WO2020007235A1

WO2020007235A1 - Parking space detection method and apparatus, and medium and device

Info

Publication number: WO2020007235A1
Application number: PCT/CN2019/093486
Authority: WO
Inventors: 陈盛军; 蒋少峰; 肖志光
Original assignee: 广州小鹏汽车科技有限公司
Priority date: 2018-07-06
Filing date: 2019-06-28
Publication date: 2020-01-09
Also published as: CN110687539A; CN110687539B

Abstract

Disclosed are a parking space detection method and apparatus, and a medium and a device. The method comprises: respectively performing clustering and piecewise fitting on points on an obstacle detected by ultrasonic radars at left and right sides of a vehicle, so as to obtain a set of obstacle line segments (l1, l2, l3) respectively corresponding to the left and right sides of the vehicle, and endowing the line segments with high and low attributes; and by means of a model matching method, determining whether there is a parking space between two adjacent line segments (l1, l2) with the high attribute. Both a driving path and a direction of a vehicle are not limited, thereby improving the universality of parking space detection.

Description

Parking space detection method, device, medium and equipment

Technical field

The invention relates to the technical field of automatic parking, in particular to a parking space detection method, device, medium and equipment.

Background technique

The first step in fully automatic parking is to detect parking spaces. In the prior art, the parking space detection is carried out by using the ultrasonic radar around the vehicle to search along the vehicle travel path, and the parkingable areas are found from the areas to be detected on both sides of the travel path.

However, in the prior art solutions, there are certain requirements for the vehicle travel path. Generally, the travel path needs to be kept parallel to the area to be detected, and the parking space cannot be detected during the turning and reversing process of the vehicle.

Summary of the invention

Embodiments of the present invention provide a method, a device, a medium, and a device for detecting a parking space, which are used to solve a problem that a vehicle travel path and direction are limited during the parking space detection process.

A parking space detection method in which an ultrasonic radar is installed at designated positions on the left and right sides of a vehicle. The method includes:

If it is determined that the vehicle has moved, the coordinate values of points on each obstacle detected by the corresponding ultrasonic radar are determined respectively for the left and right sides of the vehicle; and the left and right sides of the vehicle are respectively determined according to the obstacles. The distance from the points on the object to the corresponding ultrasonic radar to group the points on the obstacle;

According to the coordinate values of the points on each obstacle in a group, line segment fitting is performed, and according to the ratio of the points on the obstacle with the secondary echo in the group, the height attribute of the fitted line segment is determined. The high and low attributes describe whether the points on the obstacles that make up the line segment originate from high or low objects;

For the line segments corresponding to the left and right sides of the vehicle, a model matching method is used to determine whether there is a parking space between two adjacent high attribute line segments.

A parking space detection device, the device includes:

A coordinate determining module, configured to determine the coordinate values of points on each obstacle detected by a corresponding ultrasonic radar for the left and right sides of the vehicle if it is determined that the vehicle has moved;

A grouping module is used to group the points on the obstacle according to the distance from the point on the obstacle to the corresponding ultrasonic radar respectively for the left and right sides of the vehicle;

The line segment fitting module is used for line segment fitting according to the coordinate values of points on each obstacle in a group;

An attribute assignment module is used to determine the height attribute of the fitted line segment according to the proportion of the points on the obstacle having the secondary echo in a group. The height attribute describes the source of the points on the obstacle constituting the line segment. Whether it is high or low;

A recognition module is used to determine whether there is a parking space between two adjacent high attribute line segments for the corresponding line segments on the left and right sides of the vehicle through a model matching method.

The present invention also provides a non-volatile computer storage medium. The computer storage medium stores an executable program, and the executable program is executed by a processor to implement the steps of the method described above.

The invention also provides a parking space detection device, which includes a memory, a processor, and a computer program stored on the memory. When the processor executes the program, the steps of the method described above are implemented.

According to the solution provided by the embodiment of the present invention, the points on the obstacles detected by the ultrasonic radars on the left and right sides of the vehicle can be clustered and segmented to obtain the sets of obstacle line segments corresponding to the left and right sides of the vehicle, respectively. , And assign line segment high and low attributes, and determine whether there is a parking space between two adjacent high attribute line segments through a model matching method. In the solution provided by the embodiment of the present invention, there are no restrictions on the travel path and direction of the vehicle, which improves the universality of parking space detection.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly explain the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are merely These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without paying creative labor.

1 is a schematic flowchart of a parking space detection method according to a first embodiment of the present invention;

2 is a schematic diagram of a coordinate system provided by Embodiment 1 of the present invention;

3 is a schematic diagram of a sudden change in obstacle point distance provided by Embodiment 1 of the present invention;

4 is a schematic diagram of a parallel parking space provided by Embodiment 1 of the present invention;

5 is a schematic diagram of a vertical parking space provided by Embodiment 1 of the present invention;

FIG. 6 is a schematic diagram of an included angle and a lateral depth distance between two adjacent high-performance line segments provided in Embodiment 1 of the present invention; FIG.

