Parking assistance method and parking assistance device for vehicle

Abstract

The invention provides a parking assisting method for a vehicle, which comprises the following steps of: S1, when the vehicle is prepared to be parked along a lateral boundary object in a lateral constraint parking scene, estimating the adequacy of a passage space reserved for other traffic participants on one side which is not adjacent to the lateral boundary object after the vehicle is parked, the height of the lateral boundary object is enough to hinder the opening of a door of the vehicle; s2, determining a parking interval between the vehicle and a lateral boundary object in a self-adaptive manner according to the adequacy of the passing space; and S3, according to the determined parking interval, the vehicle is controlled to execute automatic parking operation in the lateral constraint parking scene. The application also proposes a parking assistance device for a vehicle, a vehicle and a computer program product. By accurately evaluating the adequacy of the passing space and intelligently setting the parking interval according to the adequacy, the traffic flow beside can be blocked as little as possible, and the user experience and the parking difficulty of the user are fully considered.

Description

Technical Field

[0001] The present application relates to a parking assistance method for a vehicle, and also relates to a parking assistance device for a vehicle, a vehicle, and a computer program product. Background Art

[0002] With the continuous progress of driving assistance technology, although vehicles equipped with an automatic parking function can greatly reduce the parking burden on drivers to a large extent, they still face many challenges in some specific scenarios.

[0003] In the urban road environment, the scenario of a vehicle parked close to a wall is very common. In this scenario, it is usually difficult to grasp the distance control between the vehicle and the wall. If the distance between the vehicle and the wall is too far, it may hinder the normal passage of oncoming traffic; while if the distance is too close, it will reduce the convenience of getting in and out of the vehicle and increase the difficulty of parking. Currently, the parking distance between the vehicle and the wall mainly depends on the subjective judgment of the human driver. However, this method has many limitations. For example, for novice drivers with insufficient experience, it may lead to poor parking results and even cause other safety hazards.

[0004] Therefore, the existing automatic parking solutions still have deficiencies. Summary of the Invention

[0005] The present application relates to a parking assistance method for a vehicle, a parking assistance device for a vehicle, a vehicle, and a computer program product, so as to at least solve some problems in the prior art.

[0006] According to a first aspect of the present application, there is provided a parking assistance method for a vehicle, the parking assistance method including the following steps:

[0007] Step S1, when the vehicle is ready to park along a lateral boundary object in a lateral constraint parking scenario, estimate the adequacy of the passing space reserved for other traffic participants on the side not adjacent to the lateral boundary object after the vehicle is parked, the height of the lateral boundary object being such that it can prevent the vehicle door from opening;

[0008] Step S2, adaptively determine the parking interval between the vehicle and the lateral boundary object according to the adequacy of the passing space; and

[0009] Step S3, control the vehicle to perform an automatic parking operation in the lateral constraint parking scenario according to the determined parking interval.

[0010] This application particularly includes the following technical concepts: By accurately evaluating the adequacy of the passing space and intelligently setting the parking intervals accordingly, it can minimize the obstruction to the adjacent traffic flow as much as possible while fully considering the user experience and the convenience of parking operations. Generally speaking, this method can provide customized parking solutions for vehicles in various complex lateral constraint scenarios, significantly improving the practicality and user acceptance of the automatic parking technology.

[0011] In an exemplary embodiment, the adequacy of the passing space is estimated in the following ways: calculating the adequacy of the passing space based on the width of the passage formed by the lateral boundary object and the other lateral boundary object opposite to it and the width of the vehicle; and / or analyzing the surrounding environment information of the vehicle by means of a machine learning model and outputting the adequacy of the passing space.

[0012] Thus, by quantifying the adequacy into specific width values, it can be calculated and characterized more intuitively and conveniently. The analysis based on the machine learning model can comprehensively consider various complex factors, thus significantly improving the accuracy and reliability of the evaluation results.

[0013] In an exemplary embodiment, the step S2 includes: at least for part of the adequacy range, as the estimated adequacy of the passing space increases, determining a larger parking interval between the vehicle and the lateral boundary object.

[0014] Thus, in this way, continuous dynamic adjustment of the parking interval can be realized, ensuring that the vehicle can obtain an appropriate parking interval in different scenarios, thereby optimizing the parking effect. Priority is given to ensuring a safe distance in narrow scenarios, and the convenience of passengers is optimized in spacious scenarios, realizing intelligent hierarchical control.

[0015] In an exemplary embodiment, the step S2 includes dynamically selecting the parking mode of the vehicle according to the adequacy, the parking mode includes a close mode and a spacious mode, the close mode and the spacious mode correspond to different parking intervals, and the step S3 includes controlling the vehicle to perform automatic parking operations according to the selected parking mode.

[0016] Thus, by dynamically selecting the close mode and the spacious mode, it is possible to meet the needs and preferences of different users while ensuring parking specifications.

[0017] In an exemplary embodiment, if the adequacy of the passing space is less than the first threshold, the close mode is selected, and the close mode corresponds to the first parking interval; if the adequacy of the passing space is greater than the first threshold, the close mode or the spacious mode is selected, the close mode corresponds to the first parking interval, and the spacious mode corresponds to the second parking interval, and the second parking interval is greater than the first parking interval.

[0018] Thus, the parking strategy can be flexibly adjusted according to different scenario characteristics: in a narrow environment, the basic safety distance is preferentially guaranteed to reduce the risk of being scratched by passing traffic, while in a spacious environment, the interval can be appropriately increased according to actual needs to improve parking safety and the convenience of users getting in and out of the vehicle.

[0019] In an exemplary embodiment, the method further includes: if the abundance of the passing space is greater than a first threshold, options of a close mode and a generous mode are output in the vehicle, and the automatic parking operation of the vehicle is determined to be controlled in the close mode or the generous mode according to the user's selection.

[0020] Thus, on the premise of ensuring compliance with safety constraint conditions, mode selection is provided to the user in the vehicle, enabling the user to independently select a parking mode according to the actual situation, enhancing the user's control over the automatic parking process, and improving the user-friendliness and applicability of the automatic parking function.

[0021] In an exemplary embodiment, the lateral boundary objects include building side walls, greening facilities, fences, and / or guardrails; and / or, the parking assistance method further includes: identifying a lateral constraint parking scenario based on the environmental perception information of the vehicle. In particular, it is recognized that a lateral constraint parking scenario is identified when the following conditions are met: an obstacle conforming to the characteristics of a lateral boundary object is identified based on the environmental perception information; the abundance of the passing space reserved for other traffic participants meets the safety conditions, and it is determined that the safety conditions are met when the abundance of the passing space is greater than a second threshold, and the second threshold is less than the first threshold; and / or, the area around the identified lateral boundary object belongs to a legal parking area.

[0022] Thus, by clarifying the type identification criteria and scenario determination logic of the lateral boundary objects, it is ensured that the automatic parking system is only activated in compliant and suitable scenarios, avoiding illegal parking or obstructing traffic in non-compliant scenarios, thereby significantly improving the reliability and safety of the system.

