A parking space error release method based on environmental information

Abstract

The present invention discloses a parking space error release method based on environmental information, and relates to the field of automatic parking technology. The present invention comprises the following steps: Step S1: Searching for parking spaces, and calculating the direct echo information of the ultrasonic sensor on the side of the vehicle in real time; Step S2: Judging whether a vertical parking space or a parallel parking space can be formed according to the current echo information, if a parking space exists, executing Step S3; if a parking space does not exist, continuing the search; Step S3: Filtering unreasonable parking space information through the parking space error release strategy; Step S4: Outputting the parking space information, and recording the current parking space type. The present invention distinguishes the types of vehicles that form the parking space through the different characteristics of ultrasonic echoes for vertical vehicles and parallel vehicles, and records the types of parking spaces that appeared within the previous driving distance, comprehensively judging the type of the parking space, improving and avoiding the safety risks of existing market technical solutions, and improving the subjective feeling of the driver's human-computer interaction.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic parking, and in particular relates to a method for preventing erroneous release of parking spaces based on environmental information, so as to solve the problem of erroneous release of parking spaces in a scenario where there are only empty parking spaces. Background Art

[0002] The increasingly complex parking environment and limited field of vision put forward higher requirements for drivers. With the development of unmanned driving technology, the intelligence of automobiles is constantly improving. Among them, autonomous parking technology can reduce the difficulty of drivers parking in narrow spaces, while also improving driver comfort and parking safety.

[0003] Automatic parking uses sensors to search the surrounding environment of the vehicle, find appropriate parking spaces or ground parking space markings (such as parking space lines, etc.) between other parked vehicles, and automatically or manually determine the target parking space according to the driver's choice, calculate the automatic parking trajectory, and send lateral and longitudinal motion control commands to guide the vehicle to park at the target parking position. Parking space detection and recognition is the premise of parking. Whether the parking space can be released correctly is an important factor affecting the driver's supervisory experience and is one of the key technologies of automatic parking.

[0004] There are two main methods for searching parking spaces. One is the pure ultrasonic method, which determines the size of parking spaces only through the echo information of ultrasonic sensors. The other is the fusion parking method, which uses the fisheye camera around the vehicle to obtain a 360-degree panoramic image, extracts visual parking space information by determining the corner features of the parking space line, and then fuses it with the spatial parking space information extracted by ultrasonic. In the case where there is no parking space line or the parking space line is blocked by obstacles, ultrasonic parking spaces are still the main method.

[0005] For the ultrasonic parking space solution, when detecting parking spaces, the vehicle drives horizontally towards the parking space at a certain speed: when the vehicle passes the parking position of car No. 1, the ultrasonic sensor installed on the side of the vehicle body begins to measure the lateral distance D between the vehicle and car No. 1. When the vehicle passes the upper edge of car No. 1, the value measured by the ultrasonic sensor will jump, and this moment will be recorded. When driving between car No. 1 and car No. 2, the depth D of the parking space is calculated. When passing the lower edge of car No. 2, the value measured by the ultrasonic sensor jumps again, and the current moment is recorded to calculate the final parking space length L. Analyze the measured parking space length L and depth D to determine whether the parking space meets the basic parking requirements and determine the parking space type. The parking space search process based on ultrasound is attached to the instruction manual. Figure 2 shown.

[0006] However, in the prior art, for ultrasonic parking solutions, the existence of parallel parking spaces or perpendicular parking spaces is usually determined by judging whether the length and depth of the parking space meet the minimum value of parallel parking spaces or the minimum value of perpendicular parking spaces. For the parking scene in the above figure, because the village is on the side of the curb or wall, only one parallel parking space is released. However, if there is no obstacle on the side, parallel parking spaces and perpendicular parking spaces will be released, as shown in the attached manual. Figure 3 But for the driver, he only wants this parking space to release the parallel parking space.

