Parking data acquisition method, vehicle-mounted system and readable storage medium

Abstract

The invention discloses a parking data acquisition method, a vehicle-mounted system and a readable storage medium. The method comprises the following steps: starting to collect parking data when a vehicle environment has available parking space; the parking data comprises vehicle environment information, positioning information and vehicle control information; in response to the fact that the vehicle is static and the vehicle gear is in the parking gear, parking data collection is stopped, and the collection stopping moment is recorded; the parking data are backtracked reversely by taking the collection stopping moment as a reference, and the initial moment when the vehicle enters the current parking space is determined; and taking the first parking data from the initial moment to the collection stopping moment as final parking data. In this way, the accuracy of parking data collection can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of parking data collection, and in particular to a parking data collection method, a vehicle-mounted system, and a readable storage medium. Background Art

[0002] With the development of intelligent driving technology, automated parking systems have gradually become a research hotspot in the automotive industry. Collecting high-quality data during the parking process is crucial to optimizing parking algorithms and improving the accuracy and reliability of the system.

[0003] At present, traditional parking data collection methods often rely on specialized data collection personnel and data collection vehicles. Data personnel conduct special tests on the required scene data and manually trigger the recording and collection of data, which is inefficient and has low accuracy in parking data collection. Summary of the invention

[0004] The present application provides a parking data collection method, a vehicle-mounted system, and a readable storage medium, which can improve the accuracy of parking data collection.

[0005] In a first aspect, the present application provides a parking data collection method, the method comprising: in response to the presence of available parking space in the environment of a vehicle, starting to collect parking data; the parking data includes environmental information, positioning information and vehicle control information of the vehicle; in response to the vehicle being stationary and the vehicle gear being in a parking gear, stopping collecting parking data and recording the time when collection stops; based on the time when collection stops, backtracking the parking data to determine the initial time when the vehicle enters the current parking space; and using the first parking data from the initial time to the time when collection stops as the final parking data.

[0006] The first parking data from the initial moment to the moment when collection stops is used as the final parking data, comprising: performing quality evaluation on the first parking data; and in response to the quality evaluation result satisfying a threshold, using the first parking data as the final parking data.

[0007] Among them, performing quality evaluation on the first parking data includes: performing path smoothness evaluation on the first parking data to obtain a first evaluation value; performing parking accuracy evaluation on the first parking data to obtain a second evaluation value; performing safety evaluation on the first parking data to obtain a third evaluation value; performing parking efficiency evaluation on the first parking data to obtain a fourth evaluation value; performing speed smoothness evaluation on the first parking data to obtain a fifth evaluation value; and obtaining a quality evaluation result according to the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value and the fifth evaluation value.

[0008] The first parking data from the initial moment to the moment when collection is stopped is used as the final parking data, including: performing a redundant path check on the first parking data and removing the redundant paths; and using the first parking data after removing the redundant paths as the final parking data.

[0009] Among them, performing a redundant path check on the first parking data and eliminating redundant paths includes: identifying whether there are two target moments with a vehicle posture difference less than a threshold in the first parking data; if so, eliminating redundant path data between the two target moments.

[0010] Among them, identifying whether there are two target moments in the first parking data whose vehicle posture difference is less than a threshold includes: traversing all vehicle postures in chronological order, determining a target hash value whose difference with the current hash value of the current vehicle posture is less than the threshold; and taking the moment of the target vehicle posture corresponding to the current vehicle posture and the target hash value as the two target moments.

[0011] The method further includes: using a fixed space to store parking data, and when the fixed space is fully occupied, discarding the parking data stored at the earliest time, and storing the parking data at the latest time.

[0012] Among them, after taking the first parking data from the initial moment to the moment of stopping collection as the final parking data, it includes: triggering the EDR mechanism, and sending the EDR recording data and the final parking data to the cloud.

[0013] In a second aspect, the present application provides a vehicle-mounted system, which includes a memory and a processor coupled to the memory, the memory storing at least one computer program, and when the at least one computer program is loaded and executed by the processor, it is used to implement the method provided in the first aspect.

[0014] In a third aspect, the present application provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the method provided in the first aspect.

