Intelligent anti-collision method for automatic parking of vehicle

Abstract

The application discloses an intelligent anti-collision method for automatic parking of a vehicle, and relates to the technical field of intelligent parking. The method comprises the following steps: acquiring environmental image information of a position where the vehicle is located according to an automatic parking instruction, constructing a parking area simulation space according to the acquired environmental image information, and identifying parking spaces contained in the environmental image information; acquiring obstacle information around the position where the vehicle is located by acquiring sensing data of the vehicle; performing parking feasibility evaluation on each parking space identified in the parking area simulation space, judging whether there is a parking space, and if there is a parking space, generating an automatic parking route from the position where the vehicle is located to the parking space; generating a corresponding safe parking space based on basic information of the vehicle in the parking area simulation space, simulating and generating a suspected obstacle identified in the parking area simulation space, judging an influence of the suspected obstacle on the parking area simulation space, and thus assisting in judging whether there is a collision risk in the automatic parking process.

Description

Technical Field

[0001] This invention relates to the field of intelligent parking technology, specifically an intelligent collision avoidance method for automatic vehicle parking. Background Technology

[0002] With the continuous increase in car ownership, the problem of parking difficulties has become increasingly prominent, leading to the emergence of automatic parking technology, which provides convenience for drivers. During automatic parking, the vehicle needs to complete complex movement operations within a limited space, making collision avoidance a key factor in ensuring safety.

[0003] How to achieve automatic parking based on demand while reducing the risk of collisions during parking is a problem we need to solve. To this end, we now provide an intelligent collision avoidance method for automatic parking of vehicles. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent collision avoidance method for automatic parking of vehicles.

[0005] The objective of this invention can be achieved through the following technical solution: an intelligent collision avoidance method for automatic parking of vehicles, comprising the following steps: The system acquires environmental image information of the vehicle's location based on the automatic parking command, constructs a simulated parking area based on the acquired environmental image information, and identifies the parking spaces contained within the environmental image information. By acquiring vehicle sensor data, information about obstacles around the vehicle's location can be obtained. The system assesses the parking feasibility of each parking space identified within the simulated parking area to determine if a parking space is available. If a parking space is available, it generates an automatic parking route from the vehicle's location to that space.

[0006] Furthermore, the process of obtaining environmental image information about the vehicle's location based on the automatic parking command includes: The user selects an automatic parking command based on the vehicle's location. The automatic parking command includes a first command, a second command, a third command, and a fourth command. Based on the automatic parking command selected by the user, images of the vehicle in the corresponding parking direction are acquired to obtain the corresponding environmental image information.

[0007] Furthermore, the process of constructing a simulated parking area space based on the acquired environmental image information and identifying the parking spaces contained within the environmental image information includes: Construct a three-dimensional spatial coordinate system based on the vehicle's location; The obtained environmental image information is rasterized, and the rasterized environmental image information is mapped to the corresponding position in the three-dimensional spatial coordinate system; The rasterized environmental image information is converted to grayscale to obtain the corresponding grayscale image. Feature extraction is performed on the grayscale image to obtain the edge features of the parking space, and a simulated parking space is generated based on the corresponding position of the obtained edge features in the three-dimensional spatial coordinate system, thereby completing the construction of the simulated parking area space.

[0008] Furthermore, the process of obtaining obstacle information around the vehicle's location by acquiring vehicle sensor data includes: Data acquisition terminals are installed at various locations on the vehicle to acquire corresponding vehicle sensor data. Based on the vehicle sensor data obtained from each data acquisition terminal, the location and distance of each obstacle are determined, and corresponding simulated obstacles are generated at their corresponding positions in the three-dimensional spatial coordinate system. The coordinates of each simulated obstacle in the three-dimensional coordinate system are labeled, and the coordinates of all simulated obstacles in the three-dimensional coordinate system are summarized to obtain the obstacle information at the vehicle's location.

