Sensor anomaly detection methods, devices, electronic equipment and storage media

Abstract

This invention discloses a sensor anomaly detection method, apparatus, electronic device, and storage medium. The method includes determining the current-time count information for each sensor based on the difference between the current-time sensing data of each sensor and a preset threshold; fusing the current-time sensing data of the multiple sensors; determining the current-time detection result based on the current-time count information of each sensor during the fusion process; and issuing a fusion anomaly warning when the current-time detection result indicates an anomaly in the fusion process. Utilizing the embodiments disclosed in this invention, anomaly detection and warning can be performed on the fusion process using the difference information between the multi-sensor sensing data and the preset threshold. This improves the efficiency of multi-sensor anomaly detection, avoids generating high additional processing load on the sensor microcontroller, and simultaneously meets the high functional safety level requirements for sensor anomaly detection coverage.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to a sensor anomaly detection method, device, electronic device, and storage medium. Background Technology

[0002] For autonomous driving systems, sensors are key components in multi-source fusion localization. Problems in the perception or fusion processing of multiple sensors can significantly impact the safety and stability of autonomous vehicles. Therefore, anomaly detection of sensors in autonomous vehicles is essential. In particular, effective sensor anomaly detection methods that meet automotive functional safety requirements are necessary.

[0003] In existing technologies, common methods involve developing and configuring special algorithms on the sensor's microcontroller to process sensor data and then perform anomaly detection; or using redundant configurations of similar sensors, employing hardware redundancy to detect anomalies by checking the consistency of sensor outputs. However, these methods impose a high additional processing load on the microcontroller and have high development and design costs. Furthermore, the hardware redundancy method for checking sensor output consistency is prone to failure and is not suitable for anomaly detection in multi-sensor data fusion. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, this invention discloses a sensor anomaly detection method, apparatus, electronic device, and storage medium. By utilizing the difference information between multi-sensor perceived data and preset thresholds to perform anomaly detection and provide early warning of fusion anomalies, this invention improves the efficiency of multi-sensor anomaly detection, avoids generating high additional processing loads on sensor microcontrollers, saves resources, reduces development and design costs, and simultaneously meets the high functional safety level requirements for sensor anomaly detection coverage, effectively performing sensor anomaly detection. The technical solution disclosed in this invention is as follows:

[0005] According to one aspect of the disclosed embodiments of the present invention, a sensor anomaly detection method is provided, comprising:

[0006] Acquire current-moment sensing data from at least two sensors;

[0007] Based on the difference between the current sensing data of each sensor and the first preset threshold, the current time count information corresponding to each sensor is determined;

[0008] The current-time sensing data of the at least two sensors are fused to obtain current-time fused data; and during the fusion processing of the current-time sensing data of the at least two sensors, the current-time detection result is determined based on the current-time count information corresponding to each sensor.

[0009] If the detection result at the current moment indicates an abnormality in the fusion process, an early warning of fusion anomaly is issued for the fused data at the current moment.

[0010] Optionally, determining the current time count information corresponding to each sensor based on the difference information between the current time sensing data of each sensor and the first preset threshold includes:

[0011] If the current sensing data of each sensor is greater than the first preset threshold, the historical time count information corresponding to each sensor is increased by a preset number to obtain the current time count information corresponding to each sensor.

[0012] If the current sensing data of each sensor is less than the first preset threshold, the historical time count information corresponding to each sensor is reduced by the preset number to obtain the current time count information corresponding to each sensor.

[0013] Optionally, the current detection result includes normal fusion processing results and abnormal fusion processing results, and determining the current detection result based on the current time count information corresponding to each sensor includes:

[0014] If the current time count information corresponding to each sensor is the same, the current time detection result is determined to be the normal result of the fusion processing;

[0015] If the current time count information corresponding to any sensor is different from the current time count information corresponding to other sensors, the current time detection result is determined to be an abnormal result of the fusion processing, and the other sensors are the sensors other than any of the at least two sensors.

[0016] Optionally, the method further includes:

[0017] The current sensing data of the at least two sensors are used to predict the next moment's predicted data of the at least two sensors.

[0018] Acquire the next-moment sensing data from the at least two sensors;

[0019] If the difference between the data sensed by any sensor at the next moment and the predicted data at the next moment of any sensor is greater than a second preset threshold, an early warning of a sensing anomaly is issued for any sensor.

