Monitoring blind area vehicle positioning method, system and device and storage medium

Abstract

The invention discloses a monitoring blind area vehicle positioning method, system and device and a storage medium, and relates to the technical field of automobile positioning. Comprising the following steps: when the positioning signal strength is lower than a signal threshold value, automatically switching to an autonomous positioning mode, and recording first signal data and first angle data at the same time; calculating a first track point set according to the first signal data and the first angle data; acquiring first speed data and second angle data, calculating a second track point set, and performing time sequence alignment on the second track point set and the first track point set to generate a deviation sequence; and establishing a trajectory deviation analysis model, calculating a correction amount according to the statistical characteristics of the deviation sequence, and if the correction amount exceeds a correction threshold value, performing real-time correction and correction positioning output by combining the current course angle. The method can ensure that the vehicle can still realize continuous positioning by means of the sensor in the vehicle under the condition that the GNSS signal cannot provide stable support.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile positioning, and in particular to a method, system, device and storage medium for positioning a vehicle in a monitoring blind area. Background Art

[0002] With the rapid development of intelligent transportation and autonomous driving technologies, vehicle positioning technology has become a key factor in improving road safety and traffic efficiency. At present, vehicle positioning systems usually rely on the global satellite positioning system as the main positioning method, using satellite signals to determine the real-time position of the vehicle. With the advancement of vehicle-mounted sensor technology, combined with information such as inertial navigation systems, vehicle speed sensors, and steering angle sensors, a variety of vehicle monitoring and positioning methods have been realized, such as wheel revolution method, landmark matching method, and lidar-assisted positioning. These methods improve positioning accuracy through multi-sensor fusion and overcome the problems of GNSS signal monitoring blind spots, loss or occlusion. Especially in environments where GNSS signals are weak or lost, such as urban high-rise dense areas, tunnels, and underground parking lots, autonomous positioning methods based on vehicle self-perception have gradually become a powerful means to solve the problem.

[0003] However, existing technologies still face some challenges. First, when GNSS signals are unstable or completely lost, most existing vehicle positioning technologies rely on inertial navigation systems for short-term autonomous positioning. Inertial navigation systems are limited by sensor error accumulation and drift problems, and their long-term accuracy is difficult to guarantee, especially in complex environments, where errors cannot be corrected in real time. In addition, existing positioning algorithms often perform poorly under high-dynamic conditions, and are prone to large trajectory deviations and excessive positioning errors. In blind areas (such as tunnels or high-rise building obstructions), especially when the signal strength is below the threshold, existing methods have failed to effectively solve the problem of maintaining and correcting positioning accuracy, often resulting in the inability to effectively control the vehicle's positioning error, affecting the safety and reliability of the autonomous driving system. Therefore, how to maintain positioning accuracy through an effective autonomous positioning method when the signal monitoring is blind or lost is a problem that needs to be urgently solved in current technology. Summary of the invention

[0004] In view of the problems existing in the existing blind spot vehicle positioning, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to accurately control and correct the positioning error.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for monitoring vehicle positioning in a blind spot, which includes, when the positioning signal strength is lower than a signal threshold, automatically switching to an autonomous positioning mode, and simultaneously recording first signal data and first angle data; calculating a first trajectory point set based on the first signal data and the first angle data; acquiring first speed data and second angle data and calculating a second trajectory point set, and performing time-series alignment on the second trajectory point set with the first trajectory point set to generate a deviation sequence; establishing a trajectory deviation analysis model, calculating a correction amount based on the statistical characteristics of the deviation sequence, and if the correction amount exceeds the correction threshold, performing real-time correction in combination with the current heading angle to correct the positioning output.

[0007] As a preferred solution of the method for monitoring blind spot vehicle positioning described in the present invention, wherein: the first signal data includes one or more of a pulse signal provided by a vehicle wheel speed sensor and dynamic measurement data provided by an inertial sensor; the first angle data is provided by a steering angle sensor or a steering wheel angle sensor of the vehicle; and the data recording frequency is dynamically adjusted according to the real-time motion state.

[0008] As a preferred solution of the method for positioning a vehicle in a monitoring blind spot described in the present invention, wherein: calculating a first trajectory point set according to the first signal data and the first angle data includes: based on the first signal data, obtaining signal timing data, and determining the constraint relationship of trajectory evolution in combination with the first angle data; determining the constraint relationship of trajectory evolution includes: performing cumulative calculation of the signal timing data in each time period to determine the distance traveled by the vehicle; at the same time, determining the trajectory point position corresponding to the vehicle at each time in combination with the deflection relationship defined by the first angle data, and generating a preliminary first trajectory point set; correlating the vehicle structure parameters to correct the preliminary first trajectory point set, including: obtaining the vehicle structure parameters, the vehicle structure parameters affecting Vehicle turning radius and motion trajectory; adjusting the relative positions of the trajectory points in the preliminary first trajectory point set based on the vehicle structural parameters, and correcting the vehicle's steering path by using the vehicle structural parameters through geometric relationships; correcting the vehicle's steering path by using the vehicle structural parameters through geometric relationships includes: calculating the turning radius of the front wheels based on the first angle data, and the front wheel steering radius determines the vehicle's turning size; according to the vehicle structural parameters, the motion trajectory of the rear wheels is offset relative to the front wheels, and the corrected path of the rear wheels is calculated based on the geometric relationship between the front wheel steering radius and the rear wheel steering path to determine the overall turning trajectory of the vehicle; the first trajectory point set is obtained by correcting the trajectory points in the preliminary first trajectory point set.

