Method, device, electronic device and storage medium for memory parking relocation

Abstract

This invention provides a method, apparatus, electronic device, and storage medium for memory-based parking repositioning. In this method, a continuous path is decomposed into multiple topological path segments. This transforms the complex continuous spatial positioning problem into a discrete path segment sequence matching problem, significantly reducing the complexity and computational load of subsequent positioning searches, resulting in low cost. Furthermore, when a vehicle is detected moving from one topological path segment to the next, the starting pose of the new topological path segment is retrieved from the database, and the vehicle's pose calculated from the trajectory is directly reset using this starting pose, achieving a "start from zero" approach. While the vehicle is traveling within a topological path segment, the lateral distance from the vehicle to the reference path of the current topological path segment is measured at preset intervals to filter and correct the vehicle's pose. This achieves the effect of periodically clearing large errors and suppressing small errors in real time, ensuring high accuracy for full-path repositioning.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and in particular to a method, apparatus, electronic device, and storage medium for memory parking repositioning. Background Technology

[0002] Memory Parking (HPP) is one of the key scenarios for the deployment of autonomous driving technology, and relocalization is the core supporting technology to ensure its reliable operation. This function solves the high-frequency and essential problem of autonomous driving in closed / semi-closed scenarios (GNSS denied scenarios) through a closed-loop logic of learning-memory-reproduction.

[0003] Relocalization refers to the process by which a vehicle determines its precise location on a map through perception or other technical means when reproducing a memorized path. Its importance is comparable to that of an "autonomous driving compass".

[0004] Current relocalization methods mainly include the following three types: The first type is visual SLAM, which has the advantages of strong environmental understanding, but the disadvantages of complex global search computation (O(n³)), low matching rate in degraded scenes, and an 80% reduction in feature points in low-light (<50 lux) scenes; The second type is laser SLAM, which has the advantages of millimeter-level accuracy and all-weather capability, but the disadvantages of initial pose dependence (<3m), high cost (>1500 USD), and point cloud distortion of ±3.2m in scenes with glass curtain wall reflections; The third type is inertial navigation (DR), which has the advantages of autonomous operation and low cost, but the disadvantages of rapid error accumulation (1.5m / minute), with an error reaching 25m after 60 seconds of loss of lock.

[0005] In summary, traditional relocation methods cannot simultaneously meet the requirements of low cost, high accuracy, and strong robustness. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a method, apparatus, electronic device and storage medium for memory parking repositioning, so as to alleviate the technical problem that traditional repositioning methods cannot simultaneously meet the requirements of low cost, high accuracy and strong robustness.

[0007] In a first aspect, embodiments of the present invention provide a method for memory-based parking relocation, comprising: When the vehicle is recording and memorizing the path, it detects turning actions in real time, and uses the pose at the time of detection of the turning action as the turning point. The turning point is used as the boundary, and then the continuous path is decomposed into multiple topological path segments to build a database of topological path segment sequences. When the vehicle travels along the memory path, the vehicle's pose is obtained by calculating the trajectory based on the inertial measurement unit, and the following layered pose correction is performed on the vehicle's pose: Inter-segment global correction: When it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the pose of the vehicle is updated using the starting pose of the new topological path segment. Intra-segment local correction: When the vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

[0008] Furthermore, real-time detection of turning motions, with the pose at the time of detection as the turning point, includes: Calculation formula based on changes in vehicle steering angle Calculate the change in the vehicle's steering angle, where, This indicates the change in the vehicle's steering angle. This represents the angular rate output by the gyroscope. This indicates the zero bias value corresponding to the gyroscope; If the change in the vehicle's steering angle is greater than a preset angle threshold and the duration is greater than a preset duration threshold, then it is determined that the vehicle has performed a turning action, and the position at the time the turning action is detected is taken as the turning point.

[0009] Furthermore, when it is detected that the vehicle is moving from one topological path segment to the next, it includes: When a turning action is detected again, it is determined that the vehicle enters the next topological path segment from one topological path segment.

[0010] Furthermore, by measuring the lateral distance from the vehicle to the reference path of the current topological path segment, the vehicle's pose is filtered and corrected, including: Using the system state calculated from the flight path as the predicted value and the lateral distance as the observation, the system state is filtered and corrected using a Kalman filter algorithm to obtain the corrected system state, wherein the system state includes the vehicle's pose.

[0011] Furthermore, the observation equation for the lateral distance is: ,in, This indicates the lateral distance. Represents the observation matrix. , This indicates the heading angle of the current topological path segment. This indicates the corrected system state. This indicates observation noise.

[0012] Furthermore, the preset interval includes: a preset distance interval or a preset time interval.

[0013] Furthermore, the system state includes: the vehicle's planar coordinates, heading angle, gyroscope zero bias, and accelerometer zero bias; The database includes: the identifier of each topological path segment, the starting pose of each topological path segment, and the length of each topological path segment.

