Long distance blind area detection sensor real-time monitoring system for complex environment

Abstract

This invention discloses a real-time monitoring system for long-distance blind zone detection sensors in complex environments, belonging to the field of blind zone detection technology. This invention solves the problem of low data transmission efficiency caused by the single communication link in existing systems. It comprehensively and accurately detects obstacles in long-distance blind zones through multi-sensor fusion technology, ensuring data richness and reliability. Furthermore, it dynamically divides the dataset according to the sensor acquisition time interval and sets corresponding data transmission time intervals for datasets of different priorities, enabling more efficient data processing and transmission, ensuring the real-time nature of high-priority data and the stability of low-priority data. By real-time monitoring of the gateway communication distance parameter of the communication link and dynamically switching the communication link when the network distance threshold is exceeded, the risk of data transmission interruption due to communication link failure or instability is reduced, thus providing more reliable technical support for long-distance blind zone monitoring in complex environments.

Description

Technical Field

[0001] This invention relates to the field of blind zone detection technology, specifically to a real-time monitoring system for long-distance blind zone detection sensors in complex environments. Background Technology

[0002] In modern complex environments, such as autonomous driving, intelligent transportation, industrial automation, and security monitoring, monitoring and early warning of long-range blind spots are crucial for ensuring safety and improving efficiency. With technological advancements, traditional monitoring systems can no longer meet the growing demands. Existing monitoring systems primarily rely on single sensors or simple sensor combinations, which have numerous limitations in data acquisition, transmission, and analysis, such as:

[0003] Traditional monitoring systems typically have only one communication link. If this link fails, data transmission will be interrupted, severely impacting the system's real-time performance and reliability. Furthermore, the corresponding data transmission time interval is fixed and cannot be dynamically adjusted according to the importance and real-time requirements of the data, resulting in low data transmission efficiency and making it difficult to achieve real-time monitoring of blind spots in complex environments.

[0004] Therefore, to address the shortcomings of existing systems, we propose a real-time monitoring system for long-range blind zone detection sensors in complex environments. Summary of the Invention

[0005] The purpose of this invention is to provide a real-time monitoring system for long-distance blind zone detection sensors in complex environments. Through multi-sensor fusion technology, it comprehensively and accurately detects obstacles in long-distance blind zones, ensuring data richness and reliability. Furthermore, it dynamically divides the dataset according to the sensor acquisition time interval and sets corresponding data transmission time intervals for datasets of different priorities, enabling more efficient data processing and transmission, ensuring the real-time nature of high-priority data and the stability of low-priority data. By real-time monitoring of the gateway communication distance parameters of the communication link and dynamically switching the communication link when the network distance threshold is exceeded, it reduces the risk of data transmission interruption due to communication link failure or instability. Therefore, it provides more reliable technical support for long-distance blind zone monitoring in complex environments and solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A real-time monitoring system for long-distance blind zone detection sensors in complex environments, the system comprising: a data acquisition unit, a data transmission unit, and a data analysis platform, wherein the data transmission unit comprises: a transmission interval module, a link setting module, and a communication transmission module;

[0008] The data acquisition unit is configured to detect obstacles in long-distance blind spots using a multi-sensor fusion method. The multi-sensor includes: ultrasonic sensor, millimeter-wave radar, lidar, and camera.

[0009] The transmission interval module is configured to set a corresponding data transmission interval based on the acquisition interval of various sensors in the data acquisition unit;

[0010] The link setting module is configured to pre-set two or more communication links and preset a normal network distance parameter between the data acquisition unit and the data analysis platform as the network distance threshold for selecting any one of the communication links.

[0011] The communication transmission module is configured to select a normal communication link and transmit the data collected by various sensors to the data analysis platform for data analysis and prediction according to a preset data transmission time interval.

[0012] The data analysis platform is configured to use a neural network model to extract features from the collected data, thereby identifying the type and location of obstacles.

[0013] Furthermore, the transmission interval module includes:

[0014] The interval acquisition module is configured to acquire the data acquisition time interval of multiple sensors in the data acquisition unit, and set a primary threshold and a secondary threshold according to the data acquisition time interval.

[0015] The interval division module is configured to divide multiple sensors according to the acquisition time interval. Data with an acquisition time interval less than the first-level threshold is divided into the first-level dataset, data with an acquisition time interval equal to or greater than the first time threshold but less than the second time threshold is divided into the second-level dataset, and data with an acquisition time interval equal to or greater than the second time threshold is divided into the third-level dataset.