[0009] As a preferred solution of the method for positioning a vehicle in a monitoring blind spot described in the present invention, the obtaining of the first speed data and the second angle data and calculating the second trajectory point set includes: obtaining vehicle posture change signal data based on the target motion state data of the vehicle, parsing the vehicle posture change signal data, and extracting target state parameters for characterizing the vehicle posture change; the target state parameters include the yaw angle, pitch angle and roll angle of the vehicle; based on an inertial navigation algorithm, calculating the posture change trajectory of the vehicle using the target state parameters, and selecting the angle data corresponding to the time to be calculated in the posture change trajectory as the second angle data; at the same time, obtaining the angle data corresponding to the vehicle at the time to be calculated based on the target motion state data The invention relates to a method for obtaining a first speed data of the vehicle by combining the vehicle acceleration sensor data and the wheel speed data; establishing a displacement evolution of the vehicle based on the first speed data and the second angle data and combining the target position state of the vehicle, and calculating the motion state of the vehicle in a continuous time segment under the constraint of the displacement evolution; wherein the motion state includes the target displacement amount and the target direction information per unit time; based on the motion state, determining the real-time displacement trajectory of the vehicle by using the time series integral calculation method, and selecting a plurality of discrete trajectory points in the real-time displacement trajectory to form the second trajectory point set.

[0010] As a preferred solution of the monitoring blind spot vehicle positioning method described in the present invention, wherein: the second trajectory point set is time-sequentially aligned with the first trajectory point set to generate a deviation sequence, including: based on the second trajectory point set and the first trajectory point set, time matching and spatial coordinate transformation are performed to obtain matching trajectory point pairs under the same time index: the time matching includes one-to-one matching according to the time index when the timestamp distribution of the first trajectory point set and the second trajectory point set is consistent; when there is a difference in the timestamp distribution of the first trajectory point set and the second trajectory point set, the timestamps are unified by an interpolation method; the spatial coordinate transformation includes: when the first trajectory point set and the second trajectory point set are represented in the same coordinate system, the global coordinates of each trajectory point are used for alignment calculation; when there is a difference in the coordinate system of the first trajectory point set and the second trajectory point set, the coordinate conversion method is used for unification; based on the matched trajectory point pairs, the trajectory deviation corresponding to each moment is calculated, including position deviation, heading deviation and cumulative deviation; the calculated trajectory deviation data is sorted according to the time index to form a complete deviation sequence.

[0011] As a preferred solution of the method for positioning a vehicle in a monitoring blind spot described in the present invention, wherein: based on trajectory deviation data, a trajectory deviation analysis model is constructed to determine the time series characteristics of a trajectory deviation sequence, wherein the trajectory deviation sequence includes lateral offset information and longitudinal displacement information of trajectory points; statistical analysis is performed on the trajectory deviation sequence to extract statistical features for correction calculation, wherein the statistical features include a deviation mean, a deviation variance, and a deviation change rate; and according to the statistical features, lateral offset and longitudinal offset correction amounts are determined.

[0012] As a preferred solution of the method for monitoring blind spot vehicle positioning described in the present invention, when the correction amount does not exceed the correction threshold, the current inertial navigation positioning data is maintained unchanged; when the correction amount exceeds the correction threshold, real-time correction is triggered.

[0013] In a second aspect, the present invention provides a vehicle remote monitoring system, which includes an autonomous positioning mode switching module, which is used to automatically switch to the autonomous positioning mode when the positioning signal strength is lower than the signal threshold, and simultaneously record the first signal data and the first angle data; a first trajectory point calculation module, which is used to calculate the first trajectory point set based on the first signal data and the first angle data; a trajectory alignment module, which is used to obtain the first speed data and the second angle data and calculate the second trajectory point set, and time-series align the second trajectory point set with the first trajectory point set to generate a deviation sequence; a trajectory deviation analysis module, which is used to establish a trajectory deviation analysis model, calculate the correction amount according to the statistical characteristics of the deviation sequence, and if the correction amount exceeds the correction threshold, perform real-time correction in combination with the current heading angle to correct the positioning output.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for positioning a vehicle in a monitored blind spot are implemented as described in the first aspect of the present invention.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for positioning a vehicle in a monitored blind spot are implemented as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: the present invention can ensure that the vehicle can still achieve continuous positioning by relying on the vehicle's internal sensors (such as wheel speed sensors and steering angle sensors) when the GNSS signal cannot provide stable support. It not only maintains the system's adaptive ability, but also continues to provide highly reliable trajectory estimation data in various environments (such as tunnels, underground garages, or gaps between high-rise buildings, etc.); and combines the vehicle's structural parameters with real-time motion data to accurately calculate the vehicle's motion trajectory, thereby reducing the impact of external interference on the positioning system. By associating with the vehicle's wheelbase parameters, this step effectively compensates for the errors that may occur when the vehicle turns, thereby improving the accuracy of trajectory prediction. And the changes in the vehicle's acceleration and angular velocity are incorporated into the trajectory calculation, making the displacement calculation more accurate. In addition, by comparing the deviation between the actual trajectory and the theoretical trajectory, the positioning error can be identified in real time to ensure high accuracy in complex dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 The present invention is a flow chart of a method for monitoring vehicle positioning in blind spots.

[0019] Figure 2 Schematic diagram of the generation of deviation sequence in the vehicle positioning method for monitoring blind spots.

[0020] Figure 3 This is a structural diagram of the vehicle positioning system for monitoring blind spots. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0024] Example 1 Reference Figure 1~Figure 3 , which is the first embodiment of the present invention, and provides a method for monitoring vehicle positioning in a blind spot, such as Figure 1 As shown, including, S1: When the positioning signal strength is lower than the signal threshold, it automatically switches to the autonomous positioning mode and records the first signal data and the first angle data at the same time.