[0014] Secondly, embodiments of the present invention also provide a memory parking repositioning device, comprising: The construction unit is used to detect turning actions in real time when the vehicle is recording and memorizing the path, and to use the pose when the turning action is detected as the turning point, and the turning point as the boundary, thereby decomposing the continuous path into multiple topological path segments and constructing a database of topological path segment sequences. The layered pose correction unit is used to calculate the vehicle's pose based on the trajectory calculation performed by the inertial measurement unit when the vehicle travels along the memory path, and to perform the following layered pose correction on the vehicle's pose: Inter-segment global correction: When it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the pose of the vehicle is updated using the starting pose of the new topological path segment. Intra-segment local correction: When the vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

[0015] Thirdly, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described in any of the first aspects above.

[0016] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the method described in any of the first aspects above.

[0017] In this embodiment of the invention, a method for memory-based parking relocation is provided, comprising: when a vehicle is recording a memory path, real-time detection of turning actions is performed, and the pose at the time of detection of the turning action is used as the turning point, the turning point is used as the boundary, and then the continuous path is decomposed into multiple topological path segments to construct a database of topological path segment sequences; when the vehicle is traveling along the memory path, trajectory calculation is performed based on the inertial measurement unit to obtain the vehicle's pose, and the following hierarchical pose correction is performed on the vehicle's pose: inter-segment global correction: when it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the vehicle's pose is updated using the starting pose of the new topological path segment; intra-segment local correction: when the vehicle is traveling inside a topological path segment, local correction is triggered at preset intervals, and the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment. As described above, the memory parking repositioning method of the present invention decomposes a continuous path into multiple topological path segments. This transforms the complex continuous spatial positioning problem into a discrete path segment sequence matching problem, greatly reducing the complexity and computational load of subsequent positioning searches, resulting in low cost. Furthermore, when a vehicle is detected entering a new topological path segment from one topological path segment, the starting pose of the new topological path segment is retrieved from the database, and the vehicle's pose calculated from the trajectory is directly reset using the starting pose of the new topological path segment, achieving "starting from zero." When the vehicle is traveling within a topological path segment, the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment at preset intervals. This achieves the effect of periodically clearing large errors and suppressing small errors in real time. The use of low-frequency absolute correction and high-frequency relative correction together ensures high accuracy and robustness of the full-path repositioning, alleviating the technical problem that traditional repositioning methods cannot simultaneously meet the requirements of low cost, high accuracy, and strong robustness. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A flowchart of a memory parking repositioning method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the generation of topology path segments provided in an embodiment of the present invention; Figure 3A schematic diagram of the hierarchical correction strategy provided in an embodiment of the present invention; Figure 4 A schematic diagram illustrating the global correction effect provided in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the comparison results of long straight sections provided in an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the comparison results of a curved road scenario provided in an embodiment of the present invention; Figure 7 A schematic diagram of a memory parking repositioning device provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Traditional relocation methods cannot simultaneously meet the requirements of low cost, high accuracy, and strong robustness.

[0022] Based on this, the memory parking relocation method of the present invention decomposes the continuous path into multiple topological path segments. In this way, the complex continuous spatial positioning problem is transformed into a discrete path segment sequence matching problem, which greatly reduces the complexity and computational load of subsequent positioning search and is low in cost. In addition, when the vehicle is detected to enter the next topological path segment from one topological path segment, the database is queried to obtain the starting pose of the new topological path segment, and the vehicle pose calculated from the trajectory is directly reset using the starting pose of the new topological path segment, realizing "starting from zero". When the vehicle is traveling inside a topological path segment, the vehicle pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment at preset intervals. This achieves the effect of periodically clearing large errors and suppressing small errors in real time. The use of low-frequency absolute correction and high-frequency relative correction together ensures high accuracy and good robustness of full path relocation.

[0023] To facilitate understanding of this embodiment, a memory parking repositioning method disclosed in this embodiment of the invention will first be described in detail.

[0024] Example 1: According to an embodiment of the present invention, an embodiment of a memory parking repositioning method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] Figure 1 This is a flowchart of a memory parking repositioning method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps: In step S102, when the vehicle is recording the memory path, it detects turning actions in real time, and uses the pose when the turning action is detected as the turning point, and uses the turning point as the boundary to decompose the continuous path into multiple topological path segments, thereby constructing a database of topological path segment sequences. In step S104, when the vehicle travels along the memory path, the trajectory is calculated based on the inertial measurement unit to obtain the vehicle's pose, and the following layered pose correction is performed on the vehicle's pose: Inter-segment global correction: When a vehicle is detected to be moving from one topological path segment to the next, global correction is triggered. The database is queried to obtain the starting pose of the new topological path segment, and the vehicle's pose is updated using the starting pose of the new topological path segment. Intra-segment local correction: When a vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