FIG. 7 is a contour description and a target posture corresponding to a parallel parking space provided in Embodiment 1 of the present invention; FIG.

FIG. 8 is a contour description and a target posture of a vertical parking space provided by Embodiment 1 of the present invention; FIG.

FIG. 9 is a schematic flowchart of a parking space identification method according to a second embodiment of the present invention; FIG.

10 is a schematic structural diagram of a parking space detection device according to a third embodiment of the present invention;

FIG. 11 is a schematic structural diagram of a parking space detection device according to a fourth embodiment of the present invention.

detailed description

In each embodiment of the present invention, a space described by a line segment composed of points on obstacles on the left and right sides of the vehicle may be referred to as an accessible space. The parking space detection scheme provided by the present invention can circulate and traverse in the accessible space, search all the parking spaces during the driving process of the vehicle, and simultaneously recognize the parking spaces on the left and right sides. Moreover, there is no excessive requirement on the driving path during the parking space detection process, and the parking space detection can also be performed during the turning and reversing of the vehicle.

At the same time, in the solution provided by the present invention, the roadside can be identified by the ultrasonic radar installed at a specified position, and the obstacle between the current vehicle and the target parking space can be described more completely.

In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms “first” and “second” in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way are interchangeable where appropriate, so that the embodiments of the invention described herein can be implemented in an order other than those illustrated or described herein. Furthermore, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units need not be limited to those explicitly listed Those steps or units may instead include other steps or units not explicitly listed or inherent to these processes, methods, products or equipment.

Example one

The first embodiment of the present invention provides a parking space detection method. The steps of the method can be shown in FIG. 1 and includes:

Step 101: Determine whether the vehicle moves.

In this embodiment, when the vehicle moves, the parking space detection may be triggered. Therefore, in this step, it can be determined whether the vehicle has moved, and if it is determined that the vehicle has moved, step 102 can be continued.

Specifically, it may be determined, but not limited to, whether the vehicle moves according to whether the posture of the vehicle changes. The vehicle posture can be understood as including vehicle center point coordinates and vehicle yaw angle.

As shown in FIG. 2, it is assumed that the global absolute plane coordinate system is oxy, and the vehicle local coordinate system is o'x'y '. The coordinates of the center point of the vehicle may be the coordinates of the center point of the rear wheel axis of the vehicle. The vehicle pose includes the center point coordinates of the rear wheel axis of the vehicle (that is, the coordinate origin o ′ point coordinate in the local coordinate system of the vehicle) and the vehicle yaw angle (θ ₀ ). The pose can be expressed as (x ₀ , y ₀ , θ ₀ ) in the global absolute plane coordinate system. The vehicle posture can be determined based on the wheel pulse, inertial measurement unit (IMU), directional turning angle and other signal inputs, using the vehicle's accumulated dead reckoning formula.

Step 102: Determine coordinate values of points on the obstacle.

In this step, if it is determined that the vehicle has moved, the coordinate values of points on each obstacle detected by the corresponding ultrasonic radar may be determined respectively for the left and right sides of the vehicle. The coordinate value of the point on each obstacle can be determined in any way. For example, the coordinate value of the point on each obstacle can be, but is not limited to, the position and posture of the vehicle. The point on the obstacle reaches the corresponding ultrasonic wave. The distance of the radar and the installation position of the ultrasonic radar on the vehicle are determined.

It should be noted that, in this embodiment, in order to detect points on an obstacle, the ultrasonic radar needs to be installed at designated positions on the left and right sides of the vehicle.

Preferably, the ultrasonic radar can be installed on the front wheel eyebrows of the vehicle, is perpendicular to the vehicle body axis, and the height above the ground is a specified height. The specified height can be set according to the characteristics of different ultrasonic radars, relative to the center of the rear wheel shaft. dimensions' of L _x, L _y. The installation position of the ultrasonic radar on the right side of the vehicle can be shown in Figure 2. The point on the obstacle detected by the ultrasonic radar, and the distance to the ultrasonic radar can be expressed by d. In order to unify the calculation formulas of the obstacle points on the left and right sides, the symbols of _Ly and d can be specified as the left side of the vehicle's forward direction is negative, the right side is positive, and the sign of L _x can be specified as both positive. Then the coordinate formula (x _p , y _p ) of the point P on the obstacle in the global absolute plane coordinate system oxy can be expressed as:

Step 103: Group the points on the obstacle.

It should be noted that step 102 and step 103 are performed in no particular order. In this embodiment, description is made by taking step 103 to be performed after step 102 as an example.

Considering that the ultrasonic radar detects the contour point of the obstacle, the contour of the obstacle does not change suddenly, so when the distance between the detected obstacle and the corresponding ultrasonic radar changes suddenly, it can be understood as the point The belonging obstacle has changed. Therefore, in this embodiment, the points belonging to different obstacles can be grouped according to whether the distance is abruptly changed, and the points on the obstacles can be clustered.