[0020] Optionally, the method further includes:

[0021] Based on the difference between the next-time sensing data of any sensor and the first preset threshold, the next-time counting information corresponding to any sensor is determined;

[0022] Based on the current time count information and the next time count information corresponding to any one of the sensors, the perception change information corresponding to any one of the sensors is determined. The perception change information is used to characterize the change information of the perception function of any one of the sensors over time.

[0023] Optionally, the step of predicting the current sensing data of the at least two sensors to obtain the next moment prediction data of the at least two sensors includes:

[0024] Based on a preset filtering algorithm, the current sensing data of the at least two sensors are predicted to obtain the next moment prediction data of the at least two sensors.

[0025] Optionally, the method further includes:

[0026] Based on the fusion anomaly warning and / or the perception anomaly warning, abnormal response processing is performed on the at least two sensors.

[0027] According to another aspect of the disclosed embodiments of the present invention, a sensor anomaly detection device is provided, comprising:

[0028] The first acquisition module is used to acquire the current sensing data from at least two sensors;

[0029] The first counting information determination module is used to determine the current time counting information corresponding to each sensor based on the difference information between the current time sensing data of each sensor and the first preset threshold.

[0030] The detection result determination module is used to perform fusion processing on the current-time sensing data of the at least two sensors to obtain current-time fused data; and to determine the current-time detection result based on the current-time count information corresponding to each sensor during the fusion processing of the current-time sensing data of the at least two sensors.

[0031] The fusion anomaly early warning module is used to provide an early warning of fusion anomalies to the fused data at the current time when the detection result at the current time indicates an anomaly in the fusion processing.

[0032] According to another aspect of the disclosed embodiments of the present invention, an electronic device is provided, characterized in that the electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded by the processor and executed by the sensor anomaly detection method described above.

[0033] According to another aspect of the disclosed embodiments of the present invention, a computer-readable storage medium is provided, characterized in that the storage medium stores at least one instruction or at least one program, wherein the at least one instruction or at least one program is loaded and executed by a processor to implement the sensor anomaly detection method described in any of the preceding claims.

[0034] According to another aspect of the disclosed embodiments of the present invention, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium, a processor of a computer device reading the computer instructions from the computer-readable storage medium, and the processor executing the computer instructions, causing the computer device to perform the sensor anomaly detection method described in any of the preceding claims.

[0035] The data processing method provided by this invention has the following technical effects:

[0036] The sensor anomaly detection method provided by this invention acquires current-time sensing data from at least two sensors, and determines the current-time count information corresponding to each sensor based on the difference information between the current-time sensing data of each sensor and a first preset threshold. Then, it fuses the current-time sensing data from the at least two sensors to obtain fused current-time data. During the fusion process, based on the current-time count information corresponding to each sensor, it determines the current-time detection result, and provides a fusion anomaly warning when the current-time detection result indicates an anomaly in the fusion process. This method improves the efficiency of multi-sensor anomaly detection by using the difference information between the multi-sensor sensing data and the preset threshold to detect fusion anomalies and provide warnings. It avoids generating high additional processing load on the sensor microcontroller, saves resources, and reduces development and design costs. Furthermore, by comparing multiple identical or similar sensors, it detects failures within the sensor's measurement range. During the acquisition and fusion of sensing data from multiple sensors, it performs synchronous detection and verification of the data to avoid the failure to detect anomalies due to dynamic changes in sensor readings. This meets the high functional safety level requirements for sensor anomaly detection coverage and effectively performs sensor anomaly detection.

[0037] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart illustrating a sensor anomaly detection method according to an exemplary embodiment;

[0040] Figure 2 This is a flowchart illustrating another sensor anomaly detection method according to an exemplary embodiment;

[0041] Figure 3 This is a schematic diagram illustrating another sensor anomaly detection method according to an exemplary embodiment;

[0042] Figure 4 This is a block diagram of a sensor anomaly detection device according to an exemplary embodiment;

[0043] Figure 5 This is a block diagram illustrating a terminal electronic device for sensor anomaly detection according to an exemplary embodiment;

[0044] Figure 6 This is a block diagram illustrating a server electronic device for sensor anomaly detection according to an exemplary embodiment. Detailed Implementation

[0045] To enable those skilled in the art to better understand the technical solutions disclosed in this invention, the technical solutions in the disclosed embodiments will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention disclosed herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0047] Sensor anomaly detection typically involves developing and configuring special algorithms on the sensor's microcontroller to process sensor data and then perform anomaly detection, or using a redundant configuration scheme for similar sensors to detect anomalies by using hardware redundancy to detect the consistency of sensor outputs. However, these methods result in a high additional processing load on the microcontroller and high development and design costs. Furthermore, the hardware redundancy method for detecting sensor output consistency can fail and is not suitable for anomaly detection in multi-sensor data fusion.