[0007] In addition, underground garages often have vertical parking spaces, and many parking spaces are parked side by side, which results in the ultrasonic solution releasing one vertical parking space and one parallel parking space. Figure 3 Although some solutions determine the parking space type by judging the length of the obstacle vehicles on the left and right sides of the parking space, thereby filtering out a parking space, this solution is not applicable to side-by-side vertical vehicles. The ultrasonic solution senses the environment through echo information, which is not applicable to the instructions attached. Figure 4 In this scenario, this solution will assume that the obstacles on the left and right are parallel parking spaces. However, this method may only place parallel parking spaces. Summary of the invention

[0008] The purpose of the present invention is to provide a method for preventing erroneous release of parking spaces based on environmental information. The method distinguishes the types of vehicles constituting parking spaces by using the different characteristics of perpendicular vehicles and parallel vehicles using ultrasonic echoes, and records the types of parking spaces that appeared within the previous driving distance. The method comprehensively determines the type of the parking space, thereby solving the problem of erroneous release of parking spaces caused by inaccurate judgment of the type of parking spaces in existing automatic parking systems.

[0009] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0010] The present invention is a method for preventing a parking space from being released by mistake based on environmental information, comprising the following steps:

[0011] Step S1: Searching for parking spaces and calculating the direct echo information of the ultrasonic sensor on the side of the vehicle in real time;

[0012] Step S2: judging whether a vertical parking space or a parallel parking space can be formed according to the current echo information, if a parking space exists, executing step S3; if no parking space exists, continuing the search;

[0013] Step S3: Filter unreasonable parking space information through parking space error release strategy;

[0014] Step S4: output parking space information and record the current parking space type;

[0015] Among them, in step S3, the specific steps of the parking space mistaken release strategy are as follows:

[0016] Step S31: according to the installation position of the ultrasonic sensor on the vehicle body and the echo information of the ultrasonic sensor, the position of the echo point in the global coordinate system is determined, and the outline of the obstacle is obtained by fitting;

[0017] Step S32: Determine the parking spaces that exist based on the obstacles on both sides of the parking spaces. If only parallel parking spaces exist, release the parallel parking spaces; if only vertical parking spaces exist, release the vertical parking spaces; if there are both parallel parking spaces and vertical parking spaces, determine the types of obstacles on both sides of the parking spaces;

[0018] Step S33: Determine the obstacle types on both sides of the parking space. If the echo characteristics of the ultrasonic sensor are stable after the vehicle passes by, then the obstacle is a parallel vehicle. If there is a gap in the echo characteristics of the ultrasonic sensor after the vehicle passes by, then the obstacle is a side-by-side perpendicular vehicle.

[0019] Step S34: Calculate the proportion of parking space types in the parking space memory module;

[0020] Step S35: Determine the current parking space type through weight calculation and release the parking space;

[0021] Step S36: output the parking space information, and record the parking space result through the parking space memory module.

[0022] As a preferred technical solution, in step S33, the echo variance is calculated to determine the type of obstacles on both sides of the parking space. If the echo variance is within a certain threshold range, it is a parallel vehicle, and if it exceeds the threshold, it is considered to be a perpendicular vehicle;

[0023] If there are parallel vehicles on both sides of the parking space, the parking space is considered to be a parallel parking space, and the weight coefficient of the parallel parking space is Parallel_w1=1, and the weight coefficient of the perpendicular parking space is Cross_w1=0;

[0024] If the obstacles on both sides of the parking space are perpendicular vehicles, the parking space is considered to be a perpendicular parking space, and the weight coefficient of the perpendicular parking space is Cross_w1=1, and the weight coefficient of the parallel parking space is Parallel_w1=0;

[0025] If the obstacle types on the left and right sides are inconsistent, the weight coefficient of the parallel parking space is Parallel_w1=0.5, and the weight coefficient of the perpendicular parking space is Cross_w1=0.5.

[0026] As a preferred technical solution, in step S34, when the driver turns on the parking button and starts searching for parking spaces, the parking space memory module will be activated, and the parking space memory module will memorize the proportion of all released parking space types in the total released parking spaces.

