A car storage method, device, storage medium and computer program product

Abstract

The application discloses a parking storage method and device, a storage medium and a computer program product. In the application, if it is detected that a current vehicle is in a parking completion state at a parking space, it is determined whether a target getting-off state of the current vehicle is an off state. If the target getting-off state is the off state, it indicates that a target such as a person in the current vehicle has gotten off. At this time, whether all the targets have left the parking space is determined through a target motion track. If all the targets have left the parking space, a parking storage process is automatically executed. It can be seen that the application can automatically execute the parking storage process after detecting that the target such as the person in the current vehicle has gotten off and left the parking space, the whole process does not need to be confirmed and triggered by a user to store the vehicle, the safety and efficiency of the parking storage are improved, the parking storage is realized without feeling, and the parking storage experience is improved.

Description

Technical Field

[0001] This application relates to the field of vehicle storage technology, and in particular to a vehicle storage method, device, storage medium, and computer program product. Background Technology

[0002] With the continuous growth of urban car ownership, multi-level parking systems have been widely used in residential communities, commercial centers, and public parking lots due to their advantages such as small footprint and high space utilization. Traditional multi-level parking systems typically employ structural forms such as lifting and lateral movement, and vertical lifting, aiming to maximize the number of parking spaces within limited land resources.

[0003] Currently, most automated parking systems rely on users manually confirming and pressing the parking button. During this process, users must determine whether the vehicle is correctly parked and whether all occupants and surrounding personnel have completely evacuated before pressing the parking button to start the system. This process essentially places the responsibility for safety confirmation entirely on the user. However, user operation is subjective, arbitrary, and uncertain, easily leading to situations where, for example, another person or unrelated individual presses the parking button while the vehicle is still occupied or nearby, causing the system to suddenly start. This poses a serious threat to the safety of those inside or near the vehicle. Such accidents caused by human error are common in existing automated parking systems and have become a key weakness restricting safety and user experience improvement.

[0004] Therefore, how to improve the safety and efficiency of parking is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] This application provides a parking method, apparatus, storage medium, and computer program product to improve the safety and efficiency of parking.

[0006] Firstly, this application provides a method for parking a vehicle, including: Detect whether the vehicle is currently in a completed parking space; If the vehicle is in a parked state, the target exit status of the current vehicle is obtained; the target exit status is used to indicate whether each target is inside the current vehicle. Obtain the target motion trajectory of the parking space; the target motion trajectory is used to represent the positional changes of each target in the parking space; If the target has disembarked and the target's trajectory indicates that the target has left the parking space, the parking process will be executed automatically.

[0007] Optionally, obtain the target alighting status of the current vehicle, including: Real-time acquisition of vehicle images of the current vehicle; Based on the vehicle image, detect the status of each door of the current vehicle; If the target door is in an open state, then it is determined whether the target is detected in the door detection area; wherein, the door detection area is the door detection area in the vehicle image corresponding to the target door; If a target is detected, and the target vehicle door is detected to be in a closed state, then the target's exit state is determined to be "has exited the vehicle".

[0008] Optionally, if the target's disembarkation state is "already disembarked," and the target's movement trajectory indicates that the target has left the parking space, this includes: Determine if the target's disembarkation status is "already disembarked"; If the vehicle has already been alighted, determine whether there is a target in the parking space based on the vehicle image; If there is no target, the target's motion trajectory is used to determine whether the last detected position of each target in the parking space is an edge detection area; the edge detection area is the detection area at the edge of the parking space in the vehicle image that does not overlap with the door detection area. If it is an edge detection zone, it is determined that the target has left the parking space.

[0009] Optionally, obtain the target alighting status of the current vehicle, including: Detect whether there is a target inside the current vehicle; If not, then the target's disembarkation status is directly determined as "disembarked".

[0010] Optionally, the automatic parking process may also include: Determine if a target is present inside the vehicle. If no target exists, determine whether the current vehicle's rearview mirror folding status, wheel status, and door status all meet the parking conditions. If all parking conditions are met, the vehicle is deemed to have passed the parking check, and the parking process is executed automatically.

[0011] Optionally, determining whether the rearview mirror folding state meets the parking conditions includes: Based on the current vehicle image, determine whether the rearview mirror is in a folded state; if it is, determine that the rearview mirror folding state meets the parking conditions. If the vehicle is not folded, determine whether the width of the current vehicle is less than the predetermined width threshold. If the width is less than the predetermined vehicle width threshold, the rearview mirror folding state is determined to meet the parking conditions; if the width is not less than the predetermined vehicle width threshold, the rearview mirror folding state is determined to not meet the parking conditions.

[0012] Optionally, determining whether the wheel condition meets the parking conditions includes: Based on the current vehicle image, calculate the first deflection angle between the left front wheel and the longitudinal centerline of the vehicle, and the second deflection angle between the right front wheel and the longitudinal centerline of the vehicle. If both the first deflection angle and the second deflection angle are less than the predetermined deflection angle threshold, then the wheel condition is determined to meet the parking conditions; otherwise, the wheel condition is determined not to meet the parking conditions.

