Parking space identification method and device, electronic equipment and storage medium

Abstract

The invention relates to a parking space identification method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining environment information collected by a plurality of sensors, and carrying out the data fusion of the environment information collected by the plurality of sensors, and obtaining a plurality of environment parameters; fuzzy processing is carried out on the depth of the to-be-recognized parking space, the width of the to-be-recognized parking space, the first obstacle attitude angle and the second obstacle attitude angle, and fuzzy subsets corresponding to the depth of the to-be-recognized parking space, the width of the to-be-recognized parking space, the first obstacle attitude angle and the second obstacle attitude angle are obtained; matching a target fuzzy rule satisfying fuzzy subsets corresponding to the depth of the to-be-identified parking space, the width of the to-be-identified parking space, the first obstacle attitude angle and the second obstacle attitude angle from a preset fuzzy rule base; and determining the parking space type of the to-be-identified parking space according to the target fuzzy rule. Therefore, the accuracy of parking space identification can be improved, and even narrow parking spaces can be accurately identified, so that the utilization rate of the parking spaces is improved.

Description

Technical Field

[0001] The present application relates to the technical field of automatic parking of vehicles, and in particular to a parking space recognition method, device, electronic device and storage medium. Background Art

[0002] ‌With the rapid growth of car ownership, the imbalance between urban land capacity and parking space supply has become increasingly prominent. In order to meet more parking needs, limited urban land resources have made parking space construction increasingly narrow.

[0003] When a vehicle is parking automatically, it must first identify the parking space. However, as parking spaces become increasingly narrow, a single sensor (such as ultrasonic radar or camera) cannot accurately identify narrow parking spaces, resulting in wasted available parking spaces. Therefore, how to improve the accuracy of parking space recognition has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present application provides a parking space recognition method, device, electronic device and storage medium to solve the problem in the prior art that a single sensor is not accurate enough in recognizing narrow parking spaces.

[0005] In a first aspect, the present application provides a parking space recognition method, the method comprising: Acquire environmental information collected by multiple sensors, and perform data fusion on the environmental information collected by the multiple sensors to obtain multiple environmental parameters, wherein the multiple environmental parameters include a depth of a parking space to be identified, a width of the parking space to be identified, a first obstacle attitude angle on the left side of the parking space to be identified, and a second obstacle attitude angle on the right side of the parking space to be identified; Performing fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle, respectively, to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle; Matching a target fuzzy rule that satisfies a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle from a preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to a plurality of parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle; According to the target fuzzy rule, the parking space type of the parking space to be identified is determined.

[0006] Optionally, the fuzzy processing is performed on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle, including: Acquire a pre-established first membership function, a second membership function, a third membership function, and a fourth membership function, wherein the first membership function is used to characterize the membership function of the parking space depth, the first membership function includes a plurality of first fuzzy subsets, and the parking space depth value ranges corresponding to different first fuzzy subsets partially overlap; the second membership function is used to characterize the membership function of the parking space width, the second membership function includes a plurality of second fuzzy subsets, and the parking space width value ranges corresponding to different second fuzzy subsets partially overlap; the third membership function is used to characterize the membership function of the first obstacle posture angle, the third membership function includes a plurality of third fuzzy subsets, and the first obstacle posture angle value ranges corresponding to different third fuzzy subsets partially overlap; the fourth membership function is used to characterize the membership function of the second obstacle posture angle, the fourth membership function includes a plurality of fourth fuzzy subsets, and the second obstacle posture angle value ranges corresponding to different fourth fuzzy subsets partially overlap; Performing fuzzy processing on the depth of the parking space to be identified by using the first membership function to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, wherein the fuzzy subset corresponding to the depth of the parking space to be identified belongs to any first fuzzy subset among the multiple first fuzzy subsets; Performing fuzzy processing on the width of the parking space to be identified by using the second membership function to obtain a fuzzy subset corresponding to the width of the parking space to be identified, wherein the fuzzy subset corresponding to the width of the parking space to be identified belongs to any second fuzzy subset among the multiple second fuzzy subsets; Performing fuzzy processing on the first obstacle attitude angle by using the third membership function to obtain a fuzzy subset corresponding to the first obstacle attitude angle, wherein the fuzzy subset corresponding to the first obstacle attitude angle belongs to any third fuzzy subset among the multiple third fuzzy subsets; The second obstacle attitude angle is fuzzified by using the fourth membership function to obtain a fuzzy subset corresponding to the second obstacle attitude angle, wherein the fuzzy subset corresponding to the second obstacle attitude angle belongs to any fourth fuzzy subset among the multiple fourth fuzzy subsets.

[0007] Optionally, the matching of a target fuzzy rule from a preset fuzzy rule library that satisfies a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle includes: Matching the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle with the fuzzy subset of the parking space depth, the fuzzy subset of the parking space width, the fuzzy subset of the first obstacle posture angle, and the fuzzy subset of the second obstacle posture angle in the preset fuzzy rule library respectively; According to the matching result, the target fuzzy rule is determined.

[0008] Optionally, the fuzzy rules in the preset fuzzy rule base include at least one of the following: When the fuzzy subset of the parking space depth is the smallest subset among the plurality of first fuzzy subsets, determining the parking space type as an unavailable parking space type; When the fuzzy subset of the parking space width is the smallest subset among the plurality of second fuzzy subsets, determining the parking space type as an unavailable parking space type; In the case where the fuzzy subset of the parking space depth is the middle subset of the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset of the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the first condition, then the parking space type is determined to be an unavailable parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the first condition, then the parking space type is determined to be a narrow parking space type, wherein the first condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset of the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the second condition, the parking space type is determined to be a regular parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the second condition, the parking space type is determined to be a narrow parking space type, wherein the second condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among the multiple fourth fuzzy subsets, or the second condition is that the fuzzy subset of the first obstacle attitude angle is the largest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the middle subset of the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset of the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the third condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the third condition, the parking space type is determined to be a regular parking space type, wherein the third condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the middle subset of the multiple fourth fuzzy subsets, or the third condition is that the fuzzy subset of the first obstacle attitude angle is the middle subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset of the multiple fourth fuzzy subsets; In the case that the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the fourth condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the fourth condition, the parking space type is determined to be a regular parking space type, wherein the fourth condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets.

[0009] Optionally, the multiple sensors include ultrasonic sensors, visual sensors and wheel speed sensors; The environmental information collected by the multiple sensors is subjected to data fusion to obtain multiple environmental parameters, including: Performing data fusion on the environmental information collected by the ultrasonic sensor, the visual sensor, and the wheel speed sensor to obtain obstacle parameter information, wherein the obstacle parameter information includes the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle posture information, and the second obstacle posture information; Performing data fusion on the environmental information collected by the ultrasonic sensor and the visual sensor to obtain the depth of the parking space to be identified; Calculating a width of the to-be-identified parking space based on the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle position information, and the second obstacle position information; The depth of the parking space to be identified, the width of the parking space to be identified and the obstacle attitude angle are determined as the multiple environmental parameters.