[0016] The interval setting module is configured to set corresponding data transmission intervals for the first-level dataset, second-level dataset, and third-level dataset, and transmit the data corresponding to the first-level dataset, second-level dataset, and third-level dataset to the data analysis platform according to the set data transmission intervals.

[0017] Furthermore, the link setting module includes:

[0018] The parameter acquisition module is configured to acquire the gateway communication distance parameter between the data acquisition unit and the data analysis platform within a certain time period, and use the median value method to calculate the average value of the gateway communication distance parameter between the data acquisition unit and the data analysis platform within that time period as the network distance threshold.

[0019] The distance judgment module is configured to obtain the gateway communication distance parameter of the communication link between the current data acquisition unit and the data analysis platform, and determine whether the gateway communication distance parameter of the current communication link is within the network distance threshold range. If it exceeds the network distance threshold range, the module obtains the gateway communication distance parameter of another communication link and determines whether it is within the network distance threshold range. If it is within the range, the module switches to that communication link for data transmission.

[0020] Furthermore, the collection frequency of the gateway communication distance parameter of the current communication link is dynamically adjusted, including:

[0021] The signal strength of the gateway communication signal is collected in real time, and the signal strength is used to determine whether there is signal attenuation.

[0022] When the signal strength is attenuating, the attenuation rate of the signal strength is monitored in real time;

[0023] The attenuation rate of the signal strength is compared with a preset attenuation rate threshold. When the attenuation rate of the signal strength exceeds the preset attenuation rate threshold, the frequency adjustment coefficient is set using the attenuation rate.

[0024] Retrieve the minimum and maximum allowable sampling frequencies of the preset gateway communication distance parameters;

[0025] The frequency of the gateway communication distance parameter is adjusted by combining the frequency adjustment coefficient with the minimum and maximum allowable sampling frequencies of the gateway communication distance parameter.

[0026] Furthermore, the frequency adjustment coefficient is obtained by means of:

[0027] Retrieve the attenuation rate of the gateway communication signal;

[0028] The rate of change of the gateway communication distance parameter between the real-time data acquisition unit and the data analysis platform;

[0029] The attenuation rate of the gateway communication signal and the change rate of the gateway communication distance parameter are normalized to obtain the normalized attenuation rate and the change rate of the gateway communication distance parameter.

[0030] The difference between the normalized attenuation rate and the normalized gateway communication distance parameter change rate is calculated to obtain the difference C = Cattenuation rate. 01 -C 02 ; where C 01 C represents the decay rate after normalization. 02This represents the rate of change of the gateway communication distance parameter after normalization.

[0031] Real-time monitoring of electromagnetic interference intensity in the current communication environment;

[0032] The electromagnetic interference intensity of the current communication environment is normalized to obtain the normalized electromagnetic interference intensity B.

[0033] The frequency adjustment coefficient corresponding to the acquisition frequency of the gateway communication distance parameter is set using the normalized electromagnetic interference intensity B and the electromagnetic interference intensity C of the current communication environment.

[0034] Furthermore, the data analysis platform includes:

[0035] The data processing module is configured to preprocess the received sensor data, including data cleaning and normalization, and to annotate each data item with information, including obstacles and their corresponding type labels and location information.

[0036] The model building module is configured to build a neural network model. The neural network model is trained based on historical labeled data. The neural network model automatically extracts feature information from the original data through the combination of multiple neurons and nonlinear transformations. A regressor is used to identify the extracted features and predict the type and location of obstacles.

[0037] Furthermore, the data analysis platform also includes:

[0038] The warning setting module is configured to preset a distance threshold based on the distance between the obstacle and the sensor, and divide different distances between the obstacle and the sensor into different intervals, with each interval corresponding to a warning level;

[0039] The alarm prompting module is configured to calculate the distance between the obstacle and the sensor after the neural network model identifies the type and location of the obstacle, compare it with a preset distance threshold to obtain the corresponding warning level, and issue a corresponding alarm prompt based on the warning level.

[0040] Furthermore, the data analysis platform also includes:

[0041] The model integration module is configured to integrate trained neural network models into the data analysis platform for identification and prediction of real-time sensor data.

[0042] Used in display terminals, configured to display obstacle location information in real time using display devices, so that users can easily view and understand the obstacle recognition results.

[0043] Furthermore, the data acquisition unit includes:

[0044] The storage management module is configured to store and manage the collected sensor data and the prediction results of the data analysis platform, providing data query, backup and recovery functions for later use;

[0045] The security encryption module is configured to encrypt transmitted and stored data, and provides user authentication and access control functions to ensure that only authorized users can access the data storage area.