[0027] As a preferred technical solution, in step S35, the weights Parallel_w1, Parallel_w2 and Cross_w1, Cross_w2 are calculated, and the process of determining the current parking space type is as follows:

[0028] If Parallel_w2>=0.8, the parking space is considered to be a parallel parking space;

[0029] If Parallel_w2>0.2&&Parallel_w2<0.8&&Parallel_w1=1, the parking space is considered to be a parallel parking space;

[0030] If Parallel_w2>0.5&&Parallel_w2<0.8&&Parallel_w1=0.5, the parking space is considered to be a parallel parking space;

[0031] Otherwise, the parking space is a vertical parking space.

[0032] The present invention has the following beneficial effects:

[0033] The present invention uses the different characteristics of ultrasonic echoes for perpendicular vehicles and parallel vehicles to distinguish the types of vehicles that make up a parking space, and records the types of parking spaces that appeared within the previous driving distance, so as to comprehensively judge the type of the parking space, improve and avoid the safety risks of existing market technical solutions, and improve the driver's subjective experience of human-computer interaction.

[0034] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0036] Figure 1 A schematic diagram of a parking space erroneous release method based on environmental information according to the present invention;

[0037] Figure 2 It is a schematic diagram of the parking space search process based on ultrasound in the prior art;

[0038] Figure 3 It is a schematic diagram of a parallel ultrasonic parking space in the prior art;

[0039] Figure 4 It is a schematic diagram of a vertical ultrasonic parking space in the prior art;

[0040] Figure 5 Schematic diagram of obstacle fitting;

[0041] Figure 6 Schematic diagram of echo characteristics of parallel vehicles;

[0042] Figure 7 Schematic diagram of echo characteristics of vehicles side by side;

[0043] Figure 8 The echo characteristics of side-by-side perpendicular vehicles in the presence of noise. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] See also Figure 1 As shown, the present invention is a method for preventing a parking space from being released by mistake based on environmental information, comprising the following steps:

[0046] Step S1: Searching for parking spaces and calculating the direct echo information of the ultrasonic sensor on the side of the vehicle in real time;

[0047] Step S2: judging whether a vertical parking space or a parallel parking space can be formed according to the current echo information, if a parking space exists, executing step S3; if no parking space exists, continuing the search;

[0048] Step S3: Filter unreasonable parking space information through parking space error release strategy;

[0049] Step S4: output parking space information and record the current parking space type;

[0050] Among them, in step S3, the specific steps of the parking space mistaken release strategy are as follows:

[0051] like Figure 5 As shown, step S31: according to the installation position of the ultrasonic sensor on the vehicle body and the echo information of the ultrasonic sensor, the position of the echo point in the global coordinate system is determined, and the outline of the obstacle is obtained by fitting;

[0052] Step S32: Determine the parking spaces that exist based on the obstacles on both sides of the parking spaces. If only parallel parking spaces exist, release the parallel parking spaces; if only vertical parking spaces exist, release the vertical parking spaces; if there are both parallel parking spaces and vertical parking spaces, determine the types of obstacles on both sides of the parking spaces;

[0053] Step S33: Determine the types of obstacles on both sides of the parking space. Since the sides of the vehicle are relatively smooth, such as Figure 6 As shown in , if the echo characteristics of the ultrasonic sensor are relatively stable after the ego vehicle passes by, it is a parallel vehicle; Figure 7 As shown in the figure, if the vehicles are parallel and perpendicular, there is a certain boundary between the vehicles. After the ego vehicle passes by, the echo characteristics of the ultrasonic sensor have a larger deviation compared to the parallel vehicles, which means they are parallel and perpendicular vehicles.

[0054] The variance is used to describe the degree of dispersion. The type of obstacles on both sides of the parking space is determined by calculating the echo variance. If the echo variance is within a certain threshold range, it is a parallel vehicle, and if it exceeds the threshold, it is considered a perpendicular vehicle. If there are parallel vehicles on both sides of the parking space, the parking space is considered to be a parallel parking space, and the weight coefficient of the parallel parking space is Parallel_w1=1, and the weight coefficient of the perpendicular parking space is Cross_w1=0; if the obstacles on both sides of the parking space are perpendicular vehicles, the parking space is considered to be a perpendicular parking space, and the weight coefficient of the perpendicular parking space is Cross_w1=1, and the weight coefficient of the parallel parking space is Parallel_w1=0; if the obstacle types on the left and right sides are inconsistent, the weight coefficient of the parallel parking space is Parallel_w1=0.5, and the weight coefficient of the perpendicular parking space is Cross_w1=0.5.