[0048] This invention utilizes simple calculations of intermediate data generated during sensor data processing to achieve anomaly detection in sensors, improving the efficiency of multi-sensor anomaly detection. It eliminates the need to design special algorithms and additional complex processing and storage capabilities on the sensor's microcontroller, reducing design costs. Furthermore, it avoids the use of redundant sensor microcontrollers, enabling anomaly detection with lower processing load and design effort while meeting functional safety requirements. The aforementioned intermediate data may include sensor sensing data at any given time, predicted data for the next time step, and sensing data for the next time step.

[0049] Furthermore, high functional safety levels place higher demands on the typical diagnostic coverage achievable by sensor safety mechanisms / measures. For example, they must be able to detect failures within their effective range. When sampling and fusing sensing data from multiple sensors, the ECU (Electronic Control Unit) needs to be as synchronized as possible to avoid failing to detect anomalies due to dynamic changes in sensor readings. This invention detects failures within the sensor's range, such as drift, offset, or jamming, by comparing multiple identical or similar sensors. During the acquisition and fusion of sensing data from multiple sensors, it performs synchronous detection and verification of the data to avoid failing to detect anomalies due to dynamic changes in sensor readings. This satisfies the high functional safety level's requirement for high coverage of sensor anomaly detection and effectively detects sensor anomalies.

[0050] Figure 1 This is a flowchart illustrating a sensor anomaly detection method according to an exemplary embodiment. This specification provides the operational steps of the method as described in the embodiments or flowchart, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only execution order. In actual system or server products, the method can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment) as shown in the embodiments or drawings. Specifically, as... Figure 1 As shown, the above method may include:

[0051] S101: Acquire current-time sensing data from at least two sensors.

[0052] In one specific embodiment, the at least two sensors can be at least two of the sensors located on the autonomous vehicle. The at least two sensors can be of the same or different types. Specifically, the sensor types can include millimeter-wave radar, lidar, and cameras, etc.

[0053] In one specific embodiment, the current-time sensing data of each sensor can be obtained from the current-time measurement data and the current-time prediction data of each sensor. Specifically, the current-time prediction data can be updated based on the current-time measurement data of each sensor using a preset filtering algorithm to obtain the current-time sensing data. Specifically, the sensing data can be used to characterize the actual state data of objects in the surrounding environment. The actual state data can include actual position data, actual velocity data, and actual acceleration data, etc. The objects in the surrounding environment can include at least vehicles, pedestrians, and obstacles in the surrounding environment. The current-time measurement data can be the surrounding environment information directly collected at the current time. The current-time prediction data can be the estimated data (estimated position data, estimated velocity data, and estimated acceleration data, etc.) of the current time obtained by predicting the sensing data of the previous time using the aforementioned preset filtering algorithm. The aforementioned preset filtering algorithm can include the Kalman filter algorithm and its extended variants (such as the asymmetric Kalman filter algorithm, the state error Kalman filter algorithm, etc.). The preset filtering algorithm can be set according to the actual application requirements.

[0054] S103: Based on the difference between the current sensing data of each sensor and the first preset threshold, determine the current time count information corresponding to each sensor.

[0055] In one specific embodiment, the first preset threshold can be determined based on the sensor sampling range, sampling accuracy, and sampling rate, and can also be determined in combination with other state information of the autonomous vehicle, such as steering wheel angle information, steering wheel angle rate, and yaw rate. Specifically, different perception data, such as actual position data, actual speed data, and actual acceleration data, can have their own corresponding first preset threshold. The first preset threshold can be set according to actual application requirements and can be calibrated through multiple tests.

[0056] In an optional embodiment, determining the current time count information corresponding to each sensor based on the difference information between the current time sensing data of each sensor and the first preset threshold may include:

[0057] If the current sensing data of each sensor is greater than the first preset threshold, the historical time count information corresponding to each sensor is increased by a preset number to obtain the current time count information corresponding to each sensor.