In this step, the points on the obstacle can be grouped for the left and right sides of the vehicle respectively according to the distance from the points on the obstacle to the corresponding ultrasonic radar.

Specifically, in this step, the distance from the point on the currently detected obstacle to the corresponding ultrasonic radar for the left and right sides of the vehicle, respectively, and the obstacle detected adjacent to the point on the obstacle before When the absolute value of the difference between the point on the object and the ultrasonic radar is greater than the set value, the points on the obstacle that are not grouped before the currently detected obstacle are divided into a group.

For example, as shown in FIG. 3, a point distance abrupt change on an obstacle is detected. Vehicles for parking space detection can pass through No. 1 vehicle, roadside and No. 2 vehicle, and can scan the obstacle contour points as shown in FIG. 3. It can be found that during the continuous travel of the vehicle, for the point concentration obtained by scanning the same obstacle, the distance between two adjacent points does not change much. For different obstacles, the distance between two adjacent points at the edge will change abruptly, such as P (m) and P (m + 1), P (n) and P (n + 1). Therefore, all the scanned points can be clustered. Taking FIG. 3 as an example, clustering can be performed into three groups.

Of course, if the obstacle disappears and the ultrasonic radar does not detect the point on the obstacle, then the distance detected by the ultrasonic radar is the maximum range value. At this time, it can also be understood as the distance of the point on the detected obstacle. A point on an obstacle was mutated.

Step 104: Perform line segment fitting.

In this step, line segment fitting can be performed according to the coordinate values of points on each obstacle in a group.

Still taking FIG. 3 as an example, line segment fitting is performed for each group, and three line segments l1, l2, and l3 describing the outline of the obstacle can be obtained, where l1, l2 are body contours, and l3 is road contour.

After performing line segment fitting, the obtained point-related information on each obstacle, such as coordinate values and distances, can be deleted to save storage space.

Step 105: Assign a line segment attribute.

The inventor found in the experiment that in addition to the direct measurement value (primary echo) of the ultrasonic radar at that time, the points on each obstacle may also have secondary echoes. When the ultrasonic radar scans a higher obstacle (such as a vehicle), the ratio of the secondary echo is higher, and when a lower obstacle (such as a road edge) is scanned, the ratio of the secondary echo is lower.

It should be noted that the ultrasonic radar involved in the embodiments of the present invention has a secondary echo detection capability. Therefore, in this step, the height attribute of the fitted line segment can be determined according to the ratio of points on the obstacle with the secondary echo in each group. The height attribute is described by the obstacles constituting the line segment. The points come from high or low objects.

When determining the high and low attributes of a line segment, the proportion of points on an obstacle with a secondary echo greater than a specific value can be grouped, and the corresponding line segment is determined to be a high attribute. The outgoing line segment is determined as a low attribute. Preferably, the specific value may be 60% through an experimental test. Of course, the specific value can be adjusted according to the actual detection capability of the ultrasonic radar.

Of course, preferably, the line segment can be further understood as including left and right attributes. The left and right attributes are described as whether the points on the obstacles constituting the line segment originate from the left side or the right side of the vehicle.

In this step, preferably, a line segment with a right attribute can also be added to a right line segment set, and a line segment with a left attribute can be added to a left line segment set, so as to facilitate subsequent parking spaces according to the right line segment set and the left line segment set. Identify.

Step 106, parking space identification.

In this step, for the line segments corresponding to the left and right sides of the vehicle, a model matching method can be used to determine whether there is a parking space between two adjacent high attribute line segments.

As shown in FIG. 4, the parking space on one side of the vehicle may be a parallel parking space. At this time, specifically, it is possible, but not limited to, to determine whether there is a parking space between two adjacent high attribute line segments by:

If a low attribute line segment is also included between two adjacent high attribute line segments, determine a lateral depth distance between the low attribute line segment and at least one of the two adjacent high attribute line segments, and Whether the difference between them is greater than the first threshold value; and determining whether the difference between the longitudinal length distance between two adjacent high-performance line segments and the vehicle length of the vehicle is greater than the second threshold value;

If the difference between the lateral depth distance and the vehicle width of the vehicle is greater than a first threshold value, and the difference between the longitudinal length distance and the vehicle length of the vehicle is greater than a second threshold value, Then determine that there is at least one parking space between two adjacent high attribute line segments; or

If a low attribute line segment is not included between two adjacent high attribute line segments, determine whether a difference between a longitudinal length distance between the adjacent two high attribute line segments and a vehicle length of the vehicle is greater than a second door The limit value, if yes, determine that there is at least one parking space between two adjacent high attribute line segments.

If two adjacent high-attribute line segments also include a low-attribute line segment, it can be understood that there is an obstacle corresponding to a low-attribute line segment between two adjacent high-attribution line segments. At this time, It is necessary to determine whether the width (dp, lateral depth distance) and length (L3, longitudinal length distance) of the space corresponding to the three line segments meet the requirements of the width and length of the vehicle for parking space detection, and whether it can be used as a parking space.

