Spatial parking space detection method and device, equipment and storage medium

Abstract

The invention relates to a space parking space detection method, device and equipment and a storage medium, and the method comprises the steps: obtaining environment point cloud data of an environment where a target vehicle is located, and determining point cloud data corresponding to a reference vehicle according to the environment point cloud data; wherein the reference vehicle is a vehicle except the target vehicle in the environment; according to the point cloud data corresponding to the reference vehicle, candidate space parking spaces in the environment where the target vehicle is located are determined; wherein the candidate spatial parking space is a spatial parking space with a preset size; adjusting the position of the candidate space parking space to obtain a target space parking space; wherein the target space parking space is used for indicating a target vehicle to park. The reference vehicle is determined firstly, then the target space parking space is obtained through the reference vehicle, and the determination precision of the space parking space is improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a method, device, equipment and storage medium for detecting a parking space. Background Art

[0002] As people's living standards improve, the number of cars has increased sharply. The number of conventional marked parking spaces can no longer meet the parking needs of all cars. More and more cars have to park in non-marked parking spaces, such as on the side of the road or in open squares. Such non-marked parking spaces can be called space parking spaces.

[0003] Since there are no markings on the parking spaces, when using the automatic parking function, the vehicle cannot accurately detect a suitable parking space in the parking area, causing parking difficulties and affecting the user's parking experience. Summary of the invention

[0004] The present invention provides a method, device, equipment and storage medium for detecting a space parking space, so as to improve the detection accuracy of the space parking space.

[0005] In a first aspect, the present invention provides a method for detecting a parking space, characterized by comprising:

[0006] Acquire environmental point cloud data of the environment where the target vehicle is located, and determine point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle;

[0007] Determine, according to the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located; wherein the candidate space parking space is a space parking space of a preset size;

[0008] The candidate parking spaces are positioned to obtain a target parking space, wherein the target parking space is used to indicate a target vehicle to park.

[0009] In a second aspect, the present invention provides a device for detecting a parking space, characterized in that it comprises:

[0010] A data acquisition module, used to acquire environmental point cloud data of the environment where the target vehicle is located, and determine point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle;

[0011] A candidate determination module, used to determine a candidate space parking space in the environment where the target vehicle is located according to the point cloud data corresponding to the reference vehicle; wherein the candidate space parking space is a space parking space of a preset size;

[0012] The target determination module is used to adjust the position of the candidate space parking space to obtain a target space parking space; wherein the target space parking space is used to indicate the target vehicle to park.

[0013] In a third aspect, the present invention provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0014] The memory stores computer-executable instructions;

[0015] The processor executes the computer-executable instructions stored in the memory to implement the method according to the first aspect.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0017] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which implements the method described in the first aspect when executed by a processor.

[0018] The present invention provides a method, device, equipment and storage medium for detecting a spatial parking space. By acquiring the environmental point cloud data of a target vehicle, a reference vehicle in the environment is determined according to the environmental point cloud data, thereby obtaining point cloud data for the reference vehicle. Based on the point cloud data of the reference vehicle, a spatial parking space that matches the reference vehicle can be generated as a candidate spatial parking space. The position of the candidate spatial parking space is adjusted to obtain the final target spatial parking space, and the target vehicle can park according to the target spatial parking space. According to the point cloud data corresponding to the reference vehicle, the state of the target vehicle after parking can be made consistent with that of the reference vehicle, thereby improving the standardization of parking in non-lined areas. Random parking between vehicles can be avoided, resulting in low space utilization, thereby improving the detection accuracy of spatial parking spaces and enhancing user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0020] Figure 1 A schematic diagram of a flow chart of a method for detecting a parking space provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of a vehicle body coordinate system provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of point cloud data corresponding to a reference vehicle provided in an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of a flow chart of a method for detecting a parking space provided by an embodiment of the present invention;

[0024] Figure 5 A schematic diagram of a flow chart of a method for detecting a parking space provided by an embodiment of the present invention;

[0025] Figure 6 A schematic diagram of projection of point cloud data provided by an embodiment of the present invention;

[0026] Figure 7 A schematic diagram of parking space types provided by an embodiment of the present invention;

[0027] Figure 8 A schematic diagram of a candidate parking space provided by an embodiment of the present invention;

[0028] Fig. 9 A schematic diagram of a flow chart of a method for detecting a parking space provided by an embodiment of the present invention;

[0029] Fig.10 A schematic diagram of candidate parking spaces and curbs provided in an embodiment of the present invention;

[0030] Fig.11 A structural block diagram of a space parking space detection device provided by an embodiment of the present invention;

[0031] Fig.12 A structural block diagram of a space parking space detection device provided by an embodiment of the present invention;

[0032] Fig.13 A structural block diagram of an electronic device provided by an embodiment of the present invention;

[0033] Fig.14 A structural block diagram of an electronic device provided by an embodiment of the present invention.

[0034] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0035] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0036] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0037] In the description of the present invention, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. In addition, in the description of the present invention, unless otherwise specified, "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0038] It should be noted that due to space limitations, this specification does not exhaustively list all optional implementations. After reading this specification, those skilled in the art should be able to think that as long as the technical features do not contradict each other, any combination of technical features can constitute an optional implementation. The following is a detailed description of each embodiment.

[0039] As people's living standards improve, more and more families own one or more cars. The surge in the number of cars makes parking more and more difficult. The number of conventional lined parking spaces can no longer meet the parking needs of all cars, so more and more cars have to park in non-lined parking spaces, such as on the side of the road or in an open square. The empty parking area between or behind these non-lined parking spaces is called a space parking space.

[0040] At present, more and more vehicles support intelligent functions, and a typical application of intelligence is the automatic parking function. The automatic parking technology for conventional lined parking spaces has gradually matured. However, the automatic parking technology for spaced parking spaces needs further research.

[0041] As a distance measurement sensor, laser radar has the advantages of long measurement distance and high measurement accuracy, so it is favored by industries such as robots and autonomous driving. In the embodiment of the present invention, in order to ensure the high reliability of spatial parking space detection, in addition to laser radar, a variety of other sensors can also be used to expand the perception range of the entire vehicle, thereby improving the detection accuracy of spatial parking spaces.

[0042] The present invention provides a method, device, equipment and storage medium for detecting a parking space, aiming to solve the above technical problems in the prior art.

[0043] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0044] Figure 1 FIG. 1 is a flow chart of a method for detecting a parking space according to an embodiment of the present invention. The method can be performed by a device for detecting a parking space. Figure 1 As shown, the method comprises the following steps:

[0045] S101, obtaining environmental point cloud data of the environment where the target vehicle is located, and determining point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle.

