CN117818587A

CN117818587A - Vehicle control method, apparatus, computer-readable storage medium, and electronic device

Info

Publication number: CN117818587A
Application number: CN202410253319.9A
Authority: CN
Inventors: 艾锐; 薛瑞彬; 刘晓; 梅崴; 顾维灏
Original assignee: Haomo Zhixing Technology Co Ltd
Current assignee: Haomo Zhixing Technology Co Ltd
Priority date: 2024-03-06
Filing date: 2024-03-06
Publication date: 2024-04-05

Abstract

The invention discloses a vehicle control method, a vehicle control device, a computer readable storage medium and electronic equipment, and relates to the field of vehicle control. The method comprises the following steps: in response to receiving an automatic parking instruction, obtaining perception information obtained by perceiving the surrounding environment of the vehicle; detecting the sensing information by a parking space and a limiter to obtain at least one detection result; performing de-duplication processing on at least one detection result based on a parking space area in the at least one detection result, and determining a target detection result; generating a parking path of the vehicle according to a target parking space area and a target limiter area contained in the target detection result, and controlling the vehicle to automatically park to the parking space based on the parking path; and constructing and generating a target limiter area in the target parking space area in the target detection result based on the target parking space area and at least one detection result, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path.

Description

Vehicle control method, apparatus, computer-readable storage medium, and electronic device

Technical Field

The present invention relates to the field of vehicle control, and in particular, to a vehicle control method, apparatus, computer-readable storage medium, and electronic device.

Background

With the development of automatic driving systems, more and more vehicles are equipped with an automatic parking function, and the vehicles can realize automatic parking by using the automatic parking function. In the related art, although a vehicle can realize automatic parking using an automatic parking function, an automatic driving system often cannot accurately control the vehicle to automatically park to a proper position in a parking space during automatic parking due to complex surrounding environments of the parking space, resulting in poor accuracy of the vehicle during automatic parking.

In view of the above problems, no effective solution has been proposed at present.

Disclosure of Invention

The embodiment of the invention provides a vehicle control method, a vehicle control device, a computer-readable storage medium and electronic equipment, which are used for at least solving the technical problem of poor accuracy of controlling a vehicle to automatically park to a parking space in the related technology.

According to an aspect of an embodiment of the present invention, there is provided a vehicle control method including: in response to receiving an automatic parking instruction, obtaining perception information obtained by perceiving the surrounding environment of the vehicle; detecting the sensing information by a parking space and a limiter to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space; performing de-duplication processing on at least one detection result based on a parking space area in the at least one detection result, and determining a target detection result; responding to the target detection result to comprise a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path; and responding to the fact that a limiter does not exist in the parking space represented by the limiter detection result in the target detection result, constructing and generating a target limiter region in the target parking space region in the target detection result based on the target parking space region and at least one detection result, generating a parking path of the vehicle according to the target parking space region and the target limiter region, and controlling the vehicle to automatically park to the parking space based on the parking path.

Optionally, in controlling the vehicle to automatically park into the parking space based on the parking path, the method further includes: determining a target point position of a target point on a current vehicle, wherein the target point is used for representing a midpoint between two preset wheels on the vehicle; based on the relative positional relationship between the target point position and the target limiter region, the parking path or the vehicle posture of the vehicle is adjusted in real time in controlling the vehicle to automatically park into the parking space based on the parking path.

Optionally, generating a parking path of the vehicle according to the target parking space area and the target limiter area includes: determining a target parking position of the vehicle in a parking space corresponding to the target parking space region based on attribute information of the vehicle, the target parking space region and the target limiter region; a parking path is generated based on a current vehicle position of the vehicle and a target parking position.

Optionally, determining the target parking position of the vehicle in the parking space corresponding to the target parking space region based on the attribute information of the vehicle, the target parking space region and the target limiter region includes: acquiring vehicle parking information of vehicles parked in other parking spaces adjacent to the parking space corresponding to the target parking space area; the target parking position is determined based on the attribute information of the vehicle, the target parking space region, the target limiter region, and the vehicle parking information.

Optionally, generating a parking path of the vehicle according to the target parking space area and the target limiter area includes: outputting a target parking space area; in response to receiving a selection instruction generated by selecting a target parking space area, determining a selected parking space area corresponding to the selection instruction; a parking path is generated based on the selected parking space region and a target limiter region within the selected parking space region.

Optionally, constructing a target limiter area inside the target parking space area in the target detection result based on the at least one detection result includes: determining whether the at least one detection result contains a limiter area; responding to at least one detection result without a limiter area, and generating a target limiter area in a target parking space area in a target detection result based on a preset position relationship between the parking space and the limiter; determining whether a parking space area in the detection result to which the limiter area belongs overlaps with a target parking space area or not in response to the fact that the limiter area is included in at least one detection result; responding to overlapping of the parking space area and the target parking space area in the detection result of the limiter area, and constructing and generating a target limiter area inside the target parking space area based on the limiter area; and constructing and generating a target stopper region in the target parking space region based on the size relationship between the parking space region and the target parking space region in the detection result of the stopper region and the target position relationship between the parking space region and the stopper region in the detection result of the stopper region in response to the fact that the parking space region and the target parking space region in the detection result of the stopper region are not overlapped.

Optionally, performing deduplication processing on at least one detection result based on the parking space region in the at least one detection result, and determining the target detection result includes: sequencing each parking space region in at least one detection result based on the detection confidence of the parking space region in the detection result to obtain a parking space region sequence; if a parking space area which is not read as a first parking space area exists in the current parking space area sequence, reading the current first parking space area from the current parking space area sequence according to the parking space area sequence, and removing a second parking space area which is positioned behind the current first parking space area and has an overlapping relation with the current first parking space area in the current parking space area sequence; and if the parking space area which is not read as the first parking space area does not exist in the current parking space area sequence, determining the detection result of the parking space area in the current parking space area sequence as a target detection result.

Optionally, determining whether there is an overlapping relationship between the first parking space region and the second parking space region includes: determining whether an overlapping region exists between the first parking space region and the second parking space region based on a positional relationship between an end point of the first parking space region and the second parking space region, a positional relationship between an end point of the second parking space region and the first parking space region, and a positional relationship between a boundary of the first parking space region and a boundary of the second parking space region; determining, in response to the existence of the overlapping region between the first parking space region and the second parking space region, region overlapping information between the first parking space region and the second parking space region, the region overlapping information characterizing a degree of region overlapping between the first parking space region and the second parking space region, based on a region shape and a relative positional relationship of the first parking space region and the second parking space region; if the region overlapping information exceeds the preset target region overlapping information, determining that an overlapping relationship exists between the first parking space region and the second parking space region; and if no overlapping area exists between the first parking space area and the second parking space area or the area overlapping information does not exceed the target area overlapping information, determining that no overlapping relation exists between the first parking space area and the second parking space area.

Optionally, determining whether an overlapping region exists between the first parking space region and the second parking space region based on a positional relationship between an end point of the first parking space region and the second parking space region, a positional relationship between an end point of the second parking space region and the first parking space region, and a positional relationship between a boundary of the first parking space region and a boundary of the second parking space region, includes: judging whether an endpoint in a second parking space area exists in the first parking space area, whether an endpoint in the first parking space area exists in the second parking space area, and whether a boundary of the first parking space area intersects with a boundary of the second parking space area; if the first parking space area has an end point positioned in the second parking space area, or the second parking space area has an end point positioned in the first parking space area, or the boundary of the first parking space area is intersected with the boundary of the second parking space area, determining that an overlapping area exists between the first parking space area and the second parking space area; if the first parking space area does not have an endpoint located in the second parking space area, the second parking space area does not have an endpoint located in the first parking space area, and the boundary of the first parking space area is not intersected with the boundary of the second parking space area, it is determined that an overlapping area does not exist between the first parking space area and the second parking space area.

Optionally, determining whether the endpoint of the second parking space region is located in the first parking space region includes: determining at least one first edge forming a first parking space area boundary, and respectively determining the pointing information of each first edge, wherein the pointing information of at least one first edge forms a clockwise direction or a counterclockwise direction; determining a relative positional relationship between the endpoint and the first parking space region based on the pointing information of the endpoint and the first edge of the second parking space region, wherein the relative positional relationship between the endpoint and the first parking space region is used for representing that the endpoint and the first parking space region are located on the same side or different sides of the first edge; determining that the endpoint is not located in the first parking space region if the endpoint is located in the second parking space region and the first parking space region are located on different sides of the first edge; in the event that there is no endpoint in the second parking space region and the first parking space region is on a different side of the first side, determining that the endpoint is located in the first parking space region.

Optionally, determining whether the boundary of the first parking space region intersects the boundary of the second parking space region includes: acquiring the endpoint coordinates of each first side in at least one first side forming the first parking space region boundary and the endpoint coordinates of each second side in at least one second side forming the second parking space region boundary, wherein the endpoint coordinates consist of endpoint abscissa and endpoint ordinate; determining that the boundary of the first parking space region and the boundary of the second parking space region do not intersect in response to the maximum endpoint abscissa of the at least one first side being less than the minimum endpoint abscissa of the at least one second side, or the minimum endpoint abscissa of the at least one first side being greater than the maximum endpoint abscissa of the at least one second side, or the maximum endpoint ordinate of the at least one first side being less than the minimum endpoint ordinate of the at least one second side, or the minimum endpoint ordinate of the at least one first side being greater than the maximum endpoint ordinate of the at least one second side; and determining that the boundary of the first parking space region intersects the boundary of the second parking space region in response to the maximum endpoint abscissa of the at least one first side being greater than or equal to the minimum endpoint abscissa of the at least one second side and the minimum endpoint abscissa of the at least one first side being less than or equal to the maximum endpoint abscissa of the at least one second side and the maximum endpoint ordinate of the at least one first side being greater than or equal to the minimum endpoint ordinate of the at least one second side and the minimum endpoint ordinate of the at least one first side being less than or equal to the maximum endpoint ordinate of the at least one second side.

