Parking control methods, devices, vehicles and storage media

Abstract

This application proposes a parking control method, device, vehicle, and storage medium. The method includes: in response to detecting the existence of a target parking space, sending multimedia data of the parking area to the cloud, wherein the multimedia data is used by the cloud to identify parking decision information of the target parking space, obtaining the parking decision information of the target parking space sent by the cloud, and performing parking control on the vehicle based on the parking decision information of the target parking space. This achieves the recognition of the target parking space by the vehicle and the identification of the parking decision information of the target parking space by the cloud. Since the cloud has stronger computing power and reasoning ability, the accuracy of the parking decision information of the target parking space is improved. Through the collaboration between the vehicle and the cloud, parking safety is improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a parking control method, device, vehicle, and storage medium. Background Technology

[0002] With the increasing popularity of autonomous driving and driver assistance systems, more and more vehicles are equipped with automatic parking technology, which helps drivers find suitable parking spots and complete parking.

[0003] However, as driving scenarios become increasingly complex, finding suitable parking locations and completing parking maneuvers in various special situations, thereby improving parking safety, are technical problems that need to be solved. Summary of the Invention

[0004] This application aims to at least partially address one of the technical problems in the related art.

[0005] Therefore, this application proposes a parking control method, device, vehicle, and storage medium to find a suitable parking location and complete parking, thereby improving parking safety.

[0006] One embodiment of this application proposes a parking control method, including:

[0007] In response to the detection of a target parking space that can be parked, multimedia data of the parking area is sent to the cloud; wherein, the multimedia data is used by the cloud to identify parking decision information of the target parking space;

[0008] Obtain the parking decision information for the target parking space sent from the cloud;

[0009] Based on the parking decision information of the target parking space, the vehicle is controlled for parking.

[0010] Another embodiment of this application provides a parking control device, including:

[0011] The sending module is used to send multimedia data of the parking area to the cloud in response to the detection of a target parking space that can be parked; wherein, the multimedia data is used by the cloud to identify parking decision information of the target parking space;

[0012] The acquisition module is used to acquire the parking decision information of the target parking space sent by the cloud.

[0013] The control module is used to control the parking of the vehicle based on the parking decision information of the target parking space.

[0014] Another embodiment of this application proposes a vehicle including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method described in the foregoing aspect.

[0015] Another embodiment of this application proposes a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the foregoing aspect.

[0016] Another embodiment of this application proposes a computer program product having a computer program stored thereon, which, when executed by a processor, implements the method described in the foregoing aspect.

[0017] The parking control method, device, vehicle, and storage medium proposed in this application, in response to detecting the existence of a target parking space, send multimedia data of the parking area to the cloud. This multimedia data is used by the cloud to identify parking decision information for the target parking space, obtain the parking decision information for the target parking space sent from the cloud, and perform parking control on the vehicle based on the parking decision information for the target parking space. This achieves the goal of identifying the target parking space at the vehicle end and obtaining the parking decision information for the target parking space through cloud identification. Because the cloud has stronger computing power and reasoning capabilities, the accuracy of the parking decision information for the target parking space is improved. Through the collaboration between the vehicle end and the cloud, parking safety is enhanced.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0020] Figure 1 A schematic flowchart illustrating a parking control method provided in an embodiment of this application;

[0021] Figure 2 A flowchart illustrating another parking control method provided in an embodiment of this application;

[0022] Figure 3 A schematic diagram of a parking scenario provided in an embodiment of this application;

[0023] Figure 4 A schematic diagram of multimedia data for a parking area provided in an embodiment of this application;

[0024] Figure 5 This is a schematic diagram of the structure of a parking control device provided in an embodiment of this application;

[0025] Figure 6 This is a schematic diagram of the structure of a vehicle proposed in an embodiment of this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The parking control method, apparatus, vehicle, and storage medium according to embodiments of this application are described below with reference to the accompanying drawings.

[0028] Figure 1 This is a schematic flowchart of a parking control method provided in an embodiment of this application.

[0029] This application example illustrates the use of a parking control method configured in a parking control device. This parking control device can be applied to any vehicle-mounted device so that the vehicle-mounted device can perform parking control functions.

[0030] like Figure 1 As shown, the method may include the following steps:

[0031] Step 101: In response to the detection of a target parking space that can be parked, the multimedia data of the parking area is sent to the cloud.