[0015] The beneficial effects of the present application are as follows: different from the prior art, the parking data collection method, vehicle-mounted system, and readable storage medium provided by the present application start collecting parking data in response to the presence of available parking space in the vehicle's environment; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and the vehicle gear being in the parking gear, the parking data collection stops and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is backtracked to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data, so that the parking data relative to the current parking space can be effectively obtained, which can not only reduce the amount of parking data and reduce the occupancy of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce the error caused by manual triggering of collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0017] Figure 1 It is a flowchart of an embodiment of a parking data collection method provided by the present application;

[0018] Figure 2 is a flow chart of another embodiment of the parking data collection method provided by the present application;

[0019] Figure 3 is a flow chart of another embodiment of the parking data collection method provided by the present application;

[0020] Figure 4 , Figure 5 and Figure 6 It is a schematic diagram of the parking area provided by this application;

[0021] Figure 7 It is a structural schematic diagram of an embodiment of a vehicle-mounted system provided by the present application;

[0022] Figure 8 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It will be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some but not all structures related to the present application are shown in the drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0024] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0025] With the development of intelligent driving technology, automated parking systems have gradually become a research hotspot in the automotive industry. Collecting high-quality data during the parking process is crucial to optimizing parking algorithms and improving the accuracy and reliability of the system.

[0026] At present, traditional parking data collection methods often rely on specialized data collection personnel and data collection vehicles. Data personnel conduct special tests on the required scene data and manually trigger the recording and collection of data, which is inefficient and has low accuracy in parking data collection.

[0027] Based on this, the present application proposes to start collecting parking data in response to the presence of parking space in the vehicle's environment; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and the vehicle gear being in the parking gear, stop collecting parking data and record the time of stopping collection; based on the time of stopping collection, reverse backtrack the parking data to determine the initial time when the vehicle enters the current parking space; use the first parking data from the initial time to the time of stopping collection as the final parking data, so as to effectively obtain the parking data relative to the current parking space, which can not only reduce the amount of parking data and reduce the occupation of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce the error caused by manual triggering of collection. Please refer to the following embodiments for details.

[0028] See also Figure 1 , Figure 1 1 is a flow chart of an embodiment of a parking data collection method provided by the present application. The method comprises:

[0029] Step 11: In response to the presence of available parking space in the environment of the vehicle, start collecting parking data; the parking data includes the vehicle's environment information, positioning information, and vehicle control information.

[0030] In some embodiments, when the vehicle speed is lower than a certain set speed threshold, such as 25 km / h, and the vehicle perception system recognizes that there is a parking space or a marked parking space in the surrounding environment, the parking data collection program starts to work and starts to collect parking data, such as real-time recording of the current environment information around the vehicle, the vehicle's positioning information, and the driver's control status information on the steering wheel and the accelerator and brake pedals.

[0031] In some embodiments, the driver of the vehicle may be a driving expert, and the parking data may be parking expert data. The driver of the vehicle may also be an ordinary driver.

[0032] Step 12: In response to the vehicle being stationary and the vehicle gear being in the parking gear, stop collecting parking data and record the time of stopping collection.

[0033] In some embodiments, when the vehicle is finally stationary and the vehicle gear is in P gear (parking gear), when the environmental perception system determines that the current vehicle position is within the parking area, the current time is obtained and recorded as the end time of the parking process interval (the time to stop collecting data).

[0034] Step 13: Based on the stop collection time, reversely trace the parking data to determine the initial time when the vehicle enters the current parking space.

[0035] In some embodiments, the current location information of the vehicle is obtained, and reverse traceback is performed according to the data storage to obtain the initial time of entering the area where the current parking space is located, and the parking-related record data from the initial time to the end time (the time when collection stops) is used as the valid parking data.