[0009] Furthermore, the process of assessing the parking feasibility of each identified parking space within the simulated parking area to determine the existence of available parking spaces includes: Based on the vehicle's basic information, a virtual vehicle is generated at the vehicle's current location, and a corresponding safe parking space is generated at the location of each simulated parking space. The coordinate range of the safe parking space is then obtained. The safe parking space is divided into peripheral and core areas; The coordinate range of each safe parking space is compared with the coordinate information of all simulated obstacles in the obstacle information. If the coordinate information in the obstacle information is within the coordinate range of any safe parking space, then the coordinate information within the coordinate range of the safe parking space is marked as overlapping coordinates. If the coordinates of the obstacle are not within a safe parking space, then the corresponding parking space is a parking space available. By summing up the overlapping coordinates within the same safe parking space, the suspected obstacle area corresponding to that safe parking space can be obtained. If there are any suspected obstruction areas within the core area of ​​the safe parking space, it means that the corresponding parking space is not a parking space. If the suspected obstacle area exists only in the edge area, the corresponding parking space will be marked as a parking space to be evaluated.

[0010] Furthermore, the vehicle's basic information includes its length, width, height, steering angle range, and orientation at its current location.

[0011] Furthermore, the process of generating an automated parking route from the vehicle's location to a available parking space includes: When there is a parking space available, mark the two ends of the entrance to the available parking space, and generate a first entrance reference point and a second entrance reference point on the two ends and the side closest to the interior of the parking space. Based on the parking direction selected by the user, two parking points are generated on the corresponding side of the virtual vehicle; Based on the vehicle's steering angle range, generate an entry curve from the entry point to the first and second entry reference points; If the curvature of the parking curve is within the vehicle's turning angle range, it means that the corresponding parking curve can meet the vehicle's parking requirements, and the generated parking curve is recorded as the automatic parking route.

[0012] Furthermore, if the curvature of the entry curve is not within the vehicle's steering angle range, a standard entry curve is generated based on the first and second entry reference points, and the optimal entry point is set on the standard entry curve. A reference line is generated based on the optimal entry point, parallel to the line connecting the first and second entry reference points. Generate a trajectory line passing through the entry point based on the vehicle's current location and orientation at that location; Obtain the intersection of the trajectory line and the standard parking curve. If the intersection point is not between the optimal parking point and the entrance reference point, mark the intersection point as the parking point. Then generate the parking curve from the parking point to the first entrance reference point and the second entrance reference point, and mark the trajectory line and the generated parking curve as the automatic parking route. If the intersection point is between the optimal parking point and the entrance reference point, adjust the vehicle's orientation so that the intersection point of the trajectory line and the standard parking curve coincides with the optimal parking point. The vehicle's route during the adjustment process, the trajectory line generated after the adjustment, and the parking curve between the optimal parking point and the first and second entrance reference points are marked as the automatic parking route. The total length of the automatic parking routes generated for each parking space is sorted from low to high, and the parking space with the highest length in the sort is selected as the target parking space. The car is then parked according to the automatic parking route corresponding to the target parking space.

[0013] Furthermore, when there are only parking spaces to be evaluated, the edge area of ​​the safe parking space corresponding to the parking space is redefined based on the suspected obstacle area in the edge area within the safe parking space. At the same time, a new safe parking space of the same size as the original safe parking space is generated based on the redefined edge area. The generated new safe parking space is evaluated to determine whether there is a suspected obstacle area within the safe parking space. If there is, it exists only in the edge area of ​​the safe parking space, and the above process is repeated. If no such space exists, a parking space is automatically drawn based on the newly generated safe parking space, and a corresponding automatic parking route is generated based on the automatically drawn parking space.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. By combining visual detection and sensor technology, a parking area simulation space is constructed, and a corresponding safe parking space is generated within the parking area simulation space based on the vehicle's basic information. The identified suspected obstacles are simulated within the parking area simulation space to determine the impact of the suspected obstacles on the parking area simulation space, thereby assisting in determining whether there is a collision risk during the automatic parking process. 2. When there are suspected obstacles in the edge area of ​​a parking space, the parking space is automatically drawn to avoid the suspected obstacles in the edge area, thereby generating a new parking space. This allows for safe parking even when there is irregular parking on both sides of the parking space. It does not rely on real-time infrared detection, but can complete automatic parking based solely on the simulated space of the pre-built parking area and the distribution of suspected obstacles, thus reducing the computing power consumption in the automatic parking process. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0017] like Figure 1 As shown, an intelligent collision avoidance method for automatic parking of vehicles includes the following steps: The system acquires environmental image information of the vehicle's location based on the automatic parking command, constructs a simulated parking area based on the acquired environmental image information, and identifies the parking spaces contained within the environmental image information. By acquiring vehicle sensor data, information about obstacles around the vehicle's location can be obtained. The system assesses the parking feasibility of each parking space identified within the simulated parking area to determine if a parking space is available. If a parking space is available, it generates an automatic parking route from the vehicle's location to that space.