[0026] Specifically, this invention proposes a hybrid positioning strategy of "segmented topology path matching + finite correction," which transforms the absolute positioning problem into a path sequence matching problem, suppresses DR (Dead Reckoning) error divergence through topological constraints, and maintains local accuracy using intra-segment correction. In the mapping stage, the continuous path is decomposed into a sequence of topology path segments with turning actions as boundaries. In the positioning stage, layered pose correction is performed, including global correction between segments (triggered during topology segment switching) and local correction within segments (triggered at preset intervals). The absolute positioning problem is transformed into determining whether the current driving path is isomorphic to the mapped path. The aforementioned reference path is the memory path recorded during memory path recording.

[0027] DR (Destination Tracking) is essentially integration. It uses sensor data (acceleration, angular velocity) from inertial measurement units (IMUs) and other sensors to calculate the vehicle's displacement and orientation.

[0028] The sensor's minute bias and random noise are amplified during integration, causing the calculated position error to increase monotonically and unbounded over time—this is known as "error divergence." The longer the run, the less accurate the position becomes.

[0029] "Topology" refers to the connectivity and structure of a vehicle's path (such as sequence and turning points). "Topological constraint" refers to the prior knowledge that the vehicle's actual path must maintain topological consistency with the pre-recorded memory path. Simply put, it's a powerful rule: "The vehicle can only follow the route drawn on the memory path; it cannot wander off." This rule provides a strong anchor for suppressing DR (Depth Reduction) errors.

[0030] Topological constraints work together to suppress error divergence through two mechanisms: global correction between segments and local correction within segments.

[0031] a) Inter-segment global correction: Error reset is performed using the "key points" of the topology (i.e., the starting point / turning point of the topological path segment (the starting point is the turning point)); Correspondence: This directly corresponds to the most core and effective part of "suppressing DR error divergence through topological constraints".

[0032] How it works: A topological path segment is composed of "straight line segments" divided by "turning points." These turning points are the key points in the topological structure. During relocalization, the system determines that the vehicle is about to enter the next topological path segment through pattern recognition (such as detecting that the vehicle is turning).

[0033] Application of constraints: At this point, the topology constraint takes effect—"the vehicle must start from the beginning of the next topology path segment." The system then queries the stored database and directly retrieves the precise coordinates (x, y, θ) of the beginning of the next topology path segment to forcibly reset the pose of the vehicle calculated by the current topology guide (DR), which has accumulated a large error.

[0034] Effect: Inter-segment global correction forces the accumulated error of DR to "zero" at the boundaries (turning points) of each topological path segment, thereby preventing the error from continuing to diverge across the topological path segment. This is a discrete, forced, and global error suppression.

[0035] b) Intra-segment local correction: Fine-tuning of errors within the topology segment; Correspondence: This is a mechanism to help maintain topological constraints within a segment.

[0036] How it works: Within a topological path segment (usually a long straight section), the DR error continues to accumulate. The topological constraint is reflected in the statement that "the vehicle should travel near this straight reference path".

[0037] Application of constraints: The system continuously measures the lateral distance by which the vehicle deviates from this straight reference path using sensors (such as cameras). If the trajectory calculated by the DR (Trajectory Determination) begins to deviate, this observation will detect the anomaly.

[0038] Effect: By using algorithms such as Kalman filtering, this lateral distance observation is used to continuously and incrementally correct the pose estimation of the DR (Depth Trajectory), ensuring that the vehicle trajectory is tightly "attached" to the reference path. This suppresses the rate and magnitude of error divergence within a single segment, providing a continuous, smooth, and local error suppression. It ensures that the DR error does not become excessive when reaching the boundary of the next topological path segment. A simple and reliable observation (the lateral distance from the vehicle to the reference path, i.e., ...) is used. The observation (specifically, the lateral distance between the current calculated position and the topological path segment saved in the memory path) is used to continuously and slightly correct the ever-accumulating errors of the inertial navigation system (DR) to prevent the vehicle from "drifting" within the lane.

[0039] "Suppressing DR error divergence through topological constraints" is a system-level design concept. It mainly achieves a qualitative leap through inter-segment global correction (periodic zeroing), while intra-segment local correction is a quantitative optimization (continuous fine-tuning) based on this. The two complement each other, transforming the DR system from an "open-loop" system where the error can diverge infinitely into a "closed-loop" system where the error is limited to fluctuating around the topological path, thereby achieving high-precision repositioning.

