Big data based automatic parking method and automated parking system

Abstract

The application provides an automatic parking method and an automatic parking system based on big data, which can improve the automatic parking efficiency of the automatic parking system. The method comprises the following steps: determining target parking data according to cloud driving data, wherein the target parking data is the driver parking data with a parking score higher than automatic parking data; obtaining parking bias information sent by the driver; optimizing the automatic parking data according to the target parking data and the parking bias information; and updating the automatic parking data to the vehicle-mounted parking software of the driver, so that the driver automatically parks through the automatic parking data.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to an automatic parking method and automated parking system based on big data. Background Technology

[0002] Automatic parking refers to a car automatically parking itself without human control. As an important functional module in intelligent driving, it is mainly used in scenarios that assist parking or park on behalf of consumers.

[0003] As consumers use vehicles more frequently and parking spaces become scarce, complex terrains and parking spaces are emerging, leading to more parking scenarios and increased demands for obstacle recognition. This places higher requirements on the performance and functionality of parking systems. Currently, the parking skills of automatic parking assist (APA) systems are still relatively poor compared to experienced drivers, resembling those of novices.

[0004] Therefore, improving APA's automatic parking skills is an urgent problem that needs to be solved. Summary of the Invention

[0005] This invention provides an automatic parking method and an automated parking system based on big data, which are used to improve the automatic parking efficiency of the automatic parking system.

[0006] A first aspect of this invention provides an automated parking method based on big data, comprising: determining target parking data based on cloud-based driving data, wherein the target parking data is driver parking data with a parking score higher than automated parking data; acquiring parking bias information sent by the driver; optimizing the automated parking data based on the target parking data and the parking bias information; and updating the automated parking data to the driver's in-vehicle parking software so that the driver can park automatically using the automated parking data.

[0007] Optionally, determining the target parking data based on cloud-based driving data includes: acquiring the cloud-based driving data from the cloud, which includes all driver parking data and all automated parking data; scoring all driver parking data and all automated parking data according to preset parking dimensions and the weights corresponding to each parking dimension; obtaining the average score for automated parking based on the scores of all automated parking data; and determining the data among all driver parking data that has a score higher than the average score for automated parking as the target parking data.

[0008] Optionally, obtaining the parking bias information sent by the driver includes: sending a parking survey instruction to the driver; receiving survey information sent by the driver in response to the parking survey instruction, the survey information including the parking bias information; and / or, after the driver completes parking using the in-vehicle parking software, receiving evaluation information sent by the driver regarding the current parking, the evaluation information including the parking bias information.

[0009] Optionally, optimizing the automated parking data based on the target parking data and the parking bias information includes: determining parking pattern information based on the target parking data and the parking bias information; determining the parking space recognition information to be collected by the parking camera based on the parking pattern information; adjusting the camera's recognition algorithm model and configuration parameters based on the parking space recognition information; and performing automated parking using the camera's recognition algorithm model and configuration parameters to obtain the optimized automated parking data.

[0010] Optionally, determining the parking space recognition information to be collected by the parking camera based on the parking pattern information includes: determining the confidence level required by the user; determining a confidence interval for the parking pattern information based on the confidence level; and determining the parking space recognition information to be collected by the camera to complete automatic parking based on the confidence interval.

[0011] Optionally, the vehicle identification information includes identification range information, and adjusting the camera's identification algorithm model and configuration parameters based on the parking space identification information includes: determining the minimum identification pixel information required for each deep learning model to output in order to achieve the identification range requirement corresponding to the identification range information; and adjusting the camera's identification algorithm model based on the minimum identification pixel information.

[0012] Optionally, the method further includes: receiving a customized scene parking instruction sent by a first driver, the customized scene parking instruction including image information of a fixed parking space and image information of adjacent parking spaces, the image information including parking space identification information; responding to the customized scene parking instruction, associating and storing the image information of the fixed parking space and the image information of adjacent parking spaces with the parking strategy corresponding to each image information and the identification information of the first driver in the cloud; when the first driver activates the customized scene parking mode during driving, detecting whether the identification information of each parking space on both sides of the vehicle includes target identification information, the target identification information being at least one of the identification information of the fixed parking space and the identification information of the adjacent parking space; if included, sending voice prompt information to the first driver, and performing automatic parking according to the parking strategy corresponding to the target identification information.