[0046] Exemplarily, the environment where the target vehicle is located may be a scene where there is no parking space marked on the ground, such as a roadside or a square. Various sensors may be installed on the vehicle, for example, a laser radar, a camera, an ultrasonic sensor, etc. In this embodiment, a plurality of laser radars may be installed on the vehicle, which may be used to obtain point cloud data in the environment as environmental point cloud data. Environmental point cloud data may be obtained in real time, or after responding to the user's parking instruction, environmental point cloud data may be obtained. Environmental point cloud data may also be automatically obtained when it is determined that the vehicle is in a waiting-to-park state. For example, several acquisition conditions for environmental point cloud data may be preset. When the current state of the vehicle satisfies these acquisition conditions, it is considered that the vehicle needs to be parked, and environmental point cloud data may be obtained. The preset acquisition conditions may include that the vehicle steering angle does not exceed a preset steering angle threshold, the vehicle speed is within a preset speed range, the vehicle is not currently parking, etc., that is, when the vehicle steering angle exceeds the steering angle threshold, there is no need to detect the space parking space, or when the vehicle is parking, there is no need to detect the space parking space, etc.

[0047] The sensor data acquired in this embodiment can be in the vehicle body coordinate system. Figure 2 is a schematic diagram of the vehicle body coordinate system. Each target vehicle has its own vehicle body coordinate system. The origin of the coordinate system is the center point of the rear axle of the vehicle. The positive direction of the X axis is perpendicular to the rear axle toward the front of the vehicle. The positive direction of the Y axis is toward the left, and the positive direction of the Z axis is vertically upward. The data obtained by the laser radar is used as the original point cloud. The original point cloud is converted to the vehicle body coordinate system using the external parameters of the laser radar to obtain the environmental point cloud data. The conversion formula can be:

[0048] p , =Rp+t;

[0049] Among them, p , is the environmental point cloud data, R is the preset three-dimensional rotation transformation matrix, p is the original point cloud, and t is the coordinate of the origin of the lidar coordinate system in the vehicle coordinate system.

[0050] The environment where the target vehicle is located may include multiple vehicles, and vehicles other than the target vehicle may be used as reference vehicles. For example, a vehicle may be randomly selected as a reference vehicle. Point cloud data corresponding to the reference vehicle is extracted from the environmental point cloud data. The point cloud data corresponding to the reference vehicle is a part of the environmental point cloud data. For example, a rectangular point cloud cluster may be identified from the environmental point cloud data, and the point cloud in the point cloud cluster may be determined as point cloud data corresponding to a reference vehicle. Figure 3 Schematic diagram of the point cloud data corresponding to the reference vehicle. Figure 3 The cube in the figure is the reference vehicle, and the point cloud data in the cube is the point cloud data corresponding to the reference vehicle.

[0051] S102. Determine, based on the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located; wherein the candidate space parking space is a space parking space of a preset size.

[0052] Exemplarily, the point cloud data corresponding to the reference vehicle can indicate the position, size, orientation and other information of the reference vehicle. Based on the point cloud data corresponding to the reference vehicle, an initial spatial parking space is planned in the environment where the target vehicle is located as a candidate spatial parking space. The length and width of the spatial parking space are predetermined, that is, the candidate spatial parking space is a spatial parking space of a preset size. For example, based on the position of the reference vehicle, a candidate spatial parking space can be planned near the reference vehicle; or, based on the orientation of the reference vehicle, a candidate spatial parking space with the same orientation as the reference vehicle can be planned.

[0053] Candidate parking spaces should be generated in places without obstacles. Therefore, the area with sparse point cloud distribution can be determined based on the environmental point cloud data. Based on the point cloud data corresponding to the reference vehicle, a candidate parking space with a preset size and the same orientation as the reference vehicle is generated in the area with sparse point cloud distribution, so that the parking direction of vehicles in the environment is relatively consistent, improving the standardization of vehicle parking and realizing effective use of space.

[0054] S103: Adjust the position of the candidate parking space to obtain a target parking space; wherein the target parking space is used to indicate the target vehicle to park.

[0055] For example, after obtaining the candidate space parking space, the candidate space parking space can also be adjusted. Since the size of the candidate space parking space is a preset size, the position of the candidate space parking space can be adjusted, and the adjusted candidate space parking space is determined as the target space parking space. The target vehicle can park according to the target space parking space, so that the target vehicle is parked in the target space parking space.

[0056] When adjusting the candidate space parking space, the candidate space parking space can be translated or rotated. For example, if there are individual point cloud data in the candidate space parking space, in order to ensure the safety of parking, the candidate space parking space can be slightly rotated so that the individual point cloud data is not in the candidate space parking space, and the rotated candidate space parking space is determined as the target space parking space.

[0057] In order to reduce the impact of roadside parking on traffic, the target space should be as close to the curb as possible. If the reference vehicle is parked far from the curb, the generated candidate space may also be far from the curb. In order to obtain a more ideal result, the position of the curb in the environment can be determined by using sensors such as cameras in the vehicle, and the candidate space can be adjusted according to the position of the curb so that the adjusted space is close to the curb.

[0058] In some specific occasions, even if the free space meets the parking requirements, parking should not be allowed, such as in aisles or elevator entrances. In order to suppress the generation of target space parking spaces in these places that are not suitable for parking, some suppression conditions can be preset. For example, the suppression condition can be that there are obstacles of a certain length such as curbs, walls, fences, etc. at the bottom of the space parking space, or there are no static obstacles between the entrance edge of the space parking space and the vehicle. If the generated candidate space parking space does not meet the suppression condition, the current candidate space parking space is deleted. Among them, the first suppression condition is to ensure that the space parking space will not be released in locations such as aisles or elevator entrances, and the second condition is to ensure that the space parking space will not be generated behind static obstacles such as pillars and fences, so as to ensure that the released space parking space can be truly parked.

[0059] If the candidate space parking space is deleted, or the target space parking space cannot be adjusted, the reference vehicle can be replaced and a new candidate space parking space can be regenerated until the target space parking space is obtained.

[0060] In an embodiment of the present invention, by acquiring the environmental point cloud data of the target vehicle, a reference vehicle in the environment is determined based on the environmental point cloud data, thereby obtaining point cloud data for the reference vehicle. Based on the point cloud data of the reference vehicle, a spatial parking space that matches the reference vehicle can be generated as a candidate spatial parking space. The position of the candidate spatial parking space is adjusted to obtain the final target spatial parking space, and the target vehicle can park according to the target spatial parking space. Based on the point cloud data corresponding to the reference vehicle, the state of the target vehicle after parking can be made consistent with that of the reference vehicle, thereby improving the standardization of parking in non-lined areas. Low space utilization caused by random parking between vehicles is avoided, thereby improving the detection accuracy of spatial parking spaces.

[0061] Figure 4 A schematic flow chart of a method for detecting a parking space provided in an embodiment of the present invention is provided. This embodiment is an optional embodiment based on the above embodiment.

[0062] In this embodiment, point cloud data corresponding to a reference vehicle is determined based on environmental point cloud data, including: determining a reference vehicle located in the environment where the target vehicle is located based on the environmental point cloud data; and extracting point cloud data corresponding to the reference vehicle from the environmental point cloud data.

[0063] like Figure 4 As shown, the method comprises the following steps:

[0064] S401, obtaining environmental point cloud data of an environment where a target vehicle is located, and determining a reference vehicle located in the environment where the target vehicle is located according to the environmental point cloud data.