[0046] Furthermore, the system also includes:

[0047] The power management unit provides AC / DC power to the data acquisition unit, data transmission unit, and data analysis platform, and automatically adjusts the power output according to the system's operating status to reduce power consumption.

[0048] The system monitoring unit monitors the system's operating status in real time, including: the working status of each sensor in the data acquisition unit, the stability of data transmission in the data transmission unit, and the health status of the communication link; it diagnoses faults or abnormalities in the system and provides corresponding solutions.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] This invention utilizes multi-sensor fusion technology to comprehensively and accurately detect obstacles in long-distance blind zones, ensuring data richness and reliability. Furthermore, it dynamically divides the dataset based on sensor acquisition time intervals and sets corresponding data transmission time intervals for datasets of different priorities, enabling more efficient data processing and transmission and ensuring the real-time performance of high-priority data and the stability of low-priority data. Simultaneously, by real-time monitoring of the gateway communication distance parameters of the communication link and dynamically switching the communication link when the network distance threshold is exceeded, the reliability and stability of communication are further enhanced, reducing the risk of data transmission interruption due to communication link failures or instability. This provides more reliable technical support for long-distance blind zone monitoring in complex environments, effectively improving the real-time performance, stability, and resource utilization efficiency of the monitoring system. Attached Figure Description

[0051] Figure 1 This is a flowchart of the real-time monitoring system for long-distance blind zone detection sensors in complex environments according to the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] To address the shortcomings of existing technologies, traditional monitoring systems typically employ only a single communication link. If this link fails, data transmission is interrupted, severely impacting the system's real-time performance and reliability. Furthermore, the fixed data transmission time intervals cannot be dynamically adjusted based on data importance and real-time requirements, leading to low data transmission efficiency and difficulty in achieving real-time monitoring of blind spots in complex environments, please refer to [the relevant documentation / reference]. Figure 1 This embodiment provides the following technical solution:

[0054] A real-time monitoring system for long-distance blind zone detection sensors in complex environments includes: a data acquisition unit, a data transmission unit, a data analysis platform, a power management unit, and a system monitoring unit.

[0055] The data acquisition unit is configured to detect obstacles in long-range blind zones using a multi-sensor fusion approach. The multi-sensor includes: ultrasonic sensors, millimeter-wave radar, lidar, and a camera. Ultrasonic sensors are used for near-range blind zone detection, capable of detecting nearby obstacles or targets, and are characterized by low cost and easy installation. Millimeter-wave radar is used for mid-to-long-range blind zone detection, capable of detecting targets at greater distances, and features strong anti-interference capabilities and high detection accuracy. LiDAR is used for mid-to-long-range blind zone detection, providing high-precision three-dimensional spatial information, suitable for target localization and identification in complex environments. The camera acquires visual information about the target and classifies and identifies the target using image recognition technology, distinguishing between different types of targets such as pedestrians, vehicles, and animals.

[0056] The data acquisition unit includes:

[0057] The storage management module is configured to store and manage the collected sensor data and the prediction results from the data analysis platform, providing subsequent data query, backup, and recovery functions to ensure data security and integrity. Specifically, the data is categorized and stored according to different types, such as sensor data, prediction results, and early warning information, for easy management and querying. Indexes are created for the stored data to facilitate quick querying and retrieval; these indexes can be based on key information such as timestamps, sensor types, and obstacle types. Users can also query historical data, including sensor data and prediction results within specific time periods, facilitating trend analysis and troubleshooting.

[0058] The security encryption module is configured to encrypt transmitted and stored data, and provides user authentication and access control functions to ensure that only authorized users can access the data storage area, thus ensuring data security.

[0059] The data transmission unit includes: an interval acquisition module, an interval division module, a transmission interval module, a link setting module, and a communication transmission module.

[0060] The transmission interval module is configured to set a corresponding data transmission interval based on the acquisition intervals of various sensors in the data acquisition unit; the transmission interval module includes:

[0061] The interval acquisition module is configured to acquire the data acquisition time intervals of various sensors in the data acquisition unit, and set a primary threshold and a secondary threshold based on the data acquisition time interval. For example, it collects the data acquisition time intervals of ultrasonic sensors, millimeter-wave radar, and lidar over a period of time, calculates the 25th percentile of the ultrasonic sensor data acquisition time interval as 80 milliseconds, and the 75th percentile as 200 milliseconds; it calculates the 25th percentile of the millimeter-wave radar data acquisition time interval as 200 milliseconds, and the 75th percentile as 600 milliseconds. The primary threshold is set to the 25th percentile of the ultrasonic sensor, i.e., 80 milliseconds; the secondary threshold is set to the 75th percentile of the millimeter-wave radar, i.e., 600 milliseconds.