[0013] Secondly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of the above-described vehicle storage method when executing the computer program.

[0014] Thirdly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described vehicle storage method.

[0015] Fourthly, this application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the above-described vehicle storage method.

[0016] Compared with the prior art, the technical solution provided in this application has the following advantages: This application provides a parking method, device, storage medium, and computer program product. In this application, if it is detected that the current vehicle is in a completed parking state in the parking space, it is determined whether the target getting out of the current vehicle is in a "getted out" state. If it is, it means that the people or other targets in the current vehicle have gotten out. At this time, it is determined whether each target has left the parking space through the target's movement trajectory. If each target has left the parking space, the parking process is automatically executed. It can be seen that this application can automatically execute the parking process after detecting that the people or other targets in the current vehicle have gotten out and left the parking space. The whole process does not require user confirmation and triggering of parking, which improves the safety and efficiency of parking, realizes seamless parking, and improves the parking experience. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

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

[0019] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0020] Figure 1 This is a schematic flowchart of a vehicle parking method provided in an embodiment of this application; Figure 2 A top view of a parking space provided in an embodiment of this application; Figure 3 This application provides a schematic diagram of the overall process for an automatic parking solution. Figure 4 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of this application.

[0022] It should be noted that, in the optional embodiments of this application, the data related to object information, when applied to specific products or technologies, requires the permission or consent of the object. Furthermore, the collection, use, and processing of this data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. In other words, if the embodiments of this application involve data related to an object, it must be obtained with the object's authorization and consent, the authorization and consent of relevant departments, and in accordance with the relevant laws, regulations, and standards of the country and region. If the embodiments involve personal information, the acquisition of all personal information requires the individual's consent. If sensitive information is involved, the separate consent of the information subject is required. The embodiments also need to be implemented with the object's authorization and consent.

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0025] In existing solutions, most automated parking systems rely on users manually confirming and triggering the parking button. During this process, users must verify that all occupants and those around the vehicle have completely evacuated before pressing the button to start the system. However, this method poses a significant safety hazard. If someone is parking or has not yet exited the vehicle, and another person or unrelated individual presses the parking button without authorization, causing the system to suddenly start, this could seriously threaten the safety of those inside or near the vehicle.

[0026] Furthermore, in actual use, most existing multi-level parking systems often add roller shutters, flip-up doors, or fence-like protective gates at the entrance of parking spaces to define parking space boundaries, prevent vehicles from falling, or restrict unauthorized personnel from entering. These systems rely on users manually confirming parking, which introduces many operational problems and safety hazards. For example, due to the protective gates, drivers' visibility and operating space are significantly limited when driving into the multi-level parking space. Compared to traditional open parking spaces, drivers must complete reversing or straight-through maneuvers within a narrow doorway, easily causing tension and fear, worrying about the vehicle scraping or colliding with the sides or top of the door. This psychological pressure not only reduces parking efficiency but also increases the risk of accidents caused by operational errors, especially for users with inexperienced driving skills.

[0027] It is evident that existing automated parking systems have significant shortcomings in terms of parking space entrance structure and parking control methods. They fail to effectively address the issues of safety and convenience during parking operations, and also lack reliable safety interlocking and anti-accidental triggering functions during the parking process. Therefore, this application provides a parking method, equipment, storage medium, and computer program product to effectively eliminate safety risks caused by human error during parking, thereby improving parking safety and efficiency. This has significant practical implications and engineering application value.

[0028] See Figure 1 The above is a schematic flowchart of a vehicle parking method provided in an embodiment of this application. The method includes the following steps: S101. Detect whether the current vehicle is in a completed parking state in the parking space.

[0029] If yes, then execute S102; otherwise, continue executing S101.

[0030] In this application, the parking completion state refers to the state where the vehicle is parked and does not need to be moved. When detecting whether the vehicle is in the parking space completion state, an image acquisition device installed in the parking space can capture images of the vehicle. If the vehicle is detected to be stationary and its parking position meets the parking requirements, then the vehicle is determined to be in the parking space completion state. When determining whether the vehicle's parking position meets the parking requirements, it can be determined whether the vehicle's wheels are accurately positioned within the designated area of ​​the parking space. If each wheel is accurately positioned within the designated area, then the vehicle's parking position meets the parking requirements.

[0031] This application can also determine whether the current vehicle is in a parked state by detecting whether a parking completion command has been received. For example, if the user actively triggers the parking completion command through the APP (Application), or the user actively triggers the parking completion command through the on-site terminal screen, or the user generates the parking completion command through voice, then it is determined that the current vehicle is in a parked state.

[0032] Furthermore, when detecting whether the current vehicle is in a parking space and is in a completed parking state, this solution can also determine that the current vehicle is in a completed parking state only after detecting that the user has actively triggered the vehicle parking completion command, and after determining that the current vehicle is stationary through the vehicle image, and that the current vehicle is parked in a parking space that meets the parking requirements. This ensures the safety of the subsequent parking process and avoids damage to the vehicle during parking.

