Radar-assisted inertial measurement unit anomaly detection method and its application

Abstract

The present invention discloses a radar-assisted inertial measurement unit abnormality detection method and its application, and relates to the field of sensor detection. The method comprises: in the process of synchronous movement of the radar and the inertial measurement unit, collecting the point cloud data of the radar and the angular velocity data of the inertial measurement unit; based on the point cloud data, calculating the relative posture change of the radar at adjacent moments; based on the relative posture change, calculating the relative rotation change of the radar; based on the relative rotation change of the radar, calculating the angular velocity of the radar; based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal. The radar-assisted inertial measurement unit abnormality detection method and its application provided by the present invention obtain the posture change of the carrier at different times through the radar to calculate the angular velocity, which is used as the reference data of the IMU measurement result; when there is a significant deviation between the IMU data and the reference data of the radar, the IMU abnormality can be determined.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensor detection, and in particular relates to a radar-assisted inertial measurement unit abnormality detection method and application thereof. Background Art

[0002] An inertial measurement unit (IMU) is a device that integrates multiple sensors, usually accelerometers, gyroscopes, and magnetometers, to measure and report the acceleration, angular velocity, and direction of an object. Because it can provide accurate motion information, IMU is widely used in various fields that require precise positioning and attitude control, including aerospace, navigation, robotics, and autonomous driving. In the aerospace field, IMU can be used for attitude control and navigation of aircraft to ensure the accuracy and stability of the flight path; in autonomous driving systems, IMU is used for vehicle attitude perception and path planning, providing basic support for the safety and accuracy of autonomous driving.

[0003] Although IMU has important application value in various fields, it also faces many technical challenges in actual use. First of all, the sensors inside the IMU will be affected by factory quality, hardware aging, environmental influences (such as temperature, humidity, vibration, etc.) and external interference, resulting in reduced measurement accuracy. This abnormality not only affects the accuracy of the sensor data, but may also cause system misjudgment or even failure. Therefore, during the operation of the IMU, it must maintain a high degree of accuracy and stability to ensure the reliability of the overall system.

[0004] However, the current detection of IMU anomalies mainly relies on sending the equipment back to the manufacturer for testing using a high-precision turntable. Although this detection method can accurately identify IMU anomalies, it has many problems such as long cycles and high costs. Usually, high-precision turntable testing requires a long preparation and calibration time, and the testing equipment is large and difficult to deploy on the production line or in daily use.

[0005] Therefore, it is necessary to provide a new solution to the above technical problems. Summary of the invention

[0006] The purpose of the present invention is to provide a radar-assisted inertial measurement unit abnormality detection method and application thereof, which can detect the abnormal state of the inertial measurement unit based on radar assistance, with a short detection cycle and low cost.

[0007] To achieve the above purpose, the technical solution provided by the present invention is as follows:

[0008] In a first aspect, the present invention provides a radar-assisted inertial measurement unit abnormality detection method, which comprises:

[0009] During the synchronous movement of the radar and the inertial measurement unit, the point cloud data of the radar and the angular velocity data of the inertial measurement unit are collected; based on the point cloud data, the relative posture change of the radar at adjacent moments is calculated; based on the relative posture change, the relative rotation change of the radar is calculated; based on the relative rotation change of the radar, the angular velocity of the radar is calculated; based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, whether the inertial measurement unit is abnormal is detected.

[0010] In one or more embodiments, based on the point cloud data, the relative posture changes of the radar at adjacent moments are calculated, including: based on the point cloud data, obtaining the motion trajectory matrix of the radar; based on the motion trajectory matrix of the radar, obtaining the posture matrix of the radar at each moment; based on the posture matrix of the radar at adjacent moments, calculating the relative posture change matrix of the radar at adjacent moments.

[0011] In one or more embodiments, the calculation formula of the relative posture change matrix of the radar at adjacent moments is:

[0012] ;

[0013] in, is the relative position change matrix of the radar at adjacent moments, For radar k The pose matrix at the moment, For radar k -1 The pose matrix at the moment.

[0014] In one or more embodiments, calculating the relative rotation change of the radar based on the relative posture change includes:

[0015] The relative position change matrix of the radar at adjacent moments is decomposed into a relative rotation change matrix and a relative translation vector.