[0010] Optionally, the calculation formula of the width of the parking space to be identified is as follows: ; ; ; in, represents the width of the parking space to be identified, Indicates the minimum distance from the first obstacle to the second obstacle, Indicates the minimum distance from the second obstacle to the first obstacle, represents the first obstacle attitude angle, represents the attitude angle of the second obstacle, ( , ) represents the coordinates of the point B1 on the first obstacle that is closest to the second obstacle, ( , ) is determined based on the first obstacle position information, ( , ) represents the coordinates of the point C2 on the second obstacle that is closest to the first obstacle, ( , ) is determined based on the second obstacle posture information.

[0011] Optionally, determining the parking space type of the to-be-identified parking space according to the target fuzzy rule includes: Get the judgment threshold of user input; Determining the probability values ​​of the parking spaces to be identified being the types of various parking spaces according to the target fuzzy rule and the judgment threshold; The parking space type of the parking space to be identified is determined according to the probability value.

[0012] In a second aspect, the present application further provides a parking space recognition device, the device comprising: an acquisition and fusion module, configured to acquire environmental information collected by a plurality of sensors, and perform data fusion on the environmental information collected by the plurality of sensors to obtain a plurality of environmental parameters, wherein the plurality of environmental parameters include a depth of a parking space to be identified, a width of the parking space to be identified, a first obstacle attitude angle on the left side of the parking space to be identified, and a second obstacle attitude angle on the right side of the parking space to be identified; a fuzzy processing module, configured to perform fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle, respectively, to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle; a matching module, configured to match a target fuzzy rule satisfying a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle from a preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to a plurality of parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle; The determination module is used to determine the parking space type of the parking space to be identified according to the target fuzzy rule.

[0013] In a third aspect, the present application further provides an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the parking space recognition method described in the first aspect when executing the program stored in the memory.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the parking space recognition method described in the first aspect is implemented.

[0015] The above-mentioned technical scheme provided by the present application has the following advantages over the prior art: the method provided by the present application obtains environmental information collected by multiple sensors and performs data fusion on the environmental information collected by the multiple sensors to obtain multiple environmental parameters, wherein the multiple environmental parameters include the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle on the left side of the parking space to be identified, and the second obstacle attitude angle on the right side of the parking space to be identified; the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle are fuzzy processed respectively to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, The fuzzy subset corresponding to the first obstacle posture angle and the fuzzy subset corresponding to the second obstacle posture angle; match the target fuzzy rule that satisfies the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle and the fuzzy subset corresponding to the second obstacle posture angle from the preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to multiple parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle or fuzzy subsets of the second obstacle posture angle; according to the target fuzzy rule, determine the parking space type of the parking space to be identified. Through the above method, the environmental information collected by various sensors can be integrated to accurately obtain the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle and the second obstacle attitude angle, and then by fuzzifying and matching the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle and the second obstacle attitude angle, the parking space type of the parking space to be identified can be accurately identified, thereby improving the accuracy of parking space identification. Even narrow parking spaces can be accurately identified, thereby improving the utilization rate of parking spaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0019] Figure 1 A schematic diagram of a flow chart of a parking space identification method provided in an embodiment of the present application; Figure 2 A schematic diagram of a detection process of an ultrasonic radar provided in an embodiment of the present application; Figure 3 A schematic diagram of a camera detection process provided in an embodiment of the present application; Figure 4 A schematic diagram of the working principle of a wheel speed sensor provided in an embodiment of the present application; Figure 5 A simplified schematic diagram of a vehicle rear axle differential steering provided in an embodiment of the present application; Figure 6 A schematic diagram of a first membership function provided in an embodiment of the present application; Figure 7 A schematic diagram of a second membership function provided in an embodiment of the present application; Figure 8 A schematic diagram of a third membership function or a fourth membership function provided in an embodiment of the present application; Fig. 9 A schematic diagram of a property function of an output variable provided in an embodiment of the present application; Fig.10 Schematic diagram of four parking scenarios provided in the embodiments of the present application; Fig.11 Schematic diagrams of two typical irregular parking postures provided in the embodiments of the present application; Fig.12 A schematic diagram of a flow chart of another parking space identification method provided in an embodiment of the present application; Fig.13 A schematic diagram of a parking space model provided in an embodiment of the present application; Fig.14 A schematic diagram of a parking space fuzzy reasoning modeling simulation process provided in an embodiment of the present application; Fig.15 A schematic structural diagram of a field of view adjustment device for an electronic rearview mirror provided in an embodiment of the present application; Fig.16 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0021] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0022] In order to solve the problem in the prior art that a single sensor is not accurate enough in identifying narrow parking spaces, the present application provides a parking space identification method, device, electronic device and storage medium, which can improve the accuracy of parking space identification.

[0023] See also Figure 1 , Figure 1 A flow chart of a parking space identification method provided in an embodiment of the present application. Figure 1 As shown, the parking space recognition method may include the following steps: Step S101, obtaining environmental information collected by multiple sensors, and performing data fusion on the environmental information collected by the multiple sensors to obtain multiple environmental parameters, wherein the multiple environmental parameters include the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle on the left side of the parking space to be identified, and the second obstacle attitude angle on the right side of the parking space to be identified.

[0024] Specifically, the above-mentioned multiple sensors may include but are not limited to ultrasonic sensors, visual sensors, and wheel speed sensors. The above-mentioned multiple environmental parameters may include but are not limited to the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle on the left side of the parking space to be identified, and the second obstacle attitude angle on the right side of the parking space to be identified. The obstacles here may be objects such as vehicles parked next to the parking space to be identified, walls, and columns.

[0025] It should be noted that the principle of using ultrasonic sensors alone to obtain environmental information and environmental parameters is: ultrasonic sensors (i.e. ultrasonic radars) use the principle of sound wave reflection to detect obstacles in the surrounding environment. Their main components include transmitting sensors, sound wave receiving sensors, controllers, and power modules. When using ultrasonic technology for parking space perception and identification, by capturing the distance signal when two large jumps occur in succession, it is calculated whether the displacement of the vehicle between the two jump times meets the parking space width requirements to determine whether a parking space is detected. The detection process is as follows: Figure 2 As shown, the ultrasonic sensor can be installed above the wheel of the vehicle, about 75 cm from the ground. Figure 2 D1 represents the actual distance between the vehicle and the obstacle, such as 0.3 meters to 1.5 meters. D2 represents the maximum threshold measured by ultrasound, such as no more than 6 meters. W represents the width between obstacles. The following distance calculation formula can be obtained based on the time interval between transmitting and receiving ultrasound: ; ; in, is the distance between the ultrasonic radar and the obstacle, v is the propagation speed of ultrasonic waves in the air, t is the time interval between transmitting and receiving ultrasonic waves, and c is the medium temperature in the air.