[0065] Exemplarily, the laser radar point cloud data in the vehicle body coordinate system is obtained as the environmental point cloud data. The environment where the target vehicle is located includes multiple vehicles, and one of the multiple vehicles is selected as the reference vehicle. The vehicles in the environment can be identified based on the environmental point cloud data, and then a vehicle other than the target vehicle is determined from these vehicles as the reference vehicle. For example, the reference vehicle can be selected randomly, or the reference vehicle can be selected according to the distance between the target vehicle and other vehicles. The vehicle closest to the target vehicle can be used as the reference vehicle.

[0066] In this embodiment, a reference vehicle located in the environment where the target vehicle is located is determined based on the environmental point cloud data, including: inputting the environmental point cloud data into a preset deep learning model to obtain attribute information of objects in the environment where the target vehicle is located; wherein the preset deep learning model is used to perform semantic segmentation processing on the environmental point cloud data, and output attribute information of the object, the attribute information including type information and location information; and determining a reference vehicle from the objects in the environment where the target vehicle is located based on the attribute information of the objects in the environment where the target vehicle is located.

[0067] Specifically, a deep learning model is pre-set, and the deep learning model is a neural network model, which can be used to perform semantic segmentation processing on environmental point cloud data and output attribute information of the object, and the attribute information may include information such as object type and location. The object refers to any object in the environment where the target vehicle is located, for example, it can be a vehicle, a tree, a building, a curb, etc. In this embodiment, the model structure of the preset deep learning model is not specifically limited.

[0068] The environmental point cloud data is input into the deep learning model, and the deep learning model performs semantic segmentation on the environmental point cloud data, and outputs the semantic objects recognized by the model. Each object contains attribute information such as ID, type, size, center point coordinates and orientation. In this embodiment, the reference vehicle is limited to a stationary vehicle, that is, the reference vehicle is determined from the stationary vehicles in the environment. According to the type information in the attribute information of each object, it can be determined which objects in the environment are vehicles, and according to the size information in the attribute information, a vehicle of the same vehicle type as the target vehicle can be determined as a candidate vehicle. Then, according to the position information in the attribute information, the candidate vehicles in a stationary state are determined, and the reference vehicle is determined from them. For example, if the target vehicle is a small car, an object of the type of a small car and a stationary state can be used as a reference vehicle. If there are multiple objects of the type of a small car and a stationary state in the environment, one can be randomly selected as a reference vehicle, or the object closest to the target vehicle can be determined as a reference vehicle according to the position information of these objects.

[0069] The beneficial effect of this setting is that the laser point cloud is semantically segmented through a deep learning model, all objects in the environment are identified, and the reference vehicle is determined, thereby improving the accuracy and efficiency of determining the reference vehicle, thereby improving the accuracy and efficiency of spatial parking space detection.

[0070] In this embodiment, a reference vehicle is determined from the objects in the environment where the target vehicle is located based on the attribute information of the objects in the environment where the target vehicle is located, including: determining the position information of the target vehicle; determining the distance value between the target vehicle and the object based on the position information of the target vehicle and the position information of the objects in the environment where the target vehicle is located; if the distance value is less than a preset first distance threshold, and the type information of the object is a preset vehicle type, then determining that the object is a reference vehicle.

[0071] Specifically, the target vehicle can determine its own position information, for example, its own position information can be determined as the origin position in the vehicle body coordinate system. The position information of each object in the environment where the target vehicle is located is determined. In this embodiment, the position information of each object in a stationary state can be determined. Based on the position information of the target vehicle and the position information of the object, the distance value between the target vehicle and the object can be determined.

[0072] A first distance threshold is preset, and each distance value is compared with the first distance threshold. If the distance value is less than the preset first distance threshold, and the type information of the object is a preset vehicle type, the object can be determined to be a reference vehicle. The preset vehicle type is the vehicle type of the target vehicle. If there are multiple distance values ​​less than the first distance threshold and are of the preset vehicle type, the object corresponding to the smallest distance value is determined to be the reference vehicle.

[0073] If the candidate parking space or the target parking space cannot be determined based on the point cloud data corresponding to the reference object, the reference vehicle can be replaced and the parking space can be re-detected. For example, the second closest stationary vehicle to the target parking space and of the same type can be used as a new reference vehicle.

[0074] The beneficial effect of such a setting is that objects of the same type and within a specified threshold distance from the self-vehicle are selected as reference vehicles, so that the target vehicle can be parked near the reference vehicle, and vehicles can park in an orderly manner, thereby achieving standardized planning of parking positions and avoiding waste of space.

[0075] S402: extracting point cloud data corresponding to a reference vehicle from the environmental point cloud data.

[0076] Exemplarily, the environmental point cloud data includes point cloud data corresponding to each object. After determining the reference object, the environmental point cloud data can be divided to obtain point cloud data corresponding to the reference vehicle. For example, the point cloud data located in the reference vehicle can be determined as the point cloud data corresponding to the reference vehicle.

[0077] In this embodiment, a reference vehicle in the environment is detected based on the environmental point cloud data, and the point cloud data of the reference vehicle is extracted, so that parking space detection is performed based on the existing vehicles in the environment, thereby avoiding random parking of the target vehicle and improving the standardization of parking space planning and the detection accuracy of parking spaces.

[0078] In this embodiment, point cloud data corresponding to a reference vehicle is extracted from environmental point cloud data, including: performing target frame selection processing on the reference vehicle in the environmental point cloud data according to a preset target frame extension range; and determining the point cloud data in the target frame as point cloud data corresponding to the reference vehicle.

[0079] Specifically, the extension range of the target frame is pre-set. The target frame refers to an external frame of a preset shape that can frame the object. In order to completely frame the object, the external frame can be expanded to a certain extent. For example, the size of the reference vehicle is first determined according to the deep learning model, and then a small expansion is performed on the basis of the determined size to obtain the target frame. The point cloud data falling within the target frame is determined as the point cloud data corresponding to the reference object.

[0080] The beneficial effect of such a setting is that the size of the reference vehicle is expanded to obtain the target frame of the reference vehicle, so that all the point cloud data corresponding to the reference vehicle can be selected, thereby improving the determination accuracy of the point cloud data corresponding to the reference vehicle and further improving the detection accuracy of the spatial parking space.

[0081] In this embodiment, point cloud data corresponding to a reference vehicle is extracted from environmental point cloud data, including: screening the environmental point cloud data according to the position information and height information of the environmental point cloud data to obtain the screened environmental point cloud data; and extracting point cloud data corresponding to the reference vehicle from the screened environmental point cloud data.

[0082] Specifically, the environmental point cloud data can be preprocessed. For example, preprocessing can be performed before determining the reference vehicle, and the reference vehicle can be determined based on the preprocessed environmental point cloud data. Alternatively, the reference vehicle can be first determined based on the environmental point cloud data, and then the environmental point cloud data can be preprocessed. Based on the preprocessed environmental point cloud data, a target box can be selected to obtain point cloud data corresponding to the reference vehicle.