[0032] The parking area refers to a designated area that provides legal parking space for motor vehicles, including at least one parking space. The multimedia data of the parking area can be at least one of image data and video data, used for cloud-based identification to obtain parking decision information for the target parking space. One implementation method is to use a cloud-based visual detection model to identify the multimedia data of the parking area and obtain parking decision information for each parking space, including the target parking space.

[0033] In one implementation of this application, an image processing algorithm is used to identify parking lines, shapes, colors, etc., in the multimedia data of the parking area, and to identify the existence of a target parking space that can be parked. A target parking space that can be parked means that the parking space is vacant and no other vehicles are parked in the parking space.

[0034] In one implementation of this application, a parking space detection model can be set on the vehicle side. This model is a deep learning model deployed on the vehicle side to perceive whether there are available vacant parking spaces in a parking scenario. As one implementation, multimedia data of the parking area in the parking scenario is collected by sensors on the vehicle side. Based on the parking space detection model set on the vehicle side, the multimedia data of the parking area is identified to obtain the target parking space that can be parked. The target parking space is determined by the vehicle; the vehicle side considers the target parking space to be available. However, due to the limited computing power of the vehicle-side platform, it is impossible to deploy complex models that require significant computing power. Therefore, for complex and special scenarios, the accuracy of parking space identification is limited; that is, the detected target parking space may not actually be available for parking. Because cloud servers possess strong computing power, they can be used to train visual detection models with reasoning capabilities based on data from complex parking scenarios. These models can then be combined with multimedia data from the parking area to identify and reason about the environmental, spatial, and route information surrounding the target parking space. This allows for the determination of parking decision information for the target parking space identified by the vehicle. The parking decision information includes the parking route and whether the target parking space is accessible. The parking route refers to the path the vehicle takes from its current location to the target parking space. If the target parking space is inaccessible, a warning message is included to indicate the presence of obstacles in the target parking space and / or the parking route. As an example, the visual detection model could be a large visual language model.

[0035] Specifically, when the vehicle detects a target parking space that can be parked, it sends multimedia data of the parking area to the cloud.

[0036] Step 102: Obtain parking decision information for the target parking space sent from the cloud.

[0037] In this embodiment, parking decision information is sent to the vehicle based on the communication link established between the vehicle and the cloud.

[0038] Step 103: Based on the parking decision information of the target parking space, perform parking control on the vehicle.

[0039] In this embodiment of the application, when the vehicle receives the parking decision information of the target parking space sent by the cloud, it will determine whether the target parking space is a real parking scenario based on the parking decision information of the target parking space, and then control the vehicle to park based on the parking decision information.

[0040] One implementation method involves determining that a target parking space is unsuitable for parking based on its parking decision information. This means that while the cloud-based visual detection model confirms the target parking space is vacant, it may not actually be suitable for parking. In other words, if the results detected by the vehicle differ from those detected by the cloud, the cloud's detection result takes precedence. Based on the target parking space's parking decision information, a parking warning message is generated. This warning message indicates that there are obstacles on the parking path and / or within the target parking space, meaning the target parking space is unsuitable for parking. This informs the driver of the reason why parking is prohibited and serves as a safety risk warning to the user.

[0041] As an example scenario, in the multimedia data of a parking area, there are obstacles in front of parking space A, such as tree stumps, irregularly shaped objects, potholes, or suspended obstacles. The parking space detection model, based on the multimedia data of the parking area, cannot identify such long-tail situations and considers the parking space a target parking space that can be parked. However, in reality, parking in this space would damage the vehicle due to the obstacle. After the multimedia data of the parking area is transmitted to the cloud, the visual detection model uses its reasoning ability to obtain parking decision information for parking space A. This parking decision information includes the fact that parking space A cannot be parked in. The prompt information is the location of the obstacle in the multimedia data that must be passed when driving into parking space A according to the parking route. The obstacle prevents the vehicle from driving to parking space A, that is, there is no parking route or the parking route has safety hazards, and parking is not possible.