[0026] When the vehicle is moving forward at a uniform low speed, assuming that the obstacle distance detected by the vehicle suddenly changes from D1 to D2, it is recorded as the moment of the first mutation. At the same time, the distance of D2 minus the distance of D1 is the depth of the parking space. When the vehicle continues to drive and the obstacle distance detected suddenly changes from D2 to D1, it is recorded as the moment of the second mutation. The parking space width can be detected based on the vehicle speed and the time of these two mutations. If the parking space-related thresholds (length and width) set by the system are met, the parking space detection can be deemed to be successful.

[0027] The principle of using visual sensors alone to obtain environmental information and environmental parameters is: by reading the parking space image information sent back by the visual sensor (such as a camera), using the relevant image processing algorithm to identify the four side lines or corner points of the parking space, and then determine the parking space information through the four side lines or corner points of the parking space. The specific detection process is as follows: Figure 3 As shown in the figure, firstly, the image uploaded by the camera is grayed, filtered and denoised to improve the clarity of the parking space line; secondly, the Canny edge detection algorithm is used to calibrate the contour of the parking space line and perform morphological filtering on it; then, the Hough straight line transform is used to locate and mark the corner points of the parking space; finally, the marking values ​​of the four corner points are used to achieve effective recognition of the parking space.

[0028] The principle of using wheel speed sensors alone to obtain environmental information and environmental parameters is that since some parking spaces are located several floors underground, GPS signals may be lost, so wheel speed sensors can be used to assist in positioning the vehicle. Figure 4 As shown in the figure, during the driving process, the wheel speed pulse signal output per unit time is analyzed, and the number of rising edges is counted, which is the number of circles the wheel rotates. From this, the formula for the distance the wheel travels can be obtained: ; Among them, n is the number of rising edges, r is the radius of the wheel, and N is the total number of teeth on the vehicle gear.

[0029] During parking, the vehicle speed is low. The center of the rear axle of the vehicle is used as the pose estimation point, and the differential steering motion of the rear axle of the vehicle is simplified as follows: Figure 5 As shown by Figure 5 The geometric relationship can be obtained: ; in, and are the rotation distances of the left and right wheels on the rear axle of the vehicle, is the rotation distance of the rear axle center, r is the rotation radius of the wheel closer to the parking space, B is the wheelbase of the vehicle, is the turning angle in rad.

[0030] Assuming that the vehicle runs steadily at a low speed for a short period of time, ignoring the effect of acceleration on it, and taking the parking starting point as the coordinate origin, the following relationship can be obtained in unit time: ; Correspondingly, the speed of the center of the vehicle's rear axle depends on the overall speed of the vehicle, that is, the average speed of the left and right wheels. , so it can be seen that the motion differential equation of the rear axle center position is: ; Assume that the original coordinates of the center of the rear axle of the vehicle are ( , ), and the original turning angle of the vehicle's rear axle center is , the estimation formula of the center coordinates can be derived by integration: ; By calculating through the above formula, the vehicle can be relatively positioned in real time. At the same time, the wheel speed sensor is combined with the ultrasonic radar. Within the time when the ultrasonic sensor undergoes two sudden changes, the width of the parking space can also be calculated based on the number of circles the vehicle has turned, thereby realizing multi-source data collection and detection.

[0031] Step S102: Fuzzy processing is performed on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle posture angle, and the second obstacle posture angle to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle.

[0032] Specifically, the fuzzification process refers to using the corresponding membership function to fuzzify the precise value into a fuzzy subset. The type of the membership function can be trapezoidal, semi-trapezoidal, parabolic, semi-parabolic, etc., which is not specifically limited in the embodiment of the present application.

[0033] In this step, the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle posture angle and the second obstacle posture angle can be fuzzy processed respectively to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle and a fuzzy subset corresponding to the second obstacle posture angle.

[0034] Step S103, matching target fuzzy rules that satisfy the fuzzy subsets corresponding to the depth of the parking space to be identified, the fuzzy subsets corresponding to the width of the parking space to be identified, the fuzzy subsets corresponding to the first obstacle posture angle, and the fuzzy subsets corresponding to the second obstacle posture angle from a preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to multiple parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle.

[0035] Specifically, the preset fuzzy rule library may include fuzzy rules corresponding to multiple parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of first obstacle attitude angles, or fuzzy subsets of second obstacle attitude angles, that is, different fuzzy rules can characterize the correspondence between different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of first obstacle attitude angles, fuzzy subsets of second obstacle attitude angles and parking space types. The target fuzzy rule is a fuzzy rule in the preset fuzzy rule library that satisfies the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle attitude angle, and the fuzzy subset corresponding to the second obstacle attitude angle.

[0036] Step S104: Determine the parking space type of the parking space to be identified according to the target fuzzy rule.

[0037] After the target fuzzy rule is determined, the parking space type of the parking space to be identified can be determined according to the target fuzzy rule. The parking space type here can be an unavailable parking space type, a narrow parking space type, a regular parking space type, etc.

[0038] In this embodiment, the environmental information collected by multiple sensors can be integrated to accurately obtain the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle and the second obstacle attitude angle, and then by fuzzifying and matching the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle and the second obstacle attitude angle, the parking space type of the parking space to be identified can be accurately identified, thereby improving the accuracy of parking space identification. Even narrow parking spaces can be accurately identified, thereby improving the utilization rate of parking spaces.

[0039] Furthermore, in step S102, the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle and the second obstacle attitude angle are fuzzy processed respectively to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle and a fuzzy subset corresponding to the second obstacle attitude angle, including: Acquire a pre-established first membership function, a second membership function, a third membership function, and a fourth membership function, wherein the first membership function is used to characterize the membership function of the parking space depth, the first membership function includes a plurality of first fuzzy subsets, and the parking space depth value ranges corresponding to different first fuzzy subsets partially overlap; the second membership function is used to characterize the membership function of the parking space width, the second membership function includes a plurality of second fuzzy subsets, and the parking space width value ranges corresponding to different second fuzzy subsets partially overlap; the third membership function is used to characterize the membership function of the first obstacle posture angle, the third membership function includes a plurality of third fuzzy subsets, and the first obstacle posture angle value ranges corresponding to different third fuzzy subsets partially overlap; the fourth membership function is used to characterize the membership function of the second obstacle posture angle, the fourth membership function includes a plurality of fourth fuzzy subsets, and the second obstacle posture angle value ranges corresponding to different fourth fuzzy subsets partially overlap; Using the first membership function to perform fuzzy processing on the depth of the parking space to be identified, a fuzzy subset corresponding to the depth of the parking space to be identified is obtained, wherein the fuzzy subset corresponding to the depth of the parking space to be identified belongs to any first fuzzy subset among the multiple first fuzzy subsets; The width of the parking space to be identified is fuzzified by using the second membership function to obtain a fuzzy subset corresponding to the width of the parking space to be identified, wherein the fuzzy subset corresponding to the width of the parking space to be identified belongs to any second fuzzy subset among the multiple second fuzzy subsets; Performing fuzzy processing on the first obstacle attitude angle by using a third membership function to obtain a fuzzy subset corresponding to the first obstacle attitude angle, wherein the fuzzy subset corresponding to the first obstacle attitude angle belongs to any third fuzzy subset among the plurality of third fuzzy subsets; The second obstacle attitude angle is fuzzified by using the fourth membership function to obtain a fuzzy subset corresponding to the second obstacle attitude angle, wherein the fuzzy subset corresponding to the second obstacle attitude angle belongs to any fourth fuzzy subset among the plurality of fourth fuzzy subsets.