[0036] In some embodiments, when the parking data collection system is started, the parking key data storage space is initialized. In order to improve data storage efficiency and reduce storage space occupancy, a fixed-length data queue FixedQueue is implemented based on an array to perform real-time data storage. The implementation principle of FixedQueue is as follows: apply for an array of fixed size in the memory, establish two pointers, respectively pointing to the starting position and the ending position of the data in the array, and the value of the pointer is specifically calculated by the corresponding virtual pointer value. Each time data is pushed into the array, the virtual position pointer of the ending position is incremented by one, and the actual pointer value of the ending position is the remainder of the virtual pointer value of the ending position to the size of the fixed-length array. When the data in the array is full, before each data is pushed in, the array is first popped out, that is, the virtual pointer of the starting position is incremented by one. Similarly, the actual pointer value of the starting position is the remainder of the corresponding virtual pointer value to the size of the fixed-length array.

[0037] The data stored in each unit in the array include but are not limited to the current vehicle position (x, y, heading), velocity v, acceleration a obtained from the vehicle positioning data, and the steering wheel angle θ and gear of the current vehicle read from the chassis data. Among them, heading represents the heading. In the actual parking process, the vehicle positioning and chassis update frequency is relatively high. During the execution of the program, the data obtained for each frame is not stored in the fixed-length data queue, but the required data is sampled at a certain time interval Δt, so as to use limited capacity to store the key data of the parking process for as long as possible. Then, the longest parking process duration that can be stored in the fixed-length data queue is related to the time interval of the queue size, specifically (N-1)·Δt.

[0038] If the vehicle position in the first frame of data, that is, the data pointed to by the starting position pointer, is located in the area where the current parking space is located, and the data is traversed in reverse order from the data pointed to by the ending position pointer, if the data position where the vehicle position is outside the area where the current parking space is located cannot be found until the starting position pointer is traversed, it indicates that the fixed-length data structure fails to save the complete parking process data, and the data recording fails. If the data position where the vehicle position is outside the area where the current parking space is located can be found, the next position of the data position when it is first found is assigned to the starting position pointer. At this time, the data between the starting position pointer and the ending position pointer is the parking interval data. By using a fixed-length data queue to store parking data, the storage system does not need to perform too many operations such as data migration and recovery, saving storage system resources for the normal operation of the entire vehicle. And by using a fixed-length data queue to store parking data, more parking data closer to the stop collection time can be retained, more parking data that starts to be collected can be eliminated, and subsequent redundant processing operations on parking data can be reduced, thereby improving the overall parking data processing efficiency.

[0039] Step 14: The first parking data from the initial moment to the moment when collection stops is used as the final parking data.

[0040] In this embodiment, in response to the presence of available parking space in the environment of the vehicle, parking data is started to be collected; the parking data includes the environment information, positioning information and vehicle control information of the vehicle; in response to the vehicle being stationary and the vehicle gear being in the parking gear, parking data collection is stopped and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is traced back in reverse to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data, so that the parking data relative to the current parking space can be effectively obtained, which can not only reduce the data volume of parking data and reduce the occupation of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce the error caused by manual triggering of collection.

[0041] See also Figure 2 , Figure 2 1 is a flow chart of another embodiment of the parking data collection method provided by the present application. The method includes:

[0042] Step 21: In response to the presence of available parking space in the environment of the vehicle, start collecting parking data; the parking data includes the vehicle's environment information, positioning information, and vehicle control information.

[0043] Step 22: In response to the vehicle being stationary and the vehicle gear being in the parking gear, stop collecting parking data and record the time of stopping collection.

[0044] Step 23: Based on the stop collection time, reversely trace the parking data to determine the initial time when the vehicle enters the current parking space.

[0045] In some embodiments, steps 21 to 23 may refer to other embodiments and will not be described in detail here.

[0046] Step 24: Perform quality evaluation on the first parking data from the initial moment to the moment when collection stops.

[0047] In some embodiments, a path smoothness evaluation is performed on the first parking data to obtain a first evaluation value.

[0048] In some embodiments, a path smoothness evaluation function may be used to evaluate the path smoothness of the first parking data to obtain a first evaluation value.

[0049] Among them, the path smoothness evaluation function Q path_smoothnessThe design is as follows, which is used to evaluate the smoothness of the trajectory during parking, measure the smoothness of the vehicle parking trajectory, and avoid abrupt direction changes. The smaller the value, the smoother the trajectory.

[0050]

[0051] Among them, θ i represents the steering wheel angle at the i-th time step, and N is the total number of time steps.