[0016] In one or more embodiments, the calculation formula of the relative rotation change matrix of the radar is:

[0017] ;

[0018] Among them, the relative rotation change matrix of the radar , the relative translation vector of the radar .

[0019] In one or more embodiments, the angular velocity of the radar is calculated based on the relative rotation change of the radar, including: converting the relative rotation change matrix of the radar into a rotation vector based on the Rodrigues formula; and calculating the angular velocity of the radar based on the rotation vector.

[0020] In one or more embodiments, the calculation formula of the angular velocity of the radar includes:

[0021] ;

[0022] ;

[0023] ;

[0024] ;

[0025] in, θ is the rotation angle of the radar, is the relative rotation transformation matrix traces, μ is the rotation axis of the radar, V is the rotation vector of the radar, ω is the angular velocity of the radar, Δ t is the time interval between adjacent moments.

[0026] In one or more embodiments, based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal includes: calculating the modulus of the angular velocity of the radar and the angular velocity of the inertial measurement unit, and constructing a time-angular velocity modulus curve of the radar and the inertial measurement unit; based on the time-angular velocity modulus curve of the radar and the inertial measurement unit, detecting whether the inertial measurement unit is abnormal.

[0027] In one or more embodiments, based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal includes: calculating the average angular velocity of the radar and the average angular velocity of the inertial measurement unit within a predetermined time length; and judging whether the inertial measurement unit is abnormal based on the difference between the average angular velocity of the radar and the average angular velocity of the inertial measurement unit.

[0028] In the second aspect, the present invention provides a radar-assisted inertial measurement unit abnormality detection system, which includes: an acquisition module, a calculation module and a detection module; the acquisition module is used to collect the point cloud data of the radar and the angular velocity data of the inertial measurement unit during the synchronous movement of the radar and the inertial measurement unit; the calculation module is used to calculate the relative posture change of the radar at adjacent moments based on the point cloud data; based on the relative posture change, calculate the relative rotation change of the radar; and based on the relative rotation change of the radar, calculate the angular velocity of the radar; the detection module is used to detect whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit.

[0029] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the radar-assisted inertial measurement unit anomaly detection method as described above is implemented.

[0030] In a fourth aspect, the present invention provides a computer-readable medium, which carries computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the radar-assisted inertial measurement unit anomaly detection method as described above.

[0031] Compared with the prior art, the radar-assisted inertial measurement unit anomaly detection method and application provided by the present invention obtain the posture changes of the carrier at different times through radar to calculate the angular velocity, which is used as reference data for the IMU measurement results; when there is a significant deviation between the IMU data and the reference data derived by the radar, the IMU anomaly can be determined; this method not only improves the real-time and convenience of detection, but also can realize continuous monitoring in conventional application scenarios, greatly reducing the dependence on specific detection equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 is a flow chart of an inertial measurement unit abnormality detection method according to an embodiment of the present invention;

[0034] Figure 2 It is a trajectory diagram of the radar and IMU in the autonomous driving test example of the present invention;

[0035] Figure 3 The present invention is a trajectory diagram of the radar and IMU in the robot arm test example;

[0036] Figure 4 It is a time-angular velocity modulus length curve diagram when both the radar and the IMU are normal in one embodiment of the present invention;

[0037] Figure 5 It is a time-angular velocity modulus length curve diagram when the radar is normal and the IMU is abnormal in one embodiment of the present invention;

[0038] Figure 6 It is a structural block diagram of an inertial measurement unit abnormality detection system in one embodiment of the present invention;

[0039] Figure 7 FIG. 4 is a structural block diagram of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] In existing inertial measurement unit (IMU) applications, its accuracy and stability are crucial to the safety and reliability of many systems. However, due to factors such as sensor factory quality, hardware aging, external interference, etc., IMU may experience reduced accuracy or abnormalities in long-term use. Traditional IMU detection methods mainly rely on high-precision turntables. Although this method can provide accurate detection results, it has many limitations, such as long detection cycle, dependence on specific equipment, and high detection cost. This makes it difficult to carry out IMU detection in real time and efficiently in the production line or actual application scenarios, especially in situations where immediate response is required, such as autonomous driving, drones, and industrial robots. This problem is particularly prominent.