[0040] In this embodiment of the invention, a method for memory-based parking relocation is provided, comprising: when a vehicle is recording a memory path, real-time detection of turning actions is performed, and the pose at the time of detection of the turning action is used as the turning point, the turning point is used as the boundary, and then the continuous path is decomposed into multiple topological path segments to construct a database of topological path segment sequences; when the vehicle is traveling along the memory path, trajectory calculation is performed based on the inertial measurement unit to obtain the vehicle's pose, and the following hierarchical pose correction is performed on the vehicle's pose: inter-segment global correction: when it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the vehicle's pose is updated using the starting pose of the new topological path segment; intra-segment local correction: when the vehicle is traveling inside a topological path segment, local correction is triggered at preset intervals, and the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment. As described above, the memory parking repositioning method of the present invention decomposes a continuous path into multiple topological path segments. This transforms the complex continuous spatial positioning problem into a discrete path segment sequence matching problem, greatly reducing the complexity and computational load of subsequent positioning searches, resulting in low cost. Furthermore, when a vehicle is detected entering a new topological path segment from one topological path segment, the starting pose of the new topological path segment is retrieved from the database, and the vehicle's pose calculated from the trajectory is directly reset using the starting pose of the new topological path segment, achieving "starting from zero." When the vehicle is traveling within a topological path segment, the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment at preset intervals. This achieves the effect of periodically clearing large errors and suppressing small errors in real time. The use of low-frequency absolute correction and high-frequency relative correction together ensures high accuracy and robustness of the full-path repositioning, alleviating the technical problem that traditional repositioning methods cannot simultaneously meet the requirements of low cost, high accuracy, and strong robustness.

[0041] The above provides a brief overview of the memory parking repositioning method of the present invention. The specific details involved are described in detail below.

[0042] In an optional embodiment of the present invention, the turning action is detected in real time, and the pose at the time of detection of the turning action is taken as the turning point, specifically including the following steps: (1) Calculation formula based on the change in vehicle steering angle Calculate the change in the vehicle's steering angle, where, Indicates the change in the vehicle's steering angle. This represents the angular rate output by the gyroscope. This indicates the zero bias value corresponding to the gyroscope; (2) If the change in the vehicle's steering angle is greater than the preset angle threshold and the duration is greater than the preset duration threshold, then the vehicle is determined to have made a turning action, and the position at the time of the detected turning action is taken as the turning point.

[0043] Specifically, in the mapping phase: the process of generating topology path segments is as follows (e.g.) Figure 2 (as shown) The vehicle's turning motion is detected in real time during its movement, and the turning point is used as the boundary of the topological path segment. Turning detection standard: The change in vehicle steering angle (i.e., the integral value of steering angular velocity) > 45° (lasting for more than 0.5 seconds). The formula for calculating the change in vehicle steering angle is as follows: ,in, This represents the angular rate output by the gyroscope. This represents the zero bias value corresponding to the gyroscope.

[0044] The path between two turning maneuvers is defined as a straight segment, and the distance of the straight segment is calculated using the following formula: ,in, Indicates the distance of a straight line segment. This indicates the amount of pulse change within the update cycle. This represents the distance corresponding to each pulse; (Pulse variation): Source: Typically from an encoder mounted on the wheel. The encoder generates a fixed number of pulse signals every time the vehicle travels a fixed distance (e.g., one rotation of the wheel).

[0045] Physical meaning: This represents the total number of pulses generated by the encoder within a data update cycle. It is directly proportional to the distance the vehicle travels during that time period.

[0046] (Calibration factor, distance corresponding to each pulse): Definition: This is a calibration constant that represents the actual physical distance corresponding to each pulse (e.g., 0.01 meters per pulse).

[0047] How to determine: It is obtained through experimental calibration. For example, let the vehicle travel precisely 100 meters and record the total number of pulses. ,but .

[0048] L (distance of the straight segment): Result: Multiplying the pulse count by the calibration factor yields the actual physical length of the straight path the vehicle traveled between the two turning points.

[0049] The distance L between straight segments plays a crucial role in topology relocation schemes and is a core attribute of the topology segment.

[0050] Map building phase: Define the "identity" attributes of topology segments; During mapping, after detecting turning points (such as P1 and P2), the system calculates the length L of the straight line segment using the aforementioned straight line distance calculation formula. Then, it stores key information about a topological path segment in the database, typically including: starting coordinates (the precise pose of turning point P1), length (L), and heading angle (the approximate direction of the straight line segment). In this way, each topological path segment is no longer an abstract line segment, but an entity with measurable geometric properties.

[0051] Relocation phase: used for matching verification and auxiliary correction; Segment Matching Verification: During vehicle repositioning, through DR calculation and turn detection, it may determine that it has entered a certain topological path segment. At this time, the system can check: "Does the travel distance calculated by DR from the start point of entering this segment roughly match the length L of this segment in the database?" If there is a significant discrepancy, a segment matching error may have occurred, and rematching is required. This increases the robustness of the system.