[0052] In some embodiments, a parking accuracy evaluation is performed on the first parking data to obtain a second evaluation value.

[0053] In some embodiments, a parking accuracy function may be used to evaluate the parking accuracy of the first parking data to obtain a second evaluation value.

[0054] Among them, the parking accuracy function Q accurancy The design is as follows, which is used to evaluate the accuracy of the final parking position and direction of the vehicle. The final evaluation score is obtained by calculating the Euclidean distance between the parking end point and the target position and considering the final direction error. The target position is obtained by calculating the optimal parking position based on the parking location information when parking is completed.

[0055]

[0056] Where (x N-1 ,y N-1 ),θ N-1 is the actual vehicle position and heading angle at the end of parking, (x target ,y target ),θ target are the target position and heading angle, and α is the weight of the heading angle error.

[0057] In some embodiments, a safety evaluation is performed on the first parking data to obtain a third evaluation value.

[0058] In some embodiments, a safety evaluation function may be used to perform safety evaluation on the first parking data to obtain a third evaluation value.

[0059] Among them, the safety evaluation function Q safety The design is as follows, which is used to evaluate whether the parking trajectory can maintain a safe distance during the entire parking process, which is evaluated by the minimum distance to obstacles at discrete time steps.

[0060]

[0061] Among them, d i Represents the distance between the vehicle and the nearest obstacle at the i-th time point.

[0062] In some embodiments, a parking efficiency evaluation is performed on the first parking data to obtain a fourth evaluation value.

[0063] In some embodiments, a parking efficiency evaluation function may be used to evaluate the parking efficiency of the first parking data to obtain a fourth evaluation value.

[0064] Among them, the parking efficiency evaluation function Q time The design is as follows, characterized by parking time, where Δt is the time interval.

[0065] Q time =(N-1)·Δt.

[0066] In some embodiments, a speed smoothness evaluation is performed on the first parking data to obtain a fifth evaluation value.

[0067] In some embodiments, a speed smoothness evaluation function may be used to evaluate the speed smoothness of the first parking data to obtain a fifth evaluation value.

[0068] Among them, the speed smoothness evaluation function Q speed_smoothness The design is as follows, which is used to measure the smoothness of the overall speed of the vehicle during parking and whether there is a sudden change in speed. The maximum value of the acceleration J is calculated by the finite difference method to evaluate it. The forward difference method is used for the initial point, the backward difference method is used for the end point, and the center difference method is used for the intermediate point to improve the calculation accuracy of the discrete data acceleration.

[0069]

[0070]

[0071] Among them, the speed smoothness evaluation function Q speed_smoothness And the path smoothness evaluation function Q path_smoothness The time discontinuity boundary values ​​after redundant paths are eliminated are ignored in the calculation.

[0072] In some embodiments, the quality evaluation result is obtained according to the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value, and the fifth evaluation value.

[0073] In some embodiments, weights may be set for the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value, and the fifth evaluation value to obtain a quality evaluation result.

[0074] In some embodiments, the overall evaluation function is designed as follows:

[0075]

[0076] Among them, C is the scene complexity factor, whose value is determined by the scene characteristics and is related to factors such as the number of obstacles, the size of the available parking area, and the size of the parking space.

[0077] For example, in a normal parking scenario (few obstacles and ample space), C=1 can be taken; in a medium-complexity scenario (many obstacles and moderate space), C=1.2 can be taken; and in a high-complexity scenario (many obstacles and narrow space), C=1.5 or higher can be taken.

[0078] Step 25: In response to the quality evaluation result satisfying the threshold, the first parking data is used as the final parking data.

[0079] In this embodiment, in response to the presence of available parking space in the environment of the vehicle, parking data is started to be collected; the parking data includes the environment information, positioning information and vehicle control information of the vehicle; in response to the vehicle being stationary and the vehicle gear being in the parking gear, parking data collection is stopped and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is traced back in reverse to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data, so that the parking data relative to the current parking space can be effectively obtained, which can not only reduce the data volume of parking data and reduce the occupation of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce the error caused by manual triggering of collection.