[0055] Step S34: Calculate the proportion of parking space types in the parking space memory module;

[0056] Due to the limitations of ultrasonic sensors, noise may occur during the measurement process. There may be a situation where the echo characteristics of parallel vehicles are highly discrete and are identified as perpendicular vehicles; or the echo characteristics of side-by-side perpendicular vehicles are less discrete and are identified as parallel vehicles. Figure 8 As shown, for this scenario, the parking space memory function is used.

[0057] When the driver turns on the parking button and starts searching for parking spaces, the parking space memory module will be activated and will memorize the proportion of all released parking space types in the total released parking spaces. If 10 parking spaces are found, 8 of which are parallel parking spaces, then Parallel_w2=0.8, Cross_w2=0.2; the parking space memory module will only memorize the most recently searched 10 parking spaces.

[0058] Step S35: Determine the current parking space type through weight calculation and release the parking space;

[0059] Based on the above weights Parallel_w1, Parallel_w2 and Cross_w1, Cross_w2, the current parking space type is determined and the parking space is released.

[0060] If Parallel_w2>=0.8, the parking space is considered to be a parallel parking space;

[0061] If Parallel_w2>0.2&&Parallel_w2<0.8&&Parallel_w1=1, the parking space is considered to be a parallel parking space;

[0062] If Parallel_w2>0.5&&Parallel_w2<0.8&&Parallel_w1=0.5, the parking space is considered to be a parallel parking space;

[0063] Otherwise, the parking space is a vertical parking space.

[0064] Step S36: output the parking space information, and record the parking space result through the parking space memory module.

[0065] Embodiment 1

[0066] An ultrasonic sensor is installed on the vehicle body, and a receiver is arranged inside the vehicle to receive echo information of the ultrasonic sensor and determine the position of the echo point in the global coordinate system;

[0067] Since the length of the vehicle is much greater than its width, when obstacle fitting is performed, the side of the vehicle is relatively smooth, and when multiple vehicles are parked in parallel, the gaps between the vehicles are small, and the smooth area in the echo features is longer and more stable; when multiple vehicles are parked side by side, the gaps between the vehicles are small, the smooth area in the echo features is shorter, and the echo features of vehicles not arranged in parallel are stable.

[0068] The system uses variance to describe the degree of discreteness and determines the type of obstacles on both sides of the parking space by calculating the echo variance. If the echo variance is within a certain threshold range, it is a parallel vehicle, and if it exceeds the threshold, it is considered to be a perpendicular vehicle.

[0069] Embodiment 2

[0070] like Figure 8 Due to the limitations of ultrasonic sensors, noise may occur during the measurement process. There may be a situation where the echo characteristics of parallel vehicles are highly discrete and are identified as perpendicular vehicles; or the echo characteristics of side-by-side perpendicular vehicles are less discrete and are identified as parallel vehicles;

[0071] In this scenario, the parking space memory function is used during automatic parking: after the driver turns on the parking button and starts searching for parking spaces, the parking space memory module will be activated. The parking space memory module will memorize the proportion of all released parking space types in the total released parking spaces; if 10 parking spaces are searched, 8 of which are parallel parking spaces, then Parallel_w2=0.8, Cross_w2=0.2; the parking space memory module will only remember the 10 most recently searched parking spaces.

[0072] Determine the above weights Parallel_w1, Parallel_w2 and Cross_w1, Cross_w2, determine the current parking space type, and release the parking space.

[0073] If Parallel_w2>=0.8, the parking space is considered to be a parallel parking space;

[0074] If Parallel_w2>0.2&&Parallel_w2<0.8&&Parallel_w1=1, the parking space is considered to be a parallel parking space;

[0075] If Parallel_w2>0.5&&Parallel_w2<0.8&&Parallel_w1=0.5, the parking space is considered to be a parallel parking space;

[0076] Otherwise, the parking space is a vertical parking space.