Optionally, determining the region overlapping information between the first parking space region and the second parking space region based on the region shape and the relative positional relationship of the first parking space region and the second parking space region includes: determining region overlapping information of the first parking space region and the second parking space region based on endpoint coordinates of each of at least one first side constituting a first parking space region boundary and endpoint coordinates of each of at least one second side constituting a second parking space region boundary in response to the region shapes of the first parking space region and the second parking space region being rectangular; and determining region overlapping information of the first parking space region and the second parking space region based on the relative position relationship between the first parking space region and the second parking space region and the number of pixel points used for representing the first parking space region and the second parking space region in the perception information respectively in response to the fact that the region shapes of the first parking space region and the second parking space region are not rectangular.

Optionally, determining the region overlapping information of the first parking space region and the second parking space region based on the relative positional relationship between the first parking space region and the second parking space region and the number of pixels used for characterizing the first parking space region and the second parking space region in the perception information respectively includes: determining the number of pixel points used for representing the overlapping area in the perception information based on the relative position relation between the first parking space area and the second parking space area; and determining region overlapping information based on the number of pixels in the perception information, which are used for representing the first parking space region, the second parking space region and the overlapping region respectively.

Optionally, detecting the sensing information by the parking space and the limiter to obtain at least one detection result, including: generating a target image based on the perception information; detecting a parking space and a limiter by using a target detection model to obtain a detection result of an opening point of the parking space; and regressing the corner points of the parking space and the end points of the limiter based on the detection results of the opening points to obtain at least one detection result.

Optionally, in response to the vehicle being provided with a plurality of sensing devices, generating the target image based on the sensing information includes: generating a plurality of partial images based on the sensing information sensed by the plurality of sensing devices; splicing the plurality of local images to obtain a spliced image; and converting the spliced image into a bird's eye view space to obtain a target image.

According to another aspect of the embodiment of the present invention, there is also provided a vehicle control apparatus including: the acquisition module is used for responding to the received automatic parking instruction and acquiring perception information obtained by perceiving the surrounding environment of the vehicle; the detection module is used for detecting the sensing information to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space; the de-duplication module is used for de-duplication processing at least one detection result based on the parking space area in the at least one detection result, and determining a target detection result; the first control module is used for responding to the target detection result and comprising a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path; the second control module is used for responding to the fact that a limiter does not exist in the parking space in the feature of the limiter detection result in the target detection result, constructing and generating a target limiter area in the target parking space area in the target detection result based on the target parking space area and at least one detection result, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to park to the parking space automatically based on the parking path.

Optionally, the apparatus further comprises: the system comprises a target point position determining module, a target point position determining module and a target point position determining module, wherein the target point position determining module is used for determining the target point position of a target point on a current vehicle, and the target point is used for representing a midpoint between two preset wheels on the vehicle; and the parameter adjustment module is used for adjusting the parking path or the vehicle posture of the vehicle in real time in the process of controlling the vehicle to automatically park to the parking space based on the parking path based on the relative position relation between the target point position and the target limiter area.

Optionally, the first control module and the second control module include: the position determining unit is used for determining a target parking position of the vehicle in a parking space corresponding to the target parking space area based on the attribute information of the vehicle, the target parking space area and the target limiter area; and the first path generation unit is used for generating a parking path based on the current vehicle position and the target parking position of the vehicle.

Optionally, the location determining unit is further configured to: acquiring vehicle parking information of vehicles parked in other parking spaces adjacent to the parking space corresponding to the target parking space area; the target parking position is determined based on the attribute information of the vehicle, the target parking space region, the target limiter region, and the vehicle parking information.

Optionally, the first control module and the second control module include: the area output unit is used for outputting a target parking space area; the first area determining unit is used for determining a selected parking space area corresponding to the selection instruction in response to receiving the selection instruction generated by selecting the target parking space area; and the second path generating unit is used for generating a parking path based on the selected parking space area and the target limiter area inside the selected parking space area.

Optionally, the second control module includes: a second area determining unit for determining whether the at least one detection result includes a limiter area; the first region generation unit is used for generating a target limiter region in the target parking space region in the target detection result based on a preset position relationship between the parking space and the limiter in response to the fact that the limiter region is not contained in at least one detection result; the overlapping detection unit is used for determining whether the parking space area and the target parking space area in the detection result to which the limiter area belongs overlap or not in response to the fact that the limiter area is included in at least one detection result; the second region generation unit is used for responding to the overlapping of the parking space region and the target parking space region in the detection result of the limiter region, and constructing and generating a target limiter region in the target parking space region based on the limiter region; and the third region generating unit is used for constructing and generating a target stopper region in the target parking space region based on the size relationship between the parking space region and the target parking space region in the detection result of the stopper region and the target position relationship between the parking space region and the stopper region in the detection result of the stopper region in response to the fact that the parking space region and the target parking space region in the detection result of the stopper region are not overlapped.

Optionally, the deduplication module comprises: the region ordering unit is used for ordering each parking space region in at least one detection result based on the detection confidence of the parking space region in the detection result to obtain a parking space region sequence; a sequence updating unit, configured to, if there is a parking space region in the current parking space region sequence that has not been read as a first parking space region, read the current first parking space region from the current parking space region sequence according to the parking space region sequence, and reject a second parking space region in the current parking space region sequence that is located behind the current first parking space region and has an overlapping relationship with the current first parking space region; and the result determining unit is used for determining the detection result of the parking space area in the current parking space area sequence as a target detection result if the parking space area which is not read as the first parking space area does not exist in the current parking space area sequence.

Optionally, the deduplication module further comprises: an overlap region determining unit configured to determine whether an overlap region exists between the first parking space region and the second parking space region based on a positional relationship between an end point of the first parking space region and the second parking space region, a positional relationship between an end point of the second parking space region and the first parking space region, and a positional relationship between a boundary of the first parking space region and a boundary of the second parking space region; an overlapping degree determining unit configured to determine, in response to the existence of an overlapping region between the first parking space region and the second parking space region, region overlapping information between the first parking space region and the second parking space region, the region overlapping information characterizing a region overlapping degree between the first parking space region and the second parking space region, based on a region shape and a relative positional relationship of the first parking space region and the second parking space region; a first overlap relation determining unit configured to determine that an overlap relation exists between the first parking space region and the second parking space region if the region overlap information exceeds the preset target region overlap information; and a second overlapping relation determining unit configured to determine that there is no overlapping relation between the first parking space region and the second parking space region if there is no overlapping region between the first parking space region and the second parking space region or the region overlapping information does not exceed the target region overlapping information.

Optionally, the overlapping region determining unit is further configured to: judging whether an endpoint in a second parking space area exists in the first parking space area, whether an endpoint in the first parking space area exists in the second parking space area, and whether a boundary of the first parking space area intersects with a boundary of the second parking space area; if the first parking space area has an end point positioned in the second parking space area, or the second parking space area has an end point positioned in the first parking space area, or the boundary of the first parking space area is intersected with the boundary of the second parking space area, determining that an overlapping area exists between the first parking space area and the second parking space area; if the first parking space area does not have an endpoint located in the second parking space area, the second parking space area does not have an endpoint located in the first parking space area, and the boundary of the first parking space area is not intersected with the boundary of the second parking space area, it is determined that an overlapping area does not exist between the first parking space area and the second parking space area.

Optionally, the overlapping region determining unit is further configured to: determining at least one first edge forming a first parking space area boundary, and respectively determining the pointing information of each first edge, wherein the pointing information of at least one first edge forms a clockwise direction or a counterclockwise direction; determining a relative positional relationship between the endpoint and the first parking space region based on the pointing information of the endpoint and the first edge of the second parking space region, wherein the relative positional relationship between the endpoint and the first parking space region is used for representing that the endpoint and the first parking space region are located on the same side or different sides of the first edge; determining that the endpoint is not located in the first parking space region if the endpoint is located in the second parking space region and the first parking space region are located on different sides of the first edge; in the event that there is no endpoint in the second parking space region and the first parking space region is on a different side of the first side, determining that the endpoint is located in the first parking space region.

Optionally, the overlapping region determining unit is further configured to: acquiring the endpoint coordinates of each first side in at least one first side forming the first parking space region boundary and the endpoint coordinates of each second side in at least one second side forming the second parking space region boundary, wherein the endpoint coordinates consist of endpoint abscissa and endpoint ordinate; determining that the boundary of the first parking space region and the boundary of the second parking space region do not intersect in response to the maximum endpoint abscissa of the at least one first side being less than the minimum endpoint abscissa of the at least one second side, or the minimum endpoint abscissa of the at least one first side being greater than the maximum endpoint abscissa of the at least one second side, or the maximum endpoint ordinate of the at least one first side being less than the minimum endpoint ordinate of the at least one second side, or the minimum endpoint ordinate of the at least one first side being greater than the maximum endpoint ordinate of the at least one second side; and determining that the boundary of the first parking space region intersects the boundary of the second parking space region in response to the maximum endpoint abscissa of the at least one first side being greater than or equal to the minimum endpoint abscissa of the at least one second side and the minimum endpoint abscissa of the at least one first side being less than or equal to the maximum endpoint abscissa of the at least one second side and the maximum endpoint ordinate of the at least one first side being greater than or equal to the minimum endpoint ordinate of the at least one second side and the minimum endpoint ordinate of the at least one first side being less than or equal to the maximum endpoint ordinate of the at least one second side.

Optionally, the overlapping degree determining unit is further configured to: determining region overlapping information of the first parking space region and the second parking space region based on endpoint coordinates of each of at least one first side constituting a first parking space region boundary and endpoint coordinates of each of at least one second side constituting a second parking space region boundary in response to the region shapes of the first parking space region and the second parking space region being rectangular; and determining region overlapping information of the first parking space region and the second parking space region based on the relative position relationship between the first parking space region and the second parking space region and the number of pixel points used for representing the first parking space region and the second parking space region in the perception information respectively in response to the fact that the region shapes of the first parking space region and the second parking space region are not rectangular.

Optionally, the overlapping degree determining unit is further configured to: determining the number of pixel points used for representing the overlapping area in the perception information based on the relative position relation between the first parking space area and the second parking space area; and determining region overlapping information based on the number of pixels in the perception information, which are used for representing the first parking space region, the second parking space region and the overlapping region respectively.