[0080] Furthermore, the parking data is evaluated to increase the quality judgment of the parking data, improve the quality issues of the parking data, avoid low-quality parking data from participating in the subsequent automatic parking function generation, and effectively improve the accuracy of the automatic parking function.

[0081] See also Figure 3 , Figure 3 1 is a flow chart of another embodiment of the parking data collection method provided by the present application. The method includes:

[0082] Step 31: In response to the presence of available parking space in the environment of the vehicle, start collecting parking data; the parking data includes the vehicle's environment information, positioning information, and vehicle control information.

[0083] Step 32: In response to the vehicle being stationary and the vehicle gear being in the parking gear, stop collecting parking data and record the time of stopping collection.

[0084] Step 33: Based on the stop collection time, reversely trace the parking data to determine the initial time when the vehicle enters the current parking space.

[0085] In some embodiments, steps 31 to 33 may refer to other embodiments and will not be described in detail here.

[0086] Step 34: Perform a redundant path check on the first parking data from the initial moment to the moment when collection stops, and remove the redundant paths.

[0087] In some embodiments, considering that repeated operations are likely to occur during parking, resulting in redundant data, it is necessary to perform a redundant path check and eliminate the redundant paths.

[0088] In some embodiments, it is identified in the first parking data whether there are two target moments whose vehicle posture difference is less than a threshold; if so, redundant path data between the two target moments is eliminated.

[0089] In some embodiments, all vehicle postures can be traversed in chronological order to determine a target hash value whose difference from the current hash value of the current vehicle posture is less than a threshold; the moments of the current vehicle posture and the target vehicle posture corresponding to the target hash value are taken as two target moments.

[0090] In some embodiments, a redundant path check is performed on the parking data, that is, to find out whether there are two moments with approximately the same vehicle posture within the interval of the parking data. After finding two moments with similar postures, the intermediate redundant path data is eliminated.

[0091] When searching, in order to improve search efficiency, the search method is as follows:

[0092] Iterate over each data point and compare it with the processed pose data. Specifically:

[0093] 1. For the current pose data point, first generate the corresponding hash value through the custom hash function, and then search the hash table for similar poses.

[0094] 2. If similar pose points that meet the judgment of the custom comparison function are found, they are determined to be similar point pairs that meet the error tolerance range, and their index positions are recorded.

[0095] 3. If no pose point that meets the conditions is found, the current pose data point is stored in the hash table for subsequent search.

[0096] In order to allow a certain error tolerance in pose matching, three error thresholds are defined: x tolerance ,y tolerance and heading tolerance, respectively applied to the x-coordinate, y-coordinate, and heading angle of the pose. These thresholds allow for small differences in pose to accommodate the deviation of floating-point precision and actual application requirements. A custom hash function is used to discretize the vehicle pose and map it to a multidimensional grid. Specifically, the x-coordinate is divided by the x- tolerance Round up, divide the y coordinate by y tolerance Round off, and divide the heading angle by the heading tolerance Round to get the corresponding discretized value x hash ,y hash and heading hash The hash function generates discrete grid coordinates (x, y, and heading) by scaling and rounding them according to the corresponding error thresholds. hash ,y hash ,heading hash ), so that poses with similar x, y, and heading values ​​are mapped to the same or adjacent grid cells. Then, these discrete values ​​are converted to integers and hashed. The hash values ​​are finally passed through x hash ,y hash and heading hash To further ensure the accuracy of posture judgment within the error range, a custom comparison function is defined to accurately determine whether two postures are equal within the error tolerance range when a hash conflict occurs. The custom comparison function is based on the comparison of the absolute difference of x, y and heading with the corresponding error threshold to ensure that each dimension meets the set error condition.

[0097] Step 35: Using the first parking data after eliminating redundant paths as the final parking data.

[0098] In some embodiments, after the redundant paths are eliminated, a quality evaluation is performed on the first parking data after the redundant paths are eliminated, and the first parking data with qualified quality evaluation is used as the final parking data.