[0042] Based on the analysis of the prior art, it can be found that the shortcomings of traditional IMU detection mainly lie in the limitations of its detection method and the lack of real-time performance. In order to address this technical defect, the present invention proposes an IMU anomaly detection method based on radar (Lidar). This method uses radar sensors to collect point cloud data in real time and compares it with IMU data, providing a detection solution that does not rely on a high-precision turntable. In this way, the radar sensor can provide a reliable reference standard for the IMU data, so that the detection process can be carried out during the daily operation of the equipment, reducing downtime and detection costs.

[0043] The core idea of ​​the present invention is to obtain the motion changes of the carrier at different times through radar, and calculate it as angular velocity, which is used as reference data for the IMU measurement results. When there is a significant deviation between the IMU data and the reference data derived from the radar, the IMU can be judged to be abnormal. This method not only improves the real-time and convenience of detection, but also can realize continuous monitoring in conventional application scenarios, greatly reducing the dependence on specific detection equipment.

[0044] Please refer to Figure 1FIG. 1 is a flow chart of a radar-assisted inertial measurement unit abnormality detection method according to an embodiment of the present invention. The radar-assisted inertial measurement unit abnormality detection method specifically comprises the following steps:

[0045] S101: During the synchronous movement of the radar and the inertial measurement unit, collecting point cloud data of the radar and angular velocity data of the inertial measurement unit.

[0046] In the specific implementation process, the radar and inertial measurement unit (IMU) can be connected to the same carrier through a rigid structure, such as installed on an autonomous vehicle, drone, industrial robot or robotic arm. The rigid connection ensures that their relative positions remain unchanged during movement, thereby achieving synchronous movement. Synchronous movement means that the data captured by the radar and IMU at the same point in time are derived from the same motion state of the carrier.

[0047] For example, in an autonomous vehicle, the onboard radar can capture the 3D point cloud information of the surrounding environment, while the IMU measures the acceleration and rotational angular velocity of the vehicle in real time. The data collection of the two is synchronized in time to ensure that each timestamp corresponds to the same motion state.

[0048] For radar, since its relative accuracy can be indirectly reflected by the constructed point cloud map, any significant error will cause the constructed point cloud to be blurred. Therefore, the accuracy of the radar odometer can be judged by checking whether the radar odometer point cloud map is clear, layered, blurred, etc. When the radar odometer point cloud has a clear outline, it can be used as a reference to evaluate the IMU.

[0049] In practical applications, the autonomous driving platform can coordinate the data collection process of multiple sensors through a unified controller or time synchronization protocol (such as PTP, NTP). For example, when an autonomous vehicle is driving, the lidar and IMU will simultaneously record the vehicle's point cloud data and angular velocity data at the same time point, so that these data can be directly compared in subsequent analysis.

[0050] In order to obtain fully stimulated data, the radar and inertial measurement unit can be stimulated by letting the carrier perform a specific motion trajectory, which can be in the shape of an "8" or in rotation. This design can ensure that the sensor works in a variety of motion states and collects comprehensive and rich motion data, providing a more reliable basis for subsequent analysis.

[0051] Specifically, fully stimulated radar and IMU data means that the two sensors must experience multiple motion states during the acquisition process, such as acceleration, deceleration, turning and rotation, so that their response performance under various conditions can be fully tested and captured. Through the "8"-shaped movement, the carrier will turn continuously in two different directions. This path design can ensure that the radar and IMU experience different actions such as left and right turns, straight-line motion, and reverse rotation, so that their sensors are fully stimulated in multiple directions and angles.

[0052] Please refer to Figure 2 The figure shows the trajectory of the radar and IMU in the autonomous driving test case, where the red one is the radar's motion trajectory and the green one is the IMU's motion trajectory. In the autonomous driving test case, the unmanned vehicle travels along an "8"-shaped trajectory, causing the vehicle to constantly change direction. The IMU can sense the frequent changes in angular velocity, while the radar will capture the changes in the surrounding environment observed from different angles, forming multi-angle and multi-directional point cloud data.

[0053] In addition, rotational motion is also an effective excitation method, especially suitable for testing how radar and IMU respond during rotation. By allowing the carrier to rotate continuously, the gyroscope sensor of the IMU and the point cloud data acquisition system of the radar can experience continuous angular velocity changes. This method can well detect the stability and accuracy of the IMU in a rotating state, and can also verify whether the radar can continuously collect clear point cloud data from different angles.