[0058] If the current sensing data of each sensor is less than a first preset threshold, the historical time count information corresponding to each sensor is reduced by a preset number to obtain the current time count information corresponding to each sensor.

[0059] In one specific embodiment, the historical time count information corresponding to each sensor can be the count information corresponding to each sensor at the previous time. Specifically, it can be determined based on the difference between the perceived data of each sensor and the first preset threshold at the previous time. The preset number can be set according to actual application requirements. For example, if the perceived data of each sensor at the current time is greater than the first preset threshold, the historical time count information corresponding to each sensor can be incremented by 1 to obtain the current time count information corresponding to each sensor; if the perceived data of each sensor at the current time is less than the first preset threshold, the historical time count information corresponding to each sensor can be decremented by 1 to obtain the current time count information corresponding to each sensor.

[0060] In the above embodiments, based on the difference between the sensing data of each sensor at the current moment and the preset threshold, the historical time count information corresponding to each sensor is increased or decreased by a preset number, thereby determining the current time count information corresponding to each sensor. This can quickly determine the count information corresponding to the sensor, thereby improving the efficiency of subsequent sensor anomaly detection.

[0061] S105: Perform fusion processing on the current-time sensing data of at least two sensors to obtain current-time fused data; and during the fusion processing of the current-time sensing data of at least two sensors, determine the current-time detection result based on the current-time count information corresponding to each sensor.

[0062] In one specific embodiment, current-time sensing data from at least two sensors can be fused based on a preset fusion algorithm to obtain current-time fused data. Specifically, the preset fusion algorithm may include the Kalman filter algorithm and its extended variants, and the preset fusion algorithm can be set according to actual application requirements.

[0063] In one specific embodiment, the detection result at the current moment may include normal fusion processing results and abnormal fusion processing results.

[0064] In an optional embodiment, determining the detection result at the current time based on the current time count information corresponding to each sensor may include:

[0065] If the current time count information of each sensor is the same, the detection result at the current time is determined to be a normal result of fusion processing;

[0066] If the current time count information corresponding to any sensor is different from the current time count information corresponding to other sensors, the detection result at the current time is determined to be an abnormal result of the fusion processing.

[0067] In one specific embodiment, during the fusion processing of current-time sensing data from at least two sensors, the fact that the current-time count information corresponding to each sensor is the same indicates that the current-time sensing data of each sensor is consistent, which can meet the data dimension requirements for fusion processing of data from at least two sensors. It can be determined that the fusion data of at least two sensors at the current time is normal; otherwise, the fusion data of at least two sensors at the current time is abnormal. Other sensors can be any of the at least two sensors except for any one of them.

[0068] In the above embodiments, the counting information corresponding to each sensor at any given time is used to determine whether the fused data obtained by the sensor in the fusion processing at that time is normal, thereby improving the efficiency of fusion anomaly detection.

[0069] Optionally, during the fusion process, the number of sensing data points exceeding a first preset threshold can be counted from the sensing data of multiple sensors at each moment. Specifically, the number of actual position data, actual velocity data, and actual acceleration data exceeding their respective first preset thresholds can be counted from the sensing data of each sensor at each moment. These individual counts are then summed at each moment to obtain the total number, which represents the number of sensing data points exceeding the first preset threshold for multiple sensors at each moment. By comparing the total number of data points corresponding to multiple sensors at adjacent moments, it can be determined whether the fusion function of multiple sensors degrades over time. Specifically, if the difference between the total number of data points corresponding to multiple sensors at the next moment and the current moment exceeds a preset difference threshold, it can be determined that the fusion function of multiple sensors has degraded over time, i.e., the fusion function of multiple sensors is abnormal. Subsequent fusion anomaly warning and response processing can be performed. The preset difference threshold can be set according to actual application requirements.

[0070] S107: If the detection result at the current moment indicates an abnormality in the fusion processing, issue an early warning for the fusion anomaly in the fusion data at the current moment.

[0071] In one specific embodiment, the anomaly warning module in the microcontroller can be used to provide an anomaly warning for the fused data at the current moment. Specifically, the anomaly warning module can generate fusion anomaly warning information to provide an anomaly warning.

[0072] In an optional embodiment, Figure 2 This is a flowchart illustrating another sensor anomaly detection method according to an exemplary embodiment, such as... Figure 2 As shown, the above method may further include:

[0073] S201: Predict the current sensing data from at least two sensors to obtain the next moment prediction data from at least two sensors.