[0040] It should be noted that according to common sense and prior knowledge, for the parking space depth L_D, when it is less than L2, it means that the parking space depth is too small; when it is greater than L1 and less than L4, it means that the parking space depth is medium; when it is greater than L3 and less than L5, it means that the parking space depth is too large. Therefore, the fuzzy subset of the parking space depth can be set to 3 levels: {DS, DM, DL}. DS represents a small depth, in which case it is likely to be an unavailable parking space; DM represents a medium depth, in which case it is likely to be a narrow parking space; DL represents a large depth, in which case it is likely to be a regular parking space. Its membership function (i.e., the first membership function) is as follows: Figure 6 It should be noted that the values ​​of L1, L2, L3, L4 and L5 can be set according to actual needs and are not limited here. As an optional implementation, L1 can represent the minimum width and length required for a vertical parking space for a vehicle, L2-L4 can represent the width and length of a narrow vertical parking space, and L3-L5 can represent the width and length of a regular vertical parking space.

[0041] Similarly, for the parking space width L_W, when it is less than L2, it means that the parking space width is too small; when it is greater than L1 and less than L4, it means that the parking space width is medium; when it is greater than L3 and less than L5, it means that the parking space width is too large. Therefore, the fuzzy subset of the parking space width can be set to 3 levels: {WS, WM, WL}. WS represents a small width, which is likely to be an unavailable parking space; WM represents a medium width, which is likely to be a narrow parking space; WL represents a large width, which is likely to be a regular parking space. Its membership function (i.e., the second membership function) is as follows: Figure 7 It should be noted that the values ​​of L1, L2, L3, L4 and L5 can be set according to actual needs and are not limited here.

[0042] For the irregular body posture angles α1 and α2 of the two cars next to the parking space, the fuzzy subsets are divided into three levels according to the deflection angle: {PS, PM, PL}, which represent small angles, medium angles, and large angles, respectively, and the domain is [0,π]. Its membership function (i.e., the third membership function and the fourth membership function) is as follows: Figure 8 shown.

[0043] In this way, the above membership function can be used to fuzzify the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle posture angle and the second obstacle posture angle, respectively, to obtain the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle and the fuzzy subset corresponding to the second obstacle posture angle. The output variable can be divided into 3 levels of fuzzy subsets according to the type of parking space: {unavailable parking space, narrow parking space, regular parking space}, and the domain is [0, +∞]. Its membership function is as follows Fig. 9 shown.

[0044] In this embodiment, the first membership function, the second membership function, the third membership function and the fourth membership function can be used to perform fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle posture angle and the second obstacle posture angle, respectively, so as to accurately obtain the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle and the fuzzy subset corresponding to the second obstacle posture angle.

[0045] Furthermore, the above step S103, matching a target fuzzy rule from a preset fuzzy rule library to satisfy the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle, comprises: The fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle are matched one by one with the fuzzy subset of the parking space depth, the fuzzy subset of the parking space width, the fuzzy subset of the first obstacle posture angle, and the fuzzy subset of the second obstacle posture angle in the preset fuzzy rule library respectively; According to the matching results, the target fuzzy rules are determined.

[0046] It should be noted that the core of fuzzy reasoning is fuzzy rules, and the set fuzzy rules are if(L_W) i and(L_D) i and(α1) i and(α2) i then(Type) i , i=1, 2, 3…M, M is the total number of fuzzy rules. After the fuzzy system is established, 81 fuzzy rules can be obtained according to the membership function of input and output and human experience, as shown in Table 1.

[0047]

[0048] Table 1 According to the fuzzy rules in Table 1, the fuzzy subsets corresponding to the depth of the parking space to be identified, the fuzzy subsets corresponding to the width of the parking space to be identified, the fuzzy subsets corresponding to the first obstacle posture angle, and the fuzzy subsets corresponding to the second obstacle posture angle can be matched one by one with the fuzzy subsets of the parking space depth, the fuzzy subsets of the parking space width, the fuzzy subsets of the first obstacle posture angle, and the fuzzy subsets of the second obstacle posture angle in each fuzzy rule in Table 1, so that the target fuzzy rules can be accurately determined, which is convenient for the subsequent accurate determination of the parking space type of the parking space to be identified according to the target fuzzy rules.

[0049] Furthermore, the fuzzy rules in the preset fuzzy rule base include at least one of the following: When the fuzzy subset of the parking space depth is the smallest subset among the multiple first fuzzy subsets, determining the parking space type as an unavailable parking space type; When the fuzzy subset of the parking space width is the smallest subset among the multiple second fuzzy subsets, determining the parking space type as an unavailable parking space type; In the case where the fuzzy subset of the parking space depth is an intermediate subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is an intermediate subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the first condition, then the parking space type is determined to be an unavailable parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the first condition, then the parking space type is determined to be a narrow parking space type, wherein the first condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, the fuzzy subset of the parking space width is the middle subset among the multiple second fuzzy subsets, the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among the multiple fourth fuzzy subsets, the parking space type is determined to be a regular parking space type; In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the second condition, then the parking space type is determined to be a regular parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the second condition, then the parking space type is determined to be a narrow parking space type, wherein the second condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among the multiple fourth fuzzy subsets, or the second condition is that the fuzzy subset of the first obstacle attitude angle is the largest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is an intermediate subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is a maximum subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the third condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the third condition, the parking space type is determined to be a regular parking space type, wherein the third condition is that the fuzzy subset of the first obstacle attitude angle is a minimum subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is an intermediate subset among the multiple fourth fuzzy subsets, or the third condition is that the fuzzy subset of the first obstacle attitude angle is an intermediate subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is a maximum subset among the multiple fourth fuzzy subsets; In the case that the fuzzy subset of the parking space depth is the largest subset among multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the fourth condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the fourth condition, the parking space type is determined to be a regular parking space type, wherein the fourth condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among multiple fourth fuzzy subsets.

[0050] It should be noted that the standard model is established for the common vertical parking space scenarios in the city, which can be divided into the following four parking space scenarios: Fig.10 As shown in the figure, obstacles refer to nearby parked vehicles, walls, columns, etc. Fig.10Among the four parking space scenarios shown, the parking space scenarios shown in (a), (b) and (c) can easily collect data information about the horizontal width and vertical depth of the parking space using relevant sensors because the obstacles are regular. In the parking space scenario shown in (d), the parking space is difficult to detect because the vehicles on the left and right are parked irregularly, so further analysis is needed. There are two types of irregular parking postures of vehicles next to the parking space, such as Fig.11 shown.