[0054] Please refer to Figure 3 The figure shows the trajectory of the radar and IMU in the robotic arm test example, where the red color is the radar's motion trajectory and the green color is the IMU's motion trajectory. The IMU and radar modules are installed on the robotic arm to make it rotate in a fixed space, thereby collecting rotation data in different directions.

[0055] S102: Calculate the relative position change of the radar at adjacent moments based on the point cloud data.

[0056] It should be noted that pose refers to the position and orientation of an object in three-dimensional space. The relative pose change is calculated to determine the motion change of the carrier between two moments, including its displacement and rotation.

[0057] In the specific implementation process, the radar sensor can capture obstacles and objects in the environment within a certain range by emitting laser pulses and receiving reflected signals to form point cloud data. These point cloud data can be used to build a detailed environmental map in three-dimensional space. During the movement of the carrier, the radar will collect different point cloud data at different time points. By comparing two frames of point clouds at adjacent moments, the position change of the carrier during this period can be deduced.

[0058] The specific implementation method can use point cloud registration algorithms, such as the iterative closest point (ICP) algorithm or the normal distribution transform (NDT) algorithm. The basic idea of ​​the ICP algorithm is to match one frame of point cloud (source point cloud) to another frame of point cloud (target point cloud) as much as possible through a series of rotation and translation operations, and determine the optimal rotation and translation parameters by minimizing the matching error, so as to obtain the relative posture change between the two moments. The NDT algorithm expresses the point cloud as a probability distribution and finds the best registration method by maximizing the probability. These algorithms can still work stably when the environment information is rich and the motion trajectory is complex.

[0059] Take autonomous driving as an example. Assume that when an autonomous vehicle is driving, the radar collects 10 frames of point cloud data per second. These point cloud data contain information about objects in the surrounding environment, such as vehicles, pedestrians, buildings, etc. When the vehicle drives from one moment to the next, by comparing the point cloud data of these two moments, the displacement and rotation changes of the vehicle can be determined. For example, if t 1 moment and t Between 2 moments, the radar point cloud shows that the position of the buildings beside the road in the field of view has shifted. The system can then deduce how far the vehicle has moved forward during this period of time and how it has rotated in direction.

[0060] In an exemplary embodiment, based on the point cloud data, the relative posture changes of the radar at adjacent moments are calculated, specifically including: based on the point cloud data, obtaining the motion trajectory matrix of the radar; based on the motion trajectory matrix of the radar, obtaining the posture matrix of the radar at each moment; based on the posture matrix of the radar at adjacent moments, calculating the relative posture change matrix of the radar at adjacent moments.

[0061] In actual operation, the radar constructs point cloud data of the surrounding environment by emitting laser beams and receiving reflected signals. These point cloud data contain the position distribution information of objects in the environment. The radar can determine its movement path in space by comparing multiple frames of point clouds. In order to obtain the motion trajectory matrix, the system can use point cloud registration algorithms such as ICP (iterative closest point algorithm) or NDT (normal distribution transformation algorithm) to match the point cloud data at adjacent moments, thereby determining the displacement and rotation changes of the radar over a period of time.

[0062] The pose matrix is ​​a mathematical expression used to describe the position and orientation of an object in space. It contains the object's translation and rotation information. By superimposing continuous motion trajectory matrices, the system can deduce the absolute pose of the radar at each moment. For example, suppose the radar is at t At time 0, the radar is at the origin. Through continuous motion trajectory calculation, the system can obtain the radar t 1. t 2, so as to draw the complete motion path of the radar. The pose matrix can be expressed as a 4×4 matrix, which includes a 3×3 rotation matrix and a 3×1 translation vector. This representation method can fully describe the spatial state of the radar.

[0063] Then, the relative change between the pose matrix at one moment and the pose matrix at the previous moment can be obtained.

[0064] Specifically, the calculation formula of the relative posture change matrix of the radar at adjacent moments is:

[0065] ;

[0066] in, is the relative position change matrix of the radar at adjacent moments, For radar k The pose matrix at the moment, For radar k -1 The pose matrix at the moment.

[0067] S103: Calculate the relative rotation change of the radar based on the relative posture change.