If the low attribute line segment is not included between two adjacent high attribute line segments, it can be understood that there are no other obstacles between the obstacles corresponding to the adjacent two high attribute line segments, then it can be understood that the width is sufficient to satisfy The width requirement of the vehicle for parking space detection only needs to determine whether the length (L3) of the space corresponding to the two line segments meets the vehicle length requirement of the vehicle for parking space detection and whether it can be used as a parking space.

Preferably, the first threshold value may be 0.3 meters, and the second threshold value may be 0.8 meters.

Further, in order to reduce the false detection of parking spaces, it is possible to further determine that the lengths of the two adjacent high-level line segments are greater than the first specified value, thereby increasing the probability that the obstacles corresponding to the two adjacent high-level line segments are vehicles, thereby increasing The detected area is the accuracy of the parking space.

Preferably, the first specified value may be 0.8 times the length of the vehicle of the vehicle for which parking space is detected.

As shown in FIG. 5, the parking space on the side of the vehicle may also be a vertical parking space. At this time, specifically, it is possible, but not limited to, to determine whether there is a parking space between two adjacent high attribute line segments by:

If a low attribute line segment is also included between two adjacent high attribute line segments, determine the lateral depth distance between the low attribute line segment and at least one of the two adjacent high attribute line segments, and the length of the vehicle Whether the difference between them is greater than the third threshold value; and determining whether the difference between the longitudinal length distance between two adjacent high attribute line segments and the width of the vehicle is greater than the fourth threshold value;

If the difference between the lateral depth distance and the vehicle length of the vehicle is greater than the third threshold value, and the difference between the longitudinal length distance and the vehicle width of the vehicle is greater than a fourth threshold value Value, determine that there is at least one parking space between two adjacent high attribute line segments; or,

If the low attribute line segment is not included between two adjacent high attribute line segments, determine whether the difference between the longitudinal length distance between the adjacent two high attribute line segments and the vehicle width of the vehicle is greater than the fourth door The limit value, if yes, determine that there is at least one parking space between two adjacent high attribute line segments.

Similar to parallel parking spaces, if there are low-attribute line segments between two adjacent high-attribute line segments, then at this time, you need to determine the depth (dp, lateral depth distance) and width (L3, longitudinal direction) of the space corresponding to the three line segments. Length distance) whether it meets the requirements of the length and width of the vehicle for parking space detection, and whether it can be used as a parking space.

If there are no low-attribute line segments between two adjacent high-attribution line segments, it can be understood at this time that the depth is sufficient to meet the vehicle length requirements of the vehicle for parking space detection, and only the width of the space corresponding to the two line segments (L3 ) Whether the vehicle width requirement of the vehicle for parking space detection is met, and whether it can be used as a parking space.

Preferably, the third threshold value may be 0.3 meters, and the fourth threshold value may be 0.6 meters.

Further, in order to reduce the false detection of the parking space, it may be further determined that the lengths of two adjacent line segments with high attributes are both larger than the second specified value. Preferably, the second specified value may be 0.8 times the vehicle width of the vehicle for which parking space is detected.

Regardless of the parallel parking space or the vertical parking space, in order to further reduce the false detection of parking spaces, it can be further determined that the included angle between two adjacent high-level line segments is less than the first threshold, and the The lateral depth distance between the line segments is less than a second threshold.

FIG. 6 is a schematic diagram of an included angle and a lateral depth distance between two adjacent high attribute line segments. The included angle θ ₁₂ between two adjacent high attribute line segments can be understood as an included angle obtained by extending L1 and L2. The angle bisector LA as the included angle passes through the midpoints of L1 and L2 as the vertical line of LA to obtain the distance d1 and d2. Then the lateral distance and depth distance d _{12 of} L1 and L2 are defined as the difference between d1 and d2. Absolute value: d ₁₂ = | d1-d2 |.

Preferably, the first threshold may be 25 degrees, and the second threshold may be 1 meter.

Of course, the above-mentioned thresholds, designated values, and thresholds can all be adjusted according to the geometric size parameters and the steering capability of the vehicle performing the parking space detection.

In addition, it should be noted that in this step, only the lower limit of each parameter is specified when performing parking space identification, so that the identified parking space can at least meet the requirements for parking a vehicle for parking space detection. According to the definition of different functional products, the upper limit of each parameter can also be set, so that the specific number of vehicles that can be parked in the identified parking space can be further determined, which will not be further described in this embodiment.

Step 107: Determine parking space information.

After the parking space is identified, a contour description corresponding to the parking space existing between two adjacent high attribute line segments may be further determined, and the contour description is used to describe the location of the parking space, and / or, the parking space is determined The corresponding target posture of the vehicle enables automatic parking in the parking space according to the determined contour description and / or target posture during subsequent automatic parking.

