the radar signal data can be filtered and denoised to obtain the corresponding processed radar signal data.
[0052] Optionally, in one embodiment of this application, before inputting environmental data and radar signal data, or environmental data, radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: acquiring the operating parameters of the vehicle radar; determining whether the operating parameters meet preset operating conditions; if the operating parameters do not meet the preset operating conditions, re-collecting radar signal data until re-collected radar signal data is obtained when the operating parameters meet the preset operating conditions, and allowing the re-collected radar signal data to be input into the pre-trained machine learning model; if the operating parameters meet the preset operating conditions, allowing the radar signal data to be input into the pre-trained machine learning model.
[0053] It is understood that, in the embodiments of this application, the operating parameters of the vehicle radar may include, but are not limited to, transmission frequency, power and receiving sensitivity, etc., and this application does not impose specific limitations.
[0054] In some embodiments, before inputting environmental data and radar signal data, or environmental data, radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the present application embodiments may also obtain the operating parameters of the vehicle radar and determine whether the operating parameters meet certain operating conditions. If not, the radar signal data is re-collected until the re-collected radar signal data when the operating parameters meet certain operating conditions is obtained, and the re-collected radar signal data is allowed to be input into the pre-trained machine learning model.
[0055] For example, in this application embodiment, the certain operating condition of the vehicle radar transmission frequency is 10-15dBm. If the actual transmission frequency is 8dBm, it is determined that the operating parameters of the vehicle radar do not meet the certain operating conditions. At this time, this application embodiment can re-acquire radar signal data until the vehicle radar transmission frequency is 10-15dBm, and allow the corresponding radar signal data to be input into the pre-trained machine learning model.
[0056] In some embodiments of this application, when the operating parameters meet certain operating conditions, radar signal data can be input into a pre-trained machine learning model.
[0057] Certain working conditions may be set by those skilled in the art according to actual circumstances, and this application does not impose specific restrictions.
[0058] The embodiments of this application can dynamically adjust the operating parameters of the ultrasonic radar, optimize radar performance through adaptive adjustment, and avoid false alarms.
[0059] Optionally, in one embodiment of this application, before inputting environmental data and radar signal data, or environmental data, radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: collecting training environment data and training radar signal data corresponding to the vehicle under different environmental factors; determining the structural information of the machine learning model based on the training environment data and training radar signal data, or training environment data, training radar signal data, and pre-collected training obstacle data, so as to use the machine learning model to learn the first training radar signal change value caused by different environmental factors and / or the second training radar signal change value caused by different obstacles; and training the machine learning model based on the structural information using the training environment data and training radar signal data, or training environment data, training radar signal data, and pre-collected training obstacle data, until the preset training conditions are met, so as to obtain the trained machine learning model.
[0060] It is understood that, in the embodiments of this application, the machine learning model may include, but is not limited to, random forest, support vector machine or deep neural network, etc., and this application does not impose specific limitations.
[0061] In some embodiments, this application can collect training environment data and training radar signal data under different environmental factors to determine the structural information of the machine learning model. The machine learning model is then trained using the training environment data and training radar signal data until certain training conditions are met, resulting in a trained machine learning model. These training conditions can be set by those skilled in the art according to actual circumstances, and this application does not impose specific limitations.
[0062] For example, in this application embodiment, the structural information of the machine learning model can be determined as a support vector machine based on training environment data and training radar signal data. Then, the support vector machine can be trained using the training environment data and training radar signal data to obtain a support vector machine that meets certain training conditions, such as a training accuracy of 90%.
[0063] In some embodiments, the present application embodiments can determine the structural information of the machine learning model based on training environment data, training radar signal data and pre-collected training obstacle data, and train the machine learning model until certain training conditions are met, thereby obtaining the trained machine learning model.
[0064] Additionally, it should be noted that in the embodiments of this application, the machine learning model can learn the complex relationship between environmental factors and ultrasonic radar performance, and can learn to predict changes in radar signals under specific environmental conditions through training data, thereby identifying situations that may lead to false alarms.
[0065] In step S102, environmental data and radar signal data, or environmental data, radar signal data and pre-collected obstacle data, are input into a pre-trained machine learning model to output the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles.
[0066] In some embodiments, the present application embodiments can input environmental data and radar signal data into a pre-trained machine learning model, and then output the first radar signal change value caused by different environmental factors of the vehicle radar.
[0067] For example, in this application embodiment, current environmental data and radar signal data can be input into a pre-trained machine learning model. Through correlation analysis, radar signal changes caused by environmental factors can be identified and distinguished, thereby obtaining the corresponding first radar signal change value.