[0033] If the vehicle is currently in a parked state, continue with the subsequent steps; if the vehicle is not currently in a parked state, continue with S101.

[0034] S102. Obtain the target exit status of the current vehicle; the target exit status is used to indicate whether each target is inside the current vehicle. In this application, if the vehicle is currently parked, it is necessary to obtain the target exit status of the vehicle. In this application, the target exit status indicates whether each target is inside the vehicle. The target exit status includes an "already exited" status and an "not yet exited" status. The "already exited" status indicates that all targets are no longer in the vehicle, while the "not yet exited" status indicates that there are still targets inside the vehicle. These targets include people, pets, and other objects that need to be detected.

[0035] The method of obtaining the target's exit status in this application can be as follows: acquiring a vehicle image containing the current vehicle through an image acquisition device, analyzing the vehicle image to detect whether a target exists inside the current vehicle; if a target exists, the target's exit status is determined to be "not exiting"; if no target is detected, the target's exit status is determined to be "exited". Alternatively, the target's exit status can be determined through thermal imaging technology, or the user can actively trigger the generation of the target's exit status on an APP or on-site terminal screen. There are no specific limitations here.

[0036] S103. Obtain the target motion trajectory of the parking space; the target motion trajectory is used to represent the positional changes of each target in the parking space. In this application, if the vehicle is currently in a parked state, it is necessary to obtain the target movement trajectory of the parking space. This target movement trajectory is used to represent the positional changes of each target in the parking space. This target movement trajectory is not static position data, but is generated based on the dynamic changes of the target position. It includes the positional change data of each target in the parking space. For example, if user A gets out of the car from the driver's side door, walks to the trunk, and then leaves the parking space from the left side, the target movement trajectory includes user A's exit position - trunk position - left side position of the parking space, that is, it includes user A's positional change data in the parking space, so as to accurately locate the target position and accurately identify whether the user has left the parking space.

[0037] When obtaining the target movement trajectory of a parking space, this application can acquire vehicle images containing the current vehicle through an image acquisition device, and generate the target movement trajectory by recognizing the target position in each frame of the vehicle image; or, it can detect the position of the target through thermal imaging technology, and then generate the target movement trajectory; or, it can generate the target movement trajectory through the real-time position of the user's movement state, without being specifically limited here.

[0038] In this application, the execution order of the two steps, S102 obtaining the target exit status of the current vehicle and S103 obtaining the target movement trajectory of the parking space, is not limited. In practical applications, it can be implemented by executing S102 first and then S103, or by executing S103 first and then S102, or by executing S102 and S103 simultaneously. There is no specific limitation here, as long as the target exit status and the target movement trajectory can be obtained.

[0039] S104. If the target's disembarkation status is "disembarked" and the target's movement trajectory indicates that the target has left the parking space, the parking process is automatically executed.

[0040] In this application, if the target's disembarkation status is detected as "disembarked" and the target's movement trajectory is determined to indicate that the target has left the parking space, it means that all targets, including personnel, have left the parking space, and the parking process can be automatically executed at this time.

[0041] It should be noted that the parking spaces in this application are those of a multi-level parking system. The layout of these spaces is identical to traditional parking spaces, without any roller shutters or other obstructions. Therefore, the parking spaces of the multi-level parking system are similar to those of a flat parking space, avoiding the difficulty of parking operations caused by limited space and eliminating driving fear. Furthermore, by detecting the target's exit status and movement trajectory, this application effectively ensures the safety of people and other targets when parking, even without roller shutters or other obstructions. The absence of roller shutters or other safety restrictions saves installation space and costs, expands the user's parking range and field of vision, making parking more convenient and providing a better parking experience.

[0042] In summary, this application automatically executes the parking process after detecting that the vehicle is parked in the parking space and that the person or other target in the vehicle has gotten off and left the parking space. The entire process does not require the user to actively confirm and trigger the parking, which improves the safety and efficiency of parking, achieves seamless parking, and enhances the parking experience.

[0043] In another embodiment of this application, obtaining the target exit status of the current vehicle includes: The system acquires real-time images of the current vehicle; based on the vehicle images, it detects the status of each door of the current vehicle; if the target door is in an open state, it determines whether the target has been detected within the door detection area; the door detection area is the door detection area in the vehicle image corresponding to the target door; if the target is detected and the target door is detected to be in a closed state, it determines that the target has alighted.

[0044] In this embodiment, when obtaining the target exit status of the current vehicle, the target exit status can be determined by detecting the vehicle image of the current vehicle. To accurately detect the target exit status, this application can divide the parking area of ​​the parking space into regions, creating a door detection area for detecting door status and the target. This door detection area can include a driver's door detection area and non-driver's door detection areas.