After determining the contour description and / or the target pose, the determined contour description and / or the target pose may be output for easy viewing.

As shown in FIG. 7, it is a contour description and a target posture diagram corresponding to parallel parking spaces. Similarly, as shown in FIG. 8, it is a contour description and a target posture diagram corresponding to a vertical parking space.

In this embodiment, the parking space outline description may be, but is not limited to, five line segments corresponding to the six points A, B, C, D, E, and F, that is, the five line segments AB, BC, CD, DE, and EF may be used to describe the parking space. Outline to facilitate path planning.

Specifically, if there are no low-attribute line segments between two adjacent high-attribution line segments, since there is actually no restriction on the positions of the C and D vertices in the parking space, the values of the coordinates of C and D can be arbitrarily determined as required. For example, in parallel parking spaces, the coordinates of point C are determined based on the length of the BC plus 0.3 meters, and the coordinates of the point D are determined based on the length of the DE plus 0.3 meters. For another example, in the vertical parking space, the coordinates of the point C are determined according to the length of the BC plus 0.3 meters, and the coordinates of the point D are determined according to the length of the DE plus 0.3 meters. After determining the coordinate values of C and D, five line segments AB, BC, CD, DE, and EF can be used to describe the parking space contour.

The target pose can be expressed in the coordinates of the target vehicle center (point P coordinates, such as the center point of the rear wheel axis of the vehicle) and the direction of the arrow passing through point P (corresponding to the target yaw angle). In the global absolute plane coordinate system, it can be expressed as (x _p , y _p , θ _p ).

Depending on the final parking strategy, the determination method of the target yaw angle θ _p may also be different. For example, for parallel parking spaces, the final parking strategy can be the direction of the arrow passing point P and any of AB, CD, and EF. The line segment remains parallel. Taking the direction of the arrow passing point P and AB as an example, then θ _p can be expressed as:

θ _p = a tan 2 (y _B -y _A , x _B -x _A )

It should be noted that atan2 (Δy, Δx) is a function for calculating the arc tangent in the C language function. Compared with arctan (Δy / Δx), a more stable calculation result is obtained.

When the absolute value of Δy is much larger than Δx, the calculation result of the arctan (Δy / Δx) function is unstable. The practice of atan2 (Δy, Δx) is to use arctan (Δy / Δx) when the absolute value of Δx is greater than the absolute value of Δy; otherwise, use arctan (Δx / Δy) to ensure the stability of the determined value.

According to the set position of the point P on the vehicle and the parking strategy of the vehicle in the parking space, the coordinates of the point P (x _p , y _p ) can be determined by, but not limited to, the four vertex coordinates of B, C, D, and E.

In the following, a specific example is used to describe step 106 in the solution provided by the first embodiment of the present invention.

Example two

The second embodiment of the present invention provides a parking space identification method. The step flow of the method can be shown in FIG. 9 and includes:

Step 201: Start searching from the first line segment of the line segment set.

In this embodiment, it is assumed that the line segment with the left attribute and the line segment with the right attribute are in the same line segment set, and the left and right attributes of the line segment can be used to determine whether the line segment is from the left side or the right side of the vehicle. At this time, it can be understood as i = 1, and i represents the number of the line segment in the line segment set.

Of course, it is also possible to identify the left parking space for the line segment corresponding to the left side of the vehicle and the right parking space for the line segment corresponding to the right side of the vehicle, that is, to identify the left and right parking spaces respectively.

Step 202: Determine whether the line segment has a high attribute.

If yes, continue to step 203; otherwise, increase the line segment number by 1 and continue with this step.

Step 203: Initialize the high attribute line segment as the first obstacle vehicle L1.

In this step, it can be assumed that the obstacle corresponding to the high-attribute line segment is a vehicle, the obstacle vehicle is recorded as L1, and the line segment number is increased by 1, and the process proceeds to step 204.

Step 204: Determine whether the line segment is a high attribute.

In this step, it can be determined whether the line segment is of a high attribute. If so, step 205 is continued, otherwise, step 205 'is performed.

Step 205: Initialize the high attribute line segment as the second obstacle vehicle L2.

In this step, it can be assumed that the obstacle corresponding to the high attribute line segment is a vehicle, and the obstacle vehicle is recorded as L2, and step 206 can be continued to determine whether there is a parking space between L1 and L2.

Step 205 ', it is determined whether the line segment corresponding to L1 is located on the same side of the vehicle.

If yes, continue to step 206 '; otherwise, increase the line segment number by 1 and return to step 202.

Step 206 ': It is determined whether the lateral depth distance dp between the line segment and L1 is greater than the vehicle width.

If not, the line segment number can be increased by 1 and the process returns to step 202. If so, step 207 'can be performed.