[0083] Preprocessing can refer to screening out discrete points, ground points, points whose height exceeds a set threshold, and points that are far away from the target vehicle. That is, the discrete points and ground points in the environmental point cloud data can be determined, as well as the height of each point in the environmental point cloud data and the distance between each point and the target vehicle can be determined, wherein the discrete points can be points that cannot be clustered into a cluster, and the ground points refer to points located on the ground surface.

[0084] The beneficial effect of this setting is that, since the sensing range of the laser radar is relatively far, in order to reduce the amount of calculation, it is necessary to limit the range of the laser point cloud currently being processed, so the laser point cloud that is more than a certain threshold away from the vehicle can be ignored. In addition, obstacles at high places will not affect parking, so points with a height exceeding a certain threshold should also be removed. The purpose of this step is to remove point cloud data that is not related to the spatial parking space as much as possible, reduce the amount of calculation for subsequent spatial parking space detection, and improve the accuracy and efficiency of detection.

[0085] S403. Determine, based on the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located; wherein the candidate space parking space is a space parking space of a preset size.

[0086] Exemplarily, this step may refer to the above-mentioned step S102 and will not be described in detail.

[0087] S404: Adjust the position of the candidate parking space to obtain a target parking space; wherein the target parking space is used to indicate the target vehicle to park.

[0088] Exemplarily, this step may refer to the above-mentioned step S103 and will not be described in detail.

[0089] In an embodiment of the present invention, by acquiring the environmental point cloud data of the target vehicle, a reference vehicle in the environment is determined based on the environmental point cloud data, thereby obtaining point cloud data for the reference vehicle. Based on the point cloud data of the reference vehicle, a spatial parking space that is consistent with the reference vehicle can be generated as a candidate spatial parking space. The position of the candidate spatial parking space is adjusted to obtain the final target spatial parking space, and the target vehicle can park according to the target spatial parking space. Based on the point cloud data corresponding to the reference vehicle, the state of the target vehicle after parking can be made consistent with the reference vehicle, thereby improving the standardization of parking in non-lined areas. Low space utilization caused by random parking between vehicles is avoided, and the detection accuracy of spatial parking spaces is improved.

[0090] Figure 5 A schematic flow chart of a method for detecting a parking space provided in an embodiment of the present invention is provided. This embodiment is an optional embodiment based on the above embodiment.

[0091] In this embodiment, candidate spatial parking spaces in the environment where the target vehicle is located are determined based on point cloud data corresponding to a reference vehicle, including: determining the coverage rate of the reference vehicle based on the point cloud data corresponding to the reference vehicle; wherein the coverage rate represents the proportion of the area of ​​the reference vehicle that is not blocked by obstacles; and determining the candidate spatial parking spaces in the environment where the target vehicle is located based on the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle.

[0092] like Figure 5 As shown, the method comprises the following steps:

[0093] S501, obtaining environmental point cloud data of the environment where the target vehicle is located, and determining point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle.

[0094] Exemplarily, this step may refer to the above-mentioned step S101 and will not be described in detail.

[0095] S502. Determine the coverage rate of the reference vehicle according to the point cloud data corresponding to the reference vehicle; wherein the coverage rate represents the proportion of the area of ​​the reference vehicle that is not blocked by obstacles.

[0096] For example, based on the point cloud data corresponding to the reference vehicle, it is determined whether the reference vehicle is blocked by obstacles. The coverage rate can be used to characterize the proportion of the area of ​​the reference vehicle that is not blocked by obstacles to the overall area of ​​the vehicle. The higher the coverage rate, the higher the proportion of the reference vehicle that is not blocked; the lower the coverage rate, the higher the proportion of the reference vehicle that is blocked.

[0097] The distance between each point cloud data and the target vehicle in the point cloud data corresponding to the reference vehicle can be determined. If the distance between the point cloud data and the target vehicle is less than the preset minimum distance threshold, it is considered that there is an obstacle between the target vehicle and the reference vehicle, and the point cloud data can be determined as occlusion data, that is, the point cloud data occludes the reference vehicle. According to the number and distribution of occlusion data, the coverage rate of the reference vehicle can be determined. For example, if there are more occlusion data in the point cloud data corresponding to the reference vehicle, it is considered that the proportion of the reference vehicle being occluded is high, that is, the coverage rate is low.

[0098] In this embodiment, the target box is a rectangular box; the coverage of the reference vehicle is determined according to the point cloud data corresponding to the reference vehicle, including: determining the target edge of the target box of the reference vehicle according to the distance between the edge of the target box of the reference vehicle and the target vehicle; projecting the point cloud data corresponding to the reference vehicle onto the target edge to obtain the projection point of the point cloud data corresponding to the reference vehicle; determining the sum of the distances between adjacent projection points, and determining the coverage of the reference vehicle according to the length of the target edge and the sum of the distances between adjacent projection points.

[0099] Specifically, the target frame may be a rectangular frame, that is, the target frame may have four sides. The distance between each side of the target frame of the reference vehicle and the target vehicle is determined, that is, four distances may be determined. According to the distance between each side of the target frame of the reference vehicle and the target vehicle, the target side of the target frame of the reference vehicle is determined, and the target side is one of the four sides. The side with the smallest distance may be used as the target side. In other words, the target side is the side closest to the target vehicle.

[0100] Project the point cloud data corresponding to the reference vehicle onto the target edge to obtain the projection points of the point cloud data corresponding to the reference vehicle, where each point cloud data corresponds to a projection point. Figure 6 Schematic diagram of the projection of point cloud data. Figure 6 In the target box, P0, P1, P2, and P3 form a target frame. There are n point cloud data in the target frame. P2P3 is the target edge. The point cloud data is projected onto the target edge to obtain p0, p1, ..., p i 、p j ,……,p n Wait for n projection points.

[0101] Determine the sum of the distances between every two adjacent projection points. For example, you can determine the distance between p0 and p1, the distance between p1 and p2, and so on. n-1 and p n The distances between the two projection points are added together to obtain the sum of the distances. The length of the target side is determined, and the coverage of the reference vehicle is determined based on the length of the target side and the sum of the distances between the adjacent projection points. The calculation formula for the coverage can be:

[0102] r=L , / L;

[0103] Among them, r is the coverage rate, L , is the sum of the distances between adjacent projection points, and L is the length of the target edge. The sum of the distances between adjacent projection points is divided by the length of the target edge to obtain the coverage rate.

[0104] The beneficial effect of this setting is that the candidate parking spaces need to be generated in places where the laser radar field of view is good enough. By projecting the point cloud data, it can be determined whether the distance between the two point clouds is too large, thereby judging whether there is an obstacle between the two point clouds. Using coverage as a measure of the laser radar field of view improves the accuracy of parking space detection and thus improves parking safety.

[0105] In this embodiment, the method further includes: if the distance between adjacent projection points is greater than a preset second distance threshold, determining that the distance between adjacent projection points is a preset value.