Optionally, the detection module includes: an image generation unit for generating a target image based on the perception information; the opening point detection unit is used for detecting the parking space and the limiter on the target image by utilizing the target detection model to obtain a detection result of the opening point of the parking space; and the opening point regression unit is used for carrying out regression on the corner points of the parking space and the end points of the limiter based on the detection result of the opening point to obtain at least one detection result.

Optionally, in response to the vehicle being provided with a plurality of sensing devices, the image generating unit is further configured to: generating a plurality of partial images based on the sensing information sensed by the plurality of sensing devices; splicing the plurality of local images to obtain a spliced image; and converting the spliced image into a bird's eye view space to obtain a target image.

According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium including a stored program, wherein the program, when run, controls a processor of a device in which the method of any one of the above is performed.

According to another aspect of the embodiment of the present invention, there is also provided an electronic device, including: one or more processors; a storage means for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to perform the method of any of the above.

In the embodiment of the invention, the sensing information obtained by sensing the surrounding environment of the vehicle is adopted; detecting the sensing information by a parking space and a limiter to obtain at least one detection result; performing de-duplication processing on at least one detection result based on a parking space area in the at least one detection result, and determining a target detection result; responding to the target detection result to comprise a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path; in response to the fact that a limiter does not exist in a parking space represented by a limiter detection result in a target detection result, a target limiter area in a target parking space area in the target detection result is constructed and generated based on at least one detection result, a parking path of a vehicle is generated according to the target parking space area and the target limiter area, and the vehicle is automatically parked to the parking space based on the parking path.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:

FIG. 1 is a flow chart illustrating a vehicle control method according to an embodiment of the present invention;

FIG. 2 is a schematic illustration of a parking space region overlap, according to an embodiment of the present invention;

FIG. 3 is a schematic diagram illustrating a parking space region overlap determination according to an embodiment of the present invention;

fig. 4 is a block diagram showing a structure of a vehicle control apparatus according to an embodiment of the invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Example 1

According to an embodiment of the present invention, a method embodiment of parking space detection is provided, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and that, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from that shown or described herein.

Fig. 1 is a flowchart illustrating a vehicle control method according to an embodiment of the present invention, as shown in fig. 1, including the steps of:

step S102, in response to receiving the automatic parking instruction, obtaining perception information obtained by perceiving the surrounding environment of the vehicle.

The sensing information may refer to information obtained by sensing the surrounding environment of the vehicle through various sensing devices (e.g., various sensors) on the vehicle, and based on the sensing information, the vehicle may be controlled to automatically park from the current position to a parking space near the vehicle, and the sensing information may be mainly classified into two types, and one type may be information related to the parking space in the surrounding environment of the vehicle, and may include, but is not limited to: information such as the use condition of a parking space in the surrounding environment of the vehicle, the position of the parking space and the like; another type may be information related to the travel of the vehicle during automatic parking of the vehicle into a parking space, which may include, but is not limited to: and information such as road surface obstacle conditions of the current road of the vehicle. It should be noted that the type and the amount of the above-mentioned sensing information may be set by the staff, and are not limited herein, for example, the type of the sensing information may include but not limited to: the image type, the electric signal type, the optical signal type, and the like, if the sensing information is the image type, the acquired sensing information can be processed by using an image processing mode, if the sensing information is the electric signal type, the sensing information can be processed by using an electric signal data processing mode, and if the sensing information is the optical signal type, the optical signal can be converted into an electric signal, and then the sensing information can be processed by using an electric signal data processing mode. If the sensing information of one angle, such as the sensing information of the environment in front of the vehicle, is acquired, the sensing information can be directly processed, and if the sensing information of a plurality of angles, such as the sensing information of the environment around the vehicle, is acquired, the sensing information of a plurality of angles can be combined into one fused sensing information, and then the fused sensing information is processed, such as the image information is spliced into one spliced image, and then the spliced image is processed.

It should be noted that the automatic parking function may be applied not only to an automatic driving vehicle but also to a conventional driver driving vehicle, and therefore, in the case where the automatic parking function is applied to an automatic driving vehicle, the above-described automatic parking instruction may be an instruction generated by an automatic driving system in the automatic driving vehicle; in the case where the automatic parking function is applied to the driver driving the vehicle, the above-described automatic parking instruction may be an instruction generated by the driver by operating a button or control on the vehicle.

In an alternative scheme of the embodiment, in order to accurately park the vehicle in the available parking space nearby, after receiving the automatic parking instruction, the automatic driving system may first perform information sensing operation on the surrounding environment of the vehicle by using various sensing devices on the vehicle to obtain sensing information capable of assisting the vehicle to automatically park in the surrounding environment of the vehicle, through which the automatic driving system can accurately locate the available parking space nearby the vehicle, and plan driving parameters capable of controlling the vehicle to safely and accurately drive to the parking space, thereby completing the operation of automatic parking of the vehicle.

In general, the information that can be sensed by various sensors on the vehicle is complex and various, and may include a lot of information with low relevance to the automatic parking of the vehicle, for example, the sensed information may include: the information such as the temperature, humidity, vehicle type of other vehicles, working conditions of working personnel in the parking lot and the like in the surrounding environment of the vehicle, and the degree of correlation between the information and the automatic parking process of the vehicle is low, which may affect the accuracy of the automatic parking control of the vehicle by the automatic driving system, so that in order to ensure that the automatic parking control of the vehicle can be accurately controlled, the automatic driving system may preprocess the information image sensed by the sensors on the vehicle, for example, remove some information with low degree of correlation between the sensing information and the automatic parking process of the vehicle, and retain some information with high degree of correlation between the sensing information and the automatic parking process of the vehicle as the sensing information obtained by sensing the surrounding environment of the vehicle finally obtained in step S102.

Step S104, detecting the sensing information to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space.

When the automatic parking instruction is received, there may be many parking spaces in the surrounding environment of the vehicle, but some parking spaces are not available (i.e., the parking spaces cannot be used for parking of the current vehicle) due to the parked vehicle or other obstacles, so the detection of the parking spaces on the sensing information in step S104 may specifically be the detection of available parking spaces on the sensing information, and the detection results obtained by detecting the parking spaces and the limiters on the sensing information specifically include the detection results of the parking spaces and the limiters in the surrounding environment of the vehicle, where the limiters are used for characterizing that no limiter exists inside the available parking spaces or the limiters inside the available parking spaces.

Specifically, all detection results obtained by detecting the available parking spaces and limiters for the sensing information may be used as at least one detection result obtained by detecting the parking spaces and limiters for the sensing information in step S104. Each of the at least one detection result includes a parking space region representing an available parking space in a surrounding environment of the vehicle and a limiter detection result inside the available parking space, the limiter detection result being used to represent a limiter region in which no limiter is present inside the available parking space or in which a limiter is present inside the available parking space.

In an alternative of this embodiment, in order to accurately control the automatic parking of the vehicle into the parking space, the autopilot system needs to accurately determine the usage of the parking space in the surrounding environment of the vehicle, and determine a parking space in which the vehicle can be parked currently according to the usage. Therefore, after the above-mentioned perception information is obtained, the automatic driving system may first detect the available parking spaces in the surrounding environment of the vehicle according to the perception information, so as to obtain a parking space detection result capable of reflecting the positions of the available parking spaces around the vehicle, where the parking space detection result may include the parking space areas of the available parking spaces in the surrounding environment of the vehicle.

Considering that most detection methods are limited to detecting parking spaces in the surrounding environment of a vehicle, for example, detecting boundary lines of the parking spaces to determine the positions of the parking spaces, other information, which is not related to the parking spaces and exists in the surrounding environment of the vehicle, may affect the detection of boundary lines due to higher detection difficulty of the boundary lines, may cause a plurality of detection results to occur when detecting one boundary line, and accordingly, the parking spaces determined according to the boundary lines may also have repeated detection, thereby causing an automatic driving system to fail to accurately determine the parking positions for automatic parking of the vehicle.

In an alternative of this embodiment, it is considered that the information having a higher degree of correlation with the parking space may include, in addition to elements of the parking space itself, such as a boundary line on the parking space, a parking space identifier, etc., a marker actively placed in the parking space by the user, such as a parking space limiter, a car stopper, a car backing pad, a tripod, etc., for limiting a specific position where the vehicle is parked inside the parking space. The automatic driving system can further detect the marker of the sensing information while detecting the parking space of the sensing information to obtain a marker detection result, so that the automatic driving system can more easily and accurately determine the parking position for automatically parking the vehicle.

In view of the fact that the limiter is relatively commonly used in various markers set by the user, in the embodiment of the application, the limiter is taken as an example for description, the automatic driving system can synchronously detect the sensing information to obtain a limiter detection result, so that whether the limiter exists in the parking space or not can be determined, and if the limiter exists in the parking space, the limiter detection result can further comprise a limiter area of the limiter in the parking space.

Illustratively, a parking space detection result of a parking space and a stopper detection result of a stopper inside the parking space constitute a detection result, the parking space detection result including a parking space region of the parking space, the stopper detection result characterizing that the inside of the parking space does not include a stopper, or a stopper region inside the parking space in case that the inside of the parking space includes a stopper. The automatic parking path is generated based on the detection result formed by the parking space detection result and the limiter detection result, so that the parking accuracy can be effectively improved.

Accordingly, if the parking space detection is performed on the sensing information and the parking space region for characterizing the parking space is not found, then step S104 detects that the sensing information is not detected by the parking space and the stopper, which indicates that no available parking space exists in the surrounding environment of the current vehicle, and parking cannot be performed.

And step S106, performing de-duplication processing on at least one detection result based on the parking space area in the at least one detection result, and determining a target detection result.

In an alternative of this embodiment, as shown in the foregoing, when detecting parking spaces with respect to the sensing information, a situation may occur in which a plurality of parking space areas corresponding to the same parking space are detected, and if at least two of the parking space areas in at least one detection result belong to the same parking space, this means that a situation in which the same parking space is repeatedly detected occurs, at this time, parameters for automatic parking of the vehicle cannot be planned directly according to the detected parking space areas, and a destination for automatic parking of the vehicle is determined.