[0025] In the embodiment of the present invention, the strength of the satellite positioning signal is first monitored, and a comparison is made with a preset signal threshold to determine whether to switch to the autonomous positioning mode. The specific process is as follows: The vehicle's GNSS receiver receives satellite signals in real time and calculates the current satellite signal strength. Signal strength usually refers to the power of the signal when it reaches the receiver, usually measured in dBm; the quality of satellite signals is not only measured by strength, but also by signal stability. Stability can be measured by the signal's volatility. If the signal strength is continuously lower than the preset threshold and the signal volatility exceeds the set fluctuation range, the signal is considered unstable.

[0026] For example, a signal stability threshold is set, and combined with the signal fluctuation rate, if the signal strength is continuously lower than the threshold and the fluctuation exceeds the specified range, the current signal is judged to be unavailable. When the signal strength is lower than the signal stability threshold and the fluctuation exceeds the set fluctuation range, the signal is confirmed to be unavailable, and the current vehicle positioning is switched to the autonomous positioning mode.

[0027] When the vehicle's satellite positioning signal cannot provide reliable positioning, it automatically switches to autonomous positioning mode and uses other sensors to provide vehicle motion information, including first signal data and first angle data. These data are crucial for estimating the vehicle's position and trajectory.

[0028] The first signal data includes one or more of a pulse signal provided by a vehicle wheel speed sensor and dynamic measurement data provided by an inertial sensor; the first angle data is provided by a steering angle sensor or a steering wheel angle sensor of the vehicle.

[0029] For example, the position of the vehicle may be estimated by relying on vehicle-mounted sensors (such as wheel speed sensors, IMUs, steering angle sensors, etc.).

[0030] In an embodiment of the present invention, the pulse signal provided by the vehicle wheel speed sensor may be a wheel speed pulse signal, and the first angle data may be steering angle data, wherein the wheel speed sensor generates a pulse signal through the rotation of the wheel, and the pulse frequency is proportional to the vehicle's driving speed. Each pulse represents a fixed angle of wheel rotation, and the number of pulses generated per second is recorded and converted into the vehicle's driving speed in combination with the tire radius or other parameters; the steering angle sensor measures the rotation angle of the vehicle's steering wheel to reflect the vehicle's steering state, and the steering angle is closely related to the change in the vehicle's driving path. The angle change data provided by the steering angle sensor is recorded as a time series. Based on the steering angle, the vehicle's turning radius can be obtained, and then the vehicle's turning path can be calculated.

[0031] For example, when the vehicle is driving, the wheel speed sensor generates 10 pulses per second and the steering angle sensor records a steering angle of 5°. These data are recorded for subsequent autonomous positioning.

[0032] Optionally, an inertial measurement unit contains an accelerometer and a gyroscope for measuring the acceleration and angular velocity of the vehicle. By integrating these data, the change in position of the vehicle can be estimated.

[0033] S2: Calculate a first trajectory point set according to the first signal data and the first angle data; the first trajectory point set is generated by the relationship between the signal data and the angle data, and is associated with the structural parameters of the vehicle.

[0034] Calculating a first trajectory point set according to the first signal data and the first angle data includes: acquiring signal timing data based on the first signal data, and determining a constraint relationship of trajectory evolution in combination with the first angle data.

[0035] In this embodiment, a set of theoretical trajectory points of the vehicle is calculated based on the wheel speed pulse signal and steering angle data of the vehicle, wherein the theoretical trajectory points are preliminary trajectory points calculated based on the relationship between the signal data and the angle data.

[0036] Specifically, signal timing data refers to the time series of pulse signals provided by the wheel speed sensor during vehicle driving. These pulse signals reflect the number of pulses generated when the vehicle travels one unit distance. Signal timing data is used to calculate the vehicle's mileage, where the driving distance is the number of pulses corresponding to each distance traveled by the vehicle. The vehicle's mileage can be calculated by multiplying it by the tire radius.

[0037] The constraint relationship for determining trajectory evolution includes: performing cumulative calculation of the signal time series data time period by time period to determine the distance traveled by the vehicle; and at the same time, combining the deflection relationship defined by the first angle data to determine the trajectory point position corresponding to the vehicle at each moment to generate a preliminary first trajectory point set.

[0038] In this embodiment, the constraint relationship of trajectory evolution refers to how the signal timing data and the angle data work together to generate trajectory points during the vehicle driving process.

[0039] Specifically, the wheel speed signal is integrated time by time to accumulate the vehicle's travel distance. This process combines the pulse frequency of each time period with the tire radius to infer the vehicle's motion state at each moment. Combined with the steering angle data, the vehicle's driving direction at each moment is determined. The steering angle is directly related to the vehicle's driving trajectory, and each turn of the vehicle will affect the evolution of the trajectory.

[0040] It should be noted that the vehicle steering angle data represents the current steering angle of the vehicle (usually in degrees or radians), and the driving direction of the vehicle at the current moment is estimated based on the steering angle of the current period.

[0041] By combining the signal timing data with the steering angle data, a series of theoretical trajectory points are generated according to the vehicle's travel distance and steering angle at each moment, and each trajectory point corresponds to the vehicle's position at a certain moment.

[0042] It can be seen that through the cumulative calculation of signal timing data and combined with the vehicle steering angle data, the system can predict the vehicle's movement trajectory in intervals with weak or no signals. This method avoids the problems of reduced accuracy and failure that may occur in traditional positioning systems that rely on satellite signals.

[0043] Traditional positioning methods, especially when high-precision positioning is required, usually rely on the strength and stability of satellite signals. However, in environments such as underground garages and high-rise buildings, satellite signals may be blocked or interfered with, resulting in reduced positioning accuracy. The present invention fills this gap through data provided by vehicle-mounted sensors (such as wheel speed sensors and steering angle sensors), ensuring that the positioning system can still operate efficiently in complex environments.