[0106] Specifically, a second distance threshold is pre-set, and after obtaining the distance between two adjacent projection points, the distance is compared with the preset second distance threshold. If the distance between adjacent projection points is less than or equal to the preset second distance threshold, the distance is not processed and is added to the sum of the distances; if the distance between adjacent projection points is greater than the preset second distance threshold, the distance is adjusted to a preset value. For example, the preset value may be 0. In other words, when the distance between two adjacent projection points exceeds the set second distance threshold, it is considered that the laser point cloud is discontinuous, and there may be occlusion between the two projection points, and the distance is not included in the sum of the distances.

[0107] The beneficial effect of such a setting is that by judging the distance between adjacent projection points and the second distance threshold, the distance can be adjusted, the situation where the reference vehicle is blocked can be inferred, and the calculation accuracy of the coverage rate can be improved.

[0108] S503: Determine candidate parking spaces in the environment where the target vehicle is located according to the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle.

[0109] For example, if it is determined based on the coverage rate that the reference vehicle is not obscured, a candidate spatial parking space can be determined based on the point cloud data corresponding to the reference vehicle. For example, a candidate spatial parking space that is consistent with the reference vehicle in size and orientation can be determined between the reference vehicle and the target vehicle.

[0110] According to the point cloud data corresponding to the reference vehicle, the location of the candidate space parking space can be determined. For example, a point cloud data can be selected from the point cloud data corresponding to the reference vehicle as a corner point of the candidate space parking space. Based on the corner point, a rectangular candidate space parking space is generated according to a preset size, and the orientation of the candidate space parking space can be consistent with the orientation of the reference vehicle. For example, if the reference vehicle is facing south, the candidate space parking space can also be a parking space in the north-south direction.

[0111] In this embodiment, the coverage rate of the reference vehicle is calculated using the point cloud data corresponding to the reference vehicle, so as to determine the candidate space parking spaces, improve the accuracy of determining the candidate space parking spaces, and further improve the parking safety and user experience.

[0112] In this embodiment, the attribute information includes an orientation angle; the candidate space parking spaces in the environment where the target vehicle is located are determined based on the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle, including: if the coverage rate of the reference vehicle is greater than a preset ratio threshold, the parking space type of the candidate space parking space is determined based on the orientation angle of the reference vehicle; wherein the parking space type is a horizontal type, a vertical type or an oblique type; according to the position information of the point cloud data corresponding to the reference vehicle and the position information of the target vehicle, the target corner point is determined from the point cloud data corresponding to the reference vehicle; wherein the target corner point is a corner point of the candidate space parking space; according to the position information of the target corner point, the parking space type of the candidate space parking space, and the preset size of the candidate space parking space, the candidate space parking space in the environment where the target vehicle is located is determined.

[0113] Specifically, a ratio threshold is set in advance, and after the coverage rate is obtained, the coverage rate is compared with the preset ratio threshold. If the coverage rate is less than or equal to the preset ratio threshold, it means that the reference vehicle is blocked and the candidate parking space cannot be generated based on the reference vehicle. A new reference vehicle is re-determined and the coverage rate is calculated again.

[0114] If the coverage rate is greater than a preset ratio threshold, the attribute information of the reference vehicle can be obtained to determine the orientation angle of the reference vehicle. According to the orientation angle of the reference vehicle, the parking space type of the candidate space is determined. The parking space type can be a horizontal type, a vertical type, or a diagonal type. Figure 7 A schematic diagram of parking space types. Figure 7 From left to right in the figure are horizontal parking spaces, vertical parking spaces, and diagonal parking spaces. Pre-set the orientation angle ranges corresponding to different types of parking spaces, determine which orientation angle range the orientation angle of the reference vehicle falls into, and obtain the parking space type of the candidate space parking space. For example, the orientation angle of the reference vehicle is limited to the range of [-90°, 90°], the orientation angle is positive counterclockwise and negative clockwise. If the absolute value of the orientation angle of the reference vehicle is less than 20°, a horizontal type candidate space parking space is generated; if the absolute value of the orientation angle of the reference vehicle is between [20°, 70°], a diagonal type candidate space parking space is generated; if the absolute value of the orientation angle of the reference vehicle is between [70°, 90°], a vertical type space parking space is generated.

[0115] According to the position information of the point cloud data corresponding to the reference vehicle and the position information of the target vehicle, a point cloud data can be determined from the point cloud data corresponding to the reference vehicle as the target corner point. The target corner point is a corner point of the candidate space parking space, that is, a vertex of the rectangle. According to the position information of the target corner point, the parking space type of the candidate space parking space, and the preset size of the candidate space parking space, the candidate space parking space in the environment where the target vehicle is located is generated. The point closest to the target vehicle in the Y direction can be first selected from the point cloud data corresponding to the reference vehicle, that is, the point with the Y-axis coordinate closest to 0 is selected as the target corner point. Then, according to the parking space type of the candidate space parking space and the preset size of the candidate space parking space, for example, three other corner points can be generated according to the reference vehicle orientation angle, and the preset space parking space length and width, so as to obtain the candidate space parking space in the environment where the target vehicle is located. Figure 8 Schematic diagram of candidate parking spaces. Figure 8 In the figure, a is the target corner point, and the parking space type is an oblique type facing the upper right. According to the distribution of the environmental point cloud data, the coordinates of the three corner points b, c, and d can be determined in the area with sparse point cloud data to obtain the candidate space parking space.

[0116] The beneficial effect of this arrangement is that the parking space type is determined according to the orientation angle of the reference vehicle, and the point closest to the vehicle in the Y direction is selected from the associated point cloud of the reference vehicle as the target corner point, thereby obtaining a candidate space parking space, making the orientation of the candidate space parking space match that of the reference vehicle, and improving parking safety.

[0117] S504: Adjust the position of the candidate parking space to obtain a target parking space; wherein the target parking space is used to indicate the target vehicle to park.

[0118] Exemplarily, this step may refer to the above-mentioned step S103 and will not be described in detail.

[0119] In an embodiment of the present invention, by acquiring the environmental point cloud data of the target vehicle, a reference vehicle in the environment is determined based on the environmental point cloud data, thereby obtaining point cloud data for the reference vehicle. Based on the point cloud data of the reference vehicle, a spatial parking space that is consistent with the reference vehicle can be generated as a candidate spatial parking space. The position of the candidate spatial parking space is adjusted to obtain the final target spatial parking space, and the target vehicle can park according to the target spatial parking space. Based on the point cloud data corresponding to the reference vehicle, the state of the target vehicle after parking can be made consistent with the reference vehicle, thereby improving the standardization of parking in non-lined areas. Low space utilization caused by random parking between vehicles is avoided, and the detection accuracy of spatial parking spaces is improved.

[0120] Fig. 9A schematic flow chart of a method for detecting a parking space provided in an embodiment of the present invention is provided. This embodiment is an optional embodiment based on the above embodiment.

[0121] In this embodiment, the position of the candidate space parking space is adjusted to obtain the target space parking space, including: obtaining environmental information of the candidate space parking space; wherein the environmental information includes surround view camera data and ultrasonic sensor data; according to the environmental information, the position of the candidate space parking space is adjusted to obtain the target space parking space.