[0074] In one specific embodiment, the next moment prediction data from at least two sensors can be estimated data (estimated position information, estimated speed information, and estimated acceleration information, etc.) obtained by predicting the next moment based on the perception data (actual position information, actual speed information, and actual acceleration information, etc.) of at least two sensors at the current moment during the operation of the autonomous vehicle.

[0075] In an optional embodiment, the above-described prediction of the current sensing data from at least two sensors to obtain the next-time prediction data from at least two sensors may include:

[0076] Based on a preset filtering algorithm, the current sensing data of at least two sensors are predicted to obtain the next moment prediction data of at least two sensors.

[0077] S203: Acquire next-moment sensing data from at least two sensors.

[0078] In one specific embodiment, the next moment sensing data can be obtained by updating the next moment prediction data based on the next moment measurement data according to the above-mentioned preset filtering algorithm. The next moment measurement data can be the surrounding environment information directly collected by the above-mentioned at least two sensors at the next moment.

[0079] S205: If the difference between the perceived data of any sensor at the next moment and the predicted data of any sensor at the next moment is greater than a second preset threshold, a perception anomaly warning is issued for any sensor.

[0080] In one specific embodiment, the difference between the next-time perceived data and the next-time predicted data of any sensor can be the residual data between the two. The second preset threshold can be determined based on the sensor's sampling range, sampling accuracy, and sampling rate, and can also be determined in conjunction with other state information of the autonomous vehicle, such as steering wheel angle information, steering wheel angle rate, and yaw rate. Specifically, different perceived data, such as actual position data, actual speed data, and actual acceleration data, can have their own corresponding second preset thresholds. The second preset threshold can be set according to actual application requirements and can be calibrated through multiple tests.

[0081] In one specific embodiment, an anomaly warning module in the microcontroller can be used to detect anomalies in any sensor. Specifically, the anomaly warning module can generate fused anomaly warning information to detect anomalies.

[0082] In the above embodiments, anomaly detection of a sensor is performed by measuring the degree of matching between the perceived data and the predicted data of any sensor at the same time. This ensures the rationality, speed, efficiency, and real-time performance of anomaly detection. Furthermore, anomaly detection of a single sensor can be achieved using only the data generated during the data tracking process of any sensor, enabling anomaly detection of the sensor with a lower processing load and design workload.

[0083] In an optional embodiment, the above method may further include:

[0084] Based on the difference between the next-time sensing data of any sensor and the first preset threshold, determine the next-time counting information corresponding to any sensor;

[0085] Based on the current time count information and the next time count information of any sensor, the perceived change information of any sensor is determined.

[0086] In one specific embodiment, the aforementioned sensing change information can be used to characterize the change in the sensing function of any sensor over time. When the current time count information and the next time count information of any sensor are different, that is, when the sensor's sensing data at the current time and the sensing data at the next time are not consistent, it can be determined that the sensor's sensing function has changed over time, thereby enabling an abnormal warning for the sensor.

[0087] Optionally, a combination of perceived data, predicted data, or other available data can be selected for sensor anomaly detection based on preset constraints. Specifically, the preset constraints may include the available processing power of the microcontroller and specific project requirements.

[0088] In an optional embodiment, the above method may further include:

[0089] Anomaly response processing is performed on at least two sensors based on fusion anomaly early warning and / or perception anomaly early warning.

[0090] In one specific embodiment, exception response processing can be performed through the exception response module in the microcontroller or by calling an additional module with exception response functionality.

[0091] As can be seen from the technical solutions provided in the embodiments of this specification above, this specification acquires the current-time sensing data of at least two sensors, and determines the current-time count information corresponding to each sensor based on the difference information between the current-time sensing data of each sensor and a first preset threshold; then, it performs fusion processing on the current-time sensing data of at least two sensors to obtain current-time fused data, and determines the current-time detection result based on the current-time count information corresponding to each sensor during the fusion processing, and provides a fusion anomaly warning when the current-time detection result indicates an anomaly in the fusion processing. This allows for anomaly detection and warning of fusion anomalies through the difference information between the multi-sensor sensing data and the preset threshold, improving the efficiency of multi-sensor anomaly detection, avoiding high additional processing load on the sensor microcontroller, saving resources, and reducing development and design costs. Furthermore, by utilizing the intermediate data generated during sensor data processing, there is no need to design special algorithms or additional complex processing and storage capabilities on the sensor microcontroller, and redundant sensor microcontrollers are avoided. This enables anomaly detection of sensors with lower processing load and design workload, meeting functional safety requirements. Furthermore, anomaly detection of a sensor can be performed by assessing the matching degree between the perceived and predicted data of any sensor at the same time, ensuring the rationality, speed, and efficiency of anomaly detection. Simultaneously, by comparing multiple identical or similar sensors, failures within the sensor's measurement range can be detected. When acquiring and fusing perceived data from multiple sensors, synchronous detection and verification of the data prevents the failure to detect anomalies due to dynamic changes in sensor readings. This meets the high functional safety level requirements for sensor anomaly detection coverage, effectively enabling sensor anomaly detection.