[0013] A second aspect of this invention provides an automated parking system, comprising: a determining unit, configured to determine target parking data based on cloud-based driving data, wherein the target parking data is driver parking data with a parking score higher than automated parking data; an acquiring unit, configured to acquire parking bias information sent by the driver; a processing unit, configured to optimize the automated parking data based on the target parking data and the parking bias information; and an updating unit, configured to update the automated parking data to the driver's in-vehicle parking software, so that the driver can automatically park using the automated parking data.

[0014] Optionally, the determining unit is specifically used to obtain the cloud-based driving data from the cloud, the cloud-based driving data including all driver parking data and all automated parking data; to score all driver parking data and all automated parking data according to preset parking dimensions and the weights corresponding to each parking dimension; to obtain the average score of automated parking based on the scores of all automated parking data; and to determine the data among all driver parking data that has a higher score than the average score of automated parking as the target parking data.

[0015] Optionally, the acquisition unit is specifically configured to send a parking survey instruction to the driver; receive survey information sent by the driver in response to the parking survey instruction, the survey information including the parking bias information; and / or, after the driver completes parking using the in-vehicle parking software, receive evaluation information sent by the driver regarding the current parking, the evaluation information including the parking bias information.

[0016] Optionally, the processing unit is specifically used to determine parking pattern information based on the target parking data and the parking bias information; determine the parking space recognition information to be collected by the parking camera based on the parking pattern information; adjust the recognition algorithm model of the camera and the configuration parameters of the camera based on the parking space recognition information; and perform automatic parking through the recognition algorithm model of the camera and the configuration parameters of the camera to obtain the optimized automated parking data.

[0017] Optionally, the processing unit is specifically used to determine the confidence level required by the user; determine a confidence interval for the parking pattern information based on the confidence level; and determine the parking space recognition information that the camera needs to collect to complete automatic parking based on the confidence interval.

[0018] Optionally, the processing unit is specifically used to determine the minimum recognition pixel information required for each deep learning model to achieve the recognition range requirement corresponding to the recognition range information; and to adjust the recognition algorithm model of the camera according to the minimum recognition pixel information.

[0019] A third aspect of the present invention provides a computer device comprising at least one connected processor, memory, and transceiver, wherein the memory is used to store program code, and the processor is used to invoke the program code in the memory to execute the steps of the big data-based automatic parking method described in the first aspect.

[0020] A fourth aspect of the present invention provides a computer storage medium including instructions that, when executed on a computer, cause the computer to perform the steps of the big data-based automatic parking method described in the first aspect.

[0021] In summary, it can be seen that in the embodiments provided by this invention, target parking data is determined based on cloud-based driving data. This target parking data refers to driver parking data where the driver's parking score is higher than that of the automated parking system. Parking bias information sent by the driver is obtained. The automated parking data is optimized based on the target parking data and the parking bias information. The automated parking data is then updated to the driver's in-vehicle parking software, enabling the driver to park automatically using the automated parking data. In these embodiments, the automated parking data is continuously optimized based on driver parking data where the driver's driving skills are superior to the automated parking technology, and the driver's parking bias information. This makes the automated parking system's proficiency in automated parking increasingly approach that of a skilled driver. Attached Figure Description

[0022] Figure 1 A flowchart illustrating the big data-based automatic parking method provided in an embodiment of the present invention;

[0023] Figure 2 is a cloud system framework diagram provided in an embodiment of the present invention;

[0024] Figure 3 This is a virtual structure diagram of an automated parking system provided in an embodiment of the present invention;

[0025] Figure 4 This is a schematic diagram of the hardware structure of a server provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0027] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. The division of modules in this invention is merely a logical division; in practical applications, other division methods may be used. For example, multiple modules may be combined or integrated into another system, or some feature vectors may be ignored or not executed. Additionally, the shown or discussed mutual couplings or direct couplings or communication connections may be through some interface, and the indirect couplings or communication connections between modules may be electrical or other similar forms, none of which are limited in this invention. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed among multiple circuit modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present invention.

[0028] The data parsing method provided in this embodiment of the invention will now be described from the perspective of a data management device. This data parsing device can be a server or a service unit within a server, and there is no specific limitation.