[0122] like Fig. 9 As shown, the method comprises the following steps:

[0123] S901. Obtain environmental point cloud data of the environment where the target vehicle is located, and determine point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle.

[0124] Exemplarily, this step may refer to the above-mentioned step S101 and will not be described in detail.

[0125] S902. Determine, based on the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located; wherein the candidate space parking space is a space parking space of a preset size.

[0126] Exemplarily, this step may refer to the above-mentioned step S102 and will not be described in detail.

[0127] S903, obtaining environmental information of the candidate parking space; wherein the environmental information includes surround view camera data and ultrasonic sensor data.

[0128] Exemplarily, a target vehicle may be equipped with a variety of sensors, such as a surround view camera and an ultrasonic sensor, etc. The surround view camera data is a semantic point cloud perceived by the surround view camera, and the ultrasonic sensor data is cluster information of the ultrasonic sensor. The semantic point cloud of the surround view camera contains the two-dimensional coordinates and semantic type of each point. In this embodiment, the surround view semantic types used include curbs and fences. The ultrasonic cluster information is mainly a set of straight line segment information obtained after clustering the ultrasonic data, including the starting point and end point of the line segment.

[0129] Environmental information of the environment where the target vehicle is located can be obtained in real time or at a fixed time. After the candidate parking space is determined, the environmental information of the candidate parking space is extracted from the environmental information of the entire environment. The candidate parking space refers to environmental information near or within the area of ​​the candidate parking space, for example, it can be environmental information within the area of ​​the candidate parking space and within a certain range outside the candidate parking space.

[0130] S904: Adjust the position of the candidate parking space according to the environmental information to obtain the target parking space.

[0131] For example, the candidate parking space can be adjusted to obtain the target parking space based on the environmental point cloud data, the surround camera data, and the ultrasonic sensor data. For example, if it is determined that there is an obstacle on the left side of the candidate parking space based on the ultrasonic sensor data, the candidate parking space can be translated to the right to obtain the target parking space.

[0132] In this embodiment, due to the limited field of view of the laser radar, even if the coverage rate meets the set ratio threshold, there may still be obstacles in the candidate parking space area that the laser radar cannot perceive. Therefore, it is necessary to use other sensor data to adjust the candidate parking space to improve the accuracy of determining the target parking space.

[0133] In this embodiment, the position of the candidate space parking space is adjusted according to the environmental information to obtain the target space parking space, including: if it is determined that there is an obstacle at the candidate space parking space according to the surround view camera data and the ultrasonic sensor data, the position of the candidate space parking space is adjusted to obtain the target space parking space.

[0134] Specifically, based on the surround camera data and ultrasonic sensor data, it can be determined whether there are obstacles that obviously intrude into the candidate parking space, that is, whether the candidate parking space is occupied. If so, the candidate parking space cannot be released directly and needs to be adjusted; if not, the candidate parking space can be determined as the target parking space.

[0135] For example, based on the surround view camera data, the semantic type of the semantic point cloud perceived by the surround view camera can be determined. If it is determined that the right side of the candidate space parking space is a fence, the candidate space parking space can be adjusted to the left to make the candidate space parking space away from the fence. If it is determined that a certain distance to the left of the candidate space parking space is a curb, the candidate space parking space can be adjusted to the left to make the candidate space parking space close to the curb. If it is determined that the candidate space parking space cannot be adjusted, for example, the candidate space parking space is surrounded by obstacles, the candidate space parking space is directly deleted, a new reference vehicle is re-determined, and the space parking space is detected again.

[0136] The beneficial effect of such a setting is that, in order to ensure that the space parking space has a high release rate, when the candidate space parking space is determined to be occupied, the parking space is not directly deleted, but the position of the candidate space parking space is adjusted. Usually, the adjusted space parking space can ensure that the obstacle is outside the bottom edge of the candidate space parking space, thereby improving the detection accuracy and efficiency of the space parking space and improving the safety of parking.

[0137] In this embodiment, the parking space type of the candidate space parking space is a vertical type or an oblique type; the method also includes: if it is determined that there is an obstacle at the candidate space parking space based on the surround view camera data and the ultrasonic sensor data, then the position information and height information of the obstacle are determined; if the position information of the obstacle is within a preset area range in the candidate space parking space, and the height information of the obstacle is less than a preset height threshold, then the candidate space parking space is determined to be a target space parking space.

[0138] Specifically, for the two types of candidate spaces, vertical or diagonal, it should be noted that there is a situation where no adjustment is required even if the candidate space is occupied. If the obstacle intrudes into the candidate space, close to the bottom edge of the candidate space, the distance from the obstacle to the candidate space is less than the rear overhang length of the vehicle, and the obstacle height is low, then no adjustment is required for the candidate space. For example, if the obstacle is a curb, the rear end of the target vehicle can extend beyond the curb when reversing without colliding with the curb. Fig.10 Schematic diagram of candidate parking spaces and curbs. Fig.10 In the figure, the oblique line part is the curb. When the vehicle backs up, even if it reaches the curb inside the parking space, it will not collide with the curb.

[0139] The beneficial effect of this setting is that it reduces the adjustment process of candidate parking spaces, improves the efficiency of parking space detection, and thus improves the efficiency of automatic parking and enhances user experience.

[0140] In an embodiment of the present invention, by acquiring the environmental point cloud data of the target vehicle, a reference vehicle in the environment is determined based on the environmental point cloud data, thereby obtaining point cloud data for the reference vehicle. Based on the point cloud data of the reference vehicle, a spatial parking space that is consistent with the reference vehicle can be generated as a candidate spatial parking space. The position of the candidate spatial parking space is adjusted to obtain the final target spatial parking space, and the target vehicle can park according to the target spatial parking space. Based on the point cloud data corresponding to the reference vehicle, the state of the target vehicle after parking can be made consistent with the reference vehicle, thereby improving the standardization of parking in non-lined areas. Low space utilization caused by random parking between vehicles is avoided, and the detection accuracy of spatial parking spaces is improved.

[0141] Fig.11 A structural block diagram of a space parking space detection device provided by an embodiment of the present invention. For ease of description, only the parts related to the embodiment of the present disclosure are shown. Fig.11 The spatial parking space detection device 1100 includes: a data acquisition module 1101, a candidate determination module 1102 and a target determination module 1103.

[0142] The data acquisition module 1101 is used to acquire the environmental point cloud data of the environment where the target vehicle is located, and determine the point cloud data corresponding to the reference vehicle according to the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle;

[0143] A candidate determination module 1102 is used to determine a candidate space parking space in the environment where the target vehicle is located according to the point cloud data corresponding to the reference vehicle; wherein the candidate space parking space is a space parking space of a preset size;

[0144] The target determination module 1103 is used to adjust the position of the candidate space parking space to obtain a target space parking space; wherein the target space parking space is used to indicate the target vehicle to park.

[0145] Fig.12 A structural block diagram of a space parking space detection device provided by an embodiment of the present disclosure, such as Fig.12 As shown, the spatial parking space detection device 1200 includes a data acquisition module 1201 , a candidate determination module 1202 and a target determination module 1203 , wherein the candidate determination module 1202 includes a coverage determination unit 12021 and a candidate determination unit 12022 .