[0099] In this embodiment, in response to the presence of available parking space in the environment of the vehicle, parking data is started to be collected; the parking data includes the environment information, positioning information and vehicle control information of the vehicle; in response to the vehicle being stationary and the vehicle gear being in the parking gear, parking data collection is stopped and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is traced back in reverse to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data, so that the parking data relative to the current parking space can be effectively obtained, which can not only reduce the data volume of parking data and reduce the occupation of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce the error caused by manual triggering of collection.

[0100] Furthermore, a redundant check and elimination operation is performed on the parking data to improve the quality of the parking data, avoid duplicate parking data from participating in the subsequent generation of the automatic parking function, and effectively improve the accuracy of the automatic parking function.

[0101] In any of the above embodiments, a fixed space is used to store parking data. When the fixed space is fully occupied, the parking data stored at the earliest time is discarded, and the parking data at the latest time is stored.

[0102] In any of the above embodiments, after the final parking data is determined, the EDR (event data recorder) mechanism is triggered to send the EDR recording data and the final parking data to the cloud. After the parking interval data is scored by the above evaluation function, if the score threshold condition is met, the EDR is triggered to transmit the data in the above data queue and the video stream data of the vehicle back.

[0103] Usually, the data recorded by EDR is short in time and it is difficult to cover the entire parking process. By publishing the fixed-length data queue message in the vehicle's DDS communication data, the DDS communication data of the data time period recorded when the EDR is triggered contains the information in the fixed-length data queue, which can indirectly obtain longer parking process trajectory data. In addition, the parking trajectory quality evaluation is strongly correlated with the scene information. The scene information should be recorded as comprehensively as possible. The video stream data recorded before and after the triggering moment, that is, the time period near the end of parking, can reflect the real parking scene. Combined with the output related to parking perception, that is, the obstacle distribution, the description of the parking space and other related information, the terminal scene information can be converted by coordinate transformation according to the relative relationship between the vehicle posture at the initial moment of parking in the fixed-length data queue and the vehicle posture near the end of parking, and the relevant environment description obtained in the scene is converted to obtain the parking scene information of the complete process, which effectively reduces the size of EDR data and reduces the load of the processor.

[0104] In some embodiments, after obtaining the parking termination time, it is determined whether the vehicle position stored in the first frame of the fixed-length data queue is within the parking space area. In order to adapt to vertical, horizontal and oblique parking spaces, the parking space area is divided as follows: Figure 4 , Figure 5 and Figure 6 , each parameter can be flexibly adjusted according to needs.

[0105] In one application scenario, in response to the presence of parking space in the vehicle's environment, parking data collection begins; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and in the parking gear, parking data collection stops and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is traced back in reverse to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is checked for redundant paths, and redundant paths are eliminated. The first parking data with redundant paths eliminated is evaluated for quality. The EDR mechanism is triggered based on the quality evaluation results, and the EDR recorded data and final parking data are sent to the cloud.

[0106] Among them, a fixed space is used to store parking data. When the fixed space is occupied, the parking data stored at the earliest time is eliminated, and the parking data at the latest time is stored.

[0107] See also Figure 7 , Figure 7 1 is a schematic diagram of the structure of an embodiment of the vehicle-mounted system provided by the present application. The vehicle-mounted system 70 includes a memory 71 and a processor 72 coupled to the memory 71. The memory 71 stores at least one computer program. When the at least one computer program is loaded and executed by the processor 72, it is used to implement the following method:

[0108] In response to the presence of available parking space in the vehicle's environment, parking data collection begins; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and the vehicle gear being in the parking gear, parking data collection stops and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is backtracked to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data.

[0109] In some embodiments, when at least one computer program is loaded and executed by the processor 72 , it is also used to implement the following method: perform quality evaluation on the first parking data; in response to the quality evaluation result satisfying a threshold, use the first parking data as the final parking data.

[0110] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: performing a path smoothness evaluation on the first parking data to obtain a first evaluation value; performing a parking accuracy evaluation on the first parking data to obtain a second evaluation value; performing a safety evaluation on the first parking data to obtain a third evaluation value; performing a parking efficiency evaluation on the first parking data to obtain a fourth evaluation value; performing a speed smoothness evaluation on the first parking data to obtain a fifth evaluation value; and obtaining a quality evaluation result based on the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value and the fifth evaluation value.