In this step, it is determined whether the distance between the line segment and L1 is sufficient to accommodate a vehicle for parking space detection. Since the vehicle width value is smaller than the vehicle length value, in this step, the lateral depth distance dp can be compared with the vehicle width to determine whether the lateral depth distance dp is sufficient to accommodate the vehicle width.

Step 207 ': Add the line segment to the roadside line segment set.

In this step, it can be assumed that the obstacle corresponding to the low-attribute line segment is a roadside, and the roadside line segment set is added, and the line segment number can be increased by 1, and the process returns to step 204 to continue searching for the high-attribute line segment.

Step 206: Determine whether θ ₁₂ is less than 25 degrees, whether d ₁₂ is less than 1 meter, and whether L3 is greater than the vehicle width + 0.6 meters.

In this step, it can be determined whether the included angle θ ₁₂ between the two adjacent high-performance line segments (that is, between L1 and L2), the lateral depth distance d ₁₂ , and the longitudinal length distance L3 meet the requirements. If the requirements are met, the judgment can be continued, and step 207 is performed; otherwise, it can be considered that there is no parking space between L1 and L2, and step 208 'is performed.

In this step, it is determined whether the distance between L1 and L2 is sufficient to accommodate a vehicle for parking space detection. Because the vertical parking space has a smaller requirement for the longitudinal length distance L3 between L1 and L2, in this step, it can be determined first whether L3 meets the vertical parking space requirement.

Step 208 ': Initialize L2 as the first obstacle vehicle L1.

If at least one of the included angle, the horizontal depth distance, and the vertical length distance between two adjacent high-performance line segments is not satisfied with the setting requirements, it can be determined that there is no parking space between L1 and L2, and the current L2 can be initialized to L1 , Re-find a parking space between the next two adjacent high-segment line segments. In addition, the line segment needs to be increased by 1 and the process returns to step 204.

Step 207: Determine whether L3 is greater than the vehicle length +0.8 meters.

If L3 meets the requirements for vertical parking spaces, further determine whether L3 meets the requirements for parallel parking spaces. If yes, proceed to step 208 to further determine the parallel parking space; otherwise, proceed to step 209 to further perform the vertical parking space judgment.

In step 208, whether the lengths of L1 and L2 are greater than 0.8 times the vehicle length.

If yes, then it can be determined that there is at least one parallel parking space between L1 and L2, and it can return to step 208 'to continue to find a parking space; otherwise, it can directly return to step 208'.

Step 209: Determine whether there are other line segments between L1 and L2.

If yes, go to step 210; otherwise, go to step 211.

Step 210: Determine whether the lateral depth distance dp of other line segments existing between L1 and L2 and at least one of L1 and L2 is greater than the vehicle length.

If L3 does not meet the requirements for parallel parking spaces, it is necessary to determine whether the distance between other line segments between L1 and L2 and L1 and L2 meets the requirements for vertical parking spaces. If yes, go to step 211; otherwise, go back to step 208 '.

In this embodiment, it can be determined whether the lateral depth distance dp between the other line segments and L1 is greater than the vehicle length, and whether the lateral depth distance dp between the other line segments and L2 is greater than the vehicle length. If both are greater than the vehicle length, then Step 211 may be performed; otherwise, step 208 ′ may be performed.

Steps 211, L1, and L2 are all longer than 0.8 times the vehicle width.

If yes, it can be determined that there is at least one vertical parking space between L1 and L2, and it can be returned to execute step 208 ', otherwise, it can be directly returned to execute step 208'.

Based on the same inventive concept as

Embodiments

1 and 2, the following devices are provided.

Example three

Embodiment 3 of the present invention provides a parking space detection device. The structure of the device may be as shown in FIG. 10 and includes:

The coordinate determination module 11 is configured to determine the coordinate values of points on each obstacle detected by the corresponding ultrasonic radar for the left and right sides of the vehicle if it is determined that the vehicle has moved;

The grouping module 12 is used to group the points on the obstacle according to the distance from the points on the obstacle to the corresponding ultrasonic radar respectively for the left and right sides of the vehicle;

The line segment fitting module 13 is configured to perform line segment fitting according to the coordinate values of points on each obstacle in a group;

The attribute assignment module 14 is used to determine the height attribute of the fitted line segment according to the proportion of the points on the obstacle having the secondary echo in a group. The height attribute describes the source of the points on the obstacle constituting the line segment Whether it is high or low;

The identification module 15 is configured to determine whether there is a parking space between two adjacent high-attribute line segments for the corresponding line segments on the left and right sides of the vehicle through a model matching method.

The device further includes a parking space output module 16 for determining a contour description corresponding to a parking space existing between two adjacent high-attribute line segments, and the contour description is used to describe the location of the parking space, and / or The target posture of the vehicle corresponding to the parking space.

The coordinate determining module 11 is configured to determine a point on each obstacle by using a position of the vehicle, a distance from a point on the obstacle to a corresponding ultrasonic radar, and a mounting position of the ultrasonic radar on the vehicle. Coordinate value.