[0044] After generating the preliminary set of trajectory points, it is necessary to make corrections based on the structural parameters of the vehicle to ensure that the calculation of the trajectory points is consistent with the physical characteristics of the vehicle: The preliminary first trajectory point set is corrected by associating vehicle structural parameters, including: obtaining vehicle structural parameters, the vehicle structural parameters affecting the vehicle turning radius and motion trajectory; adjusting the relative positions of trajectory points in the preliminary first trajectory point set based on the vehicle structural parameters, and correcting the vehicle's steering path using the vehicle structural parameters through geometric relationships.

[0045] For example, structural parameters of a vehicle, such as a turning radius and a wheelbase, directly affect the geometry of the vehicle's travel path, especially the vehicle's turning path.

[0046] The method of correcting the vehicle's turning path by using the vehicle structural parameters through geometric relationships includes: calculating the turning radius of the front wheels based on the first angle data, and the front wheel steering radius determines the turning size of the vehicle; according to the vehicle structural parameters, the movement trajectory of the rear wheels is offset relative to the front wheels, and the corrected path of the rear wheels is calculated based on the geometric relationship between the front wheel steering radius and the rear wheel steering path to determine the overall turning trajectory of the vehicle.

[0047] The turning radius of a vehicle is closely related to structural parameters such as the vehicle's wheelbase and steering angle, and the precise use of these parameters helps to more accurately correct the vehicle's driving path. By calculating the turning radius of the front and rear wheels, the present invention can accurately correct the geometric relationship of the vehicle's trajectory when turning, thereby improving the accuracy of the calculation results.

[0048] In this embodiment, the turning radius of the vehicle is related to parameters such as the steering angle and wheelbase of the vehicle. The turning radius of the front wheels can be calculated using the steering angle data of the front wheels and the wheelbase of the vehicle, and the trajectory point can be corrected.

[0049] Specifically, the following calculation method can be used: the front wheel turning radius is equal to the quotient of the vehicle wheelbase and the tangent value of the steering angle. In other words, when the vehicle turns, the turning radius of the front wheel will change with the change of the steering angle and wheelbase, which determines the turning radius and turning trajectory of the vehicle.

[0050] In most vehicles, the paths of the front and rear wheels are not exactly the same. The steering path of the rear wheels is usually offset from the path of the front wheels. This offset is closely related to the wheelbase of the vehicle and its steering angle. Based on the relationship between the front wheel steering radius and the wheelbase of the vehicle, the path of the rear wheels can be corrected.

[0051] Specifically, the steering path of the rear wheels will have an offset relative to the steering path of the front wheels, that is, the offset caused by the difference between the steering angles of the rear wheels and the front wheels. By correcting this offset, the steering path of the rear wheels can be accurately calculated, thereby obtaining the overall turning trajectory of the vehicle. The corrected path more accurately reflects the actual driving trajectory of the vehicle when turning.

[0052] The first trajectory point set is obtained by correcting the trajectory points in the preliminary first trajectory point set. The preliminary trajectory point set is geometrically corrected and combined with vehicle structural parameters to obtain a corrected trajectory point set, namely the first trajectory point set, as the final driving path of the vehicle.

[0053] S3: Acquire second angle data, calculate a second trajectory point set, and perform time alignment on the second trajectory point set and the first trajectory point set to generate a deviation sequence.

[0054] Based on the target motion state data of the vehicle, vehicle posture change signal data is obtained, the vehicle posture change signal data is analyzed, and target state parameters for characterizing the vehicle posture change are extracted; the target state parameters include the yaw angle, pitch angle and roll angle of the vehicle; based on the inertial navigation algorithm, the target state parameters are used to calculate the posture change trajectory of the vehicle, and the angle data corresponding to the time to be calculated is selected in the posture change trajectory as the second angle data.

[0055] In this embodiment, based on the target motion state data of the vehicle, three-axis acceleration data and three-axis angular velocity data of the vehicle are collected, and the three-axis acceleration data and the three-axis angular velocity data are analyzed to extract target state parameters for characterizing the change of vehicle posture, wherein the target state parameters include the yaw angle, pitch angle and roll angle of the vehicle, these angles represent the change of the orientation of the vehicle in three-dimensional space, and can reflect the posture adjustment of the vehicle during driving, for example, the yaw angle describes the rotation of the vehicle around the vertical axis, which is usually related to the steering angle of the vehicle, the pitch angle indicates that the vehicle rotates around the lateral axis (usually the front and rear axes of the vehicle), affecting the front and rear tilt of the vehicle, and the roll angle describes the rotation of the vehicle around the longitudinal axis (the left and right direction of the vehicle body), affecting the lateral tilt of the vehicle; based on the inertial navigation algorithm, the inertial measurement data is subjected to posture solution, the real-time posture change parameters of the vehicle are obtained, and the angle data corresponding to the time to be calculated is selected in the posture change trajectory as the second angle data.

[0056] The second angle data usually comes from the vehicle's posture sensor, especially data related to the vehicle's motion state and steering changes. The vehicle's posture change signal data comes from an inertial measurement unit (IMU).

[0057] Furthermore, based on the inertial navigation algorithm, the inertial measurement data is subjected to attitude calculation to obtain the real-time attitude change parameters of the vehicle, including: Use filtering methods (such as Kalman filtering and low-pass filtering) to pre-process signal data to remove high-frequency noise and interference; Based on the attitude angular velocity data, the quaternion method or direction cosine matrix (DCM) can be used for attitude analysis to avoid the singularity in the Euler angle calculation and establish the attitude evolution equation; optionally, in the attitude evolution process, constraints can also be considered, such as driving stability restrictions, and the vehicle's yaw angle, pitch angle and roll angle can be calculated.