[0052] Triggering Local Correction within a Segment: "Triggered at preset intervals" for correction within a segment. If the "preset interval" is a preset distance interval, the reference for this "preset distance interval" is the total length L of the topology segment (i.e., the topology path segment). For example, the system can be set to perform lateral observation and Kalman filter correction when traveling to positions such as L / 4, L / 2, and 3L / 4. L provides a precise distance reference scale for the correction.

[0053] Assisted DR calculation: The length L itself can be used as a distance observation value to constrain the DR's trajectory calculation and further suppress accumulated errors. Technically, this is achieved by comparing the theoretical length L (true value) of the topology segment with the cumulative travel distance (observation value) calculated by the DR within that segment, and the difference (residual) is used to correct the DR's systematic errors (such as odometer coefficients and sensor zero bias).

[0054] In an optional embodiment of the present invention, when a vehicle is detected entering the next topological path segment from one topological path segment, the following steps are included: When a turning action is detected again, the vehicle is determined to move from one topological path segment to the next.

[0055] In an optional embodiment of the present invention, the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment, including: Using the system state derived from the flight path as the predicted value and the lateral distance as the observation, the system state is filtered and corrected using the Kalman filter algorithm to obtain the corrected system state, which includes the vehicle's pose.

[0056] Specifically, the observation equation for lateral distance is: ,in, Indicates horizontal distance. Represents the observation matrix. , This indicates the heading angle of the current topological path segment. This indicates the corrected system state. This indicates observation noise.

[0057] Specifically, stratified correction strategies, such as Figure 3 As shown: Inter-segment global correction: When a new topology path segment is detected (i.e., when the vehicle enters the next topology path segment from one topology path segment, that is, when a turning action is detected again), the starting pose of the new topology path segment is obtained from the database to reset the current absolute pose (i.e., the pose of the vehicle obtained from the trajectory calculation) and eliminate the previous accumulated error.

[0058] Intra-segment local pose correction: In some application scenarios, the path segment length may exceed 100 meters or even longer. Even using a 1% accuracy rate for trajectory estimation, the error is at least 1 meter, exceeding the capabilities of visual matching correction. Therefore, an intra-segment local correction strategy is introduced, and the optimized model is shown below: ,in, Represents the vehicle's planar coordinates. Indicates the vehicle's heading angle. These represent the zero bias of the gyroscope and accelerometer, respectively. This represents process noise, and the initial value in this system is taken as... .

[0059] In the vehicle coordinate system, the lateral distance This represents the perpendicular distance from the vehicle to the reference path. The specific observation equation is shown below: The Jacobi matrix observed is ,in, This indicates the heading angle of the locked topology path segment. For the purpose of observing noise, we take 0.1m here.

[0060] The process of local correction within a segment will be described below: Step 1: DR State Prediction Action: During vehicle movement, sensors such as the inertial measurement unit (IMU) and wheel speedometers continuously operate. The DR algorithm predicts the vehicle's state vector at time k by integrating the data from these sensors (angular velocity, acceleration).

[0061] State vector ( ): This typically includes the vehicle's position, speed, attitude, and sensor error parameters.

[0062] ,in, Represents the vehicle's global coordinates. Indicates the heading angle (the direction the vehicle is facing). This indicates that the gyroscope has zero bias. This indicates that the accelerometer is at zero bias.

[0063] Output: DR outputs a predicted state vector. However, due to sensor bias and noise, this prediction will accumulate errors over time.

[0064] Step 2: Lateral Distance Observation Triggering conditions: The system triggers an observation and correction once at a preset distance interval (e.g., every 5 meters) or a preset time interval.

[0065] Action: Calculate the vertical (horizontal) distance between the current solution location and the topological path segment saved in the memory path. This value is denoted as the observation. .

[0066] importance: It is an absolute observation that does not diverge over time. If the DR calculation is accurate, the vehicle should always travel on the reference path. It should always be 0. Any The fact that ≠0 directly reveals the error in the DR calculation.

[0067] Step 3: Kalman Filter Fusion and Correction This is the core algorithmic step in the entire correction process. The Kalman filter receives the predicted value of the DR. and sensor observations To achieve optimal fusion.

[0068] Calculate the Kalman gain (Kk): Gain Kk is a weight matrix used to determine whether more trust should be placed in the DR's predictions or in the sensor's observations.

[0069] If the observations are very reliable (e.g., the camera is clear), the gain increases and the observations are trusted more; if the DR has high short-term accuracy, the predictions are trusted more.

[0070] Update state estimate ( ): The filter calculates the residual (Innovation) between the predicted and observed values: Residual = -H×

[0071] Wherein, H (i.e.) ) is the observation matrix, which represents the state vector Mapped to the predicted observation space (i.e., what should the lateral distance be based on the DR state prediction).