In view of the fact that if repeated detection occurs, i.e., a plurality of parking space areas are detected for one parking space, overlapping occurs between the plurality of parking space areas, after detecting the above-mentioned at least one detection result, the automated driving system may determine whether overlapping occurs between the parking space areas in the at least one detection result (i.e., whether an overlapping relationship exists between the parking space areas), and perform deduplication processing on the above-mentioned at least one detection result according to the overlapping condition between the parking space areas, thereby determining that the target detection result is obtained.

For example, if there is an overlapping relationship between a plurality of parking space areas in at least one detection result, it is indicated that there is a case where detection of the same parking space is repeated currently, and at this time, it is necessary to perform deduplication processing on the plurality of parking space areas to remove the repeated parking space areas. In the case that the parking space region in at least one of the detection results is repeatedly removed, the stopper detection result in each detection result is matched with the parking space region in the detection result, so that the stopper detection result in the detection result can be ignored. Thus, when the plurality of parking space areas are subjected to the deduplication processing, the detection result to which the repeated parking space area belongs can be removed from at least one detection result, and each of the remaining detection results can be regarded as one target detection result.

In an alternative of this embodiment, after performing the deduplication processing on at least one detection result based on the parking space area in at least one detection result in step S106 to obtain the target detection result, if the number of the target detection results is 1, it may indicate that only one available parking space area is detected, so that a parking path may be generated directly based on the available parking space area and a stopper area of a stopper inside the available parking space, and the vehicle may be controlled to automatically park to a parking space corresponding to the target detection result according to the parking path; if the number of the target detection results is not 1, it may be indicated that more than one available parking space area is detected, at this time, the automatic driving system may determine a target detection result in combination with driving habits of a driver, a nearby parking principle and an easy-to-exit principle, perform a response to the determined target detection result including a target parking space area and a target limiter area according to the determined target detection result, generate a parking path of the vehicle according to the target parking space area and the target limiter area, and control the vehicle to automatically park into the parking space based on the parking path; and responding to the fact that a limiter does not exist in the parking space represented by the limiter detection result in the target detection result, constructing and generating a target limiter region in the target parking space region in the target detection result based on the target parking space region and at least one detection result, generating a parking path of the vehicle according to the target parking space region and the target limiter region, and controlling the vehicle to automatically park to the parking space based on the parking path. Or the automatic driving system can feed back all target detection results to a user for viewing, the user selects one target detection result as a detection result of a parking space for parking the current vehicle and a limiter in the parking space based on own driving habit, a nearby parking principle and an easy-to-drive-out principle, the selected target detection result is used for executing a response to the target detection result to comprise a target parking space area and a target limiter area, a parking path of the vehicle is generated according to the target parking space area and the target limiter area, and the vehicle is controlled to automatically park to the parking space based on the parking path; and responding to the fact that a limiter does not exist in the parking space represented by the limiter detection result in the target detection result, constructing and generating a target limiter region in the target parking space region in the target detection result based on the target parking space region and at least one detection result, generating a parking path of the vehicle according to the target parking space region and the target limiter region, and controlling the vehicle to automatically park to the parking space based on the parking path.

According to the embodiment of the application, the parking space region including the target detection result can be called a target parking space region, and under the condition that the limiter detection result in the target detection result represents the limiter region of the internal limiter of the parking space, the internal inclusion of the limiter of the parking space represented by the target parking space region can be described based on the limiter detection result, and at the moment, the limiter region represented by the limiter detection result in the target detection result can be called a target limiter region. Correspondingly, if the limiter detection result in the target detection result indicates that the limiter does not exist in the parking space, the fact that the limiter is not included in the parking space characterized by the target parking space region can be indicated based on the limiter detection result.

And step S108, responding to the target detection result to comprise a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path.

In an alternative of this embodiment, in order to enable a vehicle to automatically park in a target parking space area accurately, after a target detection result is obtained, the autopilot system may generate a parking path for controlling the vehicle to automatically park from a current position to a parking space according to the target detection result, so as to control the vehicle to automatically park, while considering that a limiter is located in an available parking space, so that the autopilot system may generate the above-mentioned parking path according to a specific or different path generation manner according to the target parking space area and the target limiter area included in the target detection result, thereby achieving the purpose of automatically parking based on the parking path. Specifically, if the target detection result indicates that the target parking space area has a limiter, the automatic driving system may first obtain the target parking space area and the target limiter area from the target detection result, and then directly generate a corresponding parking path according to the target limiter area and the target parking space area, thereby improving the generation efficiency of the parking path, and simultaneously ensuring that the parking path has higher precision, and after the parking path is obtained, the automatic driving system may control the vehicle to automatically park into the parking space corresponding to the target parking space area according to the parking path.

Step S110, responding to the stop detection result in the target detection result to represent that no stop exists in the parking space, constructing and generating a target stop region in the target parking space region in the target detection result based on the target parking space region and at least one detection result, generating a parking path of the vehicle according to the target parking space region and the target stop region, and controlling the vehicle to automatically park to the parking space based on the parking path.

In an alternative scheme of this embodiment, if the detection result of the target limiter indicates that no limiter exists in the parking space, the autopilot system may first obtain a target parking space area from the target detection result, then combine at least one detection result with the target parking space area to construct and generate a target limiter area in the target parking space area, then generate the parking path according to the target parking space area and the generated target limiter area, so as to improve the accuracy of the generated parking path, and finally control the vehicle to automatically park in the parking space corresponding to the target parking space area according to the parking path, thereby improving the accuracy of the vehicle during automatic parking.

In an alternative of this embodiment, when generating the target limiter area, the autopilot system may first determine whether the limiter area is included in at least one detection result, that is, whether the limiter is present in the interior of the parking space in the surrounding environment of the vehicle, and if the limiter area is not included in at least one detection result, may generate a target limiter area in the interior of the target parking space area according to the positional relationship between the parking space and the limiter; if the at least one detection result includes a limiter area, the limiter area may be acquired, and considering that the detection result to which the limiter area belongs may or may not be the detection result of the same parking space, the autopilot system may generate a target limiter area inside the target parking space area according to the positional relationship between the limiter area and the parking space area in the detection result to which the limiter area belongs. Alternatively, if the detection result to which the limiter region belongs and the target detection result are the same detection result of the parking space, one way may be to generate a target limiter region inside the target parking space region directly based on the positional relationship between the limiter region and the parking space region in the detection result to which the limiter region belongs, and another way may be to generate a target limiter region inside the target parking space region based on the dimensional relationship between the parking space region and the target parking space region in the detection result to which the limiter region belongs and the positional relationship between the parking space region and the limiter region in the detection result to which the limiter region belongs. Alternatively, if the detection result to which the limiter region belongs is not the same detection result of the parking space as the target detection result, a target limiter region may be generated inside the target parking space region based on the dimensional relationship between the parking space region and the target parking space region in the detection result to which the limiter region belongs, and the positional relationship between the parking space region and the limiter region in the detection result to which the limiter region belongs.

It should be noted that the process of generating the target limiter area is only shown by way of example, and the specific generating manner may be set by the user, which is not limited herein.

The two preset wheels may refer to wheels on the vehicle that are pre-designated by the user, and may include, but are not limited to: two front wheels of the vehicle, two rear wheels of the vehicle, two wheels on the left side of the vehicle, two wheels on the right side of the vehicle, wheels on the left front side and right rear side of the vehicle, wheels on the left rear side and right front side of the vehicle, and the like. In general, the preset wheels may be two wheels in front of the vehicle or two wheels in rear of the vehicle, considering that the vehicle is usually parked in forward and backward directions when traveling into a parking space.

In an alternative scheme of the embodiment, in order to more accurately control the automatic parking of the vehicle into the parking space, in the process of controlling the automatic parking of the vehicle into the parking space, the automatic driving system can also adjust the parking path of the vehicle and the vehicle posture of the vehicle in the driving process and the parking process according to the specific point position on the vehicle and the characteristic point position on the parking space, thereby improving the accuracy of controlling the automatic parking of the vehicle into the parking space. For example, the position deviation between the center point and the center position may be determined according to the relative positional relationship between the center point of the two rear wheels on the vehicle and the center position of the target limiter region, and if the position deviation is large, the parking path may be adjusted, and if the position deviation is small, the vehicle posture of the vehicle may be adjusted to ensure that the vehicle can be safely and accurately parked inside the parking space. For another example, if the vehicle is not currently traveling inside the parking space, the automatic driving system may increase a section of parking path according to a relative positional relationship between the target point position and the target limiter region at the current time, or may delete a section of parking path to ensure that the vehicle can smoothly and automatically park into the parking space, so as to improve accuracy of controlling the vehicle to automatically park into the parking space; if the vehicle is currently running into the parking space, the automatic driving system can adjust the current posture of the vehicle according to the relative position relationship between the target point position and the target limiter so as to ensure that the adjusted vehicle can be safely parked into the parking space.

In an alternative of this embodiment, in order to ensure the normalization of the automatic parking of the vehicle into the parking space and reduce the influence of the parked vehicle on other vehicles, it is generally possible to locate the midpoint of the two wheels in front of or behind the vehicle, i.e., the above-mentioned target point position, on the midpoint of the longer boundary line on both sides of the parking space limiter while bringing the front or rear wheels of the vehicle into contact with the limiter. Based on this, in order to improve the user's experience of using the vehicle, when the parking path is added or subtracted according to the relative positional relationship between the target point position and the limiter region, the autopilot system may first determine the coordinates of the target point position in the preset coordinate system, and the coordinates of the end points of the longer boundary lines on both sides of the limiter in the preset coordinate system, where the preset coordinate system may include, but is not limited to: a vehicle coordinate system, a geodetic coordinate system, etc. After obtaining the coordinates of the end points, the autopilot system may determine an expression of the perpendicular bisector according to the coordinates and determine the target point location according to the coordinates and the expression. For example, the coordinates are respectivelyAnd->The expression isBy solving this expression, the target point position (x, y) can be obtained.