[0058] At the same time, based on the target motion state data, vehicle speed signal data is acquired, and the vehicle speed signal data is parsed to extract vehicle angular velocity data to obtain first vehicle speed data.

[0059] The target state parameters of the vehicle are mainly obtained by the inertial navigation algorithm, which uses the acceleration data and angular velocity data obtained by the IMU sensor to calculate the instantaneous attitude of the vehicle. The raw data provided by the acceleration sensor can be used to estimate the instantaneous speed of the vehicle after filtering and denoising; the angular velocity sensor is used to measure the rotation rate of the vehicle around each axis, and combined with the acceleration data, the attitude change of the vehicle can be calculated.

[0060] Based on the first speed data and the second angle data, combined with the target position state of the vehicle, the displacement evolution of the vehicle is established, and under the constraint of the displacement evolution, the motion state of the vehicle in a continuous time segment is calculated; wherein the motion state includes the target displacement amount and target direction information per unit time; based on the motion state, the real-time displacement trajectory of the vehicle is determined using a time series integral calculation method, and a plurality of discrete trajectory points are selected from the real-time displacement trajectory to form the second trajectory point set.

[0061] It should be noted that the displacement evolution of the vehicle is based on the target motion state data of the vehicle, and is combined with the first velocity data and the second angle data to calculate the displacement of the vehicle at a certain moment. In this process, the velocity data (including instantaneous velocity and acceleration data) and the attitude change angle data are used to estimate the displacement and driving direction of the vehicle.

[0062] In an embodiment of the present invention, data such as the vehicle's three-axis acceleration and angular velocity data are obtained to calculate the displacement per unit time. For example, assuming that the vehicle's speed is known, the vehicle's displacement can be approximately determined within a small time step; and under the influence of the yaw angle, the vehicle's direction of movement changes, and the lateral and longitudinal displacements of the vehicle under the action of the yaw angle can be obtained.

[0063] Optionally, the motion of the vehicle is affected not only by the longitudinal velocity, but also by the lateral angle. Assuming that the vehicle has a certain lateral acceleration (e.g. centrifugal force caused by a bend in the road), the vehicle's motion trajectory needs to be corrected based on this. Using the vehicle's roll angle and pitch angle, the vehicle's motion trajectory on uneven roads or slopes can be further adjusted. Usually, these angle changes are obtained through acceleration sensors and used to fine-tune the vehicle's driving trajectory.

[0064] In each time segment, the vehicle's current speed and angle data are used to update the vehicle's position and direction. Using the vehicle's displacement evolution, the target displacement and target direction information at each moment can be updated and calculated. For example, the new position after a time segment is obtained based on the current position and the calculated lateral and longitudinal displacements.

[0065] In order to accurately obtain the displacement trajectory of the vehicle, the time series integration calculation method can be used. This is a method of calculating the motion state of the vehicle in different time segments by step-by-step integration: For each time segment, the displacement and direction of the vehicle are calculated, and the position of the vehicle is updated according to the motion state. The updating process is as follows: calculate the speed and angle data in each time segment; update the motion state of the vehicle in the time segment based on the current speed and angle data; calculate the displacement using the aforementioned displacement evolution, and update the current position of the vehicle; gradually calculate the motion trajectory of the vehicle by accumulating the displacement of each time segment; select several discrete trajectory points in the trajectory based on the calculated real-time displacement trajectory.

[0066] It should be noted that when establishing displacement evolution, certain constraints need to be considered to ensure that the calculation results are consistent with the actual situation. Common constraints include: the acceleration and speed of the vehicle have physical maximum values, and these limitations need to be considered in the model; the steering angle of the vehicle cannot exceed a certain maximum value, and the vehicle's yaw angle, pitch angle and other attitude angles are also subject to certain restrictions; or environmental factors such as road friction and slope will affect the vehicle's motion state and need to be included in the model as constraints. Through these constraints, it can be ensured that the established displacement evolution can be effectively applied in different driving environments.

[0067] The second set of trajectory points is time-aligned with the first set of trajectory points, such as Figure 2 As shown, generating the deviation sequence includes the following steps: Based on the second trajectory point set and the first trajectory point set, time matching and space coordinate transformation are performed to obtain matching trajectory point pairs under the same time index.

[0068] It should be noted that the purpose of temporal alignment is to arrange the trajectory data from different sources (the second trajectory point set and the first trajectory point set) in time order for further deviation analysis. This process can be divided into two parts: time matching and spatial coordinate transformation.

[0069] Time matching is the core part of timing alignment. The points in the trajectory point set are matched by timestamps so that the trajectory point of each timestamp can be matched with the trajectory point in another set.

[0070] The time matching includes performing one-to-one matching according to the time index when the timestamp distributions of the first trajectory point set and the second trajectory point set are consistent; when the timestamp distributions of the first trajectory point set and the second trajectory point set are different, unifying the timestamps by using an interpolation method; In this embodiment, time matching is mainly divided into the following two situations: Matching when timestamp distribution is consistent: If the timestamp distribution of the first trajectory point set and the second trajectory point set is consistent (that is, their timestamp values ​​are strictly corresponding), they can be matched in a one-to-one correspondence manner. Each trajectory point directly corresponds to the trajectory point with the same timestamp in the other set according to its timestamp. For example, assuming the timestamp of the second trajectory point set is , the timestamp of the first trajectory point set is also ,but The trajectory point at The trajectory points at , and so on.