[0111] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: performing a redundant path check on the first parking data and eliminating the redundant paths; and using the first parking data after eliminating the redundant paths as the final parking data.

[0112] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: identifying whether there are two target moments in the first parking data whose vehicle posture difference is less than a threshold; if so, eliminating redundant path data between the two target moments.

[0113] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: traverse all vehicle postures in chronological order, determine a target hash value whose difference with the current hash value of the current vehicle posture is less than a threshold; and use the moment of the target vehicle posture corresponding to the current vehicle posture and the target hash value as two target moments.

[0114] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: using a fixed space to store parking data, when the fixed space is occupied, eliminating the parking data stored at the earliest time, and storing the parking data at the latest time.

[0115] In some embodiments, after the first parking data from the initial moment to the moment of stopping collection is used as the final parking data, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the following method: triggering the EDR mechanism, sending the EDR recording data and the final parking data to the cloud.

[0116] In some embodiments, when at least one computer program is loaded and executed by the processor 72, it is also used to implement the method of any of the above embodiments.

[0117] See also Figure 8 , Figure 8 1 is a schematic diagram of a computer-readable storage medium according to an embodiment of the present application. The computer-readable storage medium 80 stores computer-executable instructions 81, which are used to implement the following method when executed by a processor:

[0118] In response to the presence of available parking space in the vehicle's environment, parking data collection begins; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and the vehicle gear being in the parking gear, parking data collection stops and the time of stopping collection is recorded; based on the time of stopping collection, the parking data is backtracked to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time of stopping collection is used as the final parking data.

[0119] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: perform quality evaluation on the first parking data; in response to the quality evaluation result meeting a threshold, use the first parking data as the final parking data.

[0120] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: performing a path smoothness evaluation on the first parking data to obtain a first evaluation value; performing a parking accuracy evaluation on the first parking data to obtain a second evaluation value; performing a safety evaluation on the first parking data to obtain a third evaluation value; performing a parking efficiency evaluation on the first parking data to obtain a fourth evaluation value; performing a speed smoothness evaluation on the first parking data to obtain a fifth evaluation value; and obtaining a quality evaluation result based on the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value and the fifth evaluation value.

[0121] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: perform a redundant path check on the first parking data and remove the redundant paths; and use the first parking data after removing the redundant paths as the final parking data.

[0122] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: identifying whether there are two target moments in the first parking data whose vehicle posture difference is less than a threshold; if so, eliminating redundant path data between the two target moments.

[0123] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: traverse all vehicle postures in chronological order, determine a target hash value whose difference with the current hash value of the current vehicle posture is less than a threshold; and use the moment of the target vehicle posture corresponding to the current vehicle posture and the target hash value as two target moments.

[0124] In some embodiments, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: using a fixed space to store parking data, when the fixed space is occupied, eliminating the parking data stored at the earliest time, and storing the parking data at the latest time.

[0125] In some embodiments, after the first parking data from the initial moment to the moment when collection stops is used as the final parking data, when the computer execution instruction 81 is executed by the processor, it is also used to implement the following method: triggering the EDR mechanism, and sending the EDR recording data and the final parking data to the cloud.

[0126] In some embodiments, the computer-executable instruction 81 is also used to implement the method of any of the above embodiments when executed by the processor.

[0127] In summary, the parking data collection method, vehicle-mounted system, and readable storage medium provided by the present application start collecting parking data in response to the presence of available parking space in the environment of the vehicle; the parking data includes the vehicle's environmental information, positioning information, and vehicle control information; in response to the vehicle being stationary and the vehicle gear being in the parking gear, stop collecting parking data and record the time when collection stops; based on the time when collection stops, reverse backtrack the parking data to determine the initial time when the vehicle enters the current parking space; the first parking data from the initial time to the time when collection stops is used as the final parking data, thereby effectively obtaining parking data relative to the current parking space, which can not only reduce the amount of parking data and reduce the occupancy of storage space, but also improve the accuracy of parking data collection and reduce the impact of other parking spaces. Further, automatically triggering parking data collection by judging conditions can also improve the accuracy of parking data collection and reduce errors caused by manual triggering of collection.