[0023] In an exemplary embodiment, the method further includes: during the process of controlling the vehicle to perform an automatic parking operation, continuously monitoring the change in the abundance of the reserved passing space, and when the change in the abundance exceeds a predetermined limit, re-determining the parking interval and continuing to perform the automatic parking operation according to the re-determined parking interval.

[0024] Thus, it is possible to cope with the possible dynamic changes during the parking process, such as the appearance or removal of new stationary obstacles, thereby timely adjusting the parking strategy to ensure the stability and safety of the automatic parking process.

[0025] In an exemplary embodiment, the method further includes: additionally determining the parking interval according to the presence of occupants on the side of the vehicle adjacent to the lateral boundary object.

[0026] Thereby, the convenience of getting in and out of the vehicle for the occupant can be better considered, and the situation where the occupant cannot open the door smoothly to get in and out of the vehicle due to the vehicle being parked too close to the wall can be avoided, thus enhancing the user experience.

[0027] In an exemplary embodiment, the method further includes: triggering an occupant getting-off prompt in the vehicle before controlling the vehicle to perform an automatic parking operation according to the determined parking interval, particularly the first parking interval.

[0028] Thereby, through this prompt mechanism, the occupant can be prepared to get off in advance, the inconvenience caused by the difficulty in opening the door after parking can be avoided, and at the same time, the need for emergency intervention can be reduced, thus further enhancing the practicality and user satisfaction of the automatic parking system.

[0029] In an exemplary embodiment, the method further includes: acquiring historical manual parking data associated with the current or other lateral constraint parking scenarios, and additionally determining the parking interval and / or controlling the vehicle to perform an automatic parking operation according to the historical manual parking data.

[0030] Thereby, not only the parking safety in complex or difficult parking scenarios is significantly improved, but also the probability of parking failure is greatly reduced. In addition, by analyzing the personalized operation characteristics (such as parking interval preferences, etc.) included in the historical manual parking data, this method can provide a more personalized parking experience for the user, further enhancing the practicality and user satisfaction of the automatic parking system.

[0031] In an exemplary embodiment, the historical manual parking data includes: the parking pose of the vehicle relative to the lateral boundary object, the manual parking interval, the manual parking trajectory, the manual parking operation sequence, and / or the vehicle speed change sequence.

[0032] Thereby, by comprehensively recording various types of historical manual parking data, rich reference information is provided for the automatic parking system, thus significantly enhancing the smoothness and naturalness of the automatic parking action and effectively reducing the failure risk during the automatic parking process.

[0033] In an exemplary embodiment, the parking assistance method further includes: storing the determined parking interval and / or the execution parameters of the performed automatic parking operation in association with the lateral constraint parking scenario; when the vehicle is in the same lateral constraint parking scenario again next time, using the stored parking interval and / or execution parameters to control the automatic parking operation of the vehicle.

[0034] Thereby, the memory and reuse of the automatic parking strategy for a specific scenario can be realized, the repetitive calculation overhead is reduced, and the system response speed is improved.

[0035] In an exemplary embodiment, the parking assistance method further includes: recording failure event information of the vehicle during and / or after performing automatic parking in the lateral constrained parking scenario; adjusting, according to the failure event information, the stored parking interval, the calculation method and / or calculation parameters of the next parking interval for the lateral constrained parking scenario. The failure event information particularly includes at least one of the following: the cumulative number of times of user manual intervention exceeds a predetermined number threshold; a contact event occurs between the vehicle body and a lateral boundary object; and / or, on a side not adjacent to the lateral boundary object, a contact event occurs between the vehicle and a passing traffic participant.

[0036] Thereby, the performance of the automatic parking system can be continuously and autonomously optimized, the occurrence of parking failures or unsafe events can be reduced, and the reliability and stability of the system can be improved.

[0037] According to a second aspect of the present application, there is provided a parking assistance device for a vehicle, the parking assistance device including a processor and a memory, the memory storing computer program instructions, when the computer program instructions are executed by the processor, the processor is capable of executing the parking assistance method according to the first aspect of the present application.

[0038] According to a third aspect of the present application, there is provided a vehicle, the vehicle including the parking assistance device according to the second aspect of the present application.

[0039] According to a fourth aspect of the present application, there is provided a computer program product, which includes computer program instructions, wherein when the computer program instructions are executed by a processor, the processor is capable of executing the parking assistance method according to the first aspect of the present application. Description of the Drawings

[0040] Next, the present application will be described in more detail by referring to the drawings, and the principles, features and advantages of the present application can be better understood. The drawings include:

[0041] Figure 1 A block diagram of a vehicle according to an exemplary embodiment of the present application is shown, the vehicle including a parking assistance device for a vehicle;

[0042] Figure 2 A flowchart of a parking assistance method for a vehicle according to an exemplary embodiment of the present application is shown;

[0043] Figure 3 Shows Figure 2 A flowchart of a step of the shown method;

[0044] Figure 4 Shows Figure 2 Another flowchart of a step of the shown method;

[0045] Figure 5 Shows a flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application;

[0046] Figure 6 Shows a flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application;

[0047] Figure 7 Shows a flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application;

[0048] Figure 8a 、 Figure 8b and Figure 8c Shows a schematic diagram of using the parking assistance method according to the present invention to determine a parking interval in an exemplary application scenario;

[0049] Figure 9 Shows the manual parking data stored for an exemplary lateral constraint parking scenario;

[0050] Figure 10 Shows a schematic diagram of performing an automatic parking process according to the manual parking data in a lateral constraint parking scenario. Detailed Description of the Invention

[0051] In order to make the technical problems to be solved, technical solutions and beneficial technical effects of the present application clearer, the present application will be further described in detail below in conjunction with the drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the protection scope of the present application.

[0052] Figure 1 Shows a block diagram of a vehicle 1 according to an exemplary embodiment of the present application, and the vehicle 1 includes a parking assistance device 10 for the vehicle 1.

[0053] The vehicle 1 is, for example, a vehicle 1 capable of at least partially autonomous driving and is, for example, equipped with an automatic parking function. The vehicle 1 includes a parking assistance device 10 for the vehicle 1, which includes a processor and a memory (not specifically shown for simplicity), and computer program instructions are stored in the memory. These instructions can be stored in computer-readable storage media such as a hard disk, memory, flash card, etc. The processor can be a central processing unit (CPU), microcontroller unit (MCU), graphics processing unit (GPU), neural network processing unit (NPU), digital signal processor (DSP) or other general-purpose processors. When the processor executes the computer program instructions in the memory, a parking assistance method can be implemented, which will be further elaborated below.

[0054] The parking assistance device 10 can also be connected, either wired or wirelessly (e.g., via Ethernet, in-vehicle Ethernet, CAN bus, FlexRay, MOST, HSVL, etc.), to in-vehicle environmental sensors 11, which for example include cameras, lidar sensors, radar sensors, and / or ultrasonic sensors. With the aid of an in-vehicle camera, it is for example possible to capture in real time an image of the environment around the vehicle 1, thereby enabling the type of lateral boundary objects and the parking space markings on the ground to be identified, and therefrom determining whether the vehicle 1 is in a lateral constrained parking scenario. In addition, by fusing the distance detection data of lidar sensors, radar sensors, and / or ultrasonic sensors, the scenario determination can be performed more precisely.