If the determined distance between the target point position and the central line is longer, for example, greater than a preset distance, and the included angle between the straight line of the front wheel or the rear wheel of the vehicle and the central line is smaller, for example, smaller than a preset angle, the situation that the parking position is askew when the vehicle directly enters the parking space is indicated, the partial area on the vehicle is out of the parking space and further causes the driving of other vehicles and influences the parking safety of other vehicles is caused, at this moment, the automatic driving system can increase a section of parking path on the basis of the current parking path to change the gesture of the vehicle according to the distance between the target point position and the central line and the included angle between the straight line of the front wheel or the rear wheel of the vehicle and the central line, so that the target point position of the vehicle can be near the central line when the vehicle is driven, and the straight line of the front wheel or the rear wheel of the vehicle can be relatively perpendicular to the central line, and the vehicle can be stably and accurately parked in the parking space; if the determined distance from the target point position to the central line is shorter, for example, less than or equal to the preset distance, and the included angle between the straight line where the front wheel or the rear wheel of the vehicle is located and the central line is larger, for example, approaching to a right angle, or no parking space is near the parking space, or an instruction which is sent by a user and does not need to consider the specific parking position is received, the automatic driving system can directly control the vehicle to travel into the parking space without considering the parking condition of the vehicle in the parking space too much, and at the moment, the automatic driving system can delete a section of parking path on the basis of the current parking path so that the vehicle can rapidly travel into the parking space, thereby improving the automatic parking efficiency.

It should be noted that the manner of adding or subtracting parking paths is merely exemplary, and the specific adding or subtracting process may be set by a worker, which is not limited herein.

The attribute information may include at least: the model parameters of the vehicle, the size of the vehicle and the like.

In an alternative of this embodiment, considering that the area of the partially divided parking space area may be small or the shape is irregular, some vehicles have large body shapes, and it is easy to occur that the partial area of the vehicle is outside the parking space area, which may affect the safety of the vehicle when the vehicle is automatically parked, for example, collide with other vehicles, or the parking position of the vehicle may affect the running of other vehicles. Therefore, when the vehicle is controlled to automatically park to the parking space, the automatic driving system can also acquire physical data of the vehicle, such as the size, the model and the like of the vehicle, then determine the target parking position of the vehicle in the parking space by combining the attribute information of the vehicle on the basis of the target parking space area and the target limiter area so as to improve the accuracy of determining the parking position of the vehicle in the parking space, and finally the automatic driving system can generate the parking path of the vehicle according to the current position of the vehicle, namely the current position of the vehicle and the definite target parking position.

In an alternative aspect of this embodiment, in order to ensure the safety of the vehicle during automatic parking, for example, without colliding with another vehicle, or without the parking position of the vehicle affecting the traveling of another vehicle, when determining the target parking position, the automatic driving system may first acquire the parking situation of the vehicle in the vicinity of the identified target parking space area, for example, the parking information of another parking space adjacent to the parking space, and determine the target parking position according to the parking information by combining the determined target parking space area, the attribute information of the vehicle, and the target limiter area.

For example, the automatic driving system may determine, according to the target limiter area and the target parking space area, a parking position where the vehicle is parked in the target parking space area under normal conditions, and then adjust the parking position according to the parking situation of the vehicle near the parking space and the attribute information of the vehicle to obtain the target parking position, so as to ensure that after the vehicle is automatically parked to the target parking position, the vehicle cannot collide with other vehicles due to the fact that the vehicle is larger in size or other vehicles approach to the target parking space area, thereby improving the safety of the vehicle during automatic parking.

In an alternative scheme of this embodiment, after detecting the target detection results of a plurality of parking spaces, the user may actively select a parking space that needs to be automatically parked according to his driving habit, a principle of nearby parking, and a principle of easy driving out, so as to improve the user's experience of using the vehicle. Specifically, after the target parking space area and the target limiter area are acquired, the automatic driving system may output the target parking space area, for example, the target parking space area is displayed in a vehicle central control display screen, and then a selection operation of the target parking space by a user is received, for example, the user clicks one parking space area to serve as a parking space for automatic parking of the vehicle. If a user selects one of the parking space regions, the autopilot system may generate the parking path described above based on the selected parking space region in combination with a target limiter region within the parking space region.

The above-mentioned preset positional relationship may be a positional relationship between the stopper and the parking space in general, and may include a distance of the stopper with respect to a boundary of the parking space, but is not limited thereto. The preset positional relationship may be stored in the database in advance, so that the preset positional relationship may be rapidly acquired.

In an alternative of this embodiment, in response to the fact that the limiter detection result in the target detection result indicates that the limiter is not present in the parking space, in order to enable automatic parking of the vehicle according to the target parking space area and the target limiter area, the autopilot system may first determine whether the at least one detection result includes the limiter area, and if not present, that is, if the at least one detection result does not include the limiter area, the autopilot system may determine that none of the limiter is present in the parking space in the surrounding environment of the vehicle, and generate the target limiter area in the target parking space area according to the target parking space area and the preset positional relationship described above; if the parking space is present, that is, at least one detection result includes a limiter area, considering that the detection result to which the limiter area belongs may or may not be the detection result of the same parking space as the target detection result, further, if the automatic driving system determines that the parking space area in the detection result to which the limiter area belongs is the same as the target parking space area at this time, the automatic driving system may generate a target limiter area inside the target parking space area based on the positional relationship between the limiter area and the parking space area in the detection result to which the limiter area belongs; if the detection result to which the limiter region belongs is not the same detection result of the parking space as the target detection result, the automatic driving system may generate a target limiter region inside the target parking space region based on the dimensional relationship between the parking space region and the target parking space region in the detection result to which the limiter region belongs, and the positional relationship between the parking space region and the limiter region in the detection result to which the limiter region belongs.

Alternatively, whether the detection result and the target detection result are the same detection result of the parking space can be determined by judging whether the parking space region and the target parking space region in the detection result to which the limiter region belongs overlap, if so, the detection result and the target detection result have higher probability of belonging to the same parking space, and if not, the detection result and the target detection result have higher probability of not belonging to the same parking space.

The detection confidence coefficient may be a confidence coefficient corresponding to each parking space region obtained in the process of detecting the parking space by the automatic driving system on the perception information.

It should be noted that, in the present invention, the overlapping refers to the existence of an overlapping area between two parking space areas, fig. 2 is a schematic diagram showing an overlapping of parking space areas according to an embodiment of the present invention, and one includes three overlapping cases, where a refers to the overlapping of the lower right corner of parking space area 1 (shown as a solid line box) and the upper left corner of parking space area 2 (shown as a broken line box), B refers to the overlapping of parking space area 3 (shown as a solid line box) inside parking space area 4 (shown as a broken line box), and C refers to the overlapping of the middle portion of parking space area 5 (shown as a solid line box) and the middle portion of parking space area 6 (shown as a broken line box).

In an alternative scheme of this embodiment, when performing the deduplication processing on a plurality of parking space regions, two parking space regions may be selected from the plurality of parking space regions to perform the overlap determination, and whether the two parking space regions overlap as shown in a to C in fig. 2 is determined, if so, one parking space region needs to be deleted from the two parking space regions, and then one parking space region is selected from the plurality of parking space regions to overlap with the parking space region that is not deleted; if no overlap occurs, both parking space areas can be reserved and the first parking space area is reselected for overlap determination.

In consideration of different detection confidence degrees of different parking space regions, the lower the detection confidence degree is, the lower the detection accuracy of the representative parking space region is, so in order to improve the detection accuracy of the determined target parking space region, the automatic driving system may firstly perform sorting processing on the plurality of parking space regions according to the detection confidence degrees to obtain a parking space region sequence, wherein the detection confidence degree of the first parking space region in the parking space region sequence is the highest, and the detection confidence degree of the last parking space region is the lowest. Then, a first parking space area is read from the parking space area sequence and is used as a first parking space area, overlapping judgment is carried out on the first parking space area and each parking space area behind the first parking space area, if overlapping occurs, the situation that the two parking space areas correspond to the same parking space is indicated, repeated detection occurs on the parking space, and the parking space area behind the first parking space area is deleted, namely, the parking space area with lower detection confidence is deleted; if no overlapping occurs, it indicates that the two parking space areas correspond to different parking spaces, so that the two parking space areas need to be reserved, and the condition of missing detection is avoided. After all the parking space areas after the first parking space area are judged to be finished, the next parking space area after the first parking space area is read from the current parking space area sequence is used as a new first parking space area, the judgment flow is repeated, and the like until no parking space area which is not read as the first parking space area exists in the parking space area sequence.

For example, if 4 parking space regions are currently detected, the detection confidence of the parking space region 1 is 7, the detection confidence of the parking space region 2 is 6, the detection confidence of the parking space region 3 is 4, and the detection confidence of the parking space region 4 is 3, the corresponding parking space region sequence may be: the first cycle may take the parking space area 1 as the first parking space area, and the parking space area 1, the parking space area 3 and the parking space area 4 are respectively overlapped and judged, if the parking space area 1 overlaps with the parking space area 2, the parking space area 1 and the parking space area 3 are not overlapped, and the parking space area 1 and the parking space area 4 are not overlapped, the parking space area 2 is deleted, and the next cycle is started; in the second cycle, the parking space area 3 may be used as the first parking space area, the parking space area 3 and the parking space area 4 may be overlapped, if the parking space area 3 and the parking space area 4 overlap, the parking space area 4 is deleted, and since the parking space area sequence does not have a parking space area located behind the parking space area 3 at this time, it may be determined that the parking space area sequence does not have a parking space area which is not subjected to the judgment at this time, the cycle may be stopped, and it may be determined that the detection result to which the parking space area 1 belongs and the detection result to which the parking space area 3 belongs are target detection results.

In an alternative scheme of this embodiment, considering that the matching degree between the detected multiple parking space areas and the actual parking space area is different, and the matching degree may be affected by factors such as the relative position of the vehicle and the parking space, the angular point position and the area of the parking space, so before the multiple parking space areas are deduplicated, the automatic driving system may determine the detection confidence of each parking space area according to the current position of the vehicle, the size and the shape of the parking space, and the included angle between the image acquisition device and the ground.

The above-described region overlapping information may refer to an overlapping proportional relationship between the first parking space region and the second parking space region, for example, a ratio of the overlapping region between the two in the first parking space region, a ratio of the overlapping region in the second parking space region, or a ratio of the overlapping region in a total region constituted by the first parking space region and the second parking space region. If the above-mentioned region overlapping information is larger, for example, greater than or equal to the preset region overlapping information threshold, it may be determined that the overlapping region of the first parking space region and the second parking space region is larger, and both belong to the same parking space, where the above-mentioned second parking space region may be removed from the parking space region; if the above-mentioned region overlapping information is smaller, for example, smaller than a preset region overlapping information threshold, it may be determined that the overlapping region of the first parking space region and the second parking space region is smaller, and the two regions belong to different parking spaces, and the second parking space region may not be removed at this time.