[0072] The residuals are then assigned back to the state vector using Kalman gain for correction. = +Kk×(residual), Result: Outputs a corrected, optimal state estimate. This new state estimate incorporates both dynamic information from the DR and calibration information from absolute observations.

[0073] Step 4: Error Compensation and Output Action: Modify the state vector Error estimates in the gyroscope (such as gyroscope zero bias) Feedback is sent to the DR system.

[0074] Objective: Closed-loop control. The DR system uses these corrected error parameters to perform the next round of trajectory estimation, thereby improving the accuracy of the prediction model at the source, rather than simply correcting the output results afterward.

[0075] Output: The final corrected high-precision pose (x,y,θ) is sent to the vehicle control system for accurate path tracking.

[0076] Step 5: Execute repeatedly Action: Repeat steps one through four until the vehicle reaches the end of the current topology segment (the next turning point).

[0077] Mode: This is a closed-loop control process of "prediction → observation → correction → re-prediction".

[0078] Effect: Through this periodic "fine-tuning," the cumulative error of DR is effectively suppressed within a very small range, preventing it from diverging freely. Even on long straight tracks up to 100 meters long, it can maintain extremely high positioning accuracy.

[0079] It is an extremely ingenious design; its dimensions imply the state vector. (That is, the system state) is 5-dimensional, for example Its value [ , ] indicates lateral distance It is strongly dependent only on the vehicle's global position (x, y), and this relationship is related to the heading angle of the reference path. There is a geometric sine / cosine relationship; [0,0,0] represents the lateral distance. With the heading angle in the state vector gyroscope zero bias accelerometer zero bias There is no direct linear relationship. This means that through observation... These states cannot be directly corrected; their estimation mainly depends on the system model and other states.

[0080] In an optional embodiment of the present invention, the preset interval includes: a preset distance interval or a preset time interval.

[0081] In an optional embodiment of the present invention, the system states include: the vehicle's planar coordinates, heading angle, gyroscope zero bias, and accelerometer zero bias; The database includes: the identifier of each topological path segment, the starting pose of each topological path segment, and the length of each topological path segment.

[0082] The global correction effect of the present invention is as follows: Figure 4 As shown.

[0083] The overall performance comparison is shown in the table below: Table 1 Performance Quantitative Comparison

[0084] Performance analysis of long straight track scenario (80m): Test conditions: No GNSS, initial error 0m, vehicle speed 7km / h; comparison results are as follows. Figure 5 As shown.

[0085] Performance analysis of continuous S-curve scenarios: Test conditions: 3 consecutive turns (60°→90°→120°), path length 120m, comparison results as follows. Figure 6 As shown.

[0086] The quantitative comparison results of turning point improvement are shown in the table below:

[0087] Conclusion: The above experiments show that the strategy of periodically resetting the trajectory estimation error by the topological segment boundary and performing lateral correction at fixed intervals to suppress error divergence is effective.

[0088] The inventive points of this invention are as follows: Invention Point 1: A path topology segmentation modeling method based on turning motion: Technical problem to be solved: Traditional solutions attempt to perform one-time positioning in a global coordinate system, which is computationally complex and susceptible to accumulated errors.

[0089] Technical approach: In the mapping phase, the vehicle turning motion (integral of steering angular velocity > 45°) is innovatively used as the natural boundary to decompose the continuous driving path into a series of straight-line-dominated "topological segments". Each segment is abstracted into a basic unit with attributes such as starting coordinates, length, and heading.

[0090] Technical effect: It transforms the complex problem of continuous spatial positioning into a discrete path segment sequence matching problem, which greatly reduces the complexity and computational cost of subsequent positioning search.

[0091] Invention Point Two: A hierarchical correction mechanism combining global inter-segment correction and local intra-segment correction. The technical problem to be solved: Low-cost inertial navigation (DR) errors accumulate over time, and a single correction method cannot balance accuracy and frequency.

[0092] Technical means: Inter-segment global correction: At the boundary (turning point) of topology segment switching, the accumulated error of DR is reset directly with the precise pose of the new segment starting point by querying the database, realizing "starting from zero".

[0093] Intra-segment local correction: Within the segment, at fixed intervals, the lateral distance from the vehicle to the current segment reference path is measured as the observation value, and algorithms such as Kalman filtering are used to make small-range corrections to the DR pose and suppress error divergence.

[0094] Technical effect: It achieves the effect of "periodic zeroing of large errors and real-time suppression of small errors", and uses low-frequency absolute correction and high-frequency relative correction to jointly ensure high accuracy throughout the path.

[0095] Invention Point 3: Lightweight Intra-segment Correction Model Based on Lateral Distance Observation Technical problem to be solved: In scenarios such as long straight roads, the cumulative error of DR is large, which exceeds the effective matching range of vision and other sensors.