[0051] Assume that the four vertices of the irregular vehicle are A(X A ,Y A )、B(X B ,Y B )、C(X C ,Y C )、D(X D ,Y D ), where X A , X B , X C , X D is the corresponding lateral mileage value measured by the wheel speed sensor at point A, point B, point C, and point D, Y A , Y B , Y C , Y D is the longitudinal distance value detected by the ultrasonic sensor at point A, point B, point C, and point D. The angles between the vehicle body and the parallel line X-axis are α and β. To simplify the problem, the vehicle body posture information is described in mathematical language to facilitate the logical understanding and main parameter calculation of the perception decision system. When the parking space perception system perceives and detects the posture of irregular vehicles, the vehicle always moves along the X-axis direction, the detection direction is from left to right, the ultrasonic sensor detection direction is perpendicular to the X-axis, and the vehicle body posture angle is the angle between the vehicle body line and the horizontal line, ranging from 0 to π.

[0052] The calculation of the body posture angles α and β can be expressed as: ; Among them, step=1 means that the camera recognizes two wheel hubs at points B and D, and step=0 means that the camera recognizes two wheel hubs at points A and C.

[0053] It can be seen that the parking space recognition result is not only affected by the depth and width of the parking space, but also by the attitude angle of the vehicle next to the parking space. As an optional implementation, the fuzzy rules in the pre-established preset fuzzy rule library include at least one of the following fuzzy rules: When the fuzzy subset of the parking space depth is the smallest subset among the multiple first fuzzy subsets, the parking space type is determined to be an unavailable parking space type, that is, according to Table 1, if {L_D=DS}, then Result=not.

[0054] When the fuzzy subset of the parking space width is the smallest subset among the multiple second fuzzy subsets, the parking space type is determined to be an unavailable parking space type, that is, according to Table 1, if {L_W=WS}, then Result=not.

[0055] In the case where the fuzzy subset of the parking space depth is the middle subset of the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset of the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the first condition, the parking space type is determined to be an unavailable parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the first condition, the parking space type is determined to be a narrow parking space type, wherein the first condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset of the multiple fourth fuzzy subsets. That is, according to Table 1, if {L_W=WM, L_D=DM, α1=PS and α2=PL}, it means that the space at the front of the vehicle on both sides is too small to be considered as a parking space, then Result=not; otherwise, Result=narrow vertical.

[0056] In the case that the fuzzy subset of the parking space depth is the largest subset among multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset among multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the second condition, the parking space type is determined to be a regular parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the second condition, the parking space type is determined to be a narrow parking space type, wherein the second condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among multiple fourth fuzzy subsets, or the second condition is that the fuzzy subset of the first obstacle attitude angle is the largest subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among multiple fourth fuzzy subsets. That is, according to Table 1, if {L_W=WM, L_D=DL, α1=PS and α2=PS} or {L_W=WM, L_D=DL, α1=PL and α2=PL}, it means that the width of the parking space is a regular parking space, but because the left and right vehicles are tilted in the same phase, it can be regarded as a regular parallel parking space, then Result=routine vertical; otherwise, Result=narrow vertical.

[0057] In the case that the fuzzy subset of the parking space depth is the middle subset among multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the third condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the third condition, the parking space type is determined to be a regular parking space type, wherein the third condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the middle subset among multiple fourth fuzzy subsets, or the third condition is that the fuzzy subset of the first obstacle attitude angle is the middle subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among multiple fourth fuzzy subsets. That is, according to Table 1, if {L_W=WL, L_D=DM, α1=PS and α2=PM} or {L_W=WL, L_D=DM, α1=PM and α2=PL}, it means that the width and depth of one side of the vehicle are irregular, which is regarded as a narrow parking space, then Result= narrow vertical; otherwise, Result= regularvertical.

[0058] In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the fourth condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the fourth condition, the parking space type is determined to be a regular parking space type, wherein the fourth condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets. That is, according to Table 1, if {L_W=WL, L_D=DL, α1=PS and α2=PL}, it means that the parking space is limited due to irregular parking of the left and right vehicles, and is considered as a narrow parking space, then Result= narrow vertical; otherwise, Result= routine vertical.

[0059] Through the above fuzzy rules, different vertical parking space types can be determined according to different input parameters, which is consistent with the common sense experience of vertical parking spaces in daily life.

[0060] Furthermore, the above-mentioned multiple sensors include ultrasonic sensors, visual sensors and wheel speed sensors; In the above step S101, data fusion is performed on the environmental information collected by various sensors to obtain multiple environmental parameters, including: Performing data fusion on the environmental information collected by the ultrasonic sensor, the visual sensor and the wheel speed sensor to obtain obstacle parameter information, wherein the obstacle parameter information includes a first obstacle attitude angle, a second obstacle attitude angle, first obstacle position information and second obstacle position information; Perform data fusion on the environmental information collected by the ultrasonic sensor and the visual sensor to obtain the depth of the parking space to be identified; Calculate the width of the parking space to be identified based on the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle position information and the second obstacle position information; The depth of the parking space to be identified, the width of the parking space to be identified, and the obstacle posture angle are determined as multiple environmental parameters.

[0061] In one embodiment, when data fusion is performed on environmental information collected by multiple sensors, data fusion may be performed on environmental information collected by ultrasonic sensors, visual sensors, and wheel speed sensors to obtain obstacle parameter information, wherein the obstacle parameter information includes a first obstacle attitude angle, a second obstacle attitude angle, first obstacle posture information, and second obstacle posture information, and data fusion is performed on environmental information collected by ultrasonic sensors and visual sensors to obtain the depth of a parking space to be identified; then, based on the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle posture information, and the second obstacle posture information, the width of the parking space to be identified is calculated; finally, the depth of the parking space to be identified, the width of the parking space to be identified, and the obstacle attitude angle are determined as multiple environmental parameters, such as Fig.12 In this way, the four parameters collected from the parking space can be used as input, the parking experience of skilled drivers can be added as judgment rules, and accurate perception of the parking space can be finally achieved through reasoning and calculation.

[0062] Furthermore, the calculation formula of the width of the parking space to be identified is as follows: ; ; ; in, Indicates the width of the parking space to be identified. Indicates the minimum distance from the first obstacle to the second obstacle, Indicates the minimum distance from the second obstacle to the first obstacle, represents the first obstacle attitude angle, represents the attitude angle of the second obstacle, ( , ) represents the coordinates of the point B1 on the first obstacle that is closest to the second obstacle, ( , ) is determined based on the first obstacle position information, ( , ) represents the coordinates of the point C2 on the second obstacle that is closest to the first obstacle, ( , ) is determined based on the position information of the second obstacle.