[0055] The parking assistance device 10 can also be connected to a communication interface 12 to receive historical manual parking data via a communication network, or upload the manual or automatic parking data already completed by the vehicle 1 to a database in the cloud if necessary. In addition, it is also possible to interact with a road supervision platform via the communication interface 12 in order to receive the parking space delineation rules issued by the road supervision platform.

[0056] The parking assistance device 10 can also be connected to a map and positioning module 13, thereby obtaining road map information and determining in real time the position information of the vehicle 1. Based on this information, the device can determine whether the vehicle 1 is in a lateral constrained parking scenario.

[0057] In addition, the parking assistance device 10 can also be connected to a status sensor 14 of the vehicle 1 for collecting status information such as the speed, acceleration, and distance from specific environmental objects of the vehicle. This information can be used to support the vehicle 1 in performing an automatic parking operation.

[0058] In addition, the parking assistance device 10 can also be connected to various input / output modules 15 so as to be able to activate or terminate the automatic parking function, select a parking mode or a parking interval according to user input. If necessary, the parking assistance device 10 can also prompt the identified lateral constrained parking scenario through the input / output module 15 and recommend a parking interval. The input / output module 15 can for example be configured as an optical, acoustic, and / or tactile input / output module, which for example includes but is not limited to an instrument panel display, a center console screen, a head-up display (HUD), and / or an augmented reality head-up display (AR-HUD), etc., and can also include a steering wheel vibration device, a seat vibration device, and a speaker, etc.

[0059] In addition, the parking assistance device 10 can also be connected to a driving control actuator 16. The driving control actuator 16 can control the lateral and longitudinal movements of the vehicle 1 according to parking assistance information such as the parking interval generated by the parking assistance device 10, thereby realizing a specific parking operation. The driving control actuator 16 generally includes a steering actuator, a braking actuator, and an acceleration actuator, etc.

[0060] It should be understood that Figure 1 the number and type of various sensors or actuators connected to the parking assistance device 10 shown in Figure 1 are merely exemplary, and the present disclosure is not intended to limit this. In practical applications, other types or numbers of sensors and actuators may be employed in the vehicle 1 to meet specific requirements and conditions.

[0061] It should also be understood that, without departing from the core concept of the present application, the connection relationships and functional divisions of the various modules may be appropriately adjusted according to actual needs, and such adjustments are all within the protection scope of the present application.

[0062] Figure 2 The flowchart of a parking assistance method for a vehicle according to an exemplary embodiment of the present application is shown. The method includes steps S1 to S3, and also includes an optional step S0, and can be executed, for example, by means of Figure 1 the parking assistance device 10 shown in Figure 1 .

[0063] In step S0, it is determined whether the vehicle to be parked is in a lateral constraint parking scenario.

[0064] The "lateral constraint parking scenario" is generally also referred to as the "parking against a wall scenario", which means that when the vehicle is parked, one side is adjacent to a lateral boundary object with a certain height (such as a wall, fence, isolation fence, etc.), and the vehicle needs to park along this lateral boundary object. This lateral boundary object is usually a static obstacle, and the height of at least part of its area is between the lower edge and the upper edge of the vehicle door, which is sufficient to prevent the normal opening of the vehicle door. The lateral boundary object may be a continuous structure, but in some cases, there may also be a partially suspended situation.

[0065] The lateral boundary object includes, for example, the side wall of a building (such as a wall), greening facilities (such as shrubs, flower beds, green belts), fences and / or guardrails, etc. In addition, the lateral boundary object may also be a row of shared bicycles or motorcycles parked on the sidewalk, a temporary construction enclosure or other municipal facilities, etc. It should be noted that, according to the definition herein, the height of the lateral boundary object is sufficient to prevent the opening of the vehicle door in a specific situation, so the curbstone, a relatively low road isolation structure, etc. do not belong to the category of the lateral boundary object defined in the context.

[0066] In one embodiment, environmental perception information can be obtained by means of in-vehicle environmental sensors (such as cameras, lidars, radars, ultrasonic sensors, etc.). These sensors can provide information such as the position, shape, and distance of objects around the vehicle. Then, based on the environmental perception information, the lateral constraint parking scenario is identified.

[0067] Exemplarily, when the following conditions are met, it can be recognized that the lateral constraint parking scenario is identified:

[0068] ·An obstacle conforming to the characteristics of a lateral boundary object is recognized in the vehicle's surrounding environment. As described above, this process can be achieved through a pre-trained machine learning model (such as an artificial neural network). The model analyzes sensor data based on an object recognition algorithm to determine whether there is an object conforming to the characteristics of a lateral boundary object.

[0069] ·The area around the recognized lateral boundary object belongs to a legal parking area. For example, the parking space marking lines on the ground can be detected and recognized through an in-vehicle camera or by fusing other sensor data. It is also possible to confirm whether the area is marked as a no-parking area with the help of map information.

[0070] Or obtain information on whether the area is a compliant parking area from the road supervision platform.

[0071] ·The adequacy of the passing space reserved for other traffic participants meets the safety conditions. Specifically, when the adequacy of the passing space is greater than a set second threshold, it is determined that the safety conditions are met. The second threshold is set, for example, to ensure that after the vehicle is parked with a minimum safety interval from the lateral boundary object, the remaining width of the passage on the side of the vehicle not adjacent to the lateral boundary object is at least greater than a specified minimum width (such as 3 meters).

[0072] Optionally, the conditions for identifying a lateral constraint parking scenario may also include: detecting that the continuous extension length of the lateral boundary object in the longitudinal direction (the vehicle driving direction) exceeds a certain limit. This can effectively exclude the interference effects of short objects such as utility poles on the recognition process.

[0073] Optionally, the conditions for identifying a lateral constraint parking scenario may also include: detecting that the lateral boundary object is in a passage scenario. For example, if there is another lateral boundary object (such as another parked vehicle, a wall, etc.) with a certain height (such as 2 meters) within a certain lateral distance range (such as 4 meters) on a specified length (such as 20 meters) parallel to the lateral boundary object, it is determined that a passage scenario is recognized. The main function of the other lateral boundary object is to restrict the driving space, but its height does not necessarily impede the opening of the car door and may also be a low object. This condition is set based on the following consideration: Only when the vehicle needs to park against a wall on one side of the passage, is it necessary to determine the parking interval, because in this case, the parking interval has a greater impact on the surrounding traffic flow. When a passage scenario is detected, that is, there are structural constraints on both sides, one of the structural constraints can be regarded as a lateral boundary object. Usually, it can be determined which side is allowed to park close according to the parking rules of other vehicles in the front and back. In most countries, the side allowed to park close is usually the right side, but there are also cases of parking on the left side.

[0074] The above-mentioned scene recognition process can be triggered when the vehicle user gives an automatic parking instruction or the vehicle is in automatic parking mode, or it can be automatically triggered when the vehicle speed is zero and reaches the navigation destination.

[0075] In another embodiment, the lateral restraint parking scenario can be identified through user input. For example, when the vehicle approaches a parking space close to a wall, the user can issue a notification through voice command or touch control unit indicating that the vehicle needs to park very close to the wall. This user input can clearly inform the system of the current parking requirements and scenario characteristics.