[0068] In some embodiments, the present application embodiments can input environmental data, radar signal data, and pre-collected obstacle data into a pre-trained machine learning model, and then output the first radar signal change value of the vehicle radar caused by different environmental factors and the second radar signal change value of the vehicle radar caused by different obstacles.
[0069] For example, in this application embodiment, current environmental data, radar signal data, and pre-collected obstacle data can be input into a pre-trained machine learning model. Through correlation analysis, the changes in radar signals caused by environmental factors can be identified and distinguished, that is, the corresponding first radar signal change value and the changes in radar signals caused by real obstacles can be obtained, that is, the corresponding second radar signal change value.
[0070] The embodiments of this application can utilize a pre-trained machine learning model to identify and distinguish between radar signal changes caused by environmental factors and radar signal changes caused by real obstacles, predict and identify situations that may lead to false alarms in advance, and identify potential abnormal alarm patterns.
[0071] In step S103, it is detected whether the change value of the first radar signal is greater than the first change threshold, and / or whether the change value of the second radar signal is greater than the second change threshold.
[0072] In some embodiments, the present application can detect whether the change value of the first radar signal is greater than a first change threshold.
[0073] For example, in the case of changes in environmental factors, such as an increase in ambient temperature from 10°C to 30°C, the corresponding first change threshold can be set to 30%. In this case, the embodiment of the present application can detect whether the change value of the first radar signal is greater than 30%.
[0074] In some embodiments, the present application can detect whether the change value of the first radar signal is greater than a first change threshold and whether the change value of the second radar signal is greater than a second change threshold.
[0075] For example, in the embodiment of this application, when environmental factors change, such as the ambient temperature rising from 10°C to 30°C, the corresponding first change threshold can be set to 30%, and if the pre-collected obstacle data includes an adult male, the corresponding second change threshold can be set to 20%. In this case, the embodiment of this application can detect whether the change value of the first radar signal is greater than 30% and whether the change value of the second radar signal is greater than 20%. The second change threshold can be set according to the number, type, size, etc. of the obstacles, and this application does not impose specific limitations.
[0076] The first and second change thresholds can be set by those skilled in the art according to the actual situation, and this application does not impose any specific restrictions.
[0077] In step S104, if the change value of the first radar signal is greater than the first change threshold, and / or the change value of the second radar signal is greater than the second change threshold, the vehicle radar is determined to be abnormal.
[0078] In some embodiments, the present application can determine that the vehicle radar is abnormal when there are no obstacles and the change value of the first radar signal is greater than the first change threshold.
[0079] For example, in an embodiment of this application, when the ambient temperature rises from 10°C to 30°C and the change value of the first radar signal is greater than 30%, the vehicle radar is determined to be abnormal.
[0080] In some embodiments, the present application can determine that the vehicle radar is abnormal when the change value of the first radar signal is greater than a first change threshold and the change value of the second radar signal is greater than a second change threshold in the presence of an obstacle.
[0081] For example, in this application embodiment, when the ambient temperature rises from 10°C to 30°C, the change value of the first radar signal is greater than 30%, the pre-collected obstacle data contains an adult male, and the change value of the second radar signal is greater than 20%, the vehicle radar is determined to be abnormal.
[0082] Optionally, in one embodiment of this application, the method further includes: determining the anomaly level of the vehicle radar based on the change value of the first radar signal and / or the change value of the second radar signal; sending a voice broadcast to the user when the anomaly level is Level 1 to remind the user that the vehicle radar has malfunctioned; issuing an audible and visual alarm when the anomaly level is Level 2 to attract the user's attention; and issuing a vehicle vibration alarm while issuing an audible and visual alarm when the anomaly level is Level 3 to ensure the user's timely perception.
[0083] In actual implementation, the embodiments of this application can classify vehicle radar malfunctions into three levels when they occur: Level 1 is voice broadcast, Level 2 is audible and visual alarm, and Level 3 is audible and visual alarm and vehicle vibration alarm. The specific settings can be configured by those skilled in the art according to the actual situation, and this application does not impose any specific limitations.
[0084] For example, in this embodiment of the application, when the ambient temperature rises from 10°C to 30°C and the first radar signal changes by 40%, the abnormality level of the vehicle radar is determined to be Level 1, and a voice broadcast is sent to the user, the content of which may be: "The vehicle radar has a slight abnormality, please check it in time" to remind the user that the vehicle radar has an abnormality.