[0068] The relative pose change describes the integrated motion of the radar between two moments, including the movement in position (translation) and the change in direction (rotation). The main goal of step S103 is to extract the rotation feature from the overall motion information of the radar between adjacent moments, so as to help understand the rotation behavior of the radar.

[0069] In an exemplary embodiment, the relative posture change matrix of the radar at adjacent moments is decomposed into a relative rotation change matrix and a relative translation vector.

[0070] In order to extract the rotation information of the radar from the relative posture change, the system needs to decompose the posture matrix into a rotation matrix and a translation vector. This decomposition process is equivalent to decomposing the movement change of the radar into two parts: one is the rotation part, which represents the rotation direction and angle of the radar in space, and the other is the translation part, which represents the position change of the radar in space.

[0071] Rotation is a movement around an axis, which does not affect the radar's observation position relative to the environment, but affects the angle at which the radar observes the environment; while translation represents the actual movement of the radar in space. By decomposing the position change matrix, the system can analyze the radar's rotation behavior separately and then calculate its rotation angle and angular velocity.

[0072] Specifically, the calculation formula of the relative rotation change matrix of the radar is:

[0073] ;

[0074] Among them, the relative rotation change matrix of the radar , the relative translation vector of the radar .

[0075] is a 3×3 rotation matrix, which indicates the rotation of the radar at two adjacent moments ( Time and k It belongs to the special orthogonal group SO(3), which represents the rotation operation without scaling and distortion in three-dimensional space. is a 3×1 translation vector, which represents the translation of the radar between two moments, that is, the movement of position, and belongs to the three-dimensional Euclidean space .

[0076] S104: Calculate the angular velocity of the radar based on the relative rotation change of the radar.

[0077] In an exemplary embodiment, the relative rotation change matrix of the radar may be converted into a rotation vector based on the Rodrigues formula; and the angular velocity of the radar may be calculated based on the rotation vector.

[0078] The Rodrigues formula is a mathematical tool that can convert a rotation matrix into a rotation vector. Specifically, the rotation vector contains the axis of rotation information (i.e., which axis is rotated around) and the magnitude of the rotation (i.e., how much the rotation is).

[0079] Specifically, the calculation formula of the angular velocity of the radar includes:

[0080] ;

[0081] ;

[0082] ;

[0083] ;

[0084] in, θ is the rotation angle of the radar, is the relative rotation transformation matrix traces, μ is the rotation axis of the radar, V is the rotation vector of the radar, ω is the angular velocity of the radar, Δ t is the time interval between adjacent moments.

[0085] Represents the rotation matrix The trace of the rotation matrix is ​​the sum of the diagonal elements of the rotation matrix (R 11 +R 22 +R 33 ). Through the trace of the rotation matrix, the cosine value of the rotation angle can be obtained, and then the inverse cosine function can be used to extract the rotation angle θ .

[0086] S105: Detect whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit.

[0087] In the above steps, the angular velocity of the radar has been calculated through the point cloud data of the radar. At the same time, the IMU also outputs its measured angular velocity in real time. These two angular velocities can be compared to detect whether the IMU data is abnormal.

[0088] To improve detection accuracy, it is necessary to ensure that the angular velocity data of the radar and IMU are collected based on the same timestamp to ensure data consistency. Through the time synchronization protocol or data timestamp alignment, the system can synchronize the measurement results of the two sensors to ensure the accuracy of the comparison.

[0089] The difference between the angular velocity of the radar and the IMU can be calculated and the size and direction of the difference can be analyzed. If the difference is close to zero, it indicates that the measurement results of the IMU and the radar are consistent and the IMU is working normally; if the difference is large, it may mean that there is an abnormality in the IMU data.

[0090] In actual operation, the difference in angular velocity may fluctuate slightly due to sensor noise and environmental interference. For this reason, the system usually sets a predefined threshold, and only when the threshold is exceeded will the IMU be considered abnormal. This method can avoid overly sensitive misjudgments while ensuring that abnormalities can be discovered in a timely manner.

[0091] In some cases, a single data anomaly may be temporary, so multiple measurement results can be continuously monitored and abnormal marking can be used to determine whether there is a long-term anomaly. For example, if the difference between the IMU angular velocity and the radar angular velocity exceeds the threshold within a preset time window (such as 5 seconds), it can be determined that the IMU has a fault.