[0076] In another embodiment, the lateral restraint parking scenario may also be identified based on the current position of the vehicle. For example, if the current position of the vehicle matches a location corresponding to the lateral restraint parking scenario pre-stored in a map, it may be determined that the lateral restraint parking scenario is identified.

[0077] In step S1, when a vehicle is to be parked along a lateral boundary object in a lateral restraint parking scenario, the adequacy of the passage space reserved for other traffic participants on the side not adjacent to the lateral boundary object after the vehicle is parked is estimated.

[0078] Other traffic participants may include, for example, another vehicle, a pedestrian, a bicycle, a motorcycle, a tricycle, a scooter, an electric wheelchair, a baby carriage and / or a stroller, etc.

[0079] The vehicle is ready to park may refer to one of the following situations: the vehicle user issues an automatic parking command, the vehicle is in automatic parking mode, the vehicle speed is zero, and the navigation destination has been reached.

[0080] The adequacy of the passage space reflects the additional safety margin on the basis of meeting the minimum passage demand. This indicator comprehensively considers multiple factors such as the physical width of the channel and the vehicle, the safety buffer requirements of dynamic obstacles, and the minimum safe distance between the vehicle and the wall. The larger the adequacy value, the better the passage conditions. For example, a higher adequacy means that the vehicles next to it pass more smoothly, and the risk of congestion and scratches is lower; while a lower adequacy may only barely accommodate vehicles passing, and the risk of congestion and scratches is relatively high.

[0081] The clearance space adequacy estimation process can be performed simultaneously with the lateral restraint parking scene recognition process, or can be performed separately thereafter. In other words, step S0 and step S1 can be performed sequentially, alternately, or at least partially synchronously.

[0082] In one embodiment, the adequacy of the passing space can be estimated based on the width of the channel formed by a lateral boundary object and the other lateral boundary object on its opposite side, as well as the width of the vehicle itself. In this case, the adequacy can be quantitatively characterized by the space width. For example, the adequacy can be calculated by the following formula:

[0083] A = W sum - W0

[0084] Where:

[0085] A represents the basic adequacy;

[0086] W sum represents the width of the channel;

[0087] W0 represents the width of the vehicle to be parked itself, and this value can be related to the vehicle model and pre-stored in the vehicle's local database or the cloud account bound to the vehicle user.

[0088] The remaining width in the channel directly calculated therefrom can be regarded as the adequacy, or the calculation result can be mapped to a normalized value between 0 and 1.

[0089] In another embodiment, the adequacy can also be calculated by the following formula:

[0090] A = W sum - W0 - S sum

[0091] A represents the adequacy;

[0092] W sum represents the width of the channel;

[0093] S sum represents the total safety margin, which, for example, covers the minimum safety interval between the vehicle to be parked and the passing vehicle beside it, the minimum safety interval between the passing vehicle beside it and the other lateral boundary object, and the minimum safety interval between the vehicle to be parked and the lateral boundary object. These safety intervals may be related to factors such as the wall surface roughness, sensor accuracy, and system error.

[0094] Exemplarily, when a vehicle enters a passageway bounded by a building wall and a curb on both sides, the total physical width of the passageway is first obtained through a surround-view camera and ultrasonic sensors as 5.4 meters. Then, the width of the vehicle itself (e.g., 1.8 meters, including the rearview mirrors) and the necessary safety margin (e.g., 0.4 meters, including a minimum safety interval of 0.2 meters from the adjacent vehicle and a minimum safety interval of 0.2 meters from the lateral boundary object) are deducted. Finally, the available passing space for other vehicles is calculated as 3.2 meters. If this calculated result is mapped to a normalized abundance value between 0 and 1 (assuming 3 meters corresponds to 0 and above 4 meters corresponds to 1), the normalized abundance value corresponding to 3.2 meters is 0.2.

[0095] In another embodiment, a machine learning model can also be used to analyze the surrounding environment information of the vehicle and output the abundance of the passing space. In this case, the abundance is presented as a qualitative evaluation result of the spatial spaciousness, rather than a specific width value. The model is trained to automatically extract key scene features (such as boundary object types, spatial openness, passageway constraints, etc.) and make a comprehensive judgment. For example, when one side is a wall and the other side is an open multi-lane road, the model outputs "high abundance"; if it identifies a narrow passageway formed by buildings on both sides, it outputs "low abundance".

[0096] In step S2, according to the abundance of the passing space, the parking interval between the vehicle and the lateral boundary object is adaptively determined.

[0097] "Adaptive" here means dynamically adjusting the parking interval according to the abundance of the passing space detected in real time, rather than using a fixed value. This adjustment method can flexibly select the most appropriate parking interval according to different parking scenarios and environmental conditions to ensure the safety, convenience, and compliance of parking.

[0098] In one embodiment, at least for some abundance ranges, as the estimated abundance of the passing space increases, a larger parking interval between the vehicle and the lateral boundary object is determined. This can be achieved by establishing a positive correlation function relationship between the abundance and the parking interval. For example, when the abundance of the passing space increases from 0.4 to 0.7, the parking interval can be linearly adjusted from 0.36 meters to 0.42 meters. This mapping relationship may only be effective within a certain limit, for example, only when the abundance exceeds a second threshold (such as corresponding to a passageway width ≥ 3 meters). Another example is that when the abundance exceeds a certain upper limit value, other constraint conditions (such as the marked edge line of the parking space) are introduced to control the parking interval, and this mapping relationship is no longer followed.

[0099] In another embodiment, the safety level or parking mode can also be divided according to the interval range where the abundance is located. For example, for different abundance ranges, corresponding different parking intervals. This grading method will be described below in combination withFigure 3 Further elaboration.

[0100] In another embodiment, the grading method can be combined with the dynamic mapping method. For example, for different ranges of traffic space abundance, different functional relationships are respectively applied to calculate the parking interval. Specifically, for a specific range of abundance, one functional relationship is used to calculate the parking interval; while for another range of abundance, another functional relationship is adopted, so as to more accurately adapt to the parking requirements in different scenarios.

[0101] In another embodiment, the parking interval can also be additionally determined according to the presence of occupants on the side adjacent to the lateral boundary object of the vehicle. For example, the distribution of occupants can be detected by a seat pressure sensor or an in-vehicle camera. The side adjacent to the lateral boundary object can be either the co-pilot side or the driver side. If there are occupants, the parking interval determined based on the traffic space abundance is appropriately increased, or a larger value is selected within the determined reference parking interval range to improve the convenience of getting on and off the vehicle; if there are no occupants, the parking interval is appropriately decreased, or a smaller value is selected within the determined reference parking interval range to optimize space utilization.

[0102] In another embodiment, the parking interval can also be determined according to the parking space marking lines on the ground. Specifically, the parking interval should ensure that at least the vehicle wheels do not exceed the parking space marking lines on the ground, thus ensuring the compliance and safety of parking.