[0042] As another implementation method, based on parking decision information, it is determined that the target parking space is accessible, meaning the results from the cloud and vehicle-side identification are consistent. The parking route corresponding to the target parking space is then determined, and the vehicle is controlled to park in the target space according to the parking route. Specifically, for determining the parking route corresponding to the target parking space, as a first implementation method, the three-dimensional spatial information of the parking area is obtained. This information can be acquired through radar installed on the vehicle. Based on this three-dimensional spatial information, the parking route corresponding to the target parking space is determined. Specifically, by using the distribution of parking spaces and the position and attitude information of already parked vehicles included in the three-dimensional spatial information, a parking route that can be driven to the target parking space is determined, and the vehicle is controlled to automatically park in the target parking space according to the parking route. As a second implementation method, the parking route corresponding to the target parking space is determined based on the parking decision information issued by the cloud. That is, due to the strong computing power of the cloud, in the process of recognizing multimedia data through the visual detection model, the cloud can also calculate the parking route based on the location information of the target parking space and the current location information of the vehicle, and infer that the target parking space is accessible, that is, there are no obstacles on the parking path and / or within the target parking space. Based on the parking route, the vehicle is controlled to automatically drive to the target parking space.

[0043] As one approach, if a target parking space is confirmed to be available, the driver can be alerted that the space is available, allowing the driver to manually park the vehicle in the target space.

[0044] In the parking control method of this application embodiment, in response to detecting the existence of a target parking space that can be parked, multimedia data of the parking area is sent to the cloud. The multimedia data is used by the cloud to identify parking decision information of the target parking space, obtain the parking decision information of the target parking space sent by the cloud, and perform parking control on the vehicle based on the parking decision information of the target parking space. This realizes the identification of the target parking space by the vehicle and the identification of the target parking space by the cloud. Since the cloud has stronger computing power and reasoning ability, the accuracy of the parking decision information of the target parking space is improved. Through the collaboration between the vehicle and the cloud, parking safety is improved.

[0045] Based on the above embodiments, Figure 2 This is a flowchart illustrating another parking control method provided in an embodiment of this application. It shows that when the results of cloud-based detection differ from those of vehicle-side detection, the parking space detection model cannot recognize the parking scenario. This parking scenario belongs to the long-tail data of parking. Therefore, the data of this parking scenario can be used as training samples to synchronously update and train the parking space detection model, thereby improving the detection effect of the parking space detection model and enhancing parking safety. Figure 2 As shown, the method includes the following steps:

[0046] Step 201: In response to the detection of a target parking space that can be parked, the multimedia data of the parking area is sent to the cloud.

[0047] The multimedia data from the parking area is used to identify parking decision information for the target parking space in the cloud.

[0048] In one implementation of this application, the cloud-based system uses a visual detection model to identify multimedia data and obtain parking decision information for the target parking space. This visual detection model is pre-trained with a large amount of data and can accurately identify obstacle information in various scenarios to accurately determine whether a parking space is suitable for parking. As one implementation, multimedia data of parking area samples is acquired. This sample multimedia data of the parking area is labeled with the parking decision information corresponding to the parking space. The sample multimedia data of the parking area is input into the visual detection model to obtain predicted parking decision information for the parking space. Based on the difference between the predicted parking decision information and the labeled parking decision information, a loss function is determined. The parameters of the visual detection model are adjusted according to the loss function to train the visual detection model, resulting in the trained visual detection model. The training process of the visual detection model can be repeated multiple times. It is trained using multimedia data of parking areas in different scenarios. When the loss function is less than a set threshold or the training is repeated a set number of times, the visual detection model is trained and the trained visual detection model is obtained. The trained visual detection model can identify parking spaces in complex scenarios based on the multimedia data of the parking area and calculate the parking route based on the location of the parking space. It can not only indicate whether the identified parking space can be parked, but also identify the dangers in the parking route, that is, identify whether there are potential hazards in the parking route.

[0049] In this embodiment, because the visual detection model set in the cloud executes slowly and cannot obtain an immediate response, but because parking speed is usually slow, the vehicle can send multimedia data of the parking area to the cloud before parking actually begins and obtain the detection results sent by the cloud. As an example, Figure 3 This is a schematic diagram of a parking scenario provided in an embodiment of this application, taking a visual detection model as an example, such as a visual large language model. Figure 3 As shown, in this parking area, the vehicle is the one waiting to park. The vehicle collects multimedia data of the parking area at point A, and uses a parking space detection model to identify that the target parking space is available. Then, it sends the multimedia data of the parking area to the cloud. The cloud performs identification based on the multimedia data of the parking area. During the cloud identification process, the vehicle prepares for parking based on the vehicle detection results. The vehicle travels from point A to point B. During the time from point A to point B, the cloud performs identification. Usually at point B, the vehicle receives the parking decision information of the parking space returned by the cloud. Based on the parking decision information, it performs parking control. Through the collaboration between the cloud and the vehicle, even when a large visual language model cannot be deployed on the vehicle, it can accurately determine whether the parking space is available, obtain the parking path, and the reason why parking is not possible, thus improving parking safety.