In an alternative of this embodiment, different overlap determination methods may be set for different overlap situations, taking three overlap situations in fig. 2 as an example, whether or not an overlap area exists between two parking space areas may be determined according to whether or not an endpoint on each boundary of one parking space area is within another parking space area, or whether or not boundaries of two parking space areas intersect, that is, according to a positional relationship between an endpoint of a first parking space area and a second parking space area, a positional relationship between an endpoint of a second parking space area and a first parking space area, and a positional relationship between a boundary of a first parking space area and a boundary of a second parking space area.

In an alternative of this embodiment, considering that a slight overlap may occur between parking space areas belonging to different parking spaces when detecting the parking spaces, but this is only a systematic error generated during the detection, and there is no overlapping relationship between the two, the autopilot system may further determine the extent of overlapping between the first parking space area and the second parking space area according to the shape of the first parking space area, the shape of the second parking space area, and the relative positional relationship between the first parking space area and the second parking space area, that is, determine the above-mentioned area overlapping information, and determine whether there is an overlapping relationship between the first parking space area and the second parking space area according to the area overlapping information. If the region overlapping information exceeds the preset target region overlapping information, determining that an overlapping relationship exists between the first parking space region and the second parking space region; and if no repeated area exists between the first parking space area and the second parking space area or the area overlapping information does not exceed the preset target area overlapping information, determining that no overlapping relation exists between the first parking space area and the second parking space area.

For example, if the above-mentioned region overlapping information exceeds the preset target region overlapping information, for example, the target region overlapping information may be a preset proportion threshold, and the ratio of the overlapping region in the total region formed by the first parking space region and the second parking space region is greater than the preset proportion threshold, it may be determined that an overlapping relationship exists between the first parking space region and the second parking space region; if there is no overlapping area between the first parking space area and the second parking space area, or the ratio of the overlapping area in the total area formed by the first parking space area and the second parking space area does not exceed the preset proportion threshold value, it may be determined that there is no overlapping relationship between the first parking space area and the second parking space area.

In an alternative of the present embodiment, when the autopilot system makes an overlap determination on the first parking space region and the second parking space region, it may be first determined whether a plurality of endpoints of the first parking space region are within the second parking space region, whether a plurality of endpoints of the second parking space region are within the first parking space region, and whether boundaries of the first parking space region and boundaries of the second parking space region intersect, wherein the first two processes may be used to determine whether a or B in fig. 2 occurs in the first parking space region and the second parking space region, and the latter process may be used to determine whether a and C in fig. 2 occur in the first parking space region and the second parking space region. If any one of the plurality of endpoints of the first parking space region is within the second parking space region, or if any one of the plurality of endpoints of the second parking space region is within the first parking space region, or if the boundary of the first parking space region and the boundary of the second parking space region intersect, it may be determined that an overlap region exists between the first parking space region and the second parking space region; if any one of the plurality of endpoints of the first parking space region is outside the second parking space region and any one of the plurality of endpoints of the second parking space region is outside the first parking space region and the boundary of the second parking space region do not intersect, it may be determined that there is no overlap region between the first parking space region and the second parking space region.

In an alternative of this embodiment, the above-mentioned endpoints of the first parking space region and the above-mentioned endpoints of the second parking space region may be represented in terms of coordinates, for example, coordinates of each of the endpoints of the first parking space region and the endpoints of the second parking space region are determined in a vehicle coordinate system, and then the above-mentioned overlap determination process is performed according to the coordinates, so as to improve the efficiency of determining whether or not overlapping occurs in the two parking space regions.

The above-mentioned directional information may be directional information of a first side of the first parking space region, and the directional information of a plurality of first sides may form a clockwise direction or a counterclockwise direction, and the above-mentioned directional information may be set by a user or may be determined by an autopilot system according to a setting rule corresponding to the directional information.

In an alternative of this embodiment, the detected parking space area is usually represented by an endpoint coordinate, and in order to be able to determine accurately whether the endpoint of the second parking space area is located in the first parking space area, first, each first edge constituting the boundary of the first parking space area may be converted into a vector form based on the endpoint coordinate of the first parking space area, that is, pointing information of each first edge is determined. For ease of understanding, fig. 3 is a schematic diagram illustrating a parking space region overlapping determination according to an embodiment of the present invention, where the left side represents a case where parking space regions are not overlapped, and the right side represents a case where parking space regions are overlapped, as shown in fig. 3, four endpoints included in the first parking space region 301 are 311, 312, 313, and 314, respectively, and four endpoints included in the second parking space region 302 are 321, 322, 323, and 324, respectively. Taking the first parking space area 301 as an example, the pointing information of each first edge is determined based on the endpoint coordinates of the four endpoints, and assuming that at least one first edge forms a clockwise direction, it may be determined that the pointing information of the first edge formed by the endpoint 311 and the endpoint 313 is right (as shown by an arrow in fig. 3), and the first parking space area 301 is located at the lower side of the first edge, and similarly, it may be determined that the pointing information of the first edge formed by the endpoint 313 and the endpoint 314 is right (as shown by an arrow in fig. 3), and the first parking space area 301 is located at the left side of the first edge; the first side of the end points 314 and 312 is directed to the right and left (as indicated by the arrow in fig. 3), and the first parking space area 301 is located on the upper side of the first side; the first side of the first parking space area 301 is located right of the first side, and the directional information of the first side is directly above (as indicated by an arrow in fig. 3) the end points 312 and 311.

The relative positional relationship between the end points and the first parking space areas, i.e. the end points and the first parking space areas are located on the same side or on different sides of the first sides, is then determined based on the end points of the second parking space areas and the pointing information of each first side. Alternatively, two vectors may be constructed based on the end points of the second parking space region and the pointing information of each first side, the start end point and the end point of each first side may be determined according to the pointing information of each first side, and the first vector may be constructed based on the end point coordinates of the two end points, the pointing direction of the vector being the direction from the start end point to the end point; based on the end point coordinates of the start end point and the end point coordinates of the second parking space region, a second vector is constructed, the pointing direction of which is a direction from the start end point to the end point of the second parking space region. Calculating an angle between the two vectors by an existing space vector angle formula, wherein the angle is used for representing the angle rotated by the first vector when the first vector rotates to be coincident with the second vector according to the direction formed by at least one edge, for example, if at least one edge forms a clockwise direction, the angle is used for representing the angle rotated by the first vector when the first vector rotates to be coincident with the second vector according to the clockwise direction; if at least one edge constitutes a counter-clockwise direction, the included angle is used to characterize the angle through which the first vector rotates when rotated counter-clockwise to coincide with the second vector. Further, the relative position relationship between the end point of the second parking space area and the first parking space area can be determined through the included angle, for example, if the included angle between the first vector and the second vector is smaller than 180 degrees, the end point of the second parking space area and the first parking space area are indicated to be positioned on the same side of a first edge formed by the starting end point and the ending end point; if the angle between the first vector and the second vector is greater than or equal to 180 deg., it is indicated that the end point of the second parking space region and the first parking space region are located on different sides of the first side formed by the start end point and the end point.

For example, as shown in the left side of fig. 3, for the end point 321 of the second parking space area 302, first, a first vector directed from the end point 311 to the end point 313 and a second vector directed from the end point 311 to the end point 321 are constructed according to the direction information of the first side constituted by the end point 311 and the end point 313, and it is determined that the first vector is overlapped with the second vector after rotating by 0 ° clockwise, that is, an angle between the first vector and the second vector is 0 °, which is smaller than 180 °, and thus, it is determined that the end point 321 is located on the same side as the first side constituted by the end point 311 and the end point 313; similarly, it may be determined that, based on the direction information of the first edge formed by the end point 313 and the end point 314, the angle between the first vector directed from the end point 313 to the end point 314 and the second vector directed from the end point 313 to the end point 321 is 270 °, which is greater than 180 °, and thus, it is determined that the end point 321 is located on a different side from the first edge formed by the end point 313 and the end point 314 than the first parking space area 301; it may be determined that the angle between the first vector from endpoint 314 to endpoint 312 and the second vector from endpoint 314 to endpoint 321 is less than 180 ° based on the direction information of the first edge formed by endpoint 314 and endpoint 312, thus determining that endpoint 321 is on the same side of the first edge formed by endpoint 314 and endpoint 312 as first parking space region 301; it is possible to determine that the angle between the first vector directed from the end point 312 to the end point 311 and the second vector directed from the end point 312 to the end point 321 is smaller than 180 ° based on the direction information of the first side constituted by the end point 312 and the end point 311, and thus, it is determined that the end point 321 is located on the same side as the first side constituted by the end point 312 and the end point 311 as the first parking space region 301. Accordingly, it may be determined that the endpoint 321 of the second parking space region 302 is not located in the first parking space region 301. In the same way it may be determined that none of the four endpoints of the second parking space area 302 are located in the first parking space area 301, i.e. it may be determined that there are no endpoints in the second parking space area that are located in the first parking space area.

As another example, as shown on the right side of fig. 3, for endpoint 322 of second parking space region 302, first, a first vector directed from endpoint 311 to endpoint 313 and a second vector directed from endpoint 311 to endpoint 322 are constructed based on the direction information of the first side constituted by endpoint 311 and endpoint 313, it may be determined that the angle between the first vector and the second vector is less than 180 °, and thus endpoint 322 and first parking space region 301 are located on the same side of the first side constituted by endpoint 311 and endpoint 313; similarly, it may be determined that the angle between the first vector from endpoint 313 to endpoint 314 and the second vector from endpoint 313 to endpoint 322 is less than 180 ° based on the direction information of the first edge formed by endpoint 313 and endpoint 314, and thus, that endpoint 322 is on the same side of the first edge formed by endpoint 313 and endpoint 314 as first parking space region 301; it may be determined that the angle between the first vector from endpoint 314 to endpoint 312 and the second vector from endpoint 314 to endpoint 322 is less than 180 ° based on the direction information of the first edge formed by endpoint 314 and endpoint 312, thus determining that endpoint 322 is on the same side of the first edge formed by endpoint 314 and endpoint 312 as first parking region 301; it is possible to determine that the angle between the first vector directed from the end point 312 to the end point 311 and the second vector directed from the end point 312 to the end point 322 is less than 180 ° based on the direction information of the first side constituted by the end point 312 and the end point 311, and thus, it is determined that the end point 322 is located on the same side as the first side constituted by the end point 312 and the end point 311 as the first parking space region 301. Thus, it may be determined that the end point 322 of the second parking space region 302 is located in the first parking space region 301, i.e. it may be determined that there is an end point located in the first parking space region in the second parking space region.