[0092] See Figure 3 , Figure 3This is a schematic diagram illustrating another sensor anomaly detection method according to an exemplary embodiment. Time tn can represent the current time, and time tn+1 can represent the next time. The method acquires the sensing data (actual position data, actual velocity data, and actual acceleration data, etc.) of each sensor at time tn. Specifically, the sensing data of each sensor at time tn can be obtained by updating the predicted data of each sensor at time tn based on the measurement data of each sensor at time tn. Then, prediction is performed based on the sensing data of each sensor at time tn to obtain the predicted data (estimated position data, estimated velocity data, and estimated acceleration data, etc.) of each sensor at time tn+1. The predicted data at time tn+1 is updated based on the measurement data of each sensor at time tn+1 to obtain the sensing data of each sensor at time tn+1. The sensing data of each sensor at time tn+1 is then associated with the sensing data at time tn. The sensing data of each sensor can be sent to the control unit (ECU) via an end-to-end (E2E) communication protection line, and the control unit performs one or more of the following methods for sensing anomaly warning and fusion anomaly warning.

[0093] By comparing the sensing data of each sensor at time tn+1 with the sensing data of the same sensor at time tn, it can be determined whether to issue a sensing anomaly warning for each sensor. Based on the differences between the sensing data of each sensor at time tn, the sensing data at time tn+1, and the first preset threshold, the corresponding count information of the sensor at time tn and time tn+1 is determined respectively. The count information of the sensor at time tn and time tn+1 is compared. If the count information of each sensor at time tn and time tn+1 is different, that is, if there is no consistency between the sensing data of the sensor at time tn and the sensing data at time tn+1, it can be determined that the sensing function of the sensor has changed over time, and thus a sensing anomaly warning can be issued for the sensor; otherwise, if the sensing data of the sensor at time tn and the sensing data at time tn+1 are consistent, there is no need to issue a sensing anomaly warning for the sensor.

[0094] By comparing the sensing data of each sensor at time tn+1 with the predicted data of that sensor at time tn+1, it can be determined whether to issue a sensing anomaly warning for each sensor. If the difference between the sensing data and the predicted data of each sensor at time tn+1 is greater than a second preset threshold, a sensing anomaly warning is issued for that sensor; otherwise, no sensing anomaly warning is required.

[0095] Furthermore, during the fusion processing of sensing data from multiple sensors at each time step, the count information corresponding to each sensor at each time step can be determined, thereby determining whether to issue a fusion anomaly warning for multiple sensors. The fusion processing of sensing data from multiple sensors at each time step yields fused data for each sensor at that time step. If the count information for each sensor at each time step is the same, the fusion processing of multiple sensors at that time step is determined to be normal; if the count information for any sensor at any time step differs from the count information for other sensors, the fusion processing of multiple sensors at that time step is determined to be abnormal. Therefore, in the event of an abnormality in the fusion processing of multiple sensors at each time step, a fusion anomaly warning is issued for the fused data of multiple sensors at that time step.

[0096] During the fusion processing of sensing data from multiple sensors at each moment, the number of sensing data points corresponding to multiple sensors that exceed a first preset threshold at each moment can also be counted. Specifically, the number of actual position data, actual velocity data, and actual acceleration data points corresponding to each sensor that exceed their respective first preset thresholds at each moment can be counted. The total number of these data points is obtained by adding them together at each moment, which is the number of sensing data points corresponding to multiple sensors that exceed the first preset threshold at each moment. Therefore, if the difference between the total number of data points corresponding to multiple sensors at time tn+1 and time tn exceeds a preset difference threshold, it can be determined that the fusion function of multiple sensors degrades over time.

[0097] Furthermore, multiple sensors can be processed for abnormal response based on fusion anomaly warning and / or perception anomaly warning.