[0103] In addition, the parking interval can also be dynamically adjusted according to factors such as the type or material of the lateral boundary object and another lateral boundary object, the height of another lateral boundary object, the traffic flow of the parking environment, weather conditions, and ground conditions. For example, if the lateral boundary object is a soft material (such as a green belt), the vehicle can park closer appropriately; while if it is a hard wall, a larger safety interval needs to be maintained. In areas with a large traffic flow, to reduce the impact on the traffic flow, a more compact parking interval can be selected; while in areas with a small traffic flow, a more generous interval can be chosen. In addition, if the weather conditions are bad (such as raining or snowing), or the ground conditions are poor (such as slippery or having water accumulation), the parking interval can be increased to ensure safety.

[0104] In step S3, according to the determined parking interval, the vehicle is controlled to perform an automatic parking operation in the lateral constraint parking scenario. Specifically, the parking interval can be provided as a target parameter to the automatic parking module to calculate the parking trajectory and control the vehicle to perform automatic parking, ensuring that the set parking interval is followed.

[0105] Optionally, during the implementation of automatic parking, the parking interval can be re-verified or re-adjusted.

[0106] In one embodiment, if the vehicle finds that the originally set parking interval is too small or too large to complete the parking operation, it can adjust the interval on the premise of ensuring a safe passing space until the vehicle can achieve parking. If the attempt fails after multiple tries, manual takeover can be triggered and a prompt of "Automatic parking cannot be achieved" can be given.

[0107] In another embodiment, during the execution of the automatic parking operation, the adequacy change of the reserved passing space can also be continuously monitored. When the adequacy change exceeds a predetermined limit, the parking interval is re-determined. For example, if a bicycle or motorcycle originally parked on the sidewalk on one side of the vehicle leaves, making the vehicle have more ample parking space; or during the automatic parking process, traffic cones are suddenly placed on the road, resulting in the effective width of the passage being reduced from 5.2 meters to 4.8 meters. At this time, when it is detected in real time that the change amount (e.g., 0.4 meters) exceeds the tolerance threshold (e.g., 0.3 meters), the original parking plan is immediately interrupted, and a new parking interval is calculated based on the changed adequacy.

[0108] Figure 3 Shows Figure 2 A flowchart of a step of the method shown. In this embodiment, Figure 2 Step S2 of the method shown is further shown to include sub-steps S21 - S26.

[0109] In sub-step S21, it is checked whether the adequacy is greater than the second threshold T2 (e.g., corresponding to a remaining passage width of 3 meters). If it is less than this threshold, vehicle parking may obstruct traffic, so parking is prohibited in sub-step S22. If sufficient safe passing space has been used as a condition for scene recognition in the scene recognition stage (such as Figure 2 Step S0 shown), there is no need to execute sub-steps S21 and S22 anymore.

[0110] In one embodiment, the adequacy can be characterized by calculating the difference between the passage width and the width of the vehicle to be parked. Then the second threshold can be set as:

[0111] T2 = S0 + W1 + S1

[0112] Where:

[0113] S0 represents the minimum safe interval between the vehicle to be parked and the lateral boundary object;

[0114] W1 represents the width of another vehicle expected to pass by the vehicle to be parked;

[0115] S1 represents the bilateral safety interval of another vehicle (e.g., including the minimum safety intervals between another vehicle and the vehicle to be parked and between another vehicle and the other lateral boundary object of the passage).

[0116] On the premise of confirming that the abundance is greater than the second threshold T2, it is further possible to check in sub-step S23 whether the abundance is greater than the first threshold T1 (for example, corresponding to the remaining width of the passage of 3.5 meters), and the first threshold is greater than the second threshold. Next, the parking mode of the vehicle can be dynamically selected according to the abundance, and different parking modes correspond to different parking intervals. The first threshold is set, for example, as:

[0117] T1 = S0 + W1 + S1 + ΔS

[0118] Where:

[0119] S0 represents the minimum safety interval between the vehicle to be parked and the lateral boundary object;

[0120] W1 represents the width of another vehicle expected to pass by the vehicle to be parked;

[0121] S1 represents the bilateral safety interval of another vehicle (for example, including the minimum safety intervals between another vehicle and the vehicle to be parked and between another vehicle and the other lateral boundary object of the passage).

[0122] ΔS represents the safety buffer. The safety buffer comprehensively considers, for example, the space for users to get on and off the vehicle on the side of the vehicle to be parked against the wall, the smoothness of the passage of the passing vehicle, the sensor accuracy, the weather conditions, and the road surface conditions, etc.

[0123] If the abundance is less than the first threshold, the close mode can be selected in sub-step S24, and this mode corresponds to the first parking interval (for example, 30 mm to 50 mm). This indicates that although safe passage is allowed, the passage space is relatively narrow, and the vehicle must be parked closer to the wall to reduce the risk of being scratched by the traffic flow on the other side. After sub-step S24, an optional sub-step S25 can also be executed to notify the occupants to get off the vehicle by outputting optical, acoustic, or tactile cues in the vehicle, so as to avoid inconvenience to the occupants getting on and off the vehicle due to the too small parking interval.

[0124] If the abundance is greater than the first threshold, the generous mode can be selected in sub-step S26, and this mode corresponds to the second parking interval (for example, 50 mm - 70 mm). This indicates that the passage space reserved for the traffic flow after the vehicle is parked is relatively abundant, and the parking interval from the wall can be appropriately increased, so as to facilitate the occupants getting on and off the vehicle, reduce the parking operation difficulty, and improve the parking success rate.

[0125] The first parking interval corresponds, for example, to a smaller opening angle of the car door (for example, 30° to 45°), which may cause greater inconvenience to the occupants, while the second parking interval corresponds to a larger opening angle of the car door (for example, 60° to 90°), providing greater convenience for the occupants getting on and off the vehicle.

[0126] Figure 4 ShowsFigure 2 Another flowchart of a step of the method shown. In this embodiment, Figure 2 Step S2 of the method shown is shown to include sub-steps S21 - S25 and sub-steps S27 - S29. The implementation process of sub-steps S21 - S25 can refer to Figure 3 the description of the corresponding steps in Figure 4 and will not be repeated here. The following mainly elaborates on Figure 3 the differences from

[0127] If it is determined in sub-step S23 that the abundance is greater than the first threshold, instead of directly selecting the spacious mode, options for the tight mode and the spacious mode are output in the vehicle in sub-step S27, and it is checked which parking mode the user selects. For example, a comparison schematic diagram of the two modes can be displayed on the vehicle display unit to allow the user to more intuitively feel the difference in the parking intervals corresponding to the two modes.

[0128] If the user selects the tight mode, it is determined in sub-step S28 that the tight mode is selected and the first parking interval is determined. If the user selects the spacious mode, it is determined in sub-step S29 that the spacious mode is selected and the second parking interval is determined. Thus, the user can make a choice according to their own preferences and actual situations. For example, a more cautious user may be worried about the risk of scratching and will choose the tight mode even if there is a surplus, while other users may choose the spacious mode for the convenience of getting in and out of the vehicle. This design provides the user with sufficient freedom and can meet the personalized needs of different users.

[0129] In another embodiment not shown, when it is determined that the abundance of the driving space is greater than the first threshold, the "spacious mode" can be preferentially adopted by default, that is, parking is performed with a larger second parking interval. However, if there are other limiting conditions, such as the parking mode preference of the occupants, user manual intervention, or there are occupants on the side of the vehicle against the wall, etc., the system can also switch to parking with a smaller first parking interval to better meet the actual needs and user preferences.