[0029] Please see Figure 1 , Figure 1 A flowchart illustrating the big data-based automatic parking method provided in this embodiment of the invention includes:

[0030] 101. Determine target parking data based on cloud-based driving data;

[0031] In this embodiment of the invention, the driving behavior data of all drivers can be collected through the cloud and analyzed and processed. Please refer to Figure 2A, which is a possible system architecture diagram provided by this embodiment of the invention. Each vehicle, including vehicle 1, vehicle 2, and vehicle 3, is connected to the cloud for communication. That is, each vehicle communicates with the cloud. For example, under the operation of driver 1, vehicle 1 uploads its parking information to the cloud, which stores it. The parking information includes, but is not limited to: 1) vehicle control parameters such as steering wheel angle, throttle opening, vehicle speed, and front and rear wheel angles obtained through vehicle bus communication; 2) positional parameters such as the relative lateral and longitudinal distances to vehicles in adjacent parking spaces obtained through onboard sensors such as cameras, ultrasonic probes, and high-precision positioning equipment in the parking lot; 3) whether the process is manual parking or automatic parking, etc.

[0032] In this embodiment, the specific iterative optimization steps are shown in Figure 2B, including:

[0033] 201. Trajectory pre-aiming processing;

[0034] The parking trajectory is obtained through the parking planning module, the vehicle position information is obtained through the parking positioning module, and the yaw angle, steering wheel angle, and vehicle speed are obtained through the vehicle body signal module. Based on the vehicle-road relative pose relationship, the expected rotation center, rotation radius, and heading angle at the preview point are obtained. This set of expected values ​​is the average expected value between two update times. Based on the assumptions that the longitudinal velocity remains constant and the initial lateral velocity is 0 with uniform acceleration / deceleration at the lane line update time, the vehicle coordinates are calculated after a preview time tp. When selecting the preview point, the principle of minimizing error is prioritized, that is, minimizing the lateral error between the preview point and the expected trajectory, i.e., minimizing the index function.

[0035] 202. Tracking point processing;

[0036] After determining the aiming point, the optimized aiming tracking point trajectory is associated with the rotation center, rotation radius, and desired heading angle.

[0037] 203. Rear wheel steering angle feedforward control;

[0038] 204. Front wheel steering angle feedforward control;

[0039] Feedforward control is an open-loop control mechanism designed to reduce the burden on closed-loop feedback control. It primarily comprises two aspects: 1. Providing a specific front and rear wheel steering angle based on the desired curvature; 2. The desired pose at the aiming point. The pose parameters based on the trajectory curvature depend on factors such as vehicle speed, curvature, and the rate of change of curvature. Based on these factors, the final feedforward steering angle is obtained by combining the desired yaw rate at the aiming point given by trajectory planning.

[0040] 205. Trajectory error calculation;

[0041] The main components of the trajectory error in low-speed following control include lateral deviation and heading angle deviation. Taking the rear axle center point of the vehicle as the origin of the vehicle coordinate system and the geodetic coordinate system where the desired trajectory of the parking function is located as the basic reference system, the deviation between the actual position and the ideal position of the vehicle and the aiming point is calculated. The position includes information such as the rotation center position, rotation radius, and heading angle.

[0042] 206. Trajectory error feedback control;

[0043] When the planned trajectory is the current desired trajectory, and the following trajectory error is detected to be greater than the set threshold for three consecutive task cycles, the trajectory replanning mode is activated. At this time, the trajectory followed by the vehicle is the replanned following trajectory.

[0044] Starting with position constraints and slope constraints at the start and end points, the specifics are as follows:

[0045] Location constraints: planning start point and end point, where the length along the X-axis is the end point position when the replanning is completed. This value is a calibrated value and can be set differently according to the curvature. The distance between the replanning end point and the final expected trajectory can be understood as the convergence distance of the replanning, which is also a calibrated value.

[0046] Slope constraint (first derivative of trajectory): At the starting point, the vehicle's heading is the slope at that point, which is 0; at the ending point, it is designed to have the same slope as the final desired trajectory at the planned ending point, ensuring a smooth switch between the vehicle's replanned trajectory and the final desired trajectory.

[0047] 207. Control quantity constraints;

[0048] 208. Smoothing process.

[0049] After the above algorithm iteration and optimization, control constraints and data smoothing are performed. Control constraints and data smoothing are common methods in existing technologies, and will not be elaborated here.

[0050] Optionally, when the automatic parking system is upgraded, the cloud will send the upgrade information to each vehicle.