[0062] The interval segmentation module is configured to segment multiple sensors according to the acquisition time interval. Data with an acquisition time interval less than the first-level threshold is segmented into the first-level dataset, data with an acquisition time interval equal to or greater than the first time threshold but less than the second time threshold is segmented into the second-level dataset, and data with an acquisition time interval equal to or greater than the second time threshold is segmented into the third-level dataset. Through data segmentation, different transmission strategies are set for data of different priorities to ensure the real-time performance of high-priority data.

[0063] The interval setting module is configured to set corresponding data transmission intervals for the first-level, second-level, and third-level datasets, and then transmit the corresponding data of the first-level, second-level, and third-level datasets to the data analysis platform according to the set data transmission intervals. For example, the first-level dataset, due to its short collection interval, usually requires more frequent transmission to ensure real-time performance, and can be set to transmit once per second; the second-level dataset has a moderate collection interval, and the transmission frequency can be appropriately reduced, such as once every 5 seconds; the third-level dataset has a long collection interval, and the transmission frequency can be further reduced, such as once every 30 seconds.

[0064] The link setting module is configured to pre-set two or more communication links and preset network distance parameters that conform to normal network distance between the data acquisition unit and the data analysis platform, serving as the network distance threshold for selecting any one of the communication links; the link setting module includes:

[0065] The parameter acquisition module is configured to acquire gateway communication distance parameters between the data acquisition unit and the data analysis platform within a certain time period. The average value of the gateway communication distance parameters between the data acquisition unit and the data analysis platform within this time period is calculated using the median method and used as the network distance threshold. By dynamically calculating the network distance threshold, it adapts to different communication environments and ensures that the selection of communication links is more scientific and reasonable.

[0066] The distance judgment module is configured to obtain the gateway communication distance parameter of the communication link between the current data acquisition unit and the data analysis platform, and determine whether the gateway communication distance parameter of the current communication link is within the network distance threshold range. If it exceeds the network distance threshold range, the module obtains the gateway communication distance parameter of another communication link and determines whether it is within the network distance threshold range. If it is within the range, the module switches to that communication link for data transmission to ensure the stability and reliability of data transmission and avoid data transmission interruption due to communication link failure or instability.

[0067] The communication transmission module is configured to select a normal communication link and transmit data collected by various sensors to a data analysis platform for data analysis and prediction according to a preset data transmission time interval.

[0068] The beneficial effects achieved by the above content are as follows: Through the collaborative work between the above modules, the system can dynamically adjust the data transmission time interval according to the sensor's acquisition time interval, and dynamically select the optimal communication link according to the gateway communication distance parameter of the communication link. This not only improves the efficiency and real-time performance of data transmission, but also enhances the stability and reliability of the system, ensuring efficient and stable long-distance blind zone monitoring in complex environments.

[0069] Specifically, the collection frequency of the gateway communication distance parameter of the current communication link is dynamically adjusted, including:

[0070] The signal strength of the gateway communication signal is collected in real time, and the signal strength is used to determine whether there is signal attenuation.

[0071] When the signal strength is attenuating, the attenuation rate of the signal strength is monitored in real time;

[0072] The attenuation rate of the signal strength is compared with a preset attenuation rate threshold. When the attenuation rate of the signal strength exceeds the preset attenuation rate threshold, the frequency adjustment coefficient is set using the attenuation rate.

[0073] Retrieve the minimum and maximum allowable sampling frequencies of the preset gateway communication distance parameters;

[0074] The frequency of the gateway communication distance parameter is adjusted by combining the frequency adjustment coefficient with the minimum and maximum allowable sampling frequencies of the gateway communication distance parameter.

[0075] The sampling frequency of the gateway communication distance parameter of the adjusted communication link is obtained by the following formula:

[0076] f t =min[max(f*K,f min ),f max ]

[0077] Among them, f t f represents the sampling frequency of the gateway communication distance parameter of the communication link after adjustment; f represents the sampling frequency of the gateway communication distance parameter of the communication link before adjustment; K represents the frequency adjustment coefficient; f min and f max These represent the minimum and maximum allowed sampling frequencies for the gateway communication distance parameter, respectively.