[0130] In other embodiments not shown, when the abundance is lower than the first threshold, a fixed strategy can be used to determine the parking interval, that is, regardless of the abundance, it is determined as a fixed first parking interval. When the abundance is higher than the first threshold, the parking interval can be dynamically calculated according to the specific magnitude of the abundance (for example, through a preset functional relationship). In addition, different functional relationships can also be used to calculate the parking interval for the cases where the abundance is lower and higher than the first threshold respectively.

[0131] It should also be noted that Figure 3 and Figure 4Only the comparison of the abundance with the first threshold and the second threshold is shown. However, in practical applications, more thresholds (such as the third threshold, the fourth threshold, etc.) can be set according to requirements to achieve a further refined calculation of the parking interval.

[0132] In addition, in Figure 3 and Figure 4 the illustrated embodiment, the occupant getting-off prompt is triggered only in the close mode. However, in practical applications, this prompt can also be triggered under other abundance conditions as needed.

[0133] Figure 5 The flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application is shown. In this embodiment, Figure 2 the illustrated method further includes additional steps S4 and S5. Regarding the specific implementation processes of steps S1 - S3, reference can be made to the descriptions of the corresponding steps in Figure 3 and will not be elaborated here.

[0134] In step S4, the determined parking interval and / or the execution parameters of the performed automatic parking operation are stored in association with the lateral constraint parking scenario. The execution parameters include but are not limited to the parking trajectory, speed, acceleration, final parking pose, and the automatic parking operation sequence.

[0135] In one embodiment, for one lateral constraint parking scenario, only one parking interval can be stored, or a range of parking intervals can be stored. This range defines all the allowable parking interval values that can be selected, and the specific selection can be determined according to the user's preference and other relevant conditions (such as vehicle type, complexity of the parking scenario, etc.).

[0136] In one embodiment, the parking interval and the execution parameters can be stored in the local database of the vehicle, or uploaded to the cloud server for being called at any time when needed. To further optimize the data management and retrieval efficiency, this database can be designed as a database in the form of a map, and the above parameters are stored in association with the map data or location data. This storage method not only facilitates data management and retrieval, but also provides a more accurate reference for subsequent automatic parking operations, thereby improving the parking efficiency and reliability.

[0137] In step S5, when the vehicle is in the same lateral constraint parking scenario again next time, the stored parking interval and / or the execution parameters are used to control the automatic parking operation of the vehicle.

[0138] In one embodiment, the consistency of the scenario can be judged by analyzing the environmental information. For example, vehicle sensors (such as cameras, radars, etc.) can be used to collect information such as the shape and size of the parking space and the positions of surrounding obstacles, and compare it with the stored environmental information. If the similarity reaches the preset threshold, it is determined as the same lateral constraint parking scenario.

[0139] In another embodiment, it is possible to determine whether it is the same lateral constraint parking scenario by analyzing the geographical location information. For example, the vehicle obtains the current geographical location information through GPS and compares it with the stored coordinates. If the two are within a certain error range (such as a few meters), it is determined to be the same scenario. If the vehicle is equipped with a high-precision map, the system can match the current position with the parking scenarios in the map to more accurately identify the scenario.

[0140] In another embodiment, when the vehicle enters the same lateral constraint parking scenario again, the user can be asked whether they wish to adopt the previously stored historical automatic parking record or re-plan the parking interval. The user can make a choice according to the specific situation of the current scenario, and the system will execute the corresponding automatic parking process according to the user's choice.

[0141] In another embodiment, after retrieving the historically stored parking interval, the retrieved data can also be fine-tuned according to the actual environmental conditions or requirements. For example, adjustments can be made according to the occupant distribution, tire wear status, wall roughness, or the status of the other lateral boundary object on the opposite side.

[0142] Figure 6 The flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application is shown. In this embodiment, Figure 2 The method shown further includes an additional step S100, steps S2 and S3 are replaced by S2' and S3', and step S100 can be executed, for example, before step S1.

[0143] In step S100, historical manual parking data 20 associated with the current or other lateral constraint parking scenarios is obtained.

[0144] The historical manual parking data 20 includes: the parking pose of the vehicle relative to the lateral boundary object, the manual parking interval, the manual parking trajectory, the manual parking operation sequence, and / or the vehicle speed change sequence.

[0145] "Associated with the current or other lateral constraint parking scenarios" means that these historical data can come from a parking environment that is exactly the same as the current scenario, or from scenarios with similar but not exactly the same environmental information, such as similar channel physical widths or similar characteristics of parking space layouts and obstacle distributions. In this way, the system can utilize a wider range of historical data to optimize the current parking operation.

[0146] These historical manual parking data 20 can be associated with the current vehicle user and this vehicle, or with the same vehicle model as this vehicle. The data can be stored in the local database of the vehicle or in the cloud database for easy access at any time.

[0147] In a pre - performed manual driving scenario, for example, the user first completes the parking operation by themselves and parks the vehicle as close as possible to a lateral boundary object (such as a wall). Subsequently, the user sends a signal indicating that parking is completed through the vehicle's sound or touch unit. At this time, the system saves the current geographical location information and the manual parking interval data detected by relevant on - vehicle environment sensors. When the vehicle approaches the same scenario again next time, these stored manual parking interval data can be directly called.

[0148] When the vehicle approaches this position again, the user will be pre - asked in the vehicle through sound or visual cues whether they wish to adopt the historical preferred position. If the user confirms, the historical manual parking data 20 can be called in step S100 to assist the vehicle in completing the automatic parking operation. If the user does not confirm, or chooses to complete parking with other parking intervals, the historical manual parking data 20 will be ignored, and the parking interval will be recalculated and determined according to the current environmental characteristics.

[0149] In step S2', additionally determine the parking interval according to the historical manual parking data 20.

[0150] In one embodiment, an average value or a weighted average process can be taken for the manual parking interval and the parking interval determined based on the abundance to determine the final parking interval. For example, if the historical parking interval is 30 cm and the interval calculated in real - time is 35 cm, the system may choose 32 cm as the final parking interval.

[0151] In another embodiment, the parking interval determined in real - time can also be fine - tuned according to the user's historical parking preferences (such as whether the user prefers a looser or tighter parking interval when parking against the wall multiple times).

[0152] In another embodiment, if there are significant differences between the historical data and the real - time environmental characteristics, a higher weight can also be assigned to the parking interval determined in real - time.

[0153] In step S3', additionally control the vehicle to perform the automatic parking operation according to the historical manual parking data 20.

[0154] In one embodiment, during the automatic parking process, the actual movement trajectory of the vehicle is collected in real - time and least - squares fitting is performed with the historical manual parking trajectory. Through least - squares fitting, the sum of squared residuals between the actual trajectory and the target trajectory is calculated, and the vehicle control parameters are dynamically adjusted to minimize the sum of squared residuals. The vehicle control parameters can include the steering wheel angle rate, braking pressure gradient, drive torque distribution ratio, etc. For example, if the actual trajectory deviates from the target trajectory, the driving direction of the vehicle can be corrected by adjusting the steering wheel angle rate.