[0050] Step 202: Obtain parking decision information for the target parking space sent from the cloud.

[0051] Step 203: Based on the parking decision information, determine that the target parking space cannot be parked in, and generate a parking warning message.

[0052] Steps 202 and 203 can be referred to the relevant explanations in the foregoing embodiments, as the principles are the same, and will not be repeated here.

[0053] It should be noted that subsequent steps 204 and 205 may be executed before step 203, after step 203, or simultaneously with step 203. No limitation is made in this embodiment.

[0054] Step 204: In response to the parking decision information indicating that the target parking space cannot be parked, training samples for the parking space detection model are generated based on the multimedia data of the parking area and the parking decision information.

[0055] In this embodiment, the parking decision information indicates that the target parking space cannot be parked in. This indicates that the results detected by the cloud-based visual detection model differ from the results detected by the parking space detection model set on the vehicle side. Since the cloud-based visual detection model has stronger computing power and scene recognition capabilities, it is considered that the vehicle-side parking space detection model is inaccurate. Therefore, the multimedia data of the parking area and the parking decision information are used to generate training samples for the parking space detection model. As one implementation method, based on the parking decision information, the annotation information of the multimedia data of the parking area is determined. The annotation information indicates that the target parking space cannot be parked in. Based on the multimedia data of the annotated parking area, training samples for the parking space detection model are generated. The training samples are negative samples. The parking space detection model is trained using negative samples so that the parking space detection model can identify which scenarios, although the parking space is vacant, actually do not meet the conditions for parking. For example, there are obstacles on the parking path to the parking space, such as low tree stumps. Figure 4 This is a schematic diagram of multimedia data for a parking area provided in an embodiment of this application, such as... Figure 4 As shown, the parking area image was captured by a camera installed on the vehicle, such as a fisheye camera. There is a tree stump B on the path leading to parking space A. The vehicle must pass through tree stump B to park in parking space A. Tree stump B will cause damage to the vehicle's tires or chassis. Therefore, although parking space A is vacant, parking space A does not meet the parking conditions, that is, parking space A cannot be parked.

[0056] Step 205: Send the training samples to the cloud.

[0057] Among them, the training samples are used to update and train the parking space detection model in the cloud to obtain an updated parking space detection model.

[0058] In this embodiment, the cloud updates and trains the parking space detection model based on the acquired training samples, enabling the trained parking space detection model to identify more special scenarios. Then, the updated parking space detection model is redeployed on the vehicle, increasing the recognition capability of long-tail data and improving parking safety.

[0059] In the parking control method of this application embodiment, in response to the detection of a target parking space that can be parked in by using a parking space detection model to analyze the multimedia data of the parking area, the multimedia data of the parking area is sent to the cloud. The multimedia data of the parking area is used by the cloud to identify parking decision information of the target parking space based on a visual detection model. The system obtains the parking decision information of the target parking space sent from the cloud and performs parking control on the vehicle based on this information. This achieves the goal of identifying the target parking space on the vehicle side and obtaining the parking decision information of the target parking space through the visual detection model in the cloud. Because the visual detection model in the cloud has stronger computing power and reasoning ability, the accuracy of the parking decision information of the target parking space is improved. Through the collaboration between the vehicle side and the cloud, parking safety is improved.

[0060] To achieve the above embodiments, this application also proposes a parking control device.

[0061] Figure 5 This is a schematic diagram of a parking control device provided in an embodiment of this application.

[0062] like Figure 5 As shown, the device may include:

[0063] The sending module 51 is used to send multimedia data of the parking area to the cloud in response to the detection of a target parking space that can be parked; wherein, the multimedia data is used by the cloud to identify parking decision information of the target parking space.

[0064] The acquisition module 52 is used to acquire the parking decision information of the target parking space sent from the cloud.

[0065] The control module 53 is used to control the parking of the vehicle based on the parking decision information of the target parking space.