[0098] This invention also provides a sensor anomaly detection device, such as... Figure 4 As shown, the device includes:

[0099] The first acquisition module 410 is used to acquire the current sensing data from at least two sensors;

[0100] The first counting information determination module 420 is used to determine the current time counting information corresponding to each sensor based on the difference information between the current time sensing data of each sensor and the first preset threshold.

[0101] The detection result determination module 430 is used to perform fusion processing on the current time sensing data of the at least two sensors to obtain current time fused data; and to determine the current time detection result based on the current time count information corresponding to each sensor during the fusion processing of the current time sensing data of the at least two sensors.

[0102] The fusion anomaly warning module 440 is used to provide a fusion anomaly warning for the fusion data at the current time when the detection result at the current time indicates an anomaly in the fusion processing.

[0103] Optionally, the first counting information determination module 420 may include:

[0104] The first counting information determining unit is used to control the historical time counting information corresponding to each sensor to increase by a preset number when the current time sensing data of each sensor is greater than the first preset threshold, so as to obtain the current time counting information corresponding to each sensor.

[0105] The second counting information determination unit is used to control the historical time counting information corresponding to each sensor to decrease by the preset number when the current time sensing data of each sensor is less than the first preset threshold, so as to obtain the current time counting information corresponding to each sensor.

[0106] Optionally, the detection result at the current moment includes normal fusion processing results and abnormal fusion processing results, and the detection result determination module 430 may include:

[0107] The fusion processing normal result determination unit is used to determine the current time detection result as the fusion processing normal result when the current time count information corresponding to each sensor is the same;

[0108] The fusion processing abnormal result determination unit is used to determine the current time detection result as the fusion processing abnormal result when the current time count information corresponding to any sensor is different from the current time count information corresponding to other sensors, wherein the other sensors are the sensors other than any of the at least two sensors.

[0109] Optionally, the device may further include:

[0110] The prediction module is used to predict the current sensing data of the at least two sensors to obtain the next moment prediction data of the at least two sensors.

[0111] The second acquisition module is used to acquire the next moment sensing data of the at least two sensors;

[0112] The perception anomaly warning module is used to issue a perception anomaly warning to any sensor when the difference between the perceived data of any sensor at the next moment and the predicted data of any sensor at the next moment is greater than a second preset threshold.

[0113] Optionally, the device may further include:

[0114] The second counting information determination module is used to determine the next-time counting information corresponding to any sensor based on the difference information between the next-time sensing data of any sensor and the first preset threshold.

[0115] The perception change information determination module is used to determine the perception change information corresponding to any sensor based on the current time count information and the next time count information corresponding to any sensor. The perception change information is used to characterize the change information of the perception function of any sensor over time.

[0116] Optionally, the prediction module may include:

[0117] The prediction unit is used to predict the current sensing data of the at least two sensors based on a preset filtering algorithm, so as to obtain the next moment prediction data of the at least two sensors.

[0118] Optionally, the device may further include:

[0119] An anomaly response processing module is used to perform anomaly response processing on the at least two sensors based on the fusion anomaly warning and / or the perception anomaly warning.

[0120] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0121] Figure 5 This is a block diagram illustrating an electronic device for sensor anomaly detection according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the electronic device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a sensor anomaly detection method. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0122] Figure 6This is a block diagram illustrating an electronic device for sensor anomaly detection according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a sensor anomaly detection method.

[0123] Those skilled in the art will understand that Figure 5 or Figure 6 The structures shown are merely block diagrams of some structures related to the disclosed solutions of this invention, and do not constitute a limitation on the electronic devices to which the disclosed solutions of this invention are applied. Specific electronic devices may include more or fewer components than those shown in the figures, or combine certain components, or have different component arrangements.

[0124] In an exemplary embodiment, an electronic device is also provided, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the sensor anomaly detection method as disclosed in the embodiments of the present invention.

[0125] Memory can be used to store software programs and modules. The processor executes various functional applications by running the software programs and modules stored in the memory. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, application programs required for functions, etc.; the data storage area can store data created based on the use of the device, etc. Furthermore, memory can include high-speed random access memory, and can also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, memory can also include a memory controller to provide the processor with access to the memory. The processor can be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits (ASICs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0126] In an exemplary embodiment, a computer storage medium is also provided, wherein at least one instruction or at least one program is stored therein, the at least one instruction or at least one program being loaded and executed by a processor to implement the sensor anomaly detection method in the disclosed embodiments of the present invention.