[0063] In one embodiment, Fig.13 In the parking space model shown, the coordinates of the vertices A1, B1, C1, D1 and A2, B2, C2, D2 around the two cars next to the target parking space can be collected by combining visual feature point detection and ultrasonic jump data. Then the attitude angles α1 and α2 of the two cars are calculated using the parking attitude angle calculation formula. The shortest distance space from the left car to the right car is the vertical distance from the right vertex B1 to the straight line segment A2C2. The shortest space distance L_min can be calculated from the point-to-straight line distance formula as follows: ; Similarly, the shortest distance from the right car to the left car can be obtained: ; In order to select the minimum spatial distance more accurately, let the lateral distance input into the multi-information fusion perception system be L_distance, then: ; The units of L_distance and L_min are meters. When L_distance is less than or equal to zero, it means that there is overlap or no gap between the left and right vehicles. Through the above method, the width of the parking space to be identified can be accurately calculated for subsequent fuzzy rule matching, thereby improving the accuracy of fuzzy rule matching.

[0064] Furthermore, the above step S104, determining the parking space type of the parking space to be identified according to the target fuzzy rule, includes: Get the judgment threshold of user input; According to the target fuzzy rules and judgment threshold, the probability values ​​of the parking spaces to be identified are determined to be the types of parking spaces; According to the probability value, the parking space type of the parking space to be identified is determined.

[0065] Continue to see Fig.12After matching the fuzzy rule base based on the driver's experience benchmark, reasoning synthesis can be performed, and then the reasoning results can be defuzzified, and finally the parking space recognition results can be output. Specifically, the width, length and obstacle attitude angle of the parking space are collected through various sensors on the vehicle, and then vertical parking spaces of different sizes are identified through fuzzy reasoning. The relevant modules are established in Simulink in Matlab to realize the simulation and output of the fuzzy reasoning fusion algorithm. The parking space fuzzy reasoning modeling and simulation process is as follows Fig.14 As shown in the figure, a series of experimental result data are obtained through Simulink simulation. Since the simulation result is a definite value of the output, the size of the output value determines the type of the vertical parking space. Therefore, the intermediate threshold of the output result (such as 0.25) can be set, and the value exceeding the intermediate threshold can be determined as the corresponding parking space. It should be noted that the intermediate threshold is a demarcation point for judgment, and the result of fuzzy reasoning will calculate a value for unavailable parking spaces, narrow parking spaces, and regular parking spaces. For example: if the output result is [0.5 0.25 0.25], it means that this parking space is unavailable; if the output result is [0.25 0.6 0.25], it means that the parking space is a narrow parking space; if the output result is [0.25 0.25 0.7], it means that the parking space is a regular parking space, and the situation of [0.25 0.6 0.7] will also occur. At this time, the parking space is judged as a narrow parking space and a regular parking space at the same time. However, as long as the first output value is greater than 0.25, that is, [0.6 0.5 0.25], it is judged as an unavailable parking space.

[0066] The main problem that parking space perception and recognition solves is that when the car is looking for a parking space, it uses the environmental perception module carried by the vehicle to detect the depth data of the vehicle to determine whether the parking space meets the parking conditions. The environmental perception sensors currently carried by vehicles include ultrasonic radar, surround-view cameras, millimeter-wave radar, lidar, and infrared perception. This application is based on multiple (similar or heterogeneous) information sources, which are combined in space or time according to a specific standard to obtain a consistent interpretation or description of the object being measured, and to enable the information system to have better performance. Compared with a single sensor perception solution, the advantages of the parking space perception and recognition solution of this application are: 1. Improved environmental perception capabilities, provided redundancy in perception data, and better compensated for the shortcomings of a single sensor; 2. Modeling and in-depth analysis of various typical parking scenarios for vertical parking spaces were carried out. By integrating the acquisition of parking space parameters by ultrasonic radar and wheel speed sensors and the recognition of parking space feature points and obstacle feature points by visual sensors, high-precision recognition of parking space boundary information was achieved, which improved the accuracy of space detection and the credibility of data. 3. The fuzzy reasoning logic for vertical parking space recognition has been constructed, which can effectively identify narrow vertical parking spaces and regular vertical parking spaces based on the fusion perception of parking space size and attitude angle; 4. It can realize automatic parking detection for narrow vertical parking spaces, greatly improve the utilization rate of land resources, provide more parking spaces in the same parking lot area, and alleviate the current dilemma of "too many cars and few parking spaces".

[0067] See also Fig.15 , Fig.15 This is a schematic diagram of the structure of a parking space recognition device provided in an embodiment of the present application. Fig.15 As shown, the parking space recognition device 1500 includes: The acquisition and fusion module 1501 is used to acquire environmental information collected by multiple sensors, and perform data fusion on the environmental information collected by the multiple sensors to obtain multiple environmental parameters, wherein the multiple environmental parameters include the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle on the left side of the parking space to be identified, and the second obstacle attitude angle on the right side of the parking space to be identified; The fuzzy processing module 1502 is used to perform fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle posture angle, and the second obstacle posture angle, respectively, to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle; A matching module 1503 is used to match a target fuzzy rule from a preset fuzzy rule library that satisfies a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle, wherein the preset fuzzy rule library includes fuzzy rules corresponding to multiple parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle; The determination module 1504 is used to determine the parking space type of the parking space to be identified according to the target fuzzy rule.

[0068] Furthermore, the fuzzification processing module 1502 includes: The first acquisition submodule is used to acquire the pre-established first membership function, second membership function, third membership function and fourth membership function, wherein the first membership function is used to characterize the membership function of the parking space depth, the first membership function includes multiple first fuzzy subsets, and the parking space depth value ranges corresponding to different first fuzzy subsets are partially overlapped; the second membership function is used to characterize the membership function of the parking space width, the second membership function includes multiple second fuzzy subsets, and the parking space width value ranges corresponding to different second fuzzy subsets are partially overlapped; the third membership function is used to characterize the membership function of the first obstacle posture angle, the third membership function includes multiple third fuzzy subsets, and the first obstacle posture angle value ranges corresponding to different third fuzzy subsets are partially overlapped; the fourth membership function is used to characterize the membership function of the second obstacle posture angle, the fourth membership function includes multiple fourth fuzzy subsets, and the second obstacle posture angle value ranges corresponding to different fourth fuzzy subsets are partially overlapped; A first processing submodule is used to perform fuzzy processing on the depth of the parking space to be identified by using a first membership function to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, wherein the fuzzy subset corresponding to the depth of the parking space to be identified belongs to any first fuzzy subset among the multiple first fuzzy subsets; A second processing submodule is used to perform fuzzy processing on the width of the parking space to be identified by using a second membership function to obtain a fuzzy subset corresponding to the width of the parking space to be identified, wherein the fuzzy subset corresponding to the width of the parking space to be identified belongs to any second fuzzy subset among the multiple second fuzzy subsets; a third processing submodule, configured to perform fuzzy processing on the first obstacle attitude angle by using a third membership function to obtain a fuzzy subset corresponding to the first obstacle attitude angle, wherein the fuzzy subset corresponding to the first obstacle attitude angle belongs to any third fuzzy subset among the plurality of third fuzzy subsets; The fourth processing submodule is used to perform fuzzy processing on the second obstacle attitude angle by using a fourth membership function to obtain a fuzzy subset corresponding to the second obstacle attitude angle, wherein the fuzzy subset corresponding to the second obstacle attitude angle belongs to any fourth fuzzy subset among multiple fourth fuzzy subsets.