[0078] The technical effects of the above solution are as follows: When signal strength attenuates and the attenuation rate exceeds a threshold, the acquisition frequency is dynamically increased by setting a frequency adjustment coefficient. For example, in industrial automation scenarios, if signal attenuation accelerates due to equipment movement or environmental interference, the gateway communication distance parameters can be acquired more quickly, avoiding data failure due to acquisition lag. This ensures accurate decision-making based on these parameters (such as equipment coordination and safety warnings). For example, in factory AGV communication, it can reduce the risk of collisions caused by lag in distance parameters. When the link is stable (no signal attenuation or slow attenuation), a relatively low acquisition frequency is maintained; when there are fluctuations (attenuation rate exceeds the threshold), the frequency is actively increased, allowing the acquisition frequency to match the actual needs of the link. This solves the problems of redundancy in stable links and lag in fluctuating links caused by traditional fixed frequencies, improving the timeliness and accuracy of distance parameter acquisition. During stable link phases, high-frequency acquisition is unnecessary. Relying on the minimum allowable acquisition frequency limit, the data acquisition, transmission, and processing overhead of gateways and other devices is reduced. Taking intelligent transportation roadside gateways as an example, when traffic flow is stable and the communication link is stable, the acquisition frequency is reduced, reducing the computing power and bandwidth usage of the gateway and backend system, saving energy and network resources. Adjusting the sampling frequency as needed based on link status, rather than continuous high-frequency sampling, effectively reduces equipment energy consumption in the long run, extends the battery life of terminal devices such as gateways (e.g., mobile vehicle gateways), and reduces the continuous high load on overall system resources, improving system efficiency and stability. By monitoring signal attenuation and rate, the sampling frequency can be adjusted in a timely manner in complex environments with strong electromagnetic interference and multiple obstacles (e.g., mine and tunnel communication), ensuring uninterrupted and effective distance parameter acquisition, improving the system's adaptability to harsh environments, and making distance-based security monitoring and equipment control functions more reliable. Setting minimum and maximum allowable sampling frequencies defines a reasonable range for sampling frequency adjustments, avoiding excessive adjustments (too low a frequency to collect effective data, too high a frequency to cause system crashes) that could affect normal system operation, enhancing the stability and operability of the solution, and making the entire communication and monitoring system more robust.

[0079] Specifically, the frequency adjustment coefficient is obtained in the following ways:

[0080] Retrieve the attenuation rate of the gateway communication signal;

[0081] The rate of change of the gateway communication distance parameter between the real-time data acquisition unit and the data analysis platform;

[0082] The attenuation rate of the gateway communication signal and the change rate of the gateway communication distance parameter are normalized to obtain the normalized attenuation rate and the change rate of the gateway communication distance parameter.

[0083] The difference between the normalized attenuation rate and the normalized gateway communication distance parameter change rate is calculated to obtain the difference C = Cattenuation rate. 01 -C 02 ; where C 01 C represents the decay rate after normalization. 02 This represents the rate of change of the gateway communication distance parameter after normalization.

[0084] Real-time monitoring of electromagnetic interference intensity in the current communication environment;

[0085] The electromagnetic interference intensity of the current communication environment is normalized to obtain the normalized electromagnetic interference intensity B.

[0086] The frequency adjustment coefficient corresponding to the acquisition frequency of the gateway communication distance parameter is set using the normalized electromagnetic interference intensity B and the electromagnetic interference intensity C of the current communication environment.

[0087] The frequency adjustment coefficient corresponding to the acquisition frequency of the gateway communication distance parameter is obtained by the following formula:

[0088] K = 1 + λ * |B - |C||

[0089] Where K represents the frequency adjustment coefficient; B represents the normalized electromagnetic interference intensity; C represents the difference between the normalized attenuation rate and the normalized gateway communication distance parameter change rate; and λ represents the data adjustment coefficient, with a value of 0.3-0.8.