[0096] Technical approach: In intra-segment correction, instead of relying on complex global feature matching, it utilizes only the simple and readily available observation of the lateral distance from the vehicle to the reference path. By defining the observation equation and Jacobian matrix, it is deeply integrated with the DR system to achieve efficient filtering.

[0097] Technical benefits: It greatly reduces the requirements for sensors, eliminating the need for expensive LiDAR or high-performance vision systems. Effective correction can be achieved with only low-cost sensors (such as camera ranging), perfectly meeting the requirements of vehicle systems for low cost and high robustness.

[0098] Example 2: This invention also provides a memory parking repositioning device, which is mainly used to execute the memory parking repositioning method provided in Embodiment 1 of this invention. The memory parking repositioning device provided in this invention will be described in detail below.

[0099] Figure 7 This is a schematic diagram of a memory parking repositioning device according to an embodiment of the present invention, as shown below. Figure 7 As shown, the device mainly includes: a construction unit 10 and a layered pose correction unit 20, wherein: The building unit is used to detect turning actions in real time when the vehicle is recording and memorizing the path. The pose at the time of detection of the turning action is used as the turning point, and the turning point is used as the boundary. Then, the continuous path is decomposed into multiple topological path segments, and a database of topological path segment sequences is built. The layered pose correction unit is used to calculate the vehicle's pose based on the inertial measurement unit when the vehicle is traveling along the memory path, and then performs the following layered pose correction on the vehicle's pose: Inter-segment global correction: When a vehicle is detected to be moving from one topological path segment to the next, global correction is triggered. The database is queried to obtain the starting pose of the new topological path segment, and the vehicle's pose is updated using the starting pose of the new topological path segment. Intra-segment local correction: When a vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

[0100] In this embodiment of the invention, a memory parking repositioning device is provided, comprising: when a vehicle is recording a memory path, real-time detection of turning actions is performed, and the pose at the time of detection of the turning action is used as the turning point, the turning point is used as the boundary, and then the continuous path is decomposed into multiple topological path segments to construct a database of topological path segment sequences; when the vehicle is traveling along the memory path, trajectory calculation is performed based on the inertial measurement unit to obtain the vehicle's pose, and the following hierarchical pose correction is performed on the vehicle's pose: inter-segment global correction: when it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the vehicle's pose is updated using the starting pose of the new topological path segment; intra-segment local correction: when the vehicle is traveling inside a topological path segment, local correction is triggered at preset intervals, and the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment. As described above, the memory parking repositioning device of the present invention decomposes a continuous path into multiple topological path segments. This transforms the complex continuous spatial positioning problem into a discrete path segment sequence matching problem, greatly reducing the complexity and computational load of subsequent positioning searches and reducing costs. Furthermore, when a vehicle is detected entering a new topological path segment from one topological path segment, the starting pose of the new topological path segment is retrieved from the database, and the vehicle's pose calculated from the trajectory is directly reset using the starting pose of the new topological path segment, achieving "starting from zero." When the vehicle is traveling within a topological path segment, the vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment at preset intervals. This achieves the effect of periodically clearing large errors and suppressing small errors in real time. The use of low-frequency absolute correction and high-frequency relative correction together ensures high accuracy and robustness of the full-path repositioning, alleviating the technical problem that traditional repositioning methods cannot simultaneously meet the requirements of low cost, high accuracy, and strong robustness.

[0101] Optionally, the building unit is also used to: calculate formulas based on changes in the vehicle's steering angle. Calculate the change in the vehicle's steering angle, where, Indicates the change in the vehicle's steering angle. This represents the angular rate output by the gyroscope. This indicates the zero bias value corresponding to the gyroscope; if the change in the vehicle's steering angle is greater than a preset angle threshold and the duration is greater than a preset duration threshold, then it is determined that the vehicle has performed a turning action, and the pose at the time of detecting the turning action is taken as the turning point.

[0102] Optionally, the hierarchical pose correction unit is also used to: determine when a turning action is detected again, whether the vehicle is moving from one topological path segment to the next.

[0103] Optionally, the hierarchical pose correction unit is also used to: use the system state calculated from the trajectory as the predicted value and the lateral distance as the observation, and use the Kalman filter algorithm to filter and correct the system state to obtain the corrected system state, wherein the system state includes the vehicle's pose.

[0104] Alternatively, the observation equation for the lateral distance is: ,in, Indicates horizontal distance. Represents the observation matrix. , This indicates the heading angle of the current topological path segment. This indicates the corrected system state. This indicates observation noise.

[0105] Optionally, the preset interval includes: a preset distance interval or a preset time interval.

[0106] Optionally, the system status includes: the vehicle's planar coordinates, heading angle, gyroscope zero bias, and accelerometer zero bias; the database includes: the identifier of each topological path segment, the starting pose of each topological path segment, and the length of each topological path segment.