[0045] See Figure 2 This is a top view of a parking space provided in an embodiment of this application. The black frame in the figure represents the parking space lines. Figure 2In the diagram, only the driver's side door detection area is marked; the non-driver's side door detection areas are not marked. These non-driver's side door detection areas include the passenger side door detection area, the rear door detection area, and the trunk door detection area. The passenger side door detection area includes a predetermined area near the passenger side door; the rear door detection areas include predetermined areas near the left and right rear doors; and the trunk door detection area includes a predetermined area near the trunk. Each door detection area is used to detect the door status and target within that area. Figure 2 As shown, an edge detection area and an image acquisition device are also set in the parking space. The edge detection area is used to detect whether the target has left in subsequent embodiments. The image acquisition device is a network camera. The network camera is installed at the positions directly above the front, directly above the rear, above the left front, above the left rear, above the right front, and above the right rear of the parking space, respectively, to detect the target and vehicle behavior during the parking process.

[0046] In this embodiment, when determining the target's exit status, the real-time vehicle image of the current vehicle can be obtained by using different cameras installed on different doors to capture vehicle images from different angles. For example, when detecting the driver's side door, the left front camera can be used to capture the vehicle image; when detecting the passenger side door, the right front camera can be used to capture the vehicle image, and so on. After acquiring the status of each door of the current vehicle, the target door whose status is open is determined. This target door can be at least one of the driver's side door and non-driver's side doors (passenger side door and rear doors). Furthermore, after detecting that the target door is open, it is necessary to determine whether the target is detected within the target door's detection area. If the target is detected and the target door's status changes to closed, it is determined that the target exited the vehicle through the target door, and the target's exit status is determined to be "exited."

[0047] It should be noted that this application can divide the target exit state into the driver's exit state and the non-driver's exit state. Specifically, regardless of whether there are passengers in the vehicle, the driver is an essential factor for vehicle parking, so it is necessary to detect the driver's normal exit. Therefore, when detecting the door status, the driver's door can be detected first to ensure that the driver opens the door. Secondly, after the door is opened, the driver must be detected to be within the driver's door detection area to ensure the driver exits. Then, the driver's door is closed, thus completing the driver's exit action. At this point, the driver's exit state is determined to be "driver's exit state". In addition, other passengers may exit the vehicle besides the driver, so the non-driver's doors also need to be detected. When there are passengers or the driver is retrieving items, after opening the non-driver's door, the personnel or other targets must be detected within the non-driver's door detection area to ensure the target exits. After the non-driver's door is closed, the exit action or item retrieval is determined to be completed. At this point, the non-driver's exit state is determined to be "non-driver's exit state". When there are no passengers, the non-driver's doors do not need to be opened.

[0048] After obtaining the driver's seat target's exit status and the non-driver's seat target's exit status through the above process, when generating the final target exit status, within a predetermined time period, if the driver's seat target's exit status is detected as "driver's seat target has exited" and the non-driver's seat target's exit status is also detected as "non-driver's seat target has exited", then the final target exit status is determined to be the exit status. Alternatively, since there may not be a target in the non-driver's seat, or the person may not have gone to the non-driver's seat to retrieve the item, the non-driver's seat target's exit status may not be determined as the exit status within the predetermined time period. Therefore, in this embodiment, the final target exit status can be determined solely based on the driver's seat target's exit status. That is, if the driver's seat target's exit status is "driver's seat target has exited", then the final target exit status can be determined to be the exit status.

[0049] In summary, this application can determine the target's alighting status by combining the door status and the target's position in the door detection area. Only when the door is open, the target is detected in the corresponding door detection area, and the door status changes to closed can it be determined that the target has performed the alighting action, thus confirming that the target has alighted. This method can ensure the accuracy of alighting detection and ensure that all personnel in the vehicle have alighted before performing subsequent parking operations, ensuring personnel safety.

[0050] In another embodiment of this application, if the target's disembarkation state is "already disembarked," and it is determined based on the target's movement trajectory that the target has left the parking space, the method includes: Determine if the target's disembarkation status is "already disembarked"; If the vehicle has already been alighted, determine whether there is a target in the parking space based on the vehicle image; If there is no target, the target's motion trajectory is used to determine whether the last detected position of each target in the parking space is an edge detection area; the edge detection area is the detection area of ​​the parking space edge position in the vehicle image that does not overlap with the door detection area. If it is an edge detection zone, it is determined that the target has left the parking space.

[0051] In this embodiment, to prevent people from returning to the vehicle, or even children crawling under the vehicle or hiding in front of the vehicle and causing the equipment to activate and create danger, this application needs to record the target's movement trajectory after detecting people or other targets in the parking space. The target's movement trajectory is used to detect people in the edge detection area of ​​the parking space, ensuring that each target leaves the parking space from the edge detection area. This method effectively ensures that the target disappears from the direction of leaving the parking space, thereby ensuring the safety of people.

[0052] In this application, the edge detection zone is an area located at the edge of the parking space that the user must pass through when leaving the parking space, and this edge detection zone does not overlap with the door detection zone. See also Figure 2 ,because Figure 2 In some parking spaces, the front of the car is enclosed, preventing people from leaving the parking space from the front. Therefore, for... Figure 2 The edge detection zone of the parking space division is... Figure 2 The red-lined frame in the diagram represents the edge detection area, which includes the left edge of the parking space, the right edge of the parking space, and the rear edge of the parking space.