[0090] The technical effects of the above solution are as follows: By collecting signal attenuation rate, distance parameter change rate, and electromagnetic interference intensity, the dynamic characteristics of the communication link are comprehensively characterized (such as distance changes caused by equipment movement in industrial scenarios, electromagnetic interference in factories, and signal attenuation with distance). The formula integrates these physical quantities so that the frequency adjustment coefficient K can reflect the true state of the link. When the link is stable (low interference, slow attenuation and distance changes), K is close to 1, maintaining the basic acquisition frequency; when the link fluctuates (such as strong interference, rapid attenuation or distance changes), K increases, increasing the acquisition frequency to ensure timely and accurate acquisition of distance parameters, providing reliable data for subsequent data analysis (such as vehicle distance monitoring in intelligent transportation and equipment collaboration in industrial automation). Real-time adjustment based on link status avoids the drawbacks of traditional fixed frequencies. In intelligent transportation roadside gateway scenarios, during peak traffic flow (fast vehicle movement, frequent distance changes) and with high electromagnetic interference (multi-vehicle wireless communication interference), K increases to accelerate distance parameter acquisition, assisting vehicle-road collaboration and collision avoidance systems; during off-peak traffic periods, K decreases, reducing the acquisition frequency, saving resources, and achieving dynamic matching between the acquisition frequency and link requirements. When the link is stable, K does not increase significantly, and the sampling frequency will not be too high, reducing the data collection volume of the gateway, the bandwidth usage, and the computing power consumption of the backend processing. Taking security monitoring as an example, the interference is small at night and the movement of monitored targets is less, so the sampling frequency is reduced, saving gateway resources and reducing the load on the backend platform, allowing resources to be allocated to more needed time periods (such as daytime when there are many people and frequent target movement). In different environments (such as factory workshops with variable electromagnetic interference and open highways), the above technical solution adaptively adjusts K based on real-time parameters (B, C), eliminating the need for frequent manual configuration and improving the applicability and stability of the system in various scenarios. In environments with complex electromagnetic interference, where the electromagnetic interference is complex and the distance changes due to equipment movement, the system can automatically adjust to ensure the effective acquisition of distance parameters and support security functions such as personnel positioning and equipment monitoring.

[0091] The data analysis platform is configured to use a neural network model to extract features from the collected data, thereby identifying the type and location of obstacles. The data analysis platform includes:

[0092] The data processing module is configured to preprocess the received sensor data, including data cleaning and normalization to ensure the training effect and generalization ability of the neural network model, and to annotate the data with information including obstacles and their corresponding type labels and location information.

[0093] The model building module is configured to construct a neural network model. It trains the model based on historical labeled data and adjusts the model's weights and bias parameters using backpropagation to enable the model to accurately learn obstacle features and patterns. A validation set is used to evaluate the model's performance, and the model's structure and hyperparameters are adjusted based on the evaluation results to improve performance and stability. The neural network model automatically extracts feature information from raw data through the combination of multiple layers of neurons and nonlinear transformations. Examples include edge, texture, and shape features in image data, and temporal features in time series data. A regressor is used to identify the extracted features and predict the type and location of obstacles, such as pedestrians, vehicles, and animals, and their locations, such as distance, angle, and coordinates. This provides strong technical support for long-distance blind spot monitoring in complex environments.

[0094] The warning setting module is configured to preset distance thresholds based on the distance between obstacles and sensors, dividing different distances between obstacles and sensors into different intervals, with each interval corresponding to a warning level; for example: Level 1 warning distance threshold: 0 meters to 10 meters, Level 2 warning distance threshold: 10 meters to 50 meters, Level 3 warning distance threshold: more than 50 meters; the distance threshold can also be dynamically adjusted according to actual application scenarios and user needs; for example, when driving on highways, the warning distance threshold can be appropriately increased to accommodate higher vehicle speeds and longer braking distances.

[0095] The alarm module is configured to calculate the distance between the obstacle and the sensor after the neural network model identifies the type and location of the obstacle, and compare it with a preset distance threshold to obtain the corresponding warning level; and issue a corresponding alarm based on the warning level; the warning information includes the type, location, distance of the obstacle and the corresponding warning level, for example: a pedestrian is detected 5 meters ahead, triggering a level one warning.

[0096] The model integration module is configured to integrate trained neural network models into a data analysis platform for the identification and prediction of real-time sensor data. Specific application scenarios include, but are not limited to, deploying optimized models to edge devices or cloud servers of autonomous vehicles to ensure that they can process and analyze sensor data in real time.

[0097] Used for display terminals, this system is configured to display the location information of obstacles in real time, allowing users to easily view and understand the obstacle recognition results. The display interface includes various formats such as map display, image display, and data tables, using different colors or icons to distinguish different types of obstacles. Specific details can be customized according to user needs and application scenarios, enabling users to clearly see the location and type of obstacles in the image. Simultaneously, it supports image zooming, rotation, and other functions, facilitating more detailed observation of obstacles by users.

[0098] The power management unit provides AC / DC power to the data acquisition unit, data transmission unit, and data analysis platform, and automatically adjusts the power output according to the system's operating status to reduce power consumption.