[0092] In an exemplary embodiment, based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal specifically includes: calculating the modulus of the angular velocity of the radar and the angular velocity of the inertial measurement unit, and constructing a time-angular velocity modulus curve of the radar and the inertial measurement unit; based on the time-angular velocity modulus curve of the radar and the inertial measurement unit, detecting whether the inertial measurement unit is abnormal.

[0093] During the detection process, the modulus of the two can be calculated from the angular velocity data obtained synchronously by the radar and IMU. The modulus of angular velocity can be understood as the magnitude of the angular velocity vector, that is, the absolute value of the rotation speed, regardless of the direction. The method of calculating the modulus eliminates the interference of directional factors and only focuses on the magnitude of the rotation rate of the radar and IMU in the same time period.

[0094] By calculating the angular velocity modulus of the radar and IMU, these values ​​can be plotted over time as a time-angular velocity modulus curve. These two curves can intuitively reflect the rotation behavior and change trend of the radar and IMU in the entire time series. If the IMU and radar are working properly, their time-angular velocity modulus curves should be basically consistent, showing synchronized fluctuations and changes. If the IMU has an abnormality, such as drift, error, or internal sensor failure, the angular velocity modulus curve of the IMU will deviate from the radar curve, resulting in a significant difference.

[0095] Figure 4 is a time-angular velocity modulus curve diagram when both the radar and the IMU are normal in one embodiment of the present invention, Figure 5 1 is a time-angular velocity modulus curve diagram when the radar is normal and the IMU is abnormal in one embodiment of the present invention, wherein the red curve represents the radar and the green curve represents the IMU.

[0096] In an exemplary embodiment, based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal specifically includes: calculating the average angular velocity of the radar and the average angular velocity of the inertial measurement unit within a predetermined time length; based on the difference between the average angular velocity of the radar and the average angular velocity of the inertial measurement unit, determining whether the inertial measurement unit is abnormal.

[0097] The angular velocity data of the radar and IMU can be collected simultaneously within a predetermined time window, and the average angular velocity of the two during this period can be calculated respectively. This time window can be set according to the specific application scenario, such as 1 second, 5 seconds or 10 seconds, so as to smooth out the errors caused by instantaneous fluctuations.

[0098] Specifically, the formula for calculating the average angular velocity of the radar and IMU is as follows:

[0099] ;

[0100] in, Represents a time window j The average angular velocity within N Represents a time window j The number of sampling points within Indicates sampling point i angular velocity.

[0101] The above formula can be used to calculate the average angular velocity of the radar and IMU in a certain time window, and the difference between the two can be compared with the preset threshold. If it exceeds the preset threshold, it can be determined that the IMU is abnormal.

[0102] By calculating the average value within the time window, the data noise in a short period of time can be smoothed out, eliminating the impact of accidental fluctuations on the detection results. This smoothing effect makes the detection more stable and avoids excessive sensitivity to short-term errors.

[0103] Please refer to Figure 6 As shown, based on the same inventive concept as the aforementioned radar-assisted inertial measurement unit anomaly detection method, the present invention provides a radar-assisted inertial measurement unit anomaly detection system 600, which includes: an acquisition module 601, a calculation module 602 and a detection module 603.

[0104] The acquisition module 601 is used to collect the point cloud data of the radar and the angular velocity data of the inertial measurement unit during the synchronous movement of the radar and the inertial measurement unit. The calculation module 602 is used to calculate the relative posture change of the radar at adjacent moments based on the point cloud data; calculate the relative rotation change of the radar based on the relative posture change; and calculate the angular velocity of the radar based on the relative rotation change of the radar. The detection module 603 is used to detect whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit.

[0105] Please refer to Figure 7As shown, an embodiment of the present invention further provides an electronic device 700, which includes at least one processor 701, a memory 702 (such as a non-volatile memory), a memory 703, and a communication interface 704, and the at least one processor 701, the memory 702, the memory 703, and the communication interface 704 are connected together via an internal bus 705. The at least one processor 701 is used to call at least one program instruction stored or encoded in the memory 702, so that the at least one processor 701 performs various operations and functions of the radar-assisted inertial measurement unit abnormality detection method described in various embodiments of this specification.