The grouping module 12 is specifically for the left and right sides of the vehicle, respectively, the distance from the point on the currently detected obstacle to the corresponding ultrasonic radar, and the obstacle detected adjacent to the point on the obstacle before When the absolute value of the difference between the distance from the upper point and the ultrasonic radar is greater than the set value, the points on the obstacle that are not grouped before the currently detected obstacle are divided into a group.

The identification module 15 is configured to determine whether there is a parking space between two adjacent high attribute line segments through a model matching method, including:

The identification module 15 is further configured to determine that the lengths of two adjacent high-segment line segments are greater than a first specified value.

The identification module 15 is configured to determine whether there is a parking space between two adjacent high-attribute line segments by using a model matching method, including: if a low-attribute line segment is also included between two adjacent high-attribution line segments, determining the low attribute Whether the difference between the lateral depth distance between the line segment and at least one of the two adjacent high attribute line segments and the vehicle length of the vehicle is greater than a third threshold; and Whether the difference between the longitudinal length distance between the line segments and the width of the vehicle is greater than a fourth threshold value;

The identification module 15 is further configured to determine that the lengths of two adjacent high attribute line segments are greater than a second specified value.

The identification module 15 is further configured to determine that an included angle between two adjacent high-level line segments is smaller than a first threshold, and a lateral depth distance between two adjacent high-level line segments is smaller than a second threshold.

Based on the same inventive concept, embodiments of the present invention provide the following devices and media.

Example 4

Embodiment 4 of the present invention provides a parking space detection device. The structure of the device may include a memory 21, a processor 22, and a computer program stored on the memory, as shown in FIG. 11, and is implemented when the processor 22 executes the program. Steps of the method according to the first embodiment of the present invention.

Optionally, the processor 22 may specifically include a central processing unit (CPU), an application-specific integrated circuit (ASIC), may be one or more integrated circuits for controlling program execution, and may be used A hardware circuit developed by a field programmable gate array (FPGA, field programmable gate array) can be a baseband processor.

Optionally, the processor 22 may include at least one processing core.

Optionally, the memory 21 may include a read-only memory (ROM, read only memory), a random access memory (RAM, random access memory), and a magnetic disk memory. The memory 21 is configured to store data required when the at least one processor 22 is running. The number of the memories 21 may be one or more.

Embodiment 5 of the present invention provides a non-volatile computer storage medium. The computer storage medium stores an executable program. When the executable program is executed by a processor, the method provided in Embodiment 1 of the present invention is implemented.

In a specific implementation process, a computer storage medium may include: a universal serial bus flash disk (USB, Universal Serial Bus flash drive), a mobile hard disk, a read-only memory (ROM, Read-Only Memory), and a random access memory (RAM , Random Access Memory), magnetic disk or compact disc and other storage media that can store program code.

In the embodiments of the present invention, it should be understood that the disclosed device and method may be implemented in other manners. For example, the device embodiments described above are only schematic. For example, the division of the unit or unit is only a logical function division. In actual implementation, there may be another division manner. For example, multiple units or components may The combination can either be integrated into another system, or some features can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, the indirect coupling or communication connection of the device or unit, and may be electrical or other forms.

Each functional unit in the embodiment of the present invention may be integrated into one processing unit, or each unit may also be an independent physical module.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, all or part of the technical solutions of the embodiments of the present invention may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for making a computer device, for example, may be A personal computer, a server, or a network device, or a processor executes all or part of the steps of the method described in each embodiment of the present invention. The foregoing storage medium includes: a universal serial bus flash drive (universal serial flash drive), a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disc, and other media that can store program codes.

Those skilled in the art should understand that the embodiments of the present invention may be provided as a method, a system, or a computer program product. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and combinations of processes and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, so that the instructions generated by the processor of the computer or other programmable data processing device are used to generate instructions Means for implementing the functions specified in one or more flowcharts and / or one or more blocks of the block diagrams.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to work in a specific manner such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, the instructions The device implements the functions specified in one or more flowcharts and / or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of steps can be performed on the computer or other programmable device to produce a computer-implemented process, which can be executed on the computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more flowcharts and / or one or more blocks of the block diagrams.

Although the preferred embodiments of the present invention have been described, those skilled in the art can make other changes and modifications to these embodiments once they know the basic inventive concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the invention.

Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. In this way, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

A parking space detection method, characterized in that an ultrasonic radar is installed at designated positions on the left and right sides of the vehicle, and the method includes:

If it is determined that the vehicle has moved, the coordinate values of points on each obstacle detected by the corresponding ultrasonic radar are determined respectively for the left and right sides of the vehicle; and the left and right sides of the vehicle are respectively determined according to the obstacles. The distance from the points on the object to the corresponding ultrasonic radar to group the points on the obstacle;

According to the coordinate values of the points on each obstacle in a group, line segment fitting is performed, and according to the ratio of the points on the obstacle with the secondary echo in the group, the height attribute of the fitted line segment is determined. The high and low attributes describe whether the points on the obstacles that make up the line segment originate from high or low objects;

For the line segments corresponding to the left and right sides of the vehicle, a model matching method is used to determine whether there is a parking space between two adjacent high attribute line segments.
The method of claim 1, further comprising:

Determining a contour description corresponding to a parking space existing between two adjacent high attribute line segments, the contour description is used to describe a position where the parking space is located, and / or determining a target position of the vehicle corresponding to the parking space posture.
The method of claim 1, wherein the coordinate value of a point on each obstacle passes through the pose of the vehicle, the distance from the point on the obstacle to the corresponding ultrasonic radar, and the ultrasonic radar at The installation position on the vehicle is determined.
The method according to claim 1, wherein the grouping of the points on the obstacle according to the distance from the point on the obstacle to the corresponding ultrasonic radar for each of the left and right sides of the vehicle comprises:

For the left and right sides of the vehicle, the distance from the point on the currently detected obstacle to the corresponding ultrasonic radar, and the point on the obstacle detected adjacent to the point before the obstacle to the ultrasonic radar. When the absolute value of the difference between the distances is greater than the set value, the points on the obstacle that are not grouped before the currently detected obstacle are divided into a group.
The method according to claim 1, wherein determining whether a parking space exists between two adjacent high attribute line segments by using a model matching method comprises:

If a low attribute line segment is also included between two adjacent high attribute line segments, determine a lateral depth distance between the low attribute line segment and at least one of the two adjacent high attribute line segments, and Whether the difference between them is greater than the first threshold value; and determining whether the difference between the longitudinal length distance between two adjacent high-performance line segments and the vehicle length of the vehicle is greater than the second threshold value;

If the difference between the lateral depth distance and the vehicle width of the vehicle is greater than a first threshold value, and the difference between the longitudinal length distance and the vehicle length of the vehicle is greater than a second threshold value, Then determine that there is at least one parking space between two adjacent high attribute line segments; or

If a low attribute line segment is not included between two adjacent high attribute line segments, determine whether a difference between a longitudinal length distance between the adjacent two high attribute line segments and a vehicle length of the vehicle is greater than a second door The limit value, if yes, determine that there is at least one parking space between two adjacent high attribute line segments.
The method according to claim 5, further comprising: determining that the lengths of two adjacent high attribute line segments are greater than a first specified value.
The method according to claim 1, wherein determining whether a parking space exists between two adjacent high attribute line segments through a model matching method comprises:

If a low attribute line segment is also included between two adjacent high attribute line segments, determine the lateral depth distance between the low attribute line segment and at least one of the two adjacent high attribute line segments, and the length of the vehicle Whether the difference between them is greater than the third threshold value; and determining whether the difference between the longitudinal length distance between two adjacent high attribute line segments and the width of the vehicle is greater than the fourth threshold value;

If the difference between the lateral depth distance and the vehicle length of the vehicle is greater than the third threshold value, and the difference between the longitudinal length distance and the vehicle width of the vehicle is greater than a fourth threshold value Value, determine that there is at least one parking space between two adjacent high attribute line segments; or,

If the low attribute line segment is not included between two adjacent high attribute line segments, determine whether the difference between the longitudinal length distance between the adjacent two high attribute line segments and the vehicle width of the vehicle is greater than the fourth door The limit value, if yes, determine that there is at least one parking space between two adjacent high attribute line segments.
The method according to claim 5, further comprising: determining that the lengths of two adjacent high attribute line segments are greater than a second specified value.
The method according to any one of claims 5 to 8, further comprising:

It is determined that an included angle between two adjacent high-level line segments is smaller than a first threshold, and a lateral depth distance between two adjacent high-level line segments is smaller than a second threshold.
A parking space detection device, characterized in that the device includes:

A coordinate determining module, configured to determine the coordinate values of points on each obstacle detected by a corresponding ultrasonic radar for the left and right sides of the vehicle if it is determined that the vehicle has moved;

A grouping module is used to group the points on the obstacle according to the distance from the point on the obstacle to the corresponding ultrasonic radar respectively for the left and right sides of the vehicle;

The line segment fitting module is used for line segment fitting according to the coordinate values of points on each obstacle in a group;

An attribute assignment module is used to determine the height attribute of the fitted line segment according to the proportion of the points on the obstacle having the secondary echo in a group. The height attribute describes the source of the points on the obstacle constituting the line segment. Whether it is high or low;

A recognition module is used to determine whether there is a parking space between two adjacent high attribute line segments for the corresponding line segments on the left and right sides of the vehicle through a model matching method.
A non-volatile computer storage medium, characterized in that the computer storage medium stores an executable program, and the executable program is executed by a processor to implement the steps of the method according to any one of claims 1 to 9.
A parking space detection device, comprising a memory, a processor, and a computer program stored on the memory. When the processor executes the program, the steps of the method according to any one of claims 1 to 9 are implemented.