[0085] In some embodiments, when the ambient temperature rises from 10°C to 30°C and the first radar signal changes by 55%, the abnormality level of the vehicle radar is determined to be Level 2, and an audible and visual alarm is triggered, that is, while broadcasting a voice message, the vehicle interior lights are turned on to attract the user's attention.
[0086] In some embodiments, this application embodiment can determine the abnormality level of the vehicle radar as level three when the ambient temperature rises from 10°C to 30°C, the first radar signal change value is 65%, the pre-collected obstacle data contains an adult male, and the second radar signal change value is 40%. A voice broadcast can then be sent to the user, stating: "The vehicle radar has a slight abnormality; please check it promptly." Simultaneously, the vehicle interior lights are turned on, and a vehicle vibration alarm is activated to ensure timely user awareness.
[0087] The vehicle radar anomaly detection method proposed in this application will be described below with reference to a specific embodiment.
[0088] Example 1:
[0089] This application embodiment can collect environmental data on radar normal operation and radar abnormal alarm under open, sunny, windless, and suitable temperature conditions; collect radar signal data on radar normal operation and radar abnormal alarm under open, sunny, windless, and suitable temperature conditions; and collect obstacle data and radar signal data on radar normal alarm when encountering obstacles under sunny, windless, and suitable temperature conditions.
[0090] Furthermore, in this embodiment, environmental data, radar signal data, and pre-collected obstacle data can be input into a pre-trained machine learning model to output a first radar signal change value caused by different environmental factors and a second radar signal change value caused by different obstacles.
[0091] Furthermore, it detects whether the change in the first radar signal is greater than 30% when environmental factors change, such as the ambient temperature rising from 10℃ to 30℃, and whether the change in the second radar signal is greater than 20% when the pre-collected obstacle data contains an adult male. If the change in the first radar signal is greater than 30% and the change in the second radar signal is greater than 20%, the vehicle radar is determined to be abnormal.
[0092] Furthermore, in this embodiment of the application, when the change value of the first radar signal is 65% and the change value of the second radar signal is 40%, the abnormality level of the vehicle radar can be determined to be level three, and a voice broadcast can be sent to the user, the content of which can be: "The vehicle radar has a slight abnormality, please check it in time." At the same time, the vehicle interior lights are turned on and the vehicle body vibration alarm is activated to ensure that the user can perceive it in time.
[0093] The vehicle radar anomaly detection method proposed in this application can collect environmental data and radar signal data corresponding to different environmental factors of the vehicle, or use environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Then, if the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This method effectively correlates environmental data with radar data, improving the accuracy and reliability of ultrasonic radar. It can intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems. Therefore, it solves the problems in related technologies, such as limited application scenarios, low detection accuracy in certain specific situations, such as open flat ground, windless and stable weather, leading to poor user experience and even potential safety hazards.
[0094] Next, the vehicle radar anomaly detection device according to an embodiment of this application is described with reference to the accompanying drawings.
[0095] Figure 2 This is a block diagram of a vehicle radar anomaly detection device provided according to an embodiment of this application.
[0096] like Figure 2 As shown, the vehicle radar anomaly detection device 10 includes: a first acquisition module 100, an output module 200, a detection module 300, and a judgment module 400.
[0097] The first acquisition module 100 is used to acquire environmental data and radar signal data of the vehicle under different environmental factors.
[0098] The output module 200 is used to input environmental data and radar signal data, or environmental data, radar signal data and pre-collected obstacle data, into a pre-trained machine learning model to output the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles.
[0099] The detection module 300 is used to detect whether the change value of the first radar signal is greater than a first change threshold, and / or whether the change value of the second radar signal is greater than a second change threshold.
[0100] The determination module 400 is used to determine that the vehicle radar is abnormal when the change value of the first radar signal is greater than the first change threshold and / or the change value of the second radar signal is greater than the second change threshold.
[0101] Optionally, in one embodiment of this application, it further includes: a second acquisition module, a first determination module, and a training module.
[0102] The second acquisition module is used to acquire training environment data and training radar signal data corresponding to the vehicle under different environmental factors before inputting environmental data and radar signal data, or environmental data, radar signal data and pre-acquired obstacle data, into the pre-trained machine learning model.
[0103] The first determining module is used to determine the structural information of the machine learning model based on training environment data and training radar signal data, or training environment data, training radar signal data and pre-collected training obstacle data, so as to use the machine learning model to learn the first training radar signal change value caused by different environmental factors and / or the second training radar signal change value caused by different obstacles.