[0146] A coverage determination unit 12021 is used to determine the coverage of the reference vehicle according to the point cloud data corresponding to the reference vehicle; wherein the coverage represents the proportion of the area of ​​the reference vehicle that is not blocked by obstacles;

[0147] The candidate determination unit 12022 is used to determine the candidate space parking space in the environment where the target vehicle is located according to the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle.

[0148] In one example, the data acquisition module 1201 includes:

[0149] A vehicle determination unit, configured to determine a reference vehicle located in the environment where the target vehicle is located according to the environmental point cloud data;

[0150] A data determination unit is used to extract point cloud data corresponding to the reference vehicle from the environmental point cloud data.

[0151] In one example, the vehicle determination unit is specifically configured to:

[0152] Inputting the environmental point cloud data into a preset deep learning model to obtain attribute information of objects in the environment where the target vehicle is located; wherein the preset deep learning model is used to perform semantic segmentation processing on the environmental point cloud data and output attribute information of the object, wherein the attribute information includes type information and location information;

[0153] The reference vehicle is determined from the objects in the environment where the target vehicle is located according to the attribute information of the objects in the environment where the target vehicle is located.

[0154] In one example, the vehicle determination unit is specifically configured to:

[0155] Determining the location information of the target vehicle;

[0156] Determine the distance between the target vehicle and the object according to the position information of the target vehicle and the position information of the object in the environment where the target vehicle is located;

[0157] If the distance value is less than a preset first distance threshold, and the type information of the object is a preset vehicle type, it is determined that the object is the reference vehicle.

[0158] In one example, the data determination unit is specifically used to:

[0159] According to a preset target frame extension range, a target frame selection process is performed on the reference vehicle in the environmental point cloud data;

[0160] The point cloud data in the target frame is determined as the point cloud data corresponding to the reference vehicle.

[0161] In one example, the data determination unit is specifically used to:

[0162] According to the position information and height information of the environmental point cloud data, the environmental point cloud data is screened to obtain the environmental point cloud data after the screening process;

[0163] The point cloud data corresponding to the reference vehicle is extracted from the filtered environment point cloud data.

[0164] In one example, the target box is a rectangular box; the coverage determination unit 12021 is specifically used to:

[0165] Determining a target edge of the target frame of the reference vehicle according to a distance between an edge of the target frame of the reference vehicle and the target vehicle;

[0166] Projecting the point cloud data corresponding to the reference vehicle onto the target edge to obtain a projection point of the point cloud data corresponding to the reference vehicle;

[0167] The sum of distances between adjacent projection points is determined, and the coverage rate of the reference vehicle is determined according to the length of the target edge and the sum of distances between the adjacent projection points.

[0168] In one example, it also includes:

[0169] The distance comparison module is used to determine that the distance between adjacent projection points is a preset value if the distance between adjacent projection points is greater than a preset second distance threshold.

[0170] In one example, the attribute information includes an orientation angle; the candidate determination unit 12022 is specifically configured to:

[0171] If the coverage rate of the reference vehicle is greater than a preset ratio threshold, the parking space type of the candidate space parking space is determined according to the orientation angle of the reference vehicle; wherein the parking space type is a horizontal type, a vertical type or a diagonal type;

[0172] Determine a target corner point from the point cloud data corresponding to the reference vehicle according to the position information of the point cloud data corresponding to the reference vehicle and the position information of the target vehicle; wherein the target corner point is a corner point of a candidate parking space;

[0173] The candidate space parking spaces in the environment where the target vehicle is located are determined according to the position information of the target corner point, the parking space type of the candidate space parking space, and the preset size of the candidate space parking space.

[0174] In one example, the target determination module 1203 includes:

[0175] An information acquisition unit, configured to acquire environmental information of the candidate parking space; wherein the environmental information includes surround view camera data and ultrasonic sensor data;

[0176] The target determination unit is used to adjust the position of the candidate space parking space according to the environmental information to obtain the target space parking space.

[0177] In one example, the target determination unit is specifically used to:

[0178] If it is determined that there is an obstacle at the candidate space parking space according to the surround view camera data and the ultrasonic sensor data, the position of the candidate space parking space is adjusted to obtain the target space parking space.

[0179] In one example, the parking space type of the candidate parking space is a vertical type or an oblique type; the device further includes:

[0180] an obstacle determination module, configured to determine the position information and height information of the obstacle if it is determined that there is an obstacle at the candidate parking space according to the surround view camera data and the ultrasonic sensor data;

[0181] The parking space determination module is used to determine that the candidate space parking space is the target space parking space if the position information of the obstacle is within a preset area of ​​the candidate space parking space and the height information of the obstacle is less than a preset height threshold.

[0182] Fig.13 A structural block diagram of an electronic device provided in an embodiment of the present application, such as Fig.13 As shown, the electronic device includes: a memory 131 and a processor 132; the memory 131 is a memory for storing instructions executable by the processor 132.

[0183] The processor 132 is configured to execute the method provided in the above embodiment.

[0184] The electronic device further includes a receiver 133 and a transmitter 134. The receiver 133 is used to receive instructions and data sent by other devices, and the transmitter 134 is used to send instructions and data to external devices.

[0185] Fig.14 It is a block diagram of a terminal device according to an exemplary embodiment. The device may be a mobile phone, a computer, a digital broadcast terminal, a message transceiver, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

[0186] Device 1400 may include one or more of the following components: a processing component 1402 , a memory 1404 , a power component 1406 , a multimedia component 1408 , an audio component 1410 , an input / output (I / O) interface 1412 , a sensor component 1414 , and a communication component 1416 .

[0187] The processing component 1402 generally controls the overall operation of the device 1400, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 1402 may include one or more processors 1420 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 1402 may include one or more modules to facilitate the interaction between the processing component 1402 and other components. For example, the processing component 1402 may include a multimedia module to facilitate the interaction between the multimedia component 1408 and the processing component 1402.

[0188] Device 1400 may include one or more of the following components: a processing component 1402 , a memory 1404 , a power component 1406 , a multimedia component 1408 , an audio component 1410 , an input / output (I / O) interface 1412 , a sensor component 1414 , and a communication component 1416 .

[0189] The memory 1404 is configured to store various types of data to support operations on the device 1400. Examples of such data include instructions for any application or method operating on the device 1400, contact data, phone book data, messages, pictures, videos, etc. The memory 1404 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0190] The power supply component 1406 provides power to the various components of the device 1400. The power supply component 1406 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 1400.

[0191] The multimedia component 1408 includes a screen that provides an output interface between the device 1400 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 1408 includes a front camera and / or a rear camera. When the device 1400 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0192] The audio component 1410 is configured to output and / or input audio signals. For example, the audio component 1410 includes a microphone (MIC), and when the device 1400 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 1404 or sent via the communication component 1416. In some embodiments, the audio component 1410 also includes a speaker for outputting audio signals.