[0155] In one embodiment, if the historical manual parking data 20 includes a sequence of manual parking operations (such as a time series of steering wheel angles) or a sequence of vehicle speed changes, these sequences can be directly used to control the automatic parking process of the vehicle. During automatic parking, for example, the manual operation sequence can be converted into executable automatic control instructions through a trajectory planning algorithm.

[0156] In another embodiment, if there are multiple manual parking records for the same parking space, the trajectories of each record can be normalized on the time axis, and then the weighted average trajectory can be calculated and used as the reference trajectory for automatic parking. This method can effectively integrate multiple parking data, reduce trajectory deviations caused by single-operation errors or environmental changes, and thus provide a more stable and reliable reference trajectory for automatic parking.

[0157] In an embodiment not shown, it can be pre-checked in step S100 whether there is historical manual parking data 20 associated with the currently identified lateral constraint parking scenario. If such data is available, the system does not need to analyze the environmental characteristics in real time to determine the parking interval, but directly calls the parking interval parameters stored in the manual parking data 20, thereby simplifying the parking process and improving efficiency. On the contrary, if the manual parking data 20 is not available or not applicable, steps S2 and S3 are entered to identify and calculate the parking interval based on the abundance of the passing space, and the corresponding automatic parking process is executed.

[0158] Figure 7 The flowchart of a parking assistance method for a vehicle according to another exemplary embodiment of the present application is shown. In this embodiment, Figure 2 The method shown further includes additional steps S40 and S50.

[0159] In step S40, failure event information during and / or after the vehicle executes automatic parking in a lateral constraint parking scenario is recorded. The failure event information includes, for example, at least one of the following:

[0160] · The cumulative number of user manual interventions exceeds a predetermined number threshold (such as 3 times). This includes, for example, the driver actively taking over the vehicle control and changing the vehicle driving state or trajectory by operating the steering wheel, brake pedal, accelerator pedal, etc. This may indicate that the current automatic parking strategy cannot meet the actual requirements, or the scenario complexity exceeds the processing capacity of the system.

[0161] · An event of contact between the vehicle body and a lateral boundary object (such as a collision with a wall during automatic parking).

[0162] On the side not adjacent to the lateral boundary, the vehicle comes into contact with a passing traffic participant (e.g. being hit by a passing vehicle during parking). This may indicate that the parking interval is not sufficient to avoid a potential collision risk, or that the parking position of the vehicle causes inconvenience to other traffic participants.

[0163] In step S50, according to the failure event information, the parking interval stored for the lateral restraint parking scenario, the calculation method and / or calculation parameters of the next parking interval are adjusted.

[0164] In one embodiment, if the parking interval and automatic parking trajectory are initially set mainly based on the margin determined in real time, but multiple manual interventions are recorded during the automatic parking process, the confidence or weight of the historical manual parking data can be increased when the parking interval is subsequently calculated. In this way, in subsequent parking processes, the system will rely more on the historical manual parking experience data to determine the parking interval.

[0165] In one embodiment, if the vehicle comes into contact with a lateral boundary object or a passing traffic participant during automatic parking or parking, the safety margin of the parking interval can be increased. For example, if the current parking interval is 30 cm, it can be adjusted to 35 cm based on the failure event information to reduce the risk of collision. In addition, the coefficient in the functional relationship between the adequacy of the passing space and the parking interval can be adjusted to enhance or weaken the correlation between the adequacy and the parking interval to better adapt to different parking scenarios and safety requirements.

[0166] Figure 8a , Figure 8b and Figure 8c A schematic diagram showing how to determine a parking interval using the parking assistance method according to the present invention in an exemplary application scenario.

[0167] Figure 8a A lateral restraint parking scene with a relatively narrow space is shown. In this scene, the lateral boundary 81 is presented in the form of a wall, and the width W of the passage 80 formed by the wall and the other lateral boundary 82 opposite is marked. sum1 , the width W0 of the vehicle 1 to be parked, based on these data, the margin of the passage space for another vehicle 2 can be calculated. For the sake of clarity, the width W1 of another vehicle 2 passing by the vehicle 1 to be parked is also marked.

[0168] Based on the calculated abundance, for example, it is determined that the close parking mode is applicable to this scenario, that is, vehicle 1 needs to park against the wall 81 with a relatively small first parking interval d1. This mode ensures that vehicle 1 can be parked smoothly without obstructing the passage of other traffic participants 2. However, due to the small first parking interval d1, the space for the occupants on the side against the wall to get in and out of the vehicle is limited. Therefore, in this scenario, it is usually necessary to trigger an occupant getting-off prompt before the automatic parking operation.

[0169] Figure 8b and Figure 8c shows a lateral constraint parking scenario with relatively ample space. The channel width W sum1 , vehicle width W0, and another vehicle width W1 are also marked in this scenario. The corresponding abundance is calculated therefrom and it is determined, for example, that this scenario is applicable to both the close parking mode (as shown in Figure 8b ) and the ample parking mode (as shown in Figure 8b ). In this way, vehicle 1 can either choose to park against the wall with a small first parking interval d1 or choose to park against the wall with a large second parking interval d2. It can be seen that the second parking interval d2 is larger than the first parking interval d1, thus providing greater convenience for the occupants to get in and out of the vehicle (for example, allowing the door opening angle to be greater than 60°). Therefore, in this scenario, it is not necessary to trigger an occupant getting-off prompt before the automatic parking operation.

[0170] Figure 9 shows the manual parking data stored for an exemplary lateral constraint parking scenario.

[0171] In Figure 9 , the manual parking data is presented exemplarily in the form of the manual parking trajectory 90 of the driver of vehicle 1 during manual parking, the final parking position P2, and the manual parking interval d from the wall m . The manual parking is estimated to extend from the parking start point P1 to the parking end point P2, and the manual parking interval d m is subjectively judged by an experienced driver or the vehicle driver according to the environmental characteristics of the observed lateral constraint parking scenario and combined with the actual situation during the manual parking process.

[0172] These historical manual parking data are saved after the user successfully completes parking. For example, when vehicle 1 is parked at a position as close to the wall 81 as possible, the driver sends a signal to the vehicle indicating that the manual parking process is completed through a voice command or a touch command. Subsequently, these data are saved, including the position coordinates of vehicle 1 during the parking process and the data from the on-vehicle environmental sensors, for marking this lateral constraint parking scenario and storing it in association with the historical manual parking data.

[0173] Figure 10A schematic diagram showing an automatic parking process performed according to manual parking data in a lateral constraint parking scenario is shown.

[0174] In Figure 10 this scenario, vehicle 1 obtains surrounding environment information through on-vehicle environment sensors, and analyzes factors such as the type of lateral boundary objects, parking space size, and lane width, so as to identify that it is currently in a lateral constraint parking scenario. Subsequently, vehicle 1 can ask the user whether to enable the automatic parking function through sound and / or visual prompts, and directly use the historical manual parking records. If the user confirms, the historical manual parking records are used to guide the automatic parking operation.