[0104] The training module is used to train a machine learning model based on structural information, using training environment data and training radar signal data, or training environment data, training radar signal data and pre-collected training obstacle data, until the preset training conditions are met, so as to obtain the trained machine learning model.
[0105] Optionally, in one embodiment of this application, it further includes: an acquisition module, a judgment module, a first input module, and a second input module.
[0106] The acquisition module is used to acquire the operating parameters of the vehicle radar before inputting environmental data and radar signal data, or environmental data, radar signal data and pre-collected obstacle data, into a pre-trained machine learning model.
[0107] The judgment module is used to determine whether the working parameters meet the preset working conditions.
[0108] The first input module is used to reacquire radar signal data when the operating parameters do not meet the preset operating conditions, until the reacquired radar signal data when the operating parameters meet the preset operating conditions is obtained, and allows the reacquired radar signal data to be input into a pre-trained machine learning model.
[0109] The second input module is used to allow radar signal data to be input into a pre-trained machine learning model when the operating parameters meet the preset operating conditions.
[0110] Optionally, in one embodiment of this application, it further includes: a second determining module, a first alarm module, a second alarm module, and a third alarm module.
[0111] The second determining module is used to determine the anomaly level of the vehicle radar based on the change value of the first radar signal and / or the change value of the second radar signal.
[0112] The first alarm module is used to send a voice broadcast to the user when the anomaly level is Level 1, to remind the user that the vehicle radar has malfunctioned.
[0113] The second alarm module is used to trigger an audible and visual alarm when the anomaly level is level two, in order to attract the user's attention.
[0114] The third alarm module is used to issue an audible and visual alarm and a vehicle vibration alarm when the anomaly level is three, so as to ensure that the user can perceive it in time.
[0115] Optionally, in one embodiment of this application, the first acquisition module 100 includes a processing unit.
[0116] The processing unit is used to perform at least one of filtering, denoising, and feature extraction on the environmental data and radar signal data respectively, so as to obtain the corresponding processed environmental data and processed radar signal data.
[0117] It should be noted that the foregoing explanation of the vehicle radar anomaly detection method embodiment also applies to the vehicle radar anomaly detection device of this embodiment, and will not be repeated here.
[0118] The vehicle radar anomaly detection device proposed in this application can collect environmental data and radar signal data corresponding to vehicles under different environmental factors, or use environmental data, radar signal data, and pre-collected obstacle data, combined with a pre-trained machine learning model, to obtain the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Then, if the first radar signal change value is greater than a first change threshold, and / or the second radar signal change value is greater than a second change threshold, the vehicle radar is determined to be abnormal. This device effectively correlates environmental data with radar data, improving the accuracy and reliability of ultrasonic radar. It can intelligently identify and adapt to environmental changes, providing more accurate information for users and automatic control systems, thereby reducing misoperation and potential safety risks, improving user satisfaction, and enhancing the safety of intelligent vehicle assistance systems. Therefore, it solves the problems in related technologies, such as limited application scenarios, low detection accuracy in certain specific situations, such as open flat ground, windless conditions, and stable weather, leading to poor user experience and even potential safety hazards.
[0119] Figure 3 This is a schematic diagram of the structure of a vehicle according to an embodiment of this application. The vehicle may include:
[0120] The memory 301, the processor 302, and the computer program stored on the memory 301 and capable of running on the processor 302.
[0121] When the processor 302 executes the program, it implements the vehicle radar anomaly detection method provided in the above embodiments.
[0122] Furthermore, the vehicle also includes:
[0123] Communication interface 303 is used for communication between memory 301 and processor 302.
[0124] The memory 301 is used to store computer programs that can run on the processor 302.
[0125] The memory 301 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.
[0126] If the memory 301, processor 302, and communication interface 303 are implemented independently, then the communication interface 303, memory 301, and processor 302 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0127] Optionally, in a specific implementation, if the memory 301, processor 302, and communication interface 303 are integrated on a single chip, then the memory 301, processor 302, and communication interface 303 can communicate with each other through an internal interface.
[0128] Processor 302 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.
[0129] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described vehicle radar anomaly detection method.
[0130] This application also provides a computer program product, including a computer program that, when executed, implements the above-described vehicle radar anomaly detection method.