[0051] In this embodiment, cloud-based driving data corresponding to all drivers is obtained from the cloud, and target parking data is determined based on the cloud-based driving data. The target parking data refers to driver parking data whose parking score is higher than that of automated parking data. Specifically, the cloud-based driving data is obtained from the cloud, including all driver parking data and all automated parking data. The all driver parking data and all automated parking data are scored according to preset parking dimensions and the weights corresponding to each parking dimension. The preset parking dimensions include at least one or more of the following dimensions: parking speed, parking time, and position / posture at the time of parking completion. An average score for automated parking is obtained based on the scores of all automated parking data. For example, in practical applications, the preset parking dimensions include parking time and position / posture at the time of parking completion, with corresponding weights of 0.4 and 0.6, respectively. If the driver's parking speed... The parking time score is 8 points, and the position and posture at the time of parking completion is 6 points, so the corresponding score is 8*0.4+6*0.6=6.8 points. The parking time score of the automated parking data is 5 points, and the position and posture at the time of parking completion is 9 points, so the corresponding score is 5*0.4+9*0.6=7.4 points. The data that is higher than the average score of the automated parking data among all driver parking data is determined as the target parking data. That is, after obtaining all driver parking data, it is compared and analyzed with the average score of automated parking, and the data that is higher than the average score of automated parking is determined as the target parking data, thereby filtering out the driver manual parking process parameters that are better than the process parameters of automated parking.

[0052] 102. Obtain the parking deviation information sent by the driver;

[0053] During the use of the automatic parking system, after automatic parking is completed, a questionnaire will pop up on the display interface, that is, a parking survey instruction will be sent to the driver. The parking survey instruction can include the driver's satisfaction with the automatic parking system, evaluation of parking speed, parking time and parking completion posture, etc.

[0054] Receive survey information sent by the driver in response to the parking survey instruction, the survey information including the parking bias information;

[0055] And / or,

[0056] After the driver completes parking using the in-vehicle parking software, the system receives evaluation information from the driver regarding the current parking session. This evaluation information includes parking bias information, meaning the driver can proactively provide suggestions for improvement to the automatic parking system.

[0057] 103. Optimize the automated parking data based on the target parking data and the parking bias information;

[0058] After obtaining the target parking data and the parking bias information, the automated parking data is optimized, specifically including the following three steps: Step 1: Compare the unoptimized and the better parking position pre-tracking points, and calculate the deviation between them; Step 2: Based on the target parking data and the parking bias information, use a genetic algorithm or evolutionary algorithm to iteratively optimize the parameters of the pre-tracking points, and compare the score difference between the parameters of the pre-tracking points and the target parking data; Step 3: Repeat Step 2 until the score difference becomes negative, that is, the parameters of the iteratively optimized pre-tracking points are higher than the score of the target parking data.

[0059] Optionally, after obtaining the target parking data and the parking bias information, the automated parking data is optimized based on the target parking data and the parking bias information. Specifically, parking pattern information is determined based on the target parking data and the parking bias information. That is, when the parking pattern information includes parking space recognition distance pattern information, determining the parking pattern information includes calculating the recognition distance of the vehicle recognition parking space frame in the target parking data to obtain recognition distance data that follows a normal distribution; fitting the recognition distance data to obtain a recognition distance distribution function; and using the recognition distance distribution function as the parking space recognition distance pattern information. Then, the parking space recognition information that the parking camera needs to collect is determined based on the parking pattern information. Specifically, the confidence level required by the user is determined; a confidence interval for the parking pattern information is determined based on the confidence level; and the parking space recognition information that the camera needs to collect to complete automated parking is determined based on the confidence interval.

[0060] After obtaining the parking space recognition information, the camera's recognition algorithm model and configuration parameters are adjusted based on this information. For example, if the vehicle recognition information includes recognition range information, the minimum recognition pixel information required for each deep learning model to achieve the recognition range requirement corresponding to the recognition range information is determined. The camera's recognition algorithm model is then adjusted based on this minimum recognition pixel information. After adjusting the recognition algorithm model, automatic parking is performed using the camera's recognition algorithm model and configuration parameters to obtain optimized automated parking data.

[0061] 104. Update the automated parking data to the driver's in-vehicle parking software so that the driver can park automatically using the automated parking data.

[0062] After optimizing the automated parking data, the automated parking data is updated to the driver's in-vehicle parking software so that the driver can park automatically using the automated parking data.