[0127] In an exemplary embodiment, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the sensor anomaly detection method according to the disclosed embodiments of the present invention.

[0128] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), and double data rate RAM.

[0129] SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM), etc.

[0130] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles disclosed herein and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0131] It should be understood that the present invention is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A sensor anomaly detection method, characterized in that, include: Acquire current-moment sensing data from at least two sensors; If the current sensing data of each sensor is greater than a first preset threshold, the historical time count information corresponding to each sensor is increased by a preset number to obtain the current time count information corresponding to each sensor. If the current sensing data of each sensor is less than the first preset threshold, the historical time count information corresponding to each sensor is reduced by the preset number to obtain the current time count information corresponding to each sensor. The current-time sensing data from the at least two sensors are fused to obtain the current-time fused data; And during the fusion processing of the current-time sensing data from the at least two sensors, the current-time detection result is determined based on the current-time count information corresponding to each sensor; The detection results at the current moment include normal fusion processing results and abnormal fusion processing results; If the detection result at the current moment indicates an anomaly in the fusion processing, an anomaly warning will be issued for the fused data at the current moment. The step of determining the current time detection result based on the current time count information corresponding to each sensor includes: if the current time count information corresponding to each sensor is the same, determining the current time detection result as the normal result of the fusion processing; if the current time count information corresponding to any sensor is different from the current time count information corresponding to other sensors, determining the current time detection result as the abnormal result of the fusion processing; the other sensors are the sensors other than any of the at least two sensors.

2. The method according to claim 1, characterized in that, The method further includes: The current sensing data of the at least two sensors are used to predict the next moment's predicted data of the at least two sensors. Acquire the next-moment sensing data from the at least two sensors; If the difference between the data sensed by any sensor at the next moment and the predicted data at the next moment of any sensor is greater than a second preset threshold, an early warning of a sensing anomaly is issued for any sensor.

3. The method according to claim 2, characterized in that, The method further includes: Based on the difference between the next-time sensing data of any sensor and the first preset threshold, the next-time counting information corresponding to any sensor is determined; Based on the current time count information and the next time count information corresponding to any one of the sensors, the perception change information corresponding to any one of the sensors is determined. The perception change information is used to characterize the change information of the perception function of any one of the sensors over time.

4. The method according to claim 2, characterized in that, The step of predicting the current sensing data of the at least two sensors to obtain the next moment prediction data of the at least two sensors includes: Based on a preset filtering algorithm, the current sensing data of the at least two sensors are predicted to obtain the next moment prediction data of the at least two sensors.

5. The method according to claim 2, characterized in that, The method further includes: Based on the fusion anomaly warning and / or the perception anomaly warning, abnormal response processing is performed on the at least two sensors.

6. A sensor anomaly detection device, characterized in that, The device includes: The first acquisition module is used to acquire the current sensing data from at least two sensors; The first counting information determining unit is configured to, when the current sensing data of each sensor is greater than a first preset threshold, control the historical time counting information corresponding to each sensor to increase by a preset number, so as to obtain the current time counting information corresponding to each sensor; the second counting information determining unit is configured to, when the current sensing data of each sensor is less than the first preset threshold, control the historical time counting information corresponding to each sensor to decrease by the preset number, so as to obtain the current time counting information corresponding to each sensor. The detection result determination module is used to perform fusion processing on the current-time sensing data of the at least two sensors to obtain current-time fused data; and to determine the current-time detection result based on the current-time count information corresponding to each sensor during the fusion processing of the current-time sensing data of the at least two sensors; the current-time detection result includes normal fusion processing result and abnormal fusion processing result; The fusion anomaly early warning module is used to provide a fusion anomaly early warning for the fused data at the current time when the detection result at the current time indicates an anomaly in the fusion processing. The detection result determination module includes: a fusion processing normal result determination unit, used to determine the current time detection result as the fusion processing normal result when the current time count information corresponding to each sensor is the same; and a fusion processing abnormal result determination unit, used to determine the current time detection result as the fusion processing abnormal result when the current time count information corresponding to any sensor is different from the current time count information corresponding to other sensors, wherein the other sensors are sensors other than any of the at least two sensors.

7. An electronic device, characterized in that, The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded by the processor and executed as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor according to any one of claims 1 to 5.

Images