It should be noted that, the determination of whether the endpoint of the second parking space area is located in the first parking space area and the determination of whether the endpoint of the first parking space area is located in the second parking space area may adopt the same implementation scheme, in this embodiment of the present application, only the determination of whether the endpoint of the second parking space area is located in the first parking space area is taken as an example to make a detailed description, and the implementation scheme of determining whether the endpoint of the first parking space area is located in the second parking space area is not described herein.

In an alternative of this embodiment, the parking space area may be generally represented by end point coordinates, and the manner of comparing the coordinates may be directly adopted to improve the determination efficiency. Specifically, in determining whether the boundary of the first parking space region intersects the boundary of the second parking space region, the automated driving system may first determine, based on the endpoint coordinates constituting the first parking space region, a maximum endpoint abscissa, a maximum endpoint ordinate, a minimum endpoint abscissa, and a minimum endpoint ordinate of the plurality of sides constituting the first parking space region, and similarly may determine a maximum endpoint abscissa, a maximum endpoint ordinate, a minimum endpoint abscissa, and a minimum endpoint ordinate of the plurality of sides constituting the second parking space region.

After determining the endpoint coordinates, if the maximum endpoint abscissa of at least one first edge is smaller than the minimum endpoint abscissa of at least one second edge, it may be determined that the first parking space region is located at the left side of the second parking space region, and the boundaries of the first and second parking space regions may not intersect; if the minimum endpoint abscissa of the at least one first edge is greater than the maximum endpoint abscissa of the at least one second edge, it may be determined that the first parking space region is located on the right side of the second parking space region, and the boundaries of the first and second parking space regions may not intersect; if the maximum endpoint ordinate of the at least one first edge is smaller than the minimum endpoint ordinate of the at least one second edge, the first parking space area can be determined to be positioned below the second parking space area, and the boundaries of the first parking space area and the second parking space area cannot be intersected; if the minimum endpoint ordinate of the at least one first edge is greater than the maximum endpoint ordinate of the at least one second edge, it may be determined that the first parking space region is located above the second parking space region, and that the boundaries of the two are unlikely to intersect. Thus, if the maximum endpoint abscissa of the at least one first edge is smaller than the minimum endpoint abscissa of the at least one second edge, or the minimum endpoint abscissa of the at least one first edge is larger than the maximum endpoint abscissa of the at least one second edge, or the maximum endpoint ordinate of the at least one first edge is smaller than the minimum endpoint ordinate of the at least one second edge, or the minimum endpoint ordinate of the at least one first edge is larger than the maximum endpoint ordinate of the at least one second edge, it may be determined that the boundary of the first parking space region and the boundary of the second parking space region do not intersect. If the maximum endpoint abscissa of the at least one first side is greater than or equal to the minimum endpoint abscissa of the at least one second side and the minimum endpoint abscissa of the at least one first side is less than or equal to the maximum endpoint abscissa of the at least one second side and the maximum endpoint ordinate of the at least one first side is greater than or equal to the minimum endpoint ordinate of the at least one second side and the minimum endpoint ordinate of the at least one first side is less than or equal to the maximum endpoint ordinate of the at least one second side, since the first parking space region is constituted by the at least one first side and the second parking space region is constituted by the at least one second side, in this case, it may be determined that the boundary of the first parking space region and the boundary of the second parking space region intersect.

Optionally, determining the region overlapping information between the first parking space region and the second parking space region based on the region shape and the relative positional relationship of the first parking space region and the second parking space region includes: determining region overlapping information of the first parking space region and the second parking space region based on endpoint coordinates of each of at least one first side constituting a boundary of the first parking space region and endpoint coordinates of each of at least one second side constituting a boundary of the second parking space region in response to the region shapes of the first parking space region and the second parking space region being rectangular; and determining region overlapping information of the first parking space region and the second parking space region based on the relative position relationship between the first parking space region and the second parking space region and the pixel point number used for representing the first parking space region and the second parking space region in the perception information respectively in response to the fact that the region shapes of the first parking space region and the second parking space region are not rectangular.

In an alternative of this embodiment, considering that elements between different parking spaces, such as shapes of parking spaces, sizes of parking spaces, etc., may be different, and these elements may also affect determination of region overlapping information between different parking space regions, therefore, when determining region overlapping information between a first parking space region and a second parking space region, a first shape type of the first parking space region and a second shape type of the second parking space region may be determined first, if both are rectangular, an area of the first parking space region and an area of the second parking space region may be determined directly based on endpoint coordinates between the two parking space regions, and endpoint coordinates of an overlapping region where the first parking space region and the second parking space region overlap may be determined by the autopilot system according to the endpoint coordinates of the overlapping region, so that region overlapping information of the first parking space region and the second parking space region may be determined; if any one of the two parking space areas is not rectangular, the area overlapping information of the first parking space area and the second parking space area is not easy to calculate through the endpoint coordinates between the two parking space areas, in order to simplify the calculation process and improve the judging efficiency, the areas of the first parking space area and the second parking space area can be characterized according to the pixel point number used for representing the first parking space area and the second parking space area in the sensing information, therefore, the images of the first parking space area and the second parking space area can be generated according to the sensing information, the pixel point number located in the first parking space area, the pixel point number located in the second parking space area and the pixel point number located in the overlapping area can be determined, and then the area overlapping information of the first parking space area and the second parking space area is determined according to the pixel point number of the overlapping area in the image, the pixel point number located in the first parking space area and the pixel point number located in the second parking space area.

In an alternative of this embodiment, for ease of understanding, when determining the region overlapping information of the first parking space region and the second parking space region according to the number of pixels, the autopilot system may first determine the number of pixels located in the first parking space region and the number of pixels located in the second parking space region, and determine the number of pixels located in the overlapping region according to the relative positional relationship between the first parking space region and the second parking space region. And finally, according to the number of the three pixel points, determining the region overlapping information of the first parking space region and the second parking space region.

The target image may be an image that reflects the environment around the vehicle, and may include at least a parking space around the vehicle. The target detection model may be a model for realizing target detection in the field of image processing, and in order to apply the model to parking space and limiter detection, a training sample may be provided to fine tune the trained model, thereby obtaining a target detection model.

In an alternative scheme of the embodiment, considering that the image processing technology is mature, when detecting the parking space and the limiter near the vehicle by using the sensing information, in order to improve the detection efficiency, the automatic driving system may first generate the target image according to the acquired sensing information, and then detect the parking space and the limiter by using a target detection mode in the image processing technology. In general, the boundary line of a parking space generally constitutes a rectangle including four corner points, and for convenience of vehicle entry and exit, the boundary line of the parking space generally has an opening, and therefore, two end points of the opening position may be used as the opening points of the parking space, and the four corner points of the rectangle may be used as the end points of the detected parking space region. Two stoppers are usually placed in a parking space for defining parking positions of two vehicles, respectively, and the stoppers are usually in a shape similar to a rectangular parallelepiped, one of the stoppers includes two ends, one of which is close to the other stopper, and the other end is close to a boundary line of the parking space, so that two ends of the two stoppers, which are close to the boundary line of the parking space, can be used as end points of a stopper region, that is, the stopper region includes two stoppers. Specifically, the pre-trained target detection model may be used to detect a plurality of opening points of the parking space on the target image, so as to determine detection results of the plurality of opening points, that is, determine two end points of the opening position on the boundary line of the parking space. And then carrying out point location regression on the corner points of the parking spaces and the end points of the limiters in the target image according to the detection results of the plurality of opening points by a regression algorithm in the related technology, and determining the corner points of the rectangle formed by the boundary lines of the parking spaces and the two end points of the limiters close to the boundary lines of the parking spaces, thereby determining the detection results of the parking space areas and the limiters.

The plurality of sensing devices may be image acquisition devices on the vehicle for acquiring information of surrounding environment of the vehicle, and may include, but is not limited to: fisheye cameras, looking around cameras, etc. The partial image may be an image acquired by the sensing device and capable of reflecting the information of the partial environment in the vicinity of the vehicle.

In an alternative of this embodiment, since the sensing range of a single sensing device is limited, a plurality of sensing devices are typically mounted on a vehicle in different directions in order to be able to acquire environmental information within a circle of the vehicle. Therefore, the automatic driving system may first acquire information perceived by a plurality of perceiving devices on the vehicle, generate the plurality of partial images, for example, shoot by using four fisheye cameras disposed around the vehicle to obtain a plurality of images capable of reflecting environmental information around the vehicle, then splice the generated plurality of partial images, for example, splice the plurality of shot images according to internal parameters and external parameters of the fisheye cameras to obtain the spliced image, and finally convert the spliced image into a bird's eye view space, thereby obtaining the target image.

Example 2

According to another aspect of the embodiment of the present invention, corresponding to the embodiment of the vehicle control method described above, the present specification further provides a vehicle control apparatus, and fig. 4 is a block diagram of a vehicle control apparatus according to an embodiment of the present invention, as shown in fig. 4, the apparatus may include: the acquisition module 402 is used for acquiring perception information obtained by perceiving the surrounding environment of the vehicle in response to receiving the automatic parking instruction; the detection module 404 is configured to detect the perceived information on the parking space and the limiter to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space; a deduplication module 406, configured to perform deduplication processing on at least one detection result based on the parking space region in the at least one detection result, and determine a target detection result; a first control module 408, configured to generate a parking path of the vehicle according to the target parking space area and the target limiter area in response to the target detection result, and control the vehicle to automatically park into the parking space based on the parking path; the second control module 410 is configured to, in response to the stop detection result in the target detection result indicating that no stop exists in the parking space, construct and generate a target stop region in the target parking space region in the target detection result based on the target parking space region and at least one detection result, generate a parking path of the vehicle according to the target parking space region and the target stop region, and control the vehicle to automatically park into the parking space based on the parking path.