[0099] The system monitoring unit monitors the system's operating status in real time, including: the working status of each sensor in the data acquisition unit, the stability of data transmission in the data transmission unit, and the health status of the communication link; it diagnoses faults or abnormalities in the system and provides corresponding solutions, such as sensor failures and communication link interruptions.

[0100] The beneficial effects achieved by the above are as follows: Multi-sensor fusion technology comprehensively and accurately detects obstacles in long-distance blind zones, ensuring data richness and reliability; dynamically dividing the dataset according to the sensor acquisition time interval and setting corresponding data transmission time intervals for datasets of different priorities enables more efficient data processing and transmission, ensuring the real-time performance of high-priority data and the stability of low-priority data; simultaneously, by real-time monitoring of the gateway communication distance parameters of the communication link and dynamically switching the communication link when the network distance threshold is exceeded, the reliability and stability of communication are further enhanced, reducing the risk of data transmission interruption due to communication link failure or instability. This provides more reliable technical support for long-distance blind zone monitoring in complex environments, effectively improving the real-time performance, stability, and resource utilization efficiency of the monitoring system.

[0101] Working principle: Multiple sensors detect obstacles in long-distance blind zones. Data is divided into primary, secondary, and tertiary datasets based on sensor acquisition intervals. Different data transmission intervals are set according to the priority of each dataset. Data is transmitted to the data analysis platform at specified intervals, and the optimal communication link is selected to ensure data transmission stability. The data analysis platform uses a neural network model to extract features and identify obstacles, predicting the type and location of obstacles. Different levels of warning signals are issued based on the distance between the obstacle and the sensor, thus achieving real-time monitoring and early warning of obstacles in long-distance blind zones.

[0102] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "possessing," or any other variations thereof are intended to cover non-exclusive possession, 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.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that variations, modifications, substitutions and alterations may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring system for long-range blind zone detection sensors in complex environments, characterized in that, The system includes: a data acquisition unit, a data transmission unit, and a data analysis platform. The data transmission unit includes: a transmission interval module, a link setting module, and a communication transmission module. The data acquisition unit is configured to detect obstacles in long-distance blind spots using a multi-sensor fusion method. The multi-sensor includes: ultrasonic sensor, millimeter-wave radar, lidar, and camera. The transmission interval module is configured to set a corresponding data transmission interval based on the acquisition interval of various sensors in the data acquisition unit; The link setting module is configured to pre-set two or more communication links and preset network distance parameters that conform to the normal distance between the data acquisition unit and the data analysis platform, as the network distance threshold for selecting any one of the communication links. The communication transmission module is configured to select a normal communication link and transmit the data collected by multiple sensors to the data analysis platform for data analysis and prediction according to a preset data transmission time interval. The data analysis platform is configured to use a neural network model to extract features from the collected data, thereby identifying the type and location of obstacles. The frequency of collecting the gateway communication distance parameter of the current communication link is dynamically adjusted, including: The signal strength of the gateway communication signal is collected in real time, and the signal strength is used to determine whether there is signal attenuation. When the signal strength is attenuating, the attenuation rate of the signal strength is monitored in real time; The attenuation rate of the signal strength is compared with a preset attenuation rate threshold. When the attenuation rate of the signal strength exceeds the preset attenuation rate threshold, the frequency adjustment coefficient is set using the attenuation rate. Retrieve the minimum and maximum allowable sampling frequencies of the preset gateway communication distance parameters; The frequency of the gateway communication distance parameter is adjusted using the frequency adjustment coefficient in combination with the minimum and maximum allowable sampling frequencies of the gateway communication distance parameter. The sampling frequency of the gateway communication distance parameter of the adjusted communication link is obtained by the following formula: Wherein, f t represents the collection frequency of the gateway communication distance parameter of the adjusted communication link; f represents the collection frequency of the gateway communication distance parameter of the communication link before adjustment; K represents a frequency adjustment coefficient; f min and f max respectively represent the minimum allowed collection frequency and the maximum allowed collection frequency of the gateway communication distance parameter; The frequency adjustment coefficient is obtained in the following ways: Retrieve the attenuation rate of the gateway communication signal; The rate of change of the gateway communication distance parameter between the real-time data acquisition unit and the data analysis platform; The attenuation rate of the gateway communication signal and the change rate of the gateway communication distance parameter are normalized to obtain the normalized attenuation rate and the change rate of the gateway communication distance parameter. The difference between the normalized attenuation rate and the normalized gateway communication distance parameter change rate is calculated, and the difference C = Cattenuation rate is obtained. 01 -C 02 Among them, C 01 C represents the decay rate after normalization. 02 This represents the rate of change of the gateway communication distance parameter after normalization. Real-time monitoring of electromagnetic interference intensity in the current communication environment; The electromagnetic interference intensity of the current communication environment is normalized to obtain the normalized electromagnetic interference intensity B. The frequency adjustment coefficient corresponding to the acquisition frequency of the gateway communication distance parameter is set using the normalized electromagnetic interference intensity B and the electromagnetic interference intensity C of the current communication environment. The frequency adjustment coefficient corresponding to the acquisition frequency of the gateway communication distance parameter is obtained by the following formula: Where K represents the frequency adjustment coefficient; B represents the normalized electromagnetic interference intensity; C represents the difference between the normalized attenuation rate and the normalized gateway communication distance parameter change rate; and λ represents the data adjustment coefficient, with a value of 0.3-0.