[0131] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0132] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0133] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0135] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0136] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
[0137] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0138] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A method for detecting anomalies in vehicle radar, characterized in that, Includes the following steps: Collect environmental data and radar signal data of the vehicle under different environmental factors; The environmental data and the radar signal data, or the environmental data, the radar signal data and the pre-collected obstacle data, are input into a pre-trained machine learning model to output the first radar signal change value caused by different environmental factors and / or the second radar signal change value caused by different obstacles. Detect whether the change value of the first radar signal is greater than a first change threshold, and / or whether the change value of the second radar signal is greater than a second change threshold; If the change value of the first radar signal is greater than the first change threshold, and / or the change value of the second radar signal is greater than the second change threshold, the vehicle radar is determined to be abnormal.
2. The method according to claim 1, characterized in that, Before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: Collect training environment data and training radar signal data of the vehicle under different environmental factors; Based on the training environment data and the training radar signal data, or the training environment data, the training radar signal data and the pre-collected training obstacle data, the structural information of the machine learning model is determined so as to use the machine learning model to learn the first training radar signal change value caused by different environmental factors and / or the second training radar signal change value caused by different obstacles. Based on the structural information, the machine learning model is trained using the training environment data and the training radar signal data, or the training environment data, the training radar signal data, and the pre-collected training obstacle data, until the preset training conditions are met, so as to obtain the trained machine learning model.
3. The method according to claim 1, characterized in that, Before inputting the environmental data and the radar signal data, or the environmental data, the radar signal data, and pre-collected obstacle data, into a pre-trained machine learning model, the method further includes: Obtain the operating parameters of the vehicle radar; Determine whether the operating parameters meet the preset operating conditions; If the operating parameters do not meet the preset operating conditions, the radar signal data is re-acquired until the re-acquired radar signal data is obtained when the operating parameters meet the preset operating conditions, and the re-acquired radar signal data is allowed to be input into the pre-trained machine learning model. If the operating parameters meet the preset operating conditions, the radar signal data can be input into a pre-trained machine learning model.
4. The method according to claim 1, characterized in that, Also includes: The anomaly level of the vehicle radar is determined based on the change value of the first radar signal and / or the change value of the second radar signal; When the anomaly level is Level 1, a voice broadcast is sent to the user to remind the user that the vehicle radar has malfunctioned; When the anomaly level is level two, an audible and visual alarm will be triggered to attract the user's attention; When the anomaly level is level three, an audible and visual alarm will be triggered, along with a vehicle vibration alarm, to ensure that the user can perceive the anomaly in a timely manner.
5. The method according to claim 1, characterized in that, The environmental data and radar signal data of the vehicle under different environmental factors collected include: The environmental data and the radar signal data are subjected to at least one of filtering, denoising, and feature extraction to obtain the corresponding processed environmental data and processed radar signal data.
6. A vehicle radar anomaly detection device, characterized in that, include: The first acquisition module is used to collect environmental data and radar signal data of the vehicle under different environmental factors. The output module is used to input the environmental data and the radar signal data, or the environmental data, the radar signal data and the pre-collected obstacle data, into a pre-trained machine learning model to output the first radar signal change value of the vehicle radar caused by different environmental factors and / or the second radar signal change value of the vehicle radar caused by different obstacles. The detection module is used to detect whether the change value of the first radar signal is greater than a first change threshold, and / or whether the change value of the second radar signal is greater than a second change threshold; The determination module is used to determine that the vehicle radar is abnormal when the change value of the first radar signal is greater than the first change threshold, and / or the change value of the second radar signal is greater than the second change threshold.
7. The apparatus according to claim 6, characterized in that, Also includes: The second acquisition module is used to acquire training environment data and training radar signal data of the vehicle under different environmental factors before inputting the environmental data and radar signal data, or the environmental data, radar signal data and pre-acquired obstacle data, into the pre-trained machine learning model. The first determining module is used to determine the structural information of the machine learning model based on the training environment data and the training radar signal data, or the training environment data, the training radar signal data and the pre-collected training obstacle data, so as to use the machine learning model to learn the first training radar signal change value caused by different environmental factors and / or the second training radar signal change value caused by different obstacles. The training module is used to train the machine learning model based on the structural information, using the training environment data and the training radar signal data, or the training environment data, the training radar signal data and the pre-collected training obstacle data, until the preset training conditions are met, so as to obtain the trained machine learning model.
8. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the vehicle radar anomaly detection method as described in any one of claims 1-5.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the vehicle radar anomaly detection method as described in any one of claims 1-5.
10. A computer program product, characterized in that, Includes a computer program, which, when executed, is used to implement the vehicle radar anomaly detection method as described in any one of claims 1-5.