[0106] In the embodiments of the present specification, the electronic device 700 may include, but is not limited to, personal computers, server computers, workstations, desktop computers, laptop computers, notebook computers, mobile electronic devices, smart phones, tablet computers, cellular phones, personal digital assistants (PDAs), handheld devices, messaging devices, wearable electronic devices, consumer electronic devices, and the like.

[0107] An embodiment of the present invention also provides a computer-readable medium, which carries computer execution instructions. When the computer execution instructions are executed by a processor, they can be used to implement various operations and functions of the radar-assisted inertial measurement unit anomaly detection method described in the various embodiments of this specification.

[0108] The computer-readable medium in the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0109] In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0110] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0111] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, systems, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0113] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A radar-assisted inertial measurement unit anomaly detection method, characterized in that: include: During the synchronous movement of the radar and the inertial measurement unit, collecting point cloud data of the radar and angular velocity data of the inertial measurement unit; Based on the point cloud data, calculating the relative posture change matrix of the radar at adjacent moments; Decomposing the relative position change matrix of the radar at adjacent moments into a relative rotation change matrix and a relative translation vector; Based on the Rodrigues formula, the relative rotation change matrix of the radar is converted into a rotation vector; Calculating an angular velocity of the radar based on the rotation vector; Based on the angular velocity of the radar and the angular velocity of the inertial measurement unit, detecting whether the inertial measurement unit is abnormal; The calculation formula of the relative rotation change matrix of the radar is: ； is the relative position change matrix of the radar at adjacent moments, and the relative rotation change matrix of the radar , the relative translation vector of the radar ; SO(3) represents a special orthogonal group containing only rotations, represents a three-dimensional Euclidean space; The calculation formula of the angular velocity of the radar includes: ； ； ； ； θ is the rotation angle of the radar, is the relative rotation transformation matrix traces, μ is the radar's rotation axis, V is the rotation vector of the radar, ω is the angular velocity of the radar, Δ t is the time interval between adjacent moments.

2. The radar-assisted inertial measurement unit abnormality detection method according to claim 1, characterized in that: Based on the point cloud data, calculating the relative position change of the radar at adjacent moments, including: Based on the point cloud data, obtaining a motion trajectory matrix of the radar; Based on the motion trajectory matrix of the radar, obtaining the position and posture matrix of the radar at each moment; Based on the pose matrix of the radar at adjacent moments, a relative pose change matrix of the radar at adjacent moments is calculated.

3. The radar-assisted inertial measurement unit abnormality detection method according to claim 2, characterized in that: The calculation formula of the relative posture change matrix of the radar at adjacent moments is: ； in, For radar k The pose matrix at the moment, For radar k -1 The pose matrix at the moment.

4. The radar-assisted inertial measurement unit anomaly detection method according to claim 1, characterized in that: Detecting whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit includes: Calculating the modulus of the angular velocity of the radar and the angular velocity of the inertial measurement unit, and constructing time-angular velocity modulus curves of the radar and the inertial measurement unit; Based on the time-angular velocity modulus length curves of the radar and the inertial measurement unit, it is detected whether the inertial measurement unit is abnormal.

5. The radar-assisted inertial measurement unit anomaly detection method according to claim 1, characterized in that: Detecting whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit includes: Calculating an average angular velocity of the radar and an average angular velocity of the inertial measurement unit within a predetermined time length; Based on the difference between the average angular velocity of the radar and the average angular velocity of the inertial measurement unit, it is determined whether the inertial measurement unit is abnormal.

6. An inertial measurement unit abnormality detection system for the inertial measurement unit abnormality detection method according to any one of claims 1 to 5, characterized in that: include: An acquisition module, used for acquiring point cloud data of the radar and angular velocity data of the inertial measurement unit during the synchronous movement of the radar and the inertial measurement unit; A calculation module, used to calculate the relative posture change of the radar at adjacent moments based on the point cloud data; calculate the relative rotation change of the radar based on the relative posture change; and calculate the angular velocity of the radar based on the relative rotation change of the radar; The detection module is used to detect whether the inertial measurement unit is abnormal based on the angular velocity of the radar and the angular velocity of the inertial measurement unit.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the radar-assisted inertial measurement unit abnormality detection method as described in any one of claims 1 to 5 is implemented.

8. A computer-readable medium, characterized in that The computer-readable medium carries computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the radar-assisted inertial measurement unit abnormality detection method according to any one of claims 1 to 5.

Images