Example 3

Example 4

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of units may be a logic function division, and there may be another division manner in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.

Claims

1. A vehicle control method characterized by comprising:

in response to receiving an automatic parking instruction, obtaining perception information obtained by perceiving the surrounding environment of the vehicle;

detecting the sensing information by a parking space and a limiter to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space;

performing de-duplication processing on the at least one detection result based on the parking space area in the at least one detection result, and determining a target detection result;

responding to the target detection result to comprise a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path;

And responding to the fact that a limiter does not exist in the parking space represented by the limiter detection result in the target detection result, constructing and generating a target limiter area in the target parking space area in the target detection result based on the target parking space area and the at least one detection result, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path.

2. The vehicle control method according to claim 1, characterized by further comprising:

determining a target point position of a target point on the vehicle at present, wherein the target point is used for representing a midpoint between two preset wheels on the vehicle;

based on the relative positional relationship between the target point position and the target limiter area, the parking path or the vehicle posture of the vehicle is adjusted in real time in the process of controlling the vehicle to automatically park to the parking space based on the parking path.

3. The vehicle control method according to claim 1, characterized in that the generating a parking path of the vehicle from the target parking space region and the target limiter region includes:

Determining a target parking position of the vehicle in a parking space corresponding to the target parking space region based on the attribute information of the vehicle, the target parking space region and the target limiter region;

the parking path is generated based on a current vehicle position of the vehicle and the target parking position.

4. The vehicle control method according to claim 1, characterized in that generating a parking path of the vehicle from the target parking space region and the target limiter region includes:

outputting the target parking space area;

in response to receiving a selection instruction generated by selecting the target parking space area, determining a selected parking space area corresponding to the selection instruction;

and generating the parking path based on the selected parking space area and a target limiter area inside the selected parking space area.

5. The vehicle control method according to claim 1, characterized in that the constructing a target limiter area inside a target parking space area in the target detection result based on the at least one detection result includes:

determining whether the at least one detection result contains a limiter region;

Generating a target limiter region inside a target parking space region in the target detection result based on a preset position relationship between the parking space and the limiter in response to the at least one detection result not including the limiter region;

determining whether a parking space area in a detection result to which the limiter area belongs overlaps with the target parking space area or not in response to the fact that the limiter area is included in the at least one detection result;

responding to the overlapping of the parking space area and the target parking space area in the detection result of the limiter area, and constructing and generating a target limiter area inside the target parking space area based on the limiter area;

and constructing and generating a target stopper region in the target parking space region based on the size relationship between the parking space region and the target parking space region in the detection result of the stopper region and the target position relationship between the parking space region and the stopper region in the detection result of the stopper region in response to the fact that the parking space region and the target parking space region in the detection result of the stopper region are not overlapped.

6. The vehicle control method according to claim 1, characterized in that the performing the deduplication processing on the at least one detection result based on the parking space region in the at least one detection result, determining a target detection result, includes:

Sequencing each parking space region in at least one detection result based on the detection confidence of the parking space region in the detection result to obtain a parking space region sequence;

if a parking space area which is not read as a first parking space area exists in the current parking space area sequence, reading the current first parking space area from the current parking space area sequence according to the parking space area sequence, and eliminating a second parking space area which is positioned behind the current first parking space area in the current parking space area sequence and has an overlapping relation with the current first parking space area;

and if no parking space area which is not read as the first parking space area exists in the current parking space area sequence, determining the detection result of the parking space area in the current parking space area sequence as a target detection result.

7. The vehicle control method according to claim 6, characterized in that determining whether there is an overlapping relationship between the first parking space region and the second parking space region includes:

determining whether an overlapping region exists between a first parking space region and a second parking space region based on a positional relationship between an end point of the first parking space region and the second parking space region, a positional relationship between an end point of the second parking space region and the first parking space region, and a positional relationship between a boundary of the first parking space region and a boundary of the second parking space region;

Determining, in response to the existence of an overlap region between the first parking space region and the second parking space region, region overlap information between the first parking space region and the second parking space region, the region overlap information characterizing a degree of region overlap between the first parking space region and the second parking space region, based on a region shape and a relative positional relationship of the first parking space region and the second parking space region;

if the region overlapping information exceeds the preset target region overlapping information, determining that an overlapping relationship exists between the first parking space region and the second parking space region;

and if no overlapping area exists between the first parking space area and the second parking space area or the area overlapping information does not exceed the target area overlapping information, determining that no overlapping relation exists between the first parking space area and the second parking space area.

8. The vehicle control method according to claim 7, characterized in that the determining whether or not an overlap region exists between the first parking space region and the second parking space region based on a positional relationship between an end point of the first parking space region and the second parking space region, a positional relationship between an end point of the second parking space region and the first parking space region, and a positional relationship between a boundary of the first parking space region and a boundary of the second parking space region, includes:

Judging whether an endpoint in the second parking space area exists in the first parking space area, whether an endpoint in the first parking space area exists in the second parking space area, and whether a boundary of the first parking space area intersects with a boundary of the second parking space area;

determining that an overlapping region exists between the first parking space region and the second parking space region if an endpoint located in the second parking space region exists in the first parking space region or an endpoint located in the first parking space region exists in the second parking space region or a boundary of the first parking space region intersects a boundary of the second parking space region;

and if the endpoint of the second parking space area does not exist in the first parking space area, the endpoint of the first parking space area does not exist in the second parking space area, the boundary of the first parking space area is not intersected with the boundary of the second parking space area, and it is determined that an overlapping area does not exist between the first parking space area and the second parking space area.

9. The vehicle control method of claim 8, wherein determining whether an endpoint of the second parking space region is located in the first parking space region comprises:

Determining at least one first edge forming the boundary of the first parking space area, and respectively determining the pointing information of each first edge, wherein the pointing information of the at least one first edge forms a clockwise direction or a counterclockwise direction;

determining a relative positional relationship between the endpoint and the first parking space region based on the endpoint of the second parking space region and the pointing information of the first side, wherein the relative positional relationship between the endpoint and the first parking space region is used for representing that the endpoint and the first parking space region are located on the same side or different sides of the first side;

determining that the endpoint is not located in the first parking space region if the endpoint is located in the second parking space region on a different side of the first side than the first parking space region;

and determining that the endpoint is located in the first parking space region if the endpoint is not located in the second parking space region and the first parking space region is located on a different side of the first side.

10. The vehicle control method according to claim 8, characterized in that determining whether the boundary of the first parking space region intersects the boundary of the second parking space region includes:

Acquiring the endpoint coordinates of each first side in at least one first side forming a first parking space region boundary and the endpoint coordinates of each second side in at least one second side forming a second parking space region boundary, wherein the endpoint coordinates consist of endpoint abscissa and endpoint ordinate;

determining that the boundary of the first parking space region does not intersect with the boundary of the second parking space region in response to the maximum endpoint abscissa of the at least one first side being less than the minimum endpoint abscissa of the at least one second side, or the minimum endpoint abscissa of the at least one first side being greater than the maximum endpoint abscissa of the at least one second side, or the maximum endpoint ordinate of the at least one first side being less than the minimum endpoint ordinate of the at least one second side, or the minimum endpoint ordinate of the at least one first side being greater than the maximum endpoint ordinate of the at least one second side;

and determining that the boundary of the first parking space region intersects the boundary of the second parking space region in response to the maximum endpoint abscissa of the at least one first side being greater than or equal to the minimum endpoint abscissa of the at least one second side and the minimum endpoint abscissa of the at least one first side being less than or equal to the maximum endpoint abscissa of the at least one second side and the maximum endpoint ordinate of the at least one first side being greater than or equal to the minimum endpoint ordinate of the at least one second side and the minimum endpoint ordinate of the at least one first side being less than or equal to the maximum endpoint ordinate of the at least one second side.

11. The vehicle control method according to claim 7, characterized in that the determining of the region overlap information between the first parking space region and the second parking space region based on the region shape and the relative positional relationship of the first parking space region and the second parking space region includes:

determining region overlapping information of the first parking space region and the second parking space region based on end point coordinates of each of the first sides of at least one first side constituting the first parking space region boundary and end point coordinates of each of the second sides of at least one second side constituting the second parking space region boundary, in response to the region shapes of the first parking space region and the second parking space region being rectangular;

and determining the region overlapping information of the first parking space region and the second parking space region based on the relative position relation between the first parking space region and the second parking space region and the pixel point number used for representing the first parking space region and the second parking space region in the perception information respectively in response to the region shapes of the first parking space region and the second parking space region not being the rectangle.

12. A vehicle control apparatus characterized by comprising:

the acquisition module is used for responding to the received automatic parking instruction and acquiring perception information obtained by perceiving the surrounding environment of the vehicle;

the detection module is used for detecting the sensing information to obtain at least one detection result; the detection result comprises a parking space area of a parking space in the surrounding environment of the vehicle and a limiter detection result, wherein the limiter detection result is used for representing that a limiter does not exist in the parking space or a limiter area of the limiter in the parking space;

the de-duplication module is used for de-duplication processing the at least one detection result based on the parking space area in the at least one detection result, and determining a target detection result;

the first control module is used for responding to the target detection result and comprising a target parking space area and a target limiter area, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path;

and the second control module is used for responding to the fact that a limiter does not exist in the parking space in the target detection result, constructing and generating a target limiter area in the target parking space area in the target detection result based on the target parking space area and the at least one detection result, generating a parking path of the vehicle according to the target parking space area and the target limiter area, and controlling the vehicle to automatically park to the parking space based on the parking path.

13. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored program, wherein the program, when run, controls a processor of a device in which the method of any one of claims 1 to 11 is performed.

14. An electronic device, comprising:

one or more processors;

a storage means for storing one or more programs;

when executed by the one or more processors, causes the one or more processors to perform the method of any of claims 1 to 11.