[0193] I / O interface 1412 provides an interface between processing component 1402 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0194] The sensor assembly 1414 includes one or more sensors for providing various aspects of the status assessment of the device 1400. For example, the sensor assembly 1414 can detect the open / closed state of the device 1400, the relative positioning of components, such as the display and keypad of the device 1400, the sensor assembly 1414 can also detect the position change of the device 1400 or a component of the device 1400, the presence or absence of user contact with the device 1400, the orientation or acceleration / deceleration of the device 1400, and the temperature change of the device 1400. The sensor assembly 1414 can include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 1414 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 1414 can also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0195] The communication component 1416 is configured to facilitate wired or wireless communication between the device 1400 and other devices. The device 1400 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 1416 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1416 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0196] In an exemplary embodiment, the device 1400 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above methods.

[0197] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 1404 including instructions, and the instructions can be executed by a processor 1420 of the device 1400 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0198] A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a terminal device, enables the terminal device to execute the spatial parking space detection method of the terminal device.

[0199] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0200] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for detecting a parking space, characterized in that: include: Acquire environmental point cloud data of the environment where the target vehicle is located, and determine point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle; Determine, according to the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located; wherein the candidate space parking space is a space parking space of a preset size; The candidate parking spaces are positioned to obtain a target parking space, wherein the target parking space is used to indicate a target vehicle to park.

2. The method according to claim 1, characterized in that: Determining point cloud data corresponding to a reference vehicle according to the environmental point cloud data includes: Determining a reference vehicle located in the environment where the target vehicle is located according to the environmental point cloud data; The point cloud data corresponding to the reference vehicle is extracted from the environmental point cloud data.

3. The method according to claim 2, characterized in that Determining a reference vehicle located in the environment where the target vehicle is located according to the environmental point cloud data includes: Inputting the environmental point cloud data into a preset deep learning model to obtain attribute information of objects in the environment where the target vehicle is located; wherein the preset deep learning model is used to perform semantic segmentation processing on the environmental point cloud data and output attribute information of the object, wherein the attribute information includes type information and location information; The reference vehicle is determined from the objects in the environment where the target vehicle is located according to the attribute information of the objects in the environment where the target vehicle is located.

4. The method according to claim 3, characterized in that Determining the reference vehicle from the objects in the environment where the target vehicle is located according to the attribute information of the objects in the environment where the target vehicle is located includes: Determining the location information of the target vehicle; Determine the distance between the target vehicle and the object according to the position information of the target vehicle and the position information of the object in the environment where the target vehicle is located; If the distance value is less than a preset first distance threshold, and the type information of the object is a preset vehicle type, it is determined that the object is the reference vehicle.

5. The method according to claim 4, characterized in that Extracting point cloud data corresponding to the reference vehicle from the environmental point cloud data includes: According to a preset target frame extension range, a target frame selection process is performed on the reference vehicle in the environmental point cloud data; The point cloud data in the target frame is determined as the point cloud data corresponding to the reference vehicle.

6. The method according to claim 5, characterized in that Extracting point cloud data corresponding to the reference vehicle from the environmental point cloud data includes: According to the position information and height information of the environmental point cloud data, the environmental point cloud data is screened to obtain the environmental point cloud data after the screening process; The point cloud data corresponding to the reference vehicle is extracted from the filtered environment point cloud data.

7. The method according to claim 5, characterized in that Determining a candidate parking space in the environment where the target vehicle is located according to the point cloud data corresponding to the reference vehicle includes: Determine the coverage rate of the reference vehicle according to the point cloud data corresponding to the reference vehicle; wherein the coverage rate represents the proportion of the area of ​​the reference vehicle that is not blocked by obstacles; According to the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle, a candidate space parking space in the environment where the target vehicle is located is determined.

8. The method according to claim 7, characterized in that The target frame is a rectangular frame; and determining the coverage rate of the reference vehicle according to the point cloud data corresponding to the reference vehicle includes: Determining a target edge of the target frame of the reference vehicle according to a distance between an edge of the target frame of the reference vehicle and the target vehicle; Projecting the point cloud data corresponding to the reference vehicle onto the target edge to obtain a projection point of the point cloud data corresponding to the reference vehicle; The sum of distances between adjacent projection points is determined, and the coverage rate of the reference vehicle is determined according to the length of the target side and the sum of distances between the adjacent projection points.

9. The method according to claim 8, characterized in that Also includes: If the distance between adjacent projection points is greater than a preset second distance threshold, the distance between the adjacent projection points is determined to be a preset value.

10. The method according to claim 7, characterized in that The attribute information includes a heading angle; determining a candidate space parking space in the environment where the target vehicle is located according to the coverage rate of the reference vehicle and the point cloud data corresponding to the reference vehicle, including: If the coverage rate of the reference vehicle is greater than a preset ratio threshold, the parking space type of the candidate space parking space is determined according to the orientation angle of the reference vehicle; wherein the parking space type is a horizontal type, a vertical type or a diagonal type; Determine a target corner point from the point cloud data corresponding to the reference vehicle according to the position information of the point cloud data corresponding to the reference vehicle and the position information of the target vehicle; wherein the target corner point is a corner point of a candidate parking space; The candidate space parking spaces in the environment where the target vehicle is located are determined according to the position information of the target corner point, the parking space type of the candidate space parking space, and the preset size of the candidate space parking space.

11. The method according to any one of claims 1 to 10, characterized in that Adjusting the position of the candidate parking space to obtain the target parking space includes: Acquire environmental information of the candidate parking space; wherein the environmental information includes surround view camera data and ultrasonic sensor data; According to the environmental information, the position of the candidate space parking space is adjusted to obtain the target space parking space.

12. The method according to claim 11, characterized in that According to the environmental information, the position of the candidate parking space is adjusted to obtain the target parking space, including: If it is determined that there is an obstacle at the candidate space parking space according to the surround view camera data and the ultrasonic sensor data, the position of the candidate space parking space is adjusted to obtain the target space parking space.

13. The method according to claim 12, characterized in that The parking space type of the candidate space parking space is a vertical type or an oblique type; the method further includes: If it is determined that there is an obstacle at the candidate parking space according to the surround view camera data and the ultrasonic sensor data, then determining the position information and height information of the obstacle; If the position information of the obstacle is located within a preset area range in the candidate space parking space, and the height information of the obstacle is less than a preset height threshold, the candidate space parking space is determined to be the target space parking space.

14. A device for detecting a parking space, characterized in that: include: A data acquisition module, used to acquire environmental point cloud data of the environment where the target vehicle is located, and determine point cloud data corresponding to a reference vehicle based on the environmental point cloud data; wherein the reference vehicle is a vehicle in the environment other than the target vehicle; A candidate determination module, used to determine a candidate space parking space in the environment where the target vehicle is located according to the point cloud data corresponding to the reference vehicle; wherein the candidate space parking space is a space parking space of a preset size; The target determination module is used to adjust the position of the candidate space parking space to obtain a target space parking space; wherein the target space parking space is used to indicate the target vehicle to park.

15. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 13 when executed by a processor.

17. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 13 when being executed by a processor.

Images