[0071] Interpolation matching when timestamp distribution is inconsistent: If there are differences in the distribution of timestamps (for example, the timestamps of the second set of trajectory points are not synchronized with the timestamps of the first set of trajectory points, and the intervals are different), an interpolation method is needed to unify the timestamps. This is usually done through linear interpolation or other advanced interpolation algorithms (such as spline interpolation, Lagrange interpolation, etc.). The purpose of interpolation is to solve the problem of timestamp mismatch caused by inconsistent sensor sampling frequencies or external factors. For example, suppose the timestamps of the first set of trajectory points are , and the timestamp of the second trajectory point set is ,because There is no corresponding timestamp, so the linear interpolation method is used. and Insert a new trajectory point between The trajectory points at the moment, thereby achieving the alignment of timestamps.

[0072] Spatial coordinate transformation refers to aligning the coordinates of the trajectory points of two trajectory sets when the coordinate systems are different. The key to this step is to use appropriate transformation methods based on the coordinate system differences of the trajectory points to map them to the same coordinate system for matching.

[0073] Among them, the spatial coordinate transformation includes: when the first trajectory point set and the second trajectory point set are represented in the same coordinate system, the global coordinates of each trajectory point are used for alignment calculation; when there is a difference between the coordinate systems of the first trajectory point set and the second trajectory point set, a coordinate conversion method is used to unify them.

[0074] In this embodiment, if the first trajectory point set and the second trajectory point set are expressed in the same coordinate system (for example, both use the global coordinate system or the local coordinate system), the alignment calculation can be performed directly using the global coordinate system. In this case, it is only necessary to directly use the coordinates of each trajectory point for matching to ensure that the trajectory points under the same time index have the same geographical location.

[0075] When there is a difference in the coordinate systems of the first trajectory point set and the second trajectory point set, a coordinate transformation method is required to unify them. Common coordinate transformation methods include: Translation transformation: If the two coordinate systems are only different in position, they can be aligned through translation transformation. For example, the origin of one trajectory point set can be translated to the origin position of the other trajectory point set. Rotation transformation: If there is a rotation difference between the two coordinate systems, coordinate transformation is required through a rotation matrix. The rotation matrix can rotate a point in one coordinate system to another coordinate system. Affine transformation: Affine transformation can include operations such as translation, rotation, and scaling at the same time, and is suitable for more complex coordinate system transformations.

[0076] Based on the matched trajectory point pairs, the trajectory deviation corresponding to each moment is calculated, including position deviation, heading deviation and cumulative deviation; the calculated trajectory deviation data is sorted by time index to form a complete deviation sequence.

[0077] In the embodiment of the present invention, the position deviation refers to the difference in position between the second trajectory point and the first trajectory point at each timestamp. For each matching trajectory point pair, the position deviation is calculated by the Euclidean distance; the heading deviation refers to the difference in direction between the second trajectory point and the first trajectory point. The heading is usually determined by the direction of movement or the steering angle of the vehicle, so the heading deviation can be obtained by calculating the difference in heading angles between the two. The cumulative deviation is the cumulative sum of the position deviations, reflecting the overall deviation of the two trajectories. Usually, the cumulative deviation gradually increases over time, so the positioning accuracy of long-term operation can be evaluated by the cumulative deviation.

[0078] S4: Establish a trajectory deviation analysis model, calculate the correction value according to the statistical characteristics of the deviation sequence, and if the correction value exceeds the correction threshold, perform real-time correction in combination with the current heading angle to correct the positioning output.

[0079] In the vehicle positioning system based on inertial navigation, the accuracy of the trajectory is crucial. In order to ensure that the system can provide accurate positioning information, it is necessary to continuously analyze and correct the deviation. Specifically, during the trajectory generation process, due to the influence of various factors, there is often a certain error between the actual trajectory of the vehicle and the theoretical trajectory. In order to reduce the impact of these errors on the positioning accuracy, it is necessary to analyze the trajectory deviation in real time and make dynamic corrections based on the analysis results.

[0080] Based on the above trajectory deviation data, a trajectory deviation analysis model is constructed to determine the time series characteristics of the trajectory deviation sequence, wherein the trajectory deviation sequence includes the lateral offset information and the longitudinal displacement information of the trajectory point; the trajectory deviation sequence is statistically analyzed to extract statistical features for correction calculation, wherein the statistical features include the deviation mean, the deviation variance, and the deviation change rate; and the lateral offset and longitudinal offset correction amounts are determined based on the statistical features.

[0081] Among them, lateral offset refers to the lateral error between the actual driving path of the vehicle and the predetermined trajectory, which is usually caused by steering angle, wheel speed error or external environmental factors (such as road conditions); longitudinal displacement refers to the deviation of the vehicle along the driving direction.

[0082] In the embodiment of the present invention, in order to better perform deviation correction, it is first necessary to perform statistical analysis on the trajectory deviation data, including the deviation mean, deviation variance and deviation change rate.

[0083] The deviation mean represents the average value of the trajectory deviation within a certain time range, reflecting the deviation level of the vehicle's overall trajectory from the target trajectory; the deviation variance measures the volatility of the trajectory deviation, reflecting the magnitude of the change in the vehicle's trajectory deviation over a period of time; the deviation change rate represents the speed of change of the deviation per unit time, and is used to judge the changing trend of the trajectory error.

[0084] By comprehensively considering multiple statistical features such as deviation mean, variance, and rate of change, the vehicle's trajectory deviation can be identified and corrected more accurately, avoiding the deviation caused by a single statistical feature.

[0085] Based on the deviation characteristics obtained by statistical analysis, a correction calculation method can be further established. The correction reflects the cumulative effect caused by trajectory deviation and is used to dynamically adjust the output of the inertial navigation system.