[0175] In Figure 10 this scenario, the automatic parking trajectory 92 (solid line) and the manual parking trajectory 90 (dashed line) are also shown. Through the trajectory optimization algorithm, the mean square error between the automatic parking trajectory 92 and the manual parking trajectory 90 is minimized (for example, less than 0.1 meter). This indicates that when there is historical manual driving data for the current lateral constraint parking scenario, the automatic parking trajectory 90, the parking interval d m and the final parking position P2 can be mainly generated based on the manual parking data. In this way, a successful parking scheme verified by a human driver can be utilized, thereby improving the success rate of automatic parking.

[0176] Embodiments of the present application also provide a computer program product. The computer program product may include a computer program. When the computer program is executed by a processor, it can implement various operations and / or functions described above.

[0177] It should be understood that the methods of the various embodiments of the present disclosure can be implemented by computer programs / software. These software can be loaded into the working memory of the processor and used to execute the methods according to the various embodiments of the present disclosure when running.

[0178] It should be understood that the same or similar parts between the various embodiments in this specification can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, since their control logics basically correspond to the method embodiments, they are described relatively briefly, and the relevant parts can refer to the partial descriptions of the method embodiments.

[0179] Although specific embodiments of the present application are described in detail here, they are given only for the purpose of explanation and should not be considered as limiting the scope of the present application. Various substitutions, changes, and modifications can be conceived without departing from the spirit and scope of the present application.

Claims

1. A parking assistance method for a vehicle (1), the parking assistance method comprising the following steps: Step S1, when the vehicle (1) is about to park along a lateral boundary object (81) in a lateral constrained parking scenario, estimating the adequacy of the passing space reserved for other traffic participants (2) on the side not adjacent to the lateral boundary object (81) after the vehicle (1) is parked, wherein the height of the lateral boundary object (81) is such that it is sufficient to prevent the opening of the vehicle door of the vehicle (1); Step S2, adaptively determining the parking interval between the vehicle (1) and the lateral boundary object (81) according to the adequacy of the passing space; and Step S3, controlling the vehicle (1) to perform an automatic parking operation in the lateral constrained parking scenario according to the determined parking interval.

2. The parking assistance method according to claim 1, wherein, The adequacy of the passing space is estimated by the following means: Calculating the adequacy of the passing space based on the width of the passage formed by the lateral boundary object (81) and the other lateral boundary object (82) on the opposite side thereof and the width of the vehicle (1); and / or Analyzing the surrounding environment information of the vehicle (1) by means of a machine learning model and outputting the adequacy of the passing space.

3. The parking assistance method according to claim 1 or 2, wherein, The step S2 includes: At least for part of the adequacy range, as the estimated adequacy of the passing space increases, determining a larger parking interval between the vehicle (1) and the lateral boundary object (81).

4. The parking assistance method according to any one of claims 1 to 3, wherein, The step S2 includes dynamically selecting a parking mode of the vehicle (1) according to the adequacy, the parking mode including a close mode and a generous mode, the close mode and the generous mode corresponding to different parking intervals, and the step S3 includes controlling the vehicle (1) to perform an automatic parking operation according to the selected parking mode, wherein in particular: - If the adequacy of the passing space is less than a first threshold, the close mode is selected, and the close mode corresponds to a first parking interval; - If the adequacy of the passing space is greater than the first threshold, the close mode or the generous mode is selected, the close mode corresponds to a first parking interval, and the generous mode corresponds to a second parking interval, and the second parking interval is greater than the first parking interval.

5. The parking assistance method according to claim 4, wherein, The method further includes: if the adequacy of the passing space is greater than the first threshold, outputting options of the close mode and the generous mode in the vehicle (1), and determining whether to control the automatic parking operation of the vehicle (1) in the close mode or the generous mode according to the user's selection.

6. The parking assistance method according to any one of claims 1 to 5, wherein, The lateral boundary object (81) includes a building side wall, a greening facility, a fence and / or a guardrail; and / or The parking assistance method further includes: identifying a lateral constrained parking scenario based on the environmental perception information of the vehicle (1), wherein a lateral constrained parking scenario is recognized when the following conditions are met: - Based on the environmental perception information, an obstacle conforming to the characteristics of the lateral boundary object (81) is recognized; - The adequacy of the passing space reserved for other traffic participants (2) meets a safety condition, and it is determined that the safety condition is met when the adequacy of the passing space is greater than a second threshold, and the second threshold is less than the first threshold; and / or - The area around the identified lateral boundary object (81) belongs to a legal parking area.

7. The parking assistance method according to any one of claims 1 to 6, wherein, The method further includes: during the process of controlling the vehicle (1) to perform an automatic parking operation, continuously monitoring the change in the adequacy of the reserved passage space, and when the change in the adequacy exceeds a predetermined limit, re-determining the parking interval and continuing to perform the automatic parking operation according to the re-determined parking interval.

8. The parking assistance method according to any one of claims 1 to 7, wherein, The method further includes: additionally determining the parking interval according to the presence of occupants on the side of the vehicle (1) adjacent to the lateral boundary object (81).

9. The parking assistance method according to any one of claims 1 to 8, wherein, The method further includes: triggering an occupant getting-off prompt in the vehicle (1) before controlling the vehicle (1) to perform an automatic parking operation according to the determined parking interval, especially the first parking interval.

10. The parking assistance method according to any one of claims 1 to 9, wherein, The method further includes: obtaining historical manual parking data associated with the current or other lateral constraint parking scenarios, and additionally determining the parking interval and / or controlling the vehicle (1) to perform an automatic parking operation according to the historical manual parking data.

11. The parking assistance method according to claim 10, wherein, The historical manual parking data includes: the parking pose of the vehicle (1) relative to the lateral boundary object (81), the manual parking interval, the manual parking trajectory, the manual parking operation sequence, and / or the vehicle speed change sequence.

12. The parking assistance method according to any one of claims 1 to 11, wherein, The parking assistance method further includes: storing the determined parking interval and / or the execution parameters of the performed automatic parking operation in association with the lateral constraint parking scenario; when the vehicle (1) is in the same lateral constraint parking scenario again next time, using the stored parking interval and / or execution parameters to control the automatic parking operation of the vehicle (1).

13. The parking assistance method according to any one of claims 1 to 12, the parking assistance method further includes: recording the failure event information during and / or after the vehicle (1) performs automatic parking in the lateral constraint parking scenario; adjusting the stored parking interval, the calculation method of the next parking interval, and / or the calculation parameters for the lateral constraint parking scenario according to the failure event information; wherein the failure event information especially includes at least one of the following: - The cumulative number of user manual interventions exceeds a predetermined number threshold; - An event of contact between the vehicle body and the lateral boundary object (81); and / or - An event of contact between the vehicle (1) and passing traffic participants on the side not adjacent to the lateral boundary object (81).

14. A parking assistance device (10) for a vehicle (1), the parking assistance device (10) includes a processor and a memory, and the memory stores computer program instructions, when the computer program instructions are executed by the processor, the processor can execute the parking assistance method according to any one of claims 1 to 13.

15. A vehicle (1), the vehicle (1) includes the parking assistance device (10) according to claim 14.

16. A computer program product, comprising computer program instructions, wherein, When the computer program instructions are executed by the processor, the processor can execute the parking assistance method according to any one of claims 1 to 13.

Images