[0107] The device provided in this embodiment of the invention has the same implementation principle and technical effect as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0108] like Figure 8 As shown in the embodiment of this application, an electronic device 600 includes a processor 601, a memory 602, and a bus. The memory 602 stores machine-readable instructions executable by the processor 601. When the electronic device is running, the processor 601 communicates with the memory 602 via the bus, and the processor 601 executes the machine-readable instructions to perform the steps of the memory parking repositioning method described above.

[0109] Specifically, the memory 602 and processor 601 mentioned above can be general-purpose memory and processor, without any specific limitations. When the processor 601 runs the computer program stored in the memory 602, it can execute the above-mentioned memory parking repositioning method.

[0110] The processor 601 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 601 or by instructions in software form. The processor 601 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 602, and processor 601 reads the information from memory 602 and, in conjunction with its hardware, completes the steps of the above method.

[0111] Corresponding to the above-described memory parking relocation method, this application embodiment also provides a computer-readable storage medium storing machine-executable instructions. When the machine-executable instructions are invoked and executed by a processor, the machine-executable instructions cause the processor to perform the steps of the above-described memory parking relocation method.

[0112] The memory parking repositioning device provided in this application embodiment can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this application embodiment are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0113] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0114] For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0115] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0117] If the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the memory parking repositioning method described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0119] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for memory-based parking relocation, characterized in that, include: When the vehicle is recording and memorizing the path, it detects turning actions in real time, and uses the pose at the time of detection of the turning action as the turning point. The turning point is used as the boundary, and then the continuous path is decomposed into multiple topological path segments to build a database of topological path segment sequences. When the vehicle travels along the memory path, the vehicle's pose is obtained by calculating the trajectory based on the inertial measurement unit, and the following layered pose correction is performed on the vehicle's pose: Inter-segment global correction: When it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the pose of the vehicle is updated using the starting pose of the new topological path segment. Intra-segment local correction: When the vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

2. The method according to claim 1, characterized in that, Real-time detection of turning motions, and using the pose at the time of detection of the turning motion as the turning point, including: Calculation formula based on changes in vehicle steering angle Calculate the change in the vehicle's steering angle, where, This indicates the change in the vehicle's steering angle. This represents the angular rate output by the gyroscope. This indicates the zero bias value corresponding to the gyroscope; If the change in the vehicle's steering angle is greater than a preset angle threshold and the duration is greater than a preset duration threshold, then it is determined that the vehicle has performed a turning action, and the position at the time the turning action is detected is taken as the turning point.

3. The method according to claim 1, characterized in that, When it is detected that the vehicle is moving from one topology path segment to the next, the following is included: When a turning action is detected again, it is determined that the vehicle enters the next topological path segment from one topological path segment.

4. The method according to claim 1, characterized in that, The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment, including: Using the system state calculated from the flight path as the predicted value and the lateral distance as the observation, the system state is filtered and corrected using a Kalman filter algorithm to obtain the corrected system state, wherein the system state includes the vehicle's pose.

5. The method according to claim 4, characterized in that, The equation for the observation of the lateral distance is: ,in, This indicates the lateral distance. Represents the observation matrix. , This indicates the heading angle of the current topological path segment. This indicates the corrected system state. This indicates observation noise.

6. The method according to claim 1, characterized in that, The preset interval includes: preset distance interval or preset time interval.

7. The method according to claim 4, characterized in that, The system status includes: the vehicle's planar coordinates, heading angle, gyroscope zero bias, and accelerometer zero bias; The database includes: the identifier of each topological path segment, the starting pose of each topological path segment, and the length of each topological path segment.

8. A device for memory-based parking repositioning, characterized in that, include: The construction unit is used to detect turning actions in real time when the vehicle is recording and memorizing the path, and to use the pose when the turning action is detected as the turning point, and the turning point as the boundary, thereby decomposing the continuous path into multiple topological path segments and constructing a database of topological path segment sequences. The layered pose correction unit is used to calculate the vehicle's pose based on the trajectory calculation performed by the inertial measurement unit when the vehicle travels along the memory path, and to perform the following layered pose correction on the vehicle's pose: Inter-segment global correction: When it is detected that the vehicle enters the next topological path segment from one topological path segment, global correction is triggered, the database is queried to obtain the starting pose of the new topological path segment, and the pose of the vehicle is updated using the starting pose of the new topological path segment. Intra-segment local correction: When the vehicle travels within a topological path segment, local correction is triggered at preset intervals. The vehicle's pose is filtered and corrected by measuring the lateral distance from the vehicle to the reference path of the current topological path segment.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program thereon, characterized in that, The computer program is executed by the processor to perform the method of any one of claims 1 to 7.

Images