8.

2. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 1, characterized in that, The transmission interval module includes: The interval acquisition module is configured to acquire the data acquisition time interval of multiple sensors in the data acquisition unit, and set a primary threshold and a secondary threshold according to the data acquisition time interval. The interval division module is configured to divide multiple sensors according to the acquisition time interval. Data with an acquisition time interval less than the first-level threshold is divided into the first-level dataset, data with an acquisition time interval equal to or greater than the first time threshold but less than the second time threshold is divided into the second-level dataset, and data with an acquisition time interval equal to or greater than the second time threshold is divided into the third-level dataset. The interval setting module is configured to set corresponding data transmission intervals for the first-level dataset, second-level dataset, and third-level dataset, and transmit the data corresponding to the first-level dataset, second-level dataset, and third-level dataset to the data analysis platform according to the set data transmission intervals.

3. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 1, characterized in that, The link setting module includes: The parameter acquisition module is configured to acquire the gateway communication distance parameter between the data acquisition unit and the data analysis platform within a certain time period, and use the median value method to calculate the average value of the gateway communication distance parameter between the data acquisition unit and the data analysis platform within that time period as the network distance threshold. The distance judgment module is configured to obtain the gateway communication distance parameter of the communication link between the current data acquisition unit and the data analysis platform, and determine whether the gateway communication distance parameter of the current communication link is within the network distance threshold range. If it exceeds the network distance threshold range, the module obtains the gateway communication distance parameter of another communication link and determines whether it is within the network distance threshold range. If it is within the range, the module switches to that communication link for data transmission.

4. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 1, characterized in that, The data analysis platform includes: The data processing module is configured to preprocess the received sensor data, including data cleaning and normalization, and to annotate each data item with information, including obstacles and their corresponding type labels and location information. The model building module is configured to build a neural network model. The neural network model is trained based on historical labeled data. The neural network model automatically extracts feature information from the original data through the combination of multiple neurons and nonlinear transformations. A regressor is used to identify the extracted features and predict the type and location of obstacles.

5. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 4, characterized in that, The data analysis platform also includes: The warning setting module is configured to preset a distance threshold based on the distance between the obstacle and the sensor, and divide different distances between the obstacle and the sensor into different intervals, with each interval corresponding to a warning level; The alarm prompting module is configured to calculate the distance between the obstacle and the sensor after the neural network model identifies the type and location of the obstacle, compare it with a preset distance threshold to obtain the corresponding warning level, and issue a corresponding alarm prompt based on the warning level.

6. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 5, characterized in that, The data analysis platform also includes: The model integration module is configured to integrate trained neural network models into the data analysis platform for identification and prediction of real-time sensor data. Used in display terminals, configured to display obstacle location information in real time using display devices, so that users can easily view and understand the obstacle recognition results.

7. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 1, characterized in that, The data acquisition unit includes: The storage management module is configured to store and manage the collected sensor data and the prediction results of the data analysis platform, providing data query, backup and recovery functions for later use; The security encryption module is configured to encrypt transmitted and stored data, and provides user authentication and access control functions to ensure that only authorized users can access the data storage area.

8. The real-time monitoring system for long-distance blind zone detection sensors in complex environments according to claim 1, characterized in that, The system also includes: The power management unit provides AC / DC power to the data acquisition unit, data transmission unit, and data analysis platform, and automatically adjusts the power output according to the system's operating status to reduce power consumption. The system monitoring unit monitors the system's operating status in real time, including: the working status of each sensor in the data acquisition unit, the stability of data transmission in the data transmission unit, and the health status of the communication link; it diagnoses faults or abnormalities in the system and provides corresponding solutions.

Images