[0053] This application, when detecting the status of people outside the vehicle and generating the target motion trajectory, can combine... Figure 2 The system uses four cameras—left front, left rear, right front, right rear, front, and rear—to perform multi-view joint detection of people within the parking space area. Once the system detects a car door opening, it continuously tracks the movement trajectory of the person, using the YOLO deep learning target detection model to detect people and other targets in real time, generating target movement trajectories. If the system determines there are no targets in the parking space based on the vehicle image, it compares the target movement trajectory with the edge detection area to determine if the last detected position of each target in the parking space falls within the edge detection area. If it is within the edge detection area, the target is considered to have left the parking space; otherwise, it is considered not to have left, and the target may be hiding in a location not captured by the cameras, such as under the car or inside the car. In this case, the automatic parking process cannot be executed to avoid harming the target during parking.

[0054] In summary, this application can record the movement trajectory of each target in a parking space. The target can be a target getting in or out of the current vehicle, or a target entering the parking space from outside. By comparing the last detected position of each target in the target movement trajectory with the edge detection area, it can accurately detect whether each target in the parking space has left the parking space, thereby ensuring the safety of people and other targets when parking.

[0055] In another embodiment of this application, obtaining the target alighting status of the current vehicle includes: detecting whether there is a target inside the current vehicle; if not, directly determining that the target alighting status is alighting status.

[0056] In the application, when obtaining the target's exit status of the current vehicle, it can also directly detect whether there are people, pets, or other targets inside the current vehicle; if there are targets inside the current vehicle, the target's exit status is determined to be "not exited"; if there are no targets inside the current vehicle, the target's exit status is determined to be "already exited".

[0057] It should be noted that when detecting whether a target is inside the vehicle, this application can combine images of the vehicle interior acquired by an image acquisition device to further detect the status of the occupants. Specifically, a front camera in the parking space can be used to take a picture of the windshield. The outline of the windshield is then marked using the YOLO deep learning object detection model. The status of targets inside the vehicle within this outline is then detected. The status of a target inside the vehicle includes two possibilities: the presence of a target and the absence of a target. If a target is present, the target's exit status is "already exited"; if a target is not present, the target's exit status is "not exited". Alternatively, thermal imaging technology can be used to determine whether a target is inside the vehicle and thus determine the target's exit status.

[0058] In practical applications, if a driver enters a parking space using the exit parking function, the method described in the above embodiment—detecting the door open, detecting the target in the door detection area, and the door being closed—cannot be used to determine the target's exit status. In this case, the target's exit status can be directly determined by detecting whether there is a target inside the vehicle. This ensures that there are no people waiting for the target in the vehicle before parking, thus ensuring the safety of parking.

[0059] In another embodiment of this application, considering that when a user parks their vehicle, they also need to determine whether the rearview mirrors are folded and the doors are closed, if the user accidentally presses the parking button when the vehicle's condition is not up to standard due to negligence, such as when the rearview mirrors are not folded or the doors are not closed properly, this could lead to scratches during the operation of the automated parking system. Therefore, in this embodiment, the parking process is further enhanced before the automatic execution of the parking procedure by including: Determine if there is a target inside the vehicle; if there is no target, determine if the vehicle's rearview mirror folding state, wheel state, and door state all meet the parking conditions; if all conditions are met, the vehicle passes the parking check and the parking process is executed automatically.

[0060] Specifically, the vehicle inspection performed before parking in this embodiment is a static safety check, meaning that even after all personnel have left, the vehicle still poses safety hazards, such as unfolded rearview mirrors, wheels not straightened, or children inside. Therefore, in this application, after determining that the target has left the parking space through the above embodiment, it is necessary to further assess the state of the target inside the vehicle, the folded state of the rearview mirrors, the wheel status, and the door status to determine whether all parking conditions are met. Only if there is no target inside the vehicle, and the rearview mirrors are folded, the wheels are straightened, and the doors are closed, are all parking conditions considered met, and the parking process can be automatically executed. If any one or more of the parking conditions are not met, a reminder can be issued, and the inspection can be repeated. The reminder method can be: APP reminder, on-site terminal screen reminder, on-site voice reminder, etc., and is not specifically limited here.

[0061] This application can detect whether the parking conditions are met in several ways. It can acquire images through image acquisition devices installed in the parking space, and then analyze the images to determine whether there is a target inside the vehicle, and whether the rearview mirror folding state, wheel state, and door state all meet the parking conditions. Alternatively, it can collect relevant data of the current vehicle, such as: acquiring images of the vehicle's interior captured by the vehicle's in-vehicle camera, determining whether there is a target inside the vehicle based on the images, acquiring the current vehicle's status data, and determining the rearview mirror folding state, wheel state, and door state based on the status data.

[0062] It should be noted that the length of vehicles on the market ranges from approximately 3.5m to 5.5m, the width from approximately 1.6m to 2.0m, and the height from approximately 1.4m to 1.9m. Therefore, there is a large difference in vehicle size. If only a single-view camera is used to detect doors and people, problems such as blind spots, large changes in scale, and severe obstruction are likely to occur, leading to false detections or missed detections. Therefore, in this application, six cameras at the front, rear, left, and right of the parking space can be used to check the vehicle again to ensure safety.