[0086] The correction amount is calculated as follows: based on the trajectory deviation analysis model, the lateral offset correction amount and the longitudinal offset correction amount at the current moment are calculated respectively, which are used to characterize the cumulative deviation output by the inertial navigation system; when the lateral offset correction amount and the longitudinal offset correction amount do not exceed the corresponding correction threshold, the current inertial navigation positioning data is maintained unchanged; when one or more of the lateral offset correction amount and the longitudinal offset correction amount exceed the corresponding correction threshold, a real-time correction is triggered.

[0087] The correction is essentially a correction for the vehicle's current deviation.

[0088] It should be noted that lateral deviation correction usually has a higher priority because it directly affects the driving stability of the vehicle.

[0089] Furthermore, the correction threshold is usually set through statistical analysis of historical trajectory data, and the correction threshold should be adjusted according to the following factors: If the system has a higher tolerance for errors, a higher correction threshold can be set. Conversely, if the system requires high-precision positioning, a lower correction threshold should be set. Different application scenarios have different requirements for correction thresholds. For example, when driving at high speeds, even small deviations may accumulate quickly, so a lower correction threshold is required; while when driving at low speeds or parking, larger deviations can be tolerated.

[0090] Once the correction exceeds the preset correction threshold, the current heading angle is combined with real-time correction to correct the positioning output. For example, if the vehicle has a lateral deviation (such as the vehicle deviates from the predetermined lane), the heading angle is used to determine the angle difference between the actual turning direction of the vehicle and the predetermined trajectory. When the vehicle's heading angle is inconsistent with the predetermined heading, the correction of the lateral deviation will be adjusted according to the vehicle's current heading.

[0091] For example, if the vehicle deviates from the lane and the current heading angle deviates greatly from the direction angle of the lane centerline, the correction strategy will focus on adjusting the vehicle's offset position and correcting it toward the lane centerline. At this time, the correction amount needs to adjust the offset direction and correction amplitude according to the change in the heading angle.

[0092] For longitudinal offset, especially when the vehicle has deviated significantly along the road, the heading angle will affect the vehicle's driving path. In the longitudinal offset correction, if the vehicle's heading angle changes significantly (such as a sharp turn), the correction amount will be dynamically adjusted based on the change.

[0093] In order to ensure the smoothness and effectiveness of the correction process, a dynamic sliding window model is used to process the trajectory data: During the correction process, the sliding window model pays special attention to the correction of lateral offset. Lateral offset is usually caused by steering angle error or vehicle dynamics. When the correction exceeds the correction threshold, the position is corrected according to the real-time data of the inertial navigation system (INS), and the corrected position information is updated to the vehicle's real-time positioning system to ensure the accuracy of the position output.

[0094] Furthermore, this embodiment also provides a vehicle remote monitoring system, such as Figure 3 As shown, including: The autonomous positioning mode switching module 100 is used to automatically switch to the autonomous positioning mode when the positioning signal strength is lower than the signal threshold, and simultaneously record the first signal data and the first angle data; A first trajectory point calculation module 200, configured to calculate a first trajectory point set according to the first signal data and the first angle data; A trajectory alignment module 300, for acquiring first speed data and second angle data and calculating a second trajectory point set, and performing time sequence alignment on the second trajectory point set and the first trajectory point set to generate a deviation sequence; The trajectory deviation analysis module 400 is used to establish a trajectory deviation analysis model, calculate the correction value according to the statistical characteristics of the deviation sequence, and if the correction value exceeds the correction threshold, perform real-time correction in combination with the current heading angle to correct the positioning output.

[0095] This embodiment also provides a computer device, which is suitable for monitoring the blind spot vehicle positioning method, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the blind spot vehicle positioning method proposed in the above embodiment.

[0096] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0097] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for monitoring the positioning of a vehicle in a blind spot as proposed in the above embodiment is implemented.

[0098] In summary, the present invention can ensure that the vehicle can still achieve continuous positioning by relying on the vehicle's internal sensors (such as wheel speed sensors and steering angle sensors) when the GNSS signal cannot provide stable support. It not only maintains the system's adaptive ability, but also continues to provide highly reliable trajectory estimation data in various environments (such as tunnels, underground garages, or gaps between high-rise buildings, etc.); and combines the vehicle's structural parameters with real-time motion data, and accurately calculates the vehicle's motion trajectory, thereby reducing the impact of external interference on the positioning system. By associating with the vehicle's wheelbase parameters, this step effectively compensates for the errors that may occur when the vehicle turns, and improves the accuracy of trajectory prediction. And the changes in the vehicle's acceleration and angular velocity are incorporated into the trajectory calculation, making the displacement calculation more accurate. In addition, by comparing the deviation between the actual trajectory and the theoretical trajectory, the positioning error can be identified in real time to ensure high accuracy in complex dynamic environments.

[0099] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for positioning a vehicle in a monitoring blind area, characterized in that: include: When the positioning signal strength is lower than the signal threshold, it automatically switches to the autonomous positioning mode and records the first signal data and the first angle data at the same time; Calculate a first trajectory point set according to the first signal data and the first angle data; Acquire first speed data and second angle data and calculate a second trajectory point set, and perform time sequence alignment between the second trajectory point set and the first trajectory point set to generate a deviation sequence; A trajectory deviation analysis model is established, and a correction value is calculated according to the statistical characteristics of the deviation sequence. If the correction value exceeds the correction threshold, a real-time correction is performed in combination with the current heading angle to correct the positioning output.

2. The method for positioning a vehicle in a monitoring blind area according to claim 1, characterized in that: The first signal data includes one or more of a pulse signal provided by a vehicle wheel speed sensor and dynamic measurement data provided by an inertial sensor; The first angle data is provided by a steering angle sensor or a steering wheel angle sensor of the vehicle.