[0077] After the parking space information is output, the parking space result is recorded through the parking space memory module.

[0078] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0079] In addition, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0080] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for preventing parking space from being released by mistake based on environmental information, characterized in that: The steps include: Step S1: Searching for parking spaces and calculating the direct echo information of the ultrasonic sensor on the side of the vehicle in real time; Step S2: judging whether a vertical parking space or a parallel parking space can be formed according to the current echo information, if a parking space exists, executing step S3; if no parking space exists, continuing the search; Step S3: Filter unreasonable parking space information through parking space error release strategy; Step S4: output parking space information and record the current parking space type; Among them, in step S3, the specific steps of the parking space mistaken release strategy are as follows: Step S31: according to the installation position of the ultrasonic sensor on the vehicle body and the echo information of the ultrasonic sensor, the position of the echo point in the global coordinate system is determined, and the outline of the obstacle is obtained by fitting; Step S32: Determine the parking spaces that exist based on the obstacles on both sides of the parking spaces. If only parallel parking spaces exist, release the parallel parking spaces; if only vertical parking spaces exist, release the vertical parking spaces; if there are both parallel parking spaces and vertical parking spaces, determine the types of obstacles on both sides of the parking spaces; Step S33: Determine the obstacle types on both sides of the parking space. If the echo characteristics of the ultrasonic sensor are stable after the vehicle passes by, then the obstacle is a parallel vehicle. If there is a gap in the echo characteristics of the ultrasonic sensor after the vehicle passes by, then the obstacle is a side-by-side perpendicular vehicle. Step S34: calculating the proportion of parking space types in the parking space memory module, and determining the weight coefficient Parallel_w2 of the parallel parking space and the weight coefficient Cross_w2 of the vertical parking space based on the proportion of parking space types; Step S35: Determine the current parking space type through weight calculation and release the parking space; Step S36: output the parking space information, and record the parking space result through the parking space memory module; In step S33, the echo variance is calculated to determine the type of obstacles on both sides of the parking space. If the echo variance is within a certain threshold range, it is a parallel vehicle, and if it exceeds the threshold, it is considered to be a perpendicular vehicle; If there are parallel vehicles on both sides of the parking space, the parking space is considered to be a parallel parking space, and the weight coefficient of the parallel parking space is Parallel_w1=1, and the weight coefficient of the perpendicular parking space is Cross_w1=0; If the obstacles on both sides of the parking space are perpendicular vehicles, the parking space is considered to be a perpendicular parking space, and the weight coefficient of the perpendicular parking space is Cross_w1=1, and the weight coefficient of the parallel parking space is Parallel_w1=0; If the obstacle types on the left and right sides are inconsistent, the weight coefficient of the parallel parking space is Parallel_w1=0.5, and the weight coefficient of the perpendicular parking space is Cross_w1=0.

5.

2. The method for preventing a parking space from being released by mistake based on environmental information according to claim 1, characterized in that: In step S34, when the driver turns on the parking button and starts searching for parking spaces, the parking space memory module will be activated, and the parking space memory module will memorize the proportion of all released parking space types in the total released parking spaces.

3. The method for preventing a parking space from being released by mistake based on environmental information according to claim 1, characterized in that: In step S35, the weight coefficients Parallel_w1 and Parallel_w2 of the parallel parking spaces and the weight coefficients Cross_w1 and Cross_w2 of the vertical parking spaces are calculated, and the process of determining the current parking space type is as follows: If Parallel_w2≥0.8, the parking space is considered to be a parallel parking space; If Parallel_w2>0.2&&Parallel_w2<0.8&&Parallel_w1=1, the parking space is considered to be a parallel parking space; If Parallel_w2>0.5&&Parallel_w2<0.8&&Parallel_w1=0.5, the parking space is considered to be a parallel parking space; Otherwise, the parking space is a vertical parking space.

Images