[0063] For example, when acquiring the status of car doors, door status detection can be performed through dedicated door detection areas. This involves deploying network cameras at four locations: front left, rear left, front right, and rear right. Each camera corresponds to a dedicated door detection area. Simultaneously, a rear camera in the parking space captures and identifies the open / closed state of the trunk door. By using cameras at different angles, the vehicle can be monitored from all directions. The YOLO deep learning object detection model independently identifies the door status in each detection area, effectively avoiding detection area offset issues caused by vehicle size differences. Door status includes the driver's door, passenger's door, rear doors, and trunk door. Only when all doors are closed does the parking condition meet. When generating target motion trajectories, the front left, rear left, front right, and rear right cameras, along with front and rear overhead cameras, can be combined to perform multi-view joint detection of targets such as people within the parking space area, obtaining the target motion trajectory.

[0064] As can be seen from the above process, after a user drives into a parking space, the system can automatically detect whether the vehicle is parked properly by taking a picture with the camera on the parking space and using deep learning algorithms. After the person opens the car door and gets out of the car normally, the robot automatically triggers the parking operation when the vehicle meets the parking conditions. There are no buttons or other operations, and the user is unaware of the entire parking process, which ensures the safety of people and other targets, as well as the safety of the vehicle, during the parking process.

[0065] In another embodiment of this application, determining whether the rearview mirror folding state meets the parking conditions includes: Based on the current vehicle image, determine whether the rearview mirror is in a folded state; if it is, determine that the rearview mirror folding state meets the parking conditions. If the rearview mirror is not folded, determine whether the current vehicle width is less than the predetermined vehicle width threshold; if it is less than the predetermined vehicle width threshold, determine that the rearview mirror folding state meets the parking conditions; if it is not less than the predetermined vehicle width threshold, determine that the rearview mirror folding state does not meet the parking conditions.

[0066] In this application, the vehicle's rearview mirrors can be identified as folded based on the vehicle image. If both the left and right rearview mirrors are folded, the vehicle is deemed to meet the parking requirements.

[0067] However, considering that some vehicles are relatively narrow, not folding the rearview mirrors may not pose a safety hazard. Therefore, in this application, if the rearview mirror is in an open state instead of a folded state, it can be determined whether the current vehicle width is less than a predetermined vehicle width threshold. If the vehicle width is less than the predetermined threshold, it means that not folding the rearview mirrors will not pose a safety hazard, and the folded state can be directly determined to meet the parking conditions. If the vehicle width is not less than the predetermined threshold, it means that the current vehicle is relatively wide, and not folding the rearview mirrors will inevitably pose a safety hazard, and the folded state is determined not to meet the parking conditions. There are various methods for measuring vehicle width, such as using a laser rangefinder sensor, or obtaining the pixel distance of the vehicle width from a vehicle image and converting it into the actual vehicle width length, etc., and no specific limitation is made here.

[0068] The predetermined vehicle width threshold can be set according to the actual situation. In this embodiment, the predetermined vehicle width threshold can be set to 1800mm. That is, if the rearview mirror of the current vehicle is not folded and the width of the current vehicle is less than 1800mm, it can be determined that the rearview mirror folding state meets the parking conditions.

[0069] In this embodiment, the width of the current vehicle can be detected first to determine whether the width of the current vehicle is less than a predetermined width threshold. If so, it can be directly determined that the rearview mirror folding state meets the parking conditions without needing to detect the rearview mirror state. If not, it can be further determined whether the rearview mirror folding state is folded. If it is folded, it can be determined that the rearview mirror folding state meets the parking conditions. If it is not folded, it can be determined that the rearview mirror folding state does not meet the parking conditions.

[0070] When acquiring vehicle images, this application uses the left and right front cameras of the parking space for monitoring. After taking pictures containing the left and right rearview mirrors, and the right and right front cameras respectively, CNN detection is performed to obtain the folding state of the rearview mirrors. This application uses cameras on both sides to detect the folding state of the rearview mirrors separately, which can effectively avoid missing the case of a single folding rearview mirror and improve the accuracy of detection. Furthermore, this application not only determines whether the parking conditions are met by the folding state of the rearview mirrors, but also needs to make a comprehensive determination based on the current vehicle width. Therefore, when there is no safety hazard, the user does not need to adjust the rearview mirror state, simplifying the parking process and improving parking efficiency.

[0071] In another embodiment of this application, determining whether the wheel state meets the parking conditions includes: calculating the first deflection angle between the left front wheel and the longitudinal centerline of the current vehicle, and the second deflection angle between the right front wheel and the longitudinal centerline of the vehicle, based on the vehicle image of the current vehicle. If both the first deflection angle and the second deflection angle are less than the predetermined deflection angle threshold, then the wheel condition is determined to meet the parking conditions; otherwise, the wheel condition is determined not to meet the parking conditions.