3. The method for positioning a vehicle in a monitoring blind area as claimed in claim 2, characterized in that: Calculating a first trajectory point set according to the first signal data and the first angle data includes: Based on the first signal data, signal timing data is acquired, and combined with the first angle data, a constraint relationship of trajectory evolution is determined; The constraint relationship of determining trajectory evolution includes: performing cumulative calculation of the signal time series data in each time period to determine the distance traveled by the vehicle; and combining the deflection relationship defined by the first angle data to determine the trajectory point position corresponding to the vehicle at each time, and generating a preliminary first trajectory point set; The preliminary first trajectory point set is corrected by associating vehicle structural parameters, including: obtaining vehicle structural parameters, the vehicle structural parameters affecting the vehicle turning radius and motion trajectory; adjusting the relative positions of trajectory points in the preliminary first trajectory point set based on the vehicle structural parameters, and correcting the vehicle's steering path by using the vehicle structural parameters through geometric relationships; correcting the vehicle's steering path by using the vehicle structural parameters through geometric relationships includes: calculating the steering radius of the front wheels according to the first angle data, the front wheel steering radius determines the vehicle's turning size; according to the vehicle structural parameters, the motion trajectory of the rear wheels is offset relative to the front wheels, and the corrected path of the rear wheels is calculated based on the geometric relationship between the front wheel steering radius and the rear wheel steering path to determine the overall turning trajectory of the vehicle; the first trajectory point set is obtained by correcting the trajectory points in the preliminary first trajectory point set.

4. The method for positioning a vehicle in a monitoring blind area as claimed in claim 3, characterized in that: The obtaining of the first speed data and the second angle data and calculating the second trajectory point set comprises: Based on the target motion state data of the vehicle, the vehicle posture change signal data is obtained, and the vehicle posture change signal data is parsed to extract the target state parameters used to characterize the vehicle posture change; the target state parameters include the yaw angle, pitch angle and roll angle of the vehicle; based on the inertial navigation algorithm, the vehicle posture change trajectory is calculated using the target state parameters, and the angle data corresponding to the time to be calculated is selected from the posture change trajectory as the second angle data; at the same time, based on the target motion state data, the vehicle speed signal data is obtained, and the vehicle speed signal data is parsed to extract the vehicle instantaneous speed data; combined with the vehicle acceleration sensor data and the wheel speed data, the first speed data of the vehicle is obtained; Based on the first speed data and the second angle data, combined with the target position state of the vehicle, the displacement evolution of the vehicle is established, and under the constraint of the displacement evolution, the motion state of the vehicle in a continuous time segment is calculated; wherein the motion state includes the target displacement amount and target direction information per unit time; based on the motion state, the real-time displacement trajectory of the vehicle is determined using a time series integral calculation method, and a plurality of discrete trajectory points are selected from the real-time displacement trajectory to form the second trajectory point set.

5. The method for positioning a vehicle in a monitoring blind area according to claim 1, characterized in that: Performing time alignment on the second trajectory point set and the first trajectory point set to generate a deviation sequence includes: Based on the second trajectory point set and the first trajectory point set, time matching and spatial coordinate transformation are performed to obtain matching trajectory point pairs under the same time index: The time matching includes performing one-to-one matching according to time index when the timestamp distributions of the first trajectory point set and the second trajectory point set are consistent; and unifying the timestamps by using an interpolation method when the timestamp distributions of the first trajectory point set and the second trajectory point set are different; The spatial coordinate transformation includes: when the first trajectory point set and the second trajectory point set are represented in the same coordinate system, using the global coordinates of each trajectory point to perform alignment calculation; when there is a difference between the coordinate systems of the first trajectory point set and the second trajectory point set, using a coordinate conversion method to unify them; Based on the matched trajectory point pairs, the trajectory deviation corresponding to each moment is calculated; the calculated trajectory deviation data is sorted by time index to form a complete deviation sequence.

6. The method for positioning a vehicle in a monitoring blind area according to claim 1, characterized in that: Based on the trajectory deviation data, a trajectory deviation analysis model is constructed to determine the time series characteristics of the trajectory deviation sequence, wherein the trajectory deviation sequence includes the lateral offset information and the longitudinal displacement information of the trajectory point; the trajectory deviation sequence is statistically analyzed to extract statistical features for correction calculation, wherein the statistical features include the deviation mean, the deviation variance, and the deviation change rate; and the lateral offset and longitudinal offset correction amounts are determined according to the statistical features.

7. The method for positioning a vehicle in a monitoring blind area according to claim 6, characterized in that: When the correction value does not exceed the correction threshold, the current inertial navigation positioning data is maintained unchanged; when the correction value exceeds the correction threshold, real-time correction is triggered.

8. A vehicle remote monitoring system, based on the monitoring blind spot vehicle positioning method according to any one of claims 1 to 7, characterized in that: Also includes: An autonomous positioning mode switching module, used to automatically switch to the autonomous positioning mode when the positioning signal strength is lower than the signal threshold, and simultaneously record the first signal data and the first angle data; A first trajectory point calculation module, used for calculating a first trajectory point set according to the first signal data and the first angle data; A trajectory alignment module, used to obtain the first speed data and the second angle data and calculate a second trajectory point set, and perform time sequence alignment on the second trajectory point set and the first trajectory point set to generate a deviation sequence; The trajectory deviation analysis module is used to establish a trajectory deviation analysis model, calculate the correction amount according to the statistical characteristics of the deviation sequence, and if the correction amount exceeds the correction threshold, perform real-time correction in combination with the current heading angle to correct the positioning output.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for positioning a vehicle in a monitored blind spot are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for positioning a vehicle in a monitored blind spot are implemented.

Images