[0066] Furthermore, in one implementation of this application embodiment, the target parking space is identified based on the parking space detection model at the vehicle end, and the device also includes a generation module.

[0067] A generation module is used to generate training samples for the parking space detection model in response to the parking decision information indicating that the target parking space cannot be parked, based on the multimedia data of the parking area and the parking decision information.

[0068] The sending module 51 is further configured to send the training samples to the cloud; wherein the training samples are used by the cloud to update the parking space detection model in order to obtain an updated parking space detection model.

[0069] In one implementation of this application, the generation module is further configured to:

[0070] Based on the parking decision information, the annotation information of the multimedia data of the parking area is determined; wherein, the annotation information indicates that the target parking space is not suitable for parking;

[0071] Training samples for the parking space detection model are generated based on the multimedia data of the labeled parking areas; wherein, the training samples are negative samples.

[0072] In one implementation of this application embodiment, the control module 53 is further configured to:

[0073] Based on the parking decision information, it is determined that the target parking space cannot be used for parking;

[0074] Generate parking warning information; wherein, the parking warning information is used to indicate that there are obstacles on the parking path corresponding to the target parking space and / or within the target parking space, and the target parking space cannot be parked.

[0075] In one implementation of this application embodiment, the control module 53 is further configured to:

[0076] Based on the parking decision information, it is determined that the target parking space is available for parking;

[0077] Determine the parking route corresponding to the target parking space;

[0078] According to the parking route, control the vehicle to park in the target parking space.

[0079] In one implementation of this application embodiment, the control module 53 is further configured to:

[0080] Obtain the three-dimensional spatial information of the parking area;

[0081] Based on the three-dimensional spatial information, the parking route corresponding to the target parking space is determined.

[0082] In one implementation of this application embodiment, the control module 53 is further configured to:

[0083] Based on the parking decision information, determine the parking route corresponding to the target parking space.

[0084] In one implementation of this application, the training method for the visual detection model includes:

[0085] Acquire sample multimedia data of the parking area; wherein, the sample multimedia data of the parking area is labeled with parking decision information corresponding to the parking space;

[0086] The sample multimedia data of the parking area is input into the visual detection model to obtain the predicted parking decision information of the parking space;

[0087] The loss function is determined based on the difference between the predicted parking decision information and the labeled parking decision information;

[0088] The visual detection model is trained according to the loss function to obtain the trained visual detection model.

[0089] It should be noted that the foregoing explanation of the method embodiments also applies to the apparatus of this embodiment, and will not be repeated here.

[0090] In the parking control device of this application embodiment, in response to detecting the existence of a target parking space that can be parked, multimedia data of the parking area is sent to the cloud. The multimedia data is used by the cloud to identify parking decision information of the target parking space, obtain the parking decision information of the target parking space sent by the cloud, and perform parking control on the vehicle based on the parking decision information of the target parking space. This realizes the identification of the target parking space by the vehicle and the identification of the parking decision information of the target parking space by the cloud. Since the cloud has stronger computing power and reasoning ability, the accuracy of the parking decision information of the target parking space is improved. Through the collaboration between the vehicle and the cloud, parking safety is improved.

[0091] To implement the above embodiments, this application also proposes a vehicle, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the method described in the foregoing method embodiments.

[0092] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing method embodiments.

[0093] To implement the above embodiments, this application also proposes a computer program product having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the foregoing method embodiments.

[0094] Figure 6 This is a schematic diagram of the structure of a vehicle according to an embodiment of this application. For example, vehicle 600 can be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. Vehicle 600 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0095] Reference Figure 6 The vehicle 600 may include various subsystems, such as an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. The vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of the vehicle 600 can be interconnected via wired or wireless means.

[0096] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, etc.

[0097] The perception system 620 may include several sensors for sensing information about the environment surrounding the vehicle 600. For example, the perception system 620 may include a global positioning system (which may be GPS, BeiDou, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0098] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0099] The drive system 640 may include components that provide powered motion to the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0100] Some or all of the functions of vehicle 600 are controlled by computing platform 650. Computing platform 650 may include at least one processor 651 and memory 652, and processor 651 may execute instructions 653 stored in memory 652.

[0101] Processor 651 can be any conventional processor, such as a commercially available CPU. The processor may also include multimedia data processors (Graphic Processing Unit, GPU), field-programmable gate arrays (FPGAs), systems-on-chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0102] The memory 652 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0103] In addition to instruction 653, memory 652 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in memory 652 can be used by computing platform 650.