[0069] Furthermore, the matching module 1503 includes: a matching submodule, for matching the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle with the fuzzy subset of the parking space depth, the fuzzy subset of the parking space width, the fuzzy subset of the first obstacle posture angle, and the fuzzy subset of the second obstacle posture angle in a preset fuzzy rule library, respectively; The first determination submodule is used to determine the target fuzzy rule according to the matching result.

[0070] Furthermore, the fuzzy rules in the preset fuzzy rule base include at least one of the following: When the fuzzy subset of the parking space depth is the smallest subset among the multiple first fuzzy subsets, determining the parking space type as an unavailable parking space type; When the fuzzy subset of the parking space width is the smallest subset among the multiple second fuzzy subsets, determining the parking space type as an unavailable parking space type; In the case where the fuzzy subset of the parking space depth is an intermediate subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is an intermediate subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the first condition, then the parking space type is determined to be an unavailable parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the first condition, then the parking space type is determined to be a narrow parking space type, wherein the first condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the second condition, then the parking space type is determined to be a regular parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the second condition, then the parking space type is determined to be a narrow parking space type, wherein the second condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among the multiple fourth fuzzy subsets, or the second condition is that the fuzzy subset of the first obstacle attitude angle is the largest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is an intermediate subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is a maximum subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the third condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the third condition, the parking space type is determined to be a regular parking space type, wherein the third condition is that the fuzzy subset of the first obstacle attitude angle is a minimum subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is an intermediate subset among the multiple fourth fuzzy subsets, or the third condition is that the fuzzy subset of the first obstacle attitude angle is an intermediate subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is a maximum subset among the multiple fourth fuzzy subsets; In the case that the fuzzy subset of the parking space depth is the largest subset among multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the fourth condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the fourth condition, the parking space type is determined to be a regular parking space type, wherein the fourth condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among multiple fourth fuzzy subsets.

[0071] Furthermore, the multiple sensors include ultrasonic sensors, visual sensors and wheel speed sensors; the acquisition and fusion module 1501 includes: A first fusion submodule is used to perform data fusion on the environmental information collected by the ultrasonic sensor, the visual sensor and the wheel speed sensor to obtain obstacle parameter information, wherein the obstacle parameter information includes a first obstacle attitude angle, a second obstacle attitude angle, first obstacle posture information and second obstacle posture information; The second fusion submodule is used to fuse the environmental information collected by the ultrasonic sensor and the visual sensor to obtain the depth of the parking space to be identified; A calculation submodule, configured to calculate a width of a parking space to be identified based on the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle position information, and the second obstacle position information; The second determination submodule is used to determine the depth of the parking space to be identified, the width of the parking space to be identified and the obstacle posture angle as multiple environmental parameters.

[0072] Furthermore, the calculation formula of the width of the parking space to be identified is as follows: ; ; ; in, Indicates the width of the parking space to be identified. Indicates the minimum distance from the first obstacle to the second obstacle, Indicates the minimum distance from the second obstacle to the first obstacle, represents the first obstacle attitude angle, represents the attitude angle of the second obstacle, ( , ) represents the coordinates of the point B1 on the first obstacle that is closest to the second obstacle, ( , ) is determined based on the first obstacle position information, ( , ) represents the coordinates of the point C2 on the second obstacle that is closest to the first obstacle, ( , ) is determined based on the position information of the second obstacle.

[0073] Further, the determination module 1504 includes: The second acquisition submodule is used to obtain the judgment threshold input by the user; The third determination submodule is used to determine the probability values ​​of the parking spaces to be identified being the types of various parking spaces according to the target fuzzy rules and the judgment threshold; The fourth determination submodule is used to determine the parking space type of the to-be-identified parking space according to the probability value.

[0074] It should be noted that the parking space recognition device 1500 can implement the parking space recognition method provided by any of the aforementioned method embodiments and can achieve the same technical effects, which will not be described in detail here.

[0075] like Fig.16 As shown, the embodiment of the present application further provides an electronic device, including a processor 1611, a communication interface 1612, a memory 1613 and a communication bus 1614, wherein the processor 1611, the communication interface 1612, and the memory 1613 communicate with each other through the communication bus 1614. Memory 1613, used for storing computer programs; In one embodiment of the present application, the processor 1611 is used to implement the parking space recognition method provided by any one of the aforementioned method embodiments when executing the program stored in the memory 1613.

[0076] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the parking space recognition method provided by any one of the aforementioned method embodiments is implemented.

[0077] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0078] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0079] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0080] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A parking space recognition method, characterized in that: The method comprises: Acquire environmental information collected by multiple sensors, and perform data fusion on the environmental information collected by the multiple sensors to obtain multiple environmental parameters, wherein the multiple environmental parameters include a depth of a parking space to be identified, a width of the parking space to be identified, a first obstacle attitude angle on the left side of the parking space to be identified, and a second obstacle attitude angle on the right side of the parking space to be identified; Performing fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle, respectively, to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle; Matching a target fuzzy rule that satisfies a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle from a preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to a plurality of parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle; According to the target fuzzy rule, the parking space type of the parking space to be identified is determined.

2. The parking space recognition method according to claim 1, characterized in that: The fuzzy processing is performed on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle, including: Acquire a pre-established first membership function, a second membership function, a third membership function, and a fourth membership function, wherein the first membership function is used to characterize the membership function of the parking space depth, the first membership function includes a plurality of first fuzzy subsets, and the parking space depth value ranges corresponding to different first fuzzy subsets partially overlap; the second membership function is used to characterize the membership function of the parking space width, the second membership function includes a plurality of second fuzzy subsets, and the parking space width value ranges corresponding to different second fuzzy subsets partially overlap; the third membership function is used to characterize the membership function of the first obstacle posture angle, the third membership function includes a plurality of third fuzzy subsets, and the first obstacle posture angle value ranges corresponding to different third fuzzy subsets partially overlap; the fourth membership function is used to characterize the membership function of the second obstacle posture angle, the fourth membership function includes a plurality of fourth fuzzy subsets, and the second obstacle posture angle value ranges corresponding to different fourth fuzzy subsets partially overlap; Performing fuzzy processing on the depth of the parking space to be identified by using the first membership function to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, wherein the fuzzy subset corresponding to the depth of the parking space to be identified belongs to any first fuzzy subset among the multiple first fuzzy subsets; Performing fuzzy processing on the width of the parking space to be identified by using the second membership function to obtain a fuzzy subset corresponding to the width of the parking space to be identified, wherein the fuzzy subset corresponding to the width of the parking space to be identified belongs to any second fuzzy subset among the multiple second fuzzy subsets; Performing fuzzy processing on the first obstacle attitude angle by using the third membership function to obtain a fuzzy subset corresponding to the first obstacle attitude angle, wherein the fuzzy subset corresponding to the first obstacle attitude angle belongs to any third fuzzy subset among the multiple third fuzzy subsets; The second obstacle attitude angle is fuzzified by using the fourth membership function to obtain a fuzzy subset corresponding to the second obstacle attitude angle, wherein the fuzzy subset corresponding to the second obstacle attitude angle belongs to any fourth fuzzy subset among the multiple fourth fuzzy subsets.