[0072] In this application, the wheel status of the current vehicle can be identified based on the vehicle image to determine whether the parking conditions are met. In this application, wheel status refers to whether the wheels are straightened. If the wheels are straightened, the parking conditions are met; if the wheels are not straightened, the parking conditions are not met. The wheel status can be determined by combining the left and right front wheels; only when both the left and right front wheels are straightened can the wheel status be considered straightened.

[0073] In order to improve the accuracy of recognition, this application can use a left front camera and a right front camera to acquire images of the left front wheel and the right front wheel of the vehicle, respectively, and monitor the image of the left front wheel to determine whether the left front wheel is straight, and monitor the image of the right front wheel to determine whether the right front wheel is straight.

[0074] This application determines whether the left and right front wheels are straight by acquiring images of the vehicle's left and right front wheels using an image acquisition device, and then using a CNN detection algorithm and a symmetry detection algorithm to determine whether the wheels are straight. Specifically, when the steering wheel is straight, the two front wheels of the vehicle should be in a straight position, that is, the left and right front wheels should be symmetrical with respect to the vehicle's longitudinal centerline. When the steering wheel is turned, although there is an inner wheel difference in the deflection angle of the left and right wheels, it is represented by opposite morphological changes in the image coordinate system. Therefore, this application can determine whether the wheels are straight by calculating the deflection angle of the front wheels relative to the vehicle's longitudinal centerline.

[0075] Specifically, the process first uses a CNN detection algorithm to identify the left front wheel in the left front wheel image and the right front wheel in the right front wheel image. Then, a symmetry detection algorithm is used to extract features from the tire information of both wheels, calculating the first deflection angle of the left front wheel relative to the vehicle's longitudinal centerline and the second deflection angle of the right front wheel relative to the vehicle's longitudinal centerline. If the first deflection angle is less than a predetermined deflection angle threshold, the left front wheel is considered to be straight; otherwise, it is not. Similarly, if the second deflection angle is less than the predetermined deflection angle threshold, the right front wheel is considered to be straight; otherwise, it is not. When both the left and right front wheels are straight, the wheel state is considered to be in a straightened state, thus meeting the parking conditions. The predetermined deflection angle threshold can be set according to actual conditions. In this embodiment, the predetermined deflection angle threshold can be set to 20, meaning that only when the deflection angles of both wheels are greater than 20 degrees is the wheel considered to be in an unstraightened state.

[0076] In summary, this application does not rely entirely on a single image acquisition to determine whether the wheel is straight, but rather combines a symmetric detection algorithm for comprehensive determination, resulting in a low false detection rate and high robustness.

[0077] For a clear explanation of this plan, please refer to [link / reference]. Figure 3 This is a schematic diagram of an overall automatic parking scheme provided in an embodiment of this application. The automatic parking scheme includes the following: Determine if the user's vehicle is parked. If not, continue with the step of determining if the user's vehicle is parked. If parked, determine if the vehicle door is open. If open, determine if the driver's door is open. If the driver's door is open, determine if the driver's door is closed while the driver's detection area is in a state of detecting a person. If yes, determine if the driver has exited the vehicle and continue with the step of determining if the vehicle door is open. If no, determine if the driver has not exited the vehicle and continue with the step of determining if the vehicle door is open.

[0078] If the driver's door is not opened, it is determined that there are other people in the vehicle besides the driver who have not gotten out. It is then determined whether the door is closed in the non-driver's side detection area while people are detected. If so, it is determined that the people in the vehicle have gotten out, and the process continues to determine whether the vehicle door is open. If not, it is determined that the people in the vehicle have not gotten out, and the process continues to determine whether the vehicle door is open.

[0079] If the vehicle door is determined not to be open, the system checks if the driver has exited the vehicle in the previous steps. If the driver has not exited, the system continues to check if the vehicle door is open. If the driver has exited, the system checks if the parking space is unoccupied. If someone is present, the system continues to check if the vehicle door is open. If no one is present, the system checks if the last detected location of the person is within the parking space edge detection zone based on their movement trajectory. If it is not within the parking space edge detection zone, the person has not yet left the parking space, and the system continues to check if the vehicle door is open. If it is within the parking space edge detection zone, the system checks if no person is detected inside the vehicle. If someone is detected, the system continues to check if the vehicle door is open. If no person is detected, the system checks if the vehicle's rearview mirrors are folded. If they are not folded, the system continues to check if the vehicle door is open. If they are folded, the system checks if the vehicle is straightened. If it is not straightened, the system continues to check if the vehicle door is open. If it is straightened, the mechanical parking process is automatically executed, and the process ends.

[0080] The parking process described in this application allows the mechanical equipment to automatically park the vehicle once the user has parked and left the parking space. The equipment can automatically identify the vehicle's status and execute the parking process automatically when the vehicle's status meets the parking conditions, eliminating the need for the user to click a confirmation button. This achieves seamless parking, improves the user experience, and greatly ensures the safety of the multi-level parking system, avoiding the safety hazards caused by the asynchronous parking and operation of traditional equipment.