[0104] In this embodiment of the disclosure, processor 651 may execute instructions 653 to complete all or part of the steps of the above method embodiments.

[0105] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0106] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0107] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0109] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0110] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0111] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0112] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A parking control method, characterized in that, include: In response to the detection of a target parking space that can be parked, multimedia data of the parking area is sent to the cloud; wherein, the multimedia data is used by the cloud to identify parking decision information of the target parking space; the target parking space is an available parking space that can be parked based on the parking space detection model on the vehicle side; Obtain the parking decision information for the target parking space sent from the cloud; Based on the parking decision information of the target parking space, the vehicle is controlled to park. After obtaining the parking decision information of the target parking space sent by the cloud, the method further includes: In response to the parking decision information indicating that the target parking space cannot be parked, training samples for the parking space detection model are generated based on the multimedia data of the parking area and the parking decision information. The training samples are sent to the cloud; wherein, the training samples are used to update the parking space detection model in the cloud to obtain an updated parking space detection model, and the updated parking space detection model is used to redeploy on the vehicle.

2. The method as described in claim 1, characterized in that, The step of generating training samples for the parking space detection model based on the multimedia data of the parking area and the parking decision information includes: Based on the parking decision information, the annotation information of the multimedia data of the parking area is determined; wherein, the annotation information indicates that the target parking space is not suitable for parking; Training samples for the parking space detection model are generated based on the multimedia data of the labeled parking areas; wherein, the training samples are negative samples.

3. The method as described in claim 1, characterized in that, The step of controlling the parking of the vehicle based on the parking decision information of the target parking space includes: Based on the parking decision information, it is determined that the target parking space cannot be used for parking; Generate parking warning information; wherein, the parking warning information is used to indicate that there are obstacles on the parking path corresponding to the target parking space and / or within the target parking space, and the target parking space cannot be parked.

4. The method as described in claim 1, characterized in that, The step of controlling the parking of the vehicle based on the parking decision information of the target parking space includes: Based on the parking decision information, it is determined that the target parking space is available for parking; Determine the parking route corresponding to the target parking space; According to the parking route, control the vehicle to park in the target parking space.

5. The method as described in claim 4, characterized in that, Determining the parking route corresponding to the target parking space includes: Obtain the three-dimensional spatial information of the parking area; Based on the three-dimensional spatial information, the parking route corresponding to the target parking space is determined.

6. The method as described in claim 4, characterized in that, Determining the parking route corresponding to the target parking space includes: Based on the parking decision information, determine the parking route corresponding to the target parking space.

7. The method according to any one of claims 1-6, characterized in that, The cloud-based system identifies multimedia data in the parking area based on a pre-set visual detection model, thereby obtaining parking decision information for the target parking space. The training method for the visual detection model includes: Acquire sample multimedia data of the parking area; wherein, the sample multimedia data is labeled with parking decision information corresponding to the parking space; The sample multimedia data is input into the visual detection model to obtain the predicted parking decision information for the parking space; The loss function is determined based on the difference between the predicted parking decision information and the labeled parking decision information; The visual detection model is trained according to the loss function to obtain the trained visual detection model.

8. A parking control device, characterized in that, include: The sending module is used to send multimedia data of the parking area to the cloud in response to the detection of a target parking space that can be parked; wherein, the multimedia data of the parking area is used by the cloud to identify parking decision information of the target parking space based on a visual detection model; the target parking space is an available parking space that can be parked based on a parking space detection model on the vehicle side; The acquisition module is used to acquire the parking decision information of the target parking space sent by the cloud. The control module is used to control the parking of the vehicle based on the parking decision information of the target parking space; A generation module is used to generate training samples for the parking space detection model in response to the parking decision information indicating that the target parking space cannot be parked, based on the multimedia data of the parking area and the parking decision information. The sending module is further configured to send the training samples to the cloud; wherein the training samples are used to update the parking space detection model in the cloud to obtain an updated parking space detection model, and the updated parking space detection model is used to redeploy on the vehicle.

9. A vehicle, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured as follows: The steps of implementing the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of a mobile terminal, enable the mobile terminal to perform the steps of the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, It includes a computer program, which, when executed by a processor, implements the method of any one of claims 1 to 7.

Images