3. The parking space recognition method according to claim 1, characterized in that: The target fuzzy rule that matches the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle from the preset fuzzy rule library includes: Matching the fuzzy subset corresponding to the depth of the parking space to be identified, the fuzzy subset corresponding to the width of the parking space to be identified, the fuzzy subset corresponding to the first obstacle posture angle, and the fuzzy subset corresponding to the second obstacle posture angle with the fuzzy subset of the parking space depth, the fuzzy subset of the parking space width, the fuzzy subset of the first obstacle posture angle, and the fuzzy subset of the second obstacle posture angle in the preset fuzzy rule library respectively; According to the matching result, the target fuzzy rule is determined.

4. The parking space recognition method according to claim 3, characterized in that: The fuzzy rules in the preset fuzzy rule base include at least one of the following: When the fuzzy subset of the parking space depth is the smallest subset among the plurality of first fuzzy subsets, determining the parking space type as an unavailable parking space type; When the fuzzy subset of the parking space width is the smallest subset among the plurality of second fuzzy subsets, determining the parking space type as an unavailable parking space type; In a case where the fuzzy subset of the parking space depth is a middle subset of the plurality of first fuzzy subsets, and the fuzzy subset of the parking space width is a middle subset of the plurality of second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet a first condition, then determining that the parking space type is an unavailable parking space type; If the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not satisfy the first condition, determining that the parking space type is a narrow parking space type, wherein the first condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the plurality of third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the plurality of fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the middle subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the second condition, the parking space type is determined to be a regular parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the second condition, the parking space type is determined to be a narrow parking space type, wherein the second condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the smallest subset among the multiple fourth fuzzy subsets, or the second condition is that the fuzzy subset of the first obstacle attitude angle is the largest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets; In the case where the fuzzy subset of the parking space depth is the middle subset of the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset of the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the third condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the third condition, the parking space type is determined to be a regular parking space type, wherein the third condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the middle subset of the multiple fourth fuzzy subsets, or the third condition is that the fuzzy subset of the first obstacle attitude angle is the middle subset of the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset of the multiple fourth fuzzy subsets; In the case that the fuzzy subset of the parking space depth is the largest subset among the multiple first fuzzy subsets, and the fuzzy subset of the parking space width is the largest subset among the multiple second fuzzy subsets, if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle meet the fourth condition, the parking space type is determined to be a narrow parking space type; if the fuzzy subset of the first obstacle attitude angle and the fuzzy subset of the second obstacle attitude angle do not meet the fourth condition, the parking space type is determined to be a regular parking space type, wherein the fourth condition is that the fuzzy subset of the first obstacle attitude angle is the smallest subset among the multiple third fuzzy subsets, and the fuzzy subset of the second obstacle attitude angle is the largest subset among the multiple fourth fuzzy subsets.

5. The parking space recognition method according to claim 1, characterized in that: The multiple sensors include ultrasonic sensors, visual sensors and wheel speed sensors; The environmental information collected by the multiple sensors is subjected to data fusion to obtain multiple environmental parameters, including: Performing data fusion on the environmental information collected by the ultrasonic sensor, the visual sensor, and the wheel speed sensor to obtain obstacle parameter information, wherein the obstacle parameter information includes the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle posture information, and the second obstacle posture information; Performing data fusion on the environmental information collected by the ultrasonic sensor and the visual sensor to obtain the depth of the parking space to be identified; Calculating a width of the to-be-identified parking space based on the first obstacle attitude angle, the second obstacle attitude angle, the first obstacle position information, and the second obstacle position information; The depth of the parking space to be identified, the width of the parking space to be identified and the obstacle attitude angle are determined as the multiple environmental parameters.

6. The parking space recognition method according to claim 5, characterized in that: The calculation formula of the width of the parking space to be identified is as follows: ； ； ； in, represents the width of the parking space to be identified, Indicates the minimum distance from the first obstacle to the second obstacle, Indicates the minimum distance from the second obstacle to the first obstacle, represents the first obstacle attitude angle, represents the attitude angle of the second obstacle, ( , ) represents the coordinates of the point B1 on the first obstacle that is closest to the second obstacle, ( , ) is determined based on the first obstacle position information, ( , ) represents the coordinates of the point C2 on the second obstacle that is closest to the first obstacle, ( , ) is determined based on the second obstacle posture information.

7. The parking space recognition method according to claim 1, characterized in that: Determining the parking space type of the to-be-identified parking space according to the target fuzzy rule includes: Get the judgment threshold of user input; Determining the probability values ​​of the parking spaces to be identified being the types of various parking spaces according to the target fuzzy rule and the judgment threshold; The parking space type of the parking space to be identified is determined according to the probability value.

8. A parking space recognition device, characterized in that: The device comprises: an acquisition and fusion module, configured to acquire environmental information collected by a plurality of sensors, and perform data fusion on the environmental information collected by the plurality of sensors to obtain a plurality of environmental parameters, wherein the plurality of environmental parameters include a depth of a parking space to be identified, a width of the parking space to be identified, a first obstacle attitude angle on the left side of the parking space to be identified, and a second obstacle attitude angle on the right side of the parking space to be identified; a fuzzy processing module, configured to perform fuzzy processing on the depth of the parking space to be identified, the width of the parking space to be identified, the first obstacle attitude angle, and the second obstacle attitude angle, respectively, to obtain a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle attitude angle, and a fuzzy subset corresponding to the second obstacle attitude angle; a matching module, configured to match a target fuzzy rule satisfying a fuzzy subset corresponding to the depth of the parking space to be identified, a fuzzy subset corresponding to the width of the parking space to be identified, a fuzzy subset corresponding to the first obstacle posture angle, and a fuzzy subset corresponding to the second obstacle posture angle from a preset fuzzy rule library, wherein the preset fuzzy rule library includes fuzzy rules corresponding to a plurality of parking space types, and different fuzzy rules are determined based on different fuzzy subsets of parking space depths, fuzzy subsets of parking space widths, fuzzy subsets of the first obstacle posture angle, or fuzzy subsets of the second obstacle posture angle; The determination module is used to determine the parking space type of the parking space to be identified according to the target fuzzy rule.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the parking space recognition method described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the parking space recognition method according to any one of claims 1 to 7 is implemented.

Images