[0063] It should be noted that in this embodiment of the invention, customization is also possible based on the scenario. For example, in practical applications, many drivers have fixed parking spaces, and a faster parking solution can be achieved through customized scenarios. Specifically, a customized scenario parking instruction sent by a first driver is received. The customized scenario parking instruction includes image information of the fixed parking space and image information of adjacent parking spaces, and the image information includes the identification information of the parking spaces. In response to the customized scenario parking instruction, the image information of the fixed parking space and the image information of adjacent parking spaces are associated with the parking strategy corresponding to each image information and the identification information of the first driver and stored in the cloud. When the first driver activates the customized scenario parking mode while driving, it is detected whether the identification information of each parking space on both sides of the vehicle includes target identification information. The target identification information is at least one of the identification information of the fixed parking space and the identification information of the adjacent parking space. If it is included, a voice prompt is sent to the first driver, and automatic parking is performed according to the parking strategy corresponding to the target identification information.

[0064] In this embodiment of the invention, based on parking data of drivers whose driving skills are superior to those of automatic parking technology and the parking preference information of drivers, the automatic parking data is continuously optimized, so that the automatic parking system becomes more and more proficient in automatic parking than a skilled driver.

[0065] The embodiments of the present invention have been described above from the perspective of an automatic parking method based on big data. The embodiments of the present invention will now be described below from the perspective of an automatic parking system.

[0066] Please see Figure 3 , Figure 3 This is a virtual structural diagram of an automated parking system 300 provided in an embodiment of the present invention. The automated parking system 300 includes:

[0067] The determining unit 301 is used to determine target parking data based on cloud-based driving data, wherein the target parking data is driver parking data with a parking score higher than that of automated parking data;

[0068] Acquisition unit 302 is used to acquire parking deviation information sent by the driver;

[0069] Processing unit 303 is used to optimize the automated parking data based on the target parking data and the parking bias information;

[0070] The updating unit 304 is used to update the automated parking data to the driver's in-vehicle parking software so that the driver can park automatically using the automated parking data.

[0071] Optionally, the determining unit 301 is specifically used to obtain the cloud-based driving data from the cloud, the cloud-based driving data including all driver parking data and all automated parking data; to score all driver parking data and all automated parking data according to preset parking dimensions and the weights corresponding to each parking dimension; to obtain the average score of automated parking based on the scores of all automated parking data; and to determine the data among all driver parking data that has a higher score than the average score of automated parking as the target parking data.

[0072] Optionally, the acquisition unit 302 is specifically used to send a parking survey instruction to the driver; receive survey information sent by the driver in response to the parking survey instruction, the survey information including the parking bias information; and / or, after the driver completes parking using the in-vehicle parking software, receive evaluation information sent by the driver on the current parking, the evaluation information including the parking bias information.

[0073] Optionally, the processing unit 303 is specifically used to determine parking pattern information based on the target parking data and the parking bias information; determine the parking space recognition information to be collected by the parking camera based on the parking pattern information; adjust the recognition algorithm model of the camera and the configuration parameters of the camera based on the parking space recognition information; and perform automatic parking through the recognition algorithm model of the camera and the configuration parameters of the camera to obtain the optimized automated parking data.

[0074] Optionally, the processing unit 303 is specifically used to determine the confidence level required by the user; determine the confidence interval for the parking pattern information based on the confidence level; and determine the parking space recognition information that the camera needs to collect to complete automatic parking based on the confidence interval.

[0075] Optionally, the processing unit 303 is specifically used to determine the minimum recognition pixel information required for each deep learning model to achieve the recognition range requirement corresponding to the recognition range information; and to adjust the recognition algorithm model of the camera according to the minimum recognition pixel information.

[0076] In this embodiment of the invention, based on parking data of drivers whose driving skills are superior to those of automatic parking technology and the parking preference information of drivers, the automatic parking data is continuously optimized, so that the automatic parking system becomes more and more proficient in automatic parking than a skilled driver.

[0077] Figure 4 This is a schematic diagram of the server structure of the present invention, as shown below. Figure 4As shown, the server 400 in this embodiment includes at least one processor 401, at least one network interface 404 or other user interface 403, a memory 405, and at least one communication bus 402. The server 400 may optionally include the user interface 403, including a display, keyboard, or clicking device. The memory 405 may include high-speed RAM, or it may also include non-volatile memory, such as at least one disk storage device. The memory 405 stores execution instructions. When the server 400 is running, the processor 401 communicates with the memory 405, and the processor 401 calls the instructions stored in the memory 405 to execute the aforementioned method for pushing product review data. The operating system 404 contains various programs for implementing various basic business operations and handling hardware-based tasks.