[0128] Furthermore, this application has the following advantages:

[0129] Intelligent trigger mechanism: Accurately judge the parking scene, record relevant data only in the parking scene and transmit it back through EDR. This intelligent trigger mechanism significantly improves the efficiency and pertinence of data collection, ensuring the practicality of the collected data.

[0130] Acquisition of high-quality data: Setting trajectory evaluation standards, and systematically evaluating the trajectory (data) during parking effectively eliminates low-quality data, improving the overall reliability and accuracy of the data set. At the same time, it is not limited by drivers and data collection scenarios, and collects highly representative and diverse expert trajectory data, overcoming the problem of insufficient generalization of scenarios and trajectories in traditional methods.

[0131] Real-time data feedback: Once EDR is triggered, it will quickly transmit the required data to provide timely information support for subsequent analysis and improve the ability to obtain information on the parking process.

[0132] EDR data optimization: Considering the hardware memory limitations of the on-board processor, the data recorded by EDR is optimized to cope with the limitations of recording time and data volume. This ensures efficient storage and return of key data under limited memory resources, and enhances the availability and value of data.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only illustrative, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0134] If the integrated units in the above other embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor 10 (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0135] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A parking data collection method, characterized in that: The method comprises: In response to the presence of available parking space in the environment of the vehicle, start collecting parking data; the parking data includes environmental information, positioning information and vehicle control information of the vehicle; In response to the vehicle being stationary and the vehicle gear being in the parking gear, stopping collecting the parking data and recording the time of stopping collecting; Taking the stop collection time as a reference, the parking data is traced back in reverse order to determine the initial time when the vehicle enters the current parking space; The first parking data from the initial moment to the stop collection moment is used as the final parking data.

2. The method according to claim 1, characterized in that The taking the first parking data from the initial moment to the stop collection moment as the final parking data comprises: performing a quality evaluation on the first parking data; In response to the quality evaluation result satisfying a threshold, the first parking data is used as final parking data.

3. The method according to claim 2, characterized in that The performing quality evaluation on the first parking data includes: performing a path smoothness evaluation on the first parking data to obtain a first evaluation value; performing a parking accuracy evaluation on the first parking data to obtain a second evaluation value; performing a safety evaluation on the first parking data to obtain a third evaluation value; performing parking efficiency evaluation on the first parking data to obtain a fourth evaluation value; performing speed smoothness evaluation on the first parking data to obtain a fifth evaluation value; The quality evaluation result is obtained according to the first evaluation value, the second evaluation value, the third evaluation value, the fourth evaluation value, and the fifth evaluation value.

4. The method according to claim 1, characterized in that: The taking the first parking data from the initial moment to the stop collection moment as the final parking data comprises: Performing a redundant path check on the first parking data and eliminating redundant paths; The first parking data after eliminating redundant paths is used as final parking data.

5. The method according to claim 4, characterized in that The performing redundant path check on the first parking data and eliminating redundant paths includes: Identifying in the first parking data whether there are two target moments at which the difference in vehicle posture is less than a threshold; If so, the redundant path data between the two target times are removed.

6. The method according to claim 4, characterized in that The step of identifying whether there are two target moments in the first parking data at which the difference in vehicle posture is less than a threshold value comprises: Traverse all vehicle positions in chronological order and determine the target hash value whose difference with the current hash value of the current vehicle position is less than the threshold; The current vehicle posture and the moment of the target vehicle posture corresponding to the target hash value are taken as the two target moments.

7. The method according to claim 1, characterized in that The method further includes: using a fixed space to store the parking data, and when the fixed space is fully occupied, removing the parking data stored at the earliest time, and storing the parking data at the latest time.

8. The method according to any one of claims 1 to 7, characterized in that: After taking the first parking data from the initial moment to the stop collection moment as the final parking data, the method includes: The EDR mechanism is triggered to send the EDR recording data and the final parking data to the cloud.

9. A vehicle-mounted system, characterized in that: The vehicle-mounted system includes a memory and a processor coupled to the memory, the memory stores at least one computer program, and when the at least one computer program is loaded and executed by the processor, it is used to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1 to 8.

Images