[0081] Figure 4 A structural diagram of an electronic device provided in an embodiment of the present invention, such as... Figure 4 As shown, it includes: Memory 10 is used to store computer programs; The processor 11 is configured to execute a computer program to implement the steps of the vehicle storage method as described in the above embodiments.

[0082] The electronic devices provided in this embodiment may include, but are not limited to, smartphones, tablets, laptops, or desktop computers.

[0083] The processor 11 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 11 may be implemented using at least one hardware form selected from Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 11 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 11 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 11 may also include an Artificial Intelligence (AI) processor, which handles computational operations related to machine learning.

[0084] The memory 10 may include one or more computer-readable storage media, which may be non-transitory. The memory 10 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 10 is used to store at least the following computer program 101, which, after being loaded and executed by the processor 11, is capable of implementing the relevant steps of the storage method disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 10 may also include an operating system 102 and data 103, and the storage method may be temporary storage or permanent storage. The operating system 102 may include Windows, Unix, Linux, etc.

[0085] In some embodiments, the electronic device may further include a display screen 12, an input / output interface 13, a communication interface 14, a power supply 15, and a communication bus 16.

[0086] Those skilled in the art will understand that Figure 4 The structures shown do not constitute a limitation on electronic devices and may include more or fewer components than those shown.

[0087] In another exemplary embodiment, a computer storage medium is also provided, wherein the program instructions, when executed by a processor, implement the steps of the vehicle storage method described in any of the above method embodiments.

[0088] In another exemplary embodiment, a computer program product is also provided, including a computer program / instructions that, when executed by a processor, implement the steps of the vehicle storage method described in any of the above method embodiments.

[0089] It is understood that if the vehicle storage method in the above embodiments is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the current technology, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, magnetic disk or optical disk, and other media capable of storing program code.

[0090] The various embodiments described in this specification are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” used herein may also mean the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a specific order described or illustrated, unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0091] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0092] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for parking a vehicle, characterized in that, include: Detect whether the vehicle is currently in a completed parking space; If the vehicle is in a parked state, then obtain the target exit status of the current vehicle. The target disembarkation status is used to indicate whether each target is currently inside the vehicle; Obtain the target movement trajectory of the parking space; The target motion trajectory is used to represent the positional changes of each target in the parking space; If the target has disembarked and the target's trajectory indicates that the target has left the parking space, the parking process will be executed automatically.

2. The parking method according to claim 1, characterized in that, Obtain the target exit status of the current vehicle, including: Real-time acquisition of vehicle images of the current vehicle; Based on the vehicle image, detect the status of each door of the current vehicle; If the target door is in an open state, then it is determined whether the target is detected in the door detection area; wherein, the door detection area is the door detection area in the vehicle image corresponding to the target door; If a target is detected, and the target vehicle door is detected to be in a closed state, then the target's exit state is determined to be "has exited the vehicle".

3. The parking method according to claim 2, characterized in that, If the target's disembarkation status is "already disembarked," and it is determined from the target's movement trajectory that the target has left the parking space, this includes: Determine if the target's disembarkation status is "already disembarked"; If the vehicle has already been alighted, determine whether there is a target in the parking space based on the vehicle image; If there is no target, the target's motion trajectory is used to determine whether the last detected position of each target in the parking space is an edge detection area; the edge detection area is the detection area at the edge of the parking space in the vehicle image that does not overlap with the door detection area. If it is an edge detection zone, it is determined that the target has left the parking space.

4. The parking method according to claim 1, characterized in that, Obtain the target exit status of the current vehicle, including: Detect whether there is a target inside the current vehicle; If not, then the target's disembarkation status is directly determined as "disembarked".

5. The parking method according to any one of claims 1 to 4, characterized in that, Before the automatic parking process is executed, the following also applies: Determine if there is a target inside the vehicle. If no target exists, determine whether the current vehicle's rearview mirror folding status, wheel status, and door status all meet the parking conditions. If all parking conditions are met, the vehicle is deemed to have passed the parking check, and the parking process is executed automatically.

6. The parking method according to claim 5, characterized in that, Determining whether the rearview mirror folding position meets the parking requirements includes: Based on the current vehicle image, determine whether the rearview mirror is in a folded state; if it is, determine that the rearview mirror folding state meets the parking conditions. If the vehicle is not folded, determine whether the width of the current vehicle is less than the predetermined width threshold. If the width is less than the predetermined vehicle width threshold, the rearview mirror folding state is determined to meet the parking conditions; if the width is not less than the predetermined vehicle width threshold, the rearview mirror folding state is determined to not meet the parking conditions.

7. The parking method according to claim 6, characterized in that, Determine whether the wheel condition meets the parking conditions, including: Based on the current vehicle image, calculate the first deflection angle between the left front wheel and the longitudinal centerline of the vehicle, and the second deflection angle between the right front wheel and the longitudinal centerline of the vehicle. If both the first deflection angle and the second deflection angle are less than the predetermined deflection angle threshold, then the wheel condition is determined to meet the parking conditions; otherwise, the wheel condition is determined not to meet the parking conditions.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the vehicle storage method as described in any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the vehicle storage method as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the vehicle storage method according to any one of claims 1 to 7.

Images