[0078] The server provided in this embodiment of the invention has a processor 401 that can execute the operations performed by the automated parking system to realize the automated parking method based on big data. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0079] This invention also provides a computer-readable medium containing computer-executable instructions. The computer-executable instructions enable a server to execute the big data-based automatic parking method described in the above embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0080] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the accompanying drawings of the device embodiments provided by this invention, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0081] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware, and of course, it can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memory, special components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for the present invention, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, portable hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0082] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0083] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state disk (SSD)).

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not deviate from the essence of the corresponding technical solutions.

Claims

1. A big data based automatic parking method, characterized by, The method comprises the following steps: determining target parking data from cloud driving data, the target parking data being driver parking data with a parking score higher than that of automated parking data, wherein the steps include: obtaining the cloud driving data from the cloud, the cloud driving data including all driver parking data and all automated parking data; scoring the all driver parking data and all automated parking data according to preset parking dimensions and weights corresponding to each parking dimension; obtaining an average score of automated parking from the scores of the all automated parking data; determining data in the all driver parking data higher than the average score of automated parking as the target parking data; obtaining parking preference information sent by the driver; optimizing the automated parking data according to the target parking data and the parking preference information; updating the automated parking data to the driver's vehicle-mounted parking software, so that the driver automatically parks by the automated parking data.

2. The automatic parking method according to claim 1, characterized by, The step of obtaining the parking preference information sent by the driver comprises the following steps: sending a parking survey instruction to the driver; receiving survey information sent by the driver in response to the parking survey instruction, the survey information including the parking preference information; and / or, when the driver completes parking using the vehicle-mounted parking software, receiving evaluation information sent by the driver on the current parking, the evaluation information including the parking preference information.

3. The automatic parking method according to claim 1, characterized by, The step of optimizing the automated parking data according to the target parking data and the parking preference information comprises the following steps: determining parking rule information according to the target parking data and the parking preference information; determining parking lot recognition information required to be collected by a parking camera according to the parking rule information; adjusting an identification algorithm model of the camera and configuration parameters of the camera according to the parking lot recognition information; performing automatic parking through the identification algorithm model of the camera and the configuration parameters of the camera to obtain optimized automated parking data.

4. The automatic parking method according to claim 3, characterized by, The step of determining parking lot recognition information required to be collected by a parking camera according to the parking rule information comprises the following steps: determining a confidence level required by a user; determining a confidence interval for the parking rule information according to the confidence level; determining the parking lot recognition information required to be collected by the camera to complete automatic parking according to the confidence interval.

5. The automatic parking method according to claim 3, characterized by, The parking lot recognition information includes recognition range information, and the steps of adjusting the identification algorithm model of the camera and the configuration parameters of the camera according to the parking lot recognition information comprise the following steps: determining minimum recognition pixel information required to be output by each deep learning model to meet a recognition range requirement corresponding to the recognition range information; adjusting the identification algorithm model of the camera according to the minimum recognition pixel information.

6. The automatic parking method according to claim 1, characterized by, The method further comprises the following steps: receiving a customized scene parking instruction sent by a first driver, the customized scene parking instruction including image information of a fixed parking lot and image information of adjacent parking lots, the image information including identification information of the parking lots; In response to the customized scene parking instruction, the image information of the fixed parking space and the image information of the adjacent parking space are stored in the cloud in association with the parking strategy corresponding to each image information and the identification information of the first driver; When the first driver starts the customized scene parking mode during driving, it is detected whether the target identification information is included in the identification information of each parking space on both sides of the vehicle, the target identification information being at least one of the identification information of the fixed parking space and the identification information of the adjacent parking space; If yes, voice prompt information is sent to the first driver, and automatic parking is performed according to the parking strategy corresponding to the target identification information.

7. An automated parking system, characterized in that Comprise: The determination unit is configured to determine target parking data according to cloud driving data, the target parking data being driver parking data with a parking score higher than that of automatic parking data; The acquisition unit is configured to acquire parking preference information sent by the driver; The processing unit is configured to optimize the automatic parking data according to the target parking data and the parking preference information; The updating unit is configured to update the automatic parking data to the vehicle-mounted parking software of the driver, so that the driver automatically parks by the automatic parking data.

8. A computer device, comprising: Comprise: At least one connected processor, memory and transceiver, wherein the memory is used to store program code, and the processor is used to call the program code in the memory to execute the big data-based automatic parking method in any one of claims 1 to 6.

9. A computer storage medium, characterized in that Comprise: Instructions, when the instructions run on a computer, make the computer execute the big data-based automatic parking method in any one of claims 1 to 6.

Images