[0018] It should be further explained that, in the specific implementation process, the process of obtaining environmental image information of the vehicle's location based on the automatic parking command includes: The user selects an automatic parking command based on the vehicle's location. The automatic parking command includes a first command, a second command, a third command, and a fourth command. It should be noted that the first command, the second command, the third command, and the fourth command correspond to different parking directions of the vehicle, specifically rear parking, front parking, left-side parking, and right-side parking. Based on the automatic parking command selected by the user, images are acquired for the vehicle in the corresponding parking direction to obtain the corresponding environmental image information. It should be noted that the vehicle is equipped with multiple cameras, and the camera corresponding to the parking direction is selected to capture environmental image information according to the automatic parking command.

[0019] It should be further explained that, in the specific implementation process, the process of constructing a simulated parking area space based on the acquired environmental image information and identifying the parking spaces contained within the environmental image information includes: Construct a three-dimensional spatial coordinate system based on the vehicle's location; The obtained environmental image information is rasterized, and the rasterized environmental image information is mapped to the corresponding position in the three-dimensional spatial coordinate system; The rasterized environmental image information is converted to grayscale to obtain the corresponding grayscale image. Feature extraction is performed on the grayscale image to obtain the edge features of the parking space, and a simulated parking space is generated based on the corresponding position of the obtained edge features in the three-dimensional spatial coordinate system, thereby completing the construction of the simulated parking area space; in the specific implementation process, identifying parking spaces in environmental image information is a common technical means used by those skilled in the art, and will not be elaborated here. It should be noted that in the specific implementation process, when the edge features of the identified parking space cannot be used to construct a complete parking space, a corresponding simulated parking space is generated based on the location of the identified edge features and the size standard of the parking space.

[0020] It should be further explained that, in the specific implementation process, the process of obtaining obstacle information around the vehicle's location by acquiring vehicle sensor data includes: Data acquisition terminals are installed at various locations on the vehicle to acquire corresponding vehicle sensor data. It should be noted that acquiring vehicle sensor data by the data acquisition terminals means that the data acquisition terminals obtain the vehicle sensor data corresponding to each data acquisition terminal by emitting infrared signals and obtaining the reflected infrared signals. Based on the vehicle sensor data obtained from each data acquisition terminal, the location and distance of each obstacle are determined, and corresponding simulated obstacles are generated at their corresponding positions in the three-dimensional spatial coordinate system. The coordinates of each simulated obstacle in the three-dimensional coordinate system are labeled, and the coordinates of all simulated obstacles in the three-dimensional coordinate system are summarized to obtain the obstacle information at the vehicle's location.

[0021] It should be further explained that, in the specific implementation process, the process of assessing the parking feasibility of each parking space identified within the simulated parking area to determine whether a parking space exists includes: Based on the vehicle's basic information, a virtual vehicle is generated at the vehicle's current location, and a corresponding safe parking space is generated at the location of each simulated parking space. The coordinate range of the safe parking space is obtained. The vehicle's basic information includes the vehicle's length, width, height, vehicle steering angle range, and the vehicle's orientation at its current location. The safe parking space is divided into peripheral and core areas; The coordinate range of each safe parking space is compared with the coordinate information of all simulated obstacles in the obstacle information. If the coordinate information in the obstacle information is within the coordinate range of any safe parking space, then the coordinate information within the coordinate range of the safe parking space is marked as overlapping coordinates. If the coordinates of the obstacle are not within a safe parking space, then the corresponding parking space is a parking space available. By summing up the overlapping coordinates within the same safe parking space, the suspected obstacle area corresponding to that safe parking space can be obtained. If there are any suspected obstruction areas within the core area of ​​the safe parking space, it means that the corresponding parking space is not a parking space. If the suspected obstacle area exists only in the edge area, the corresponding parking space will be marked as a parking space to be evaluated.

[0022] It should be further explained that, in the specific implementation process, the process of generating the automatic parking route from the vehicle's location to the available parking space includes: When there is a parking space available, mark the two ends of the entrance to the available parking space, and generate a first entrance reference point and a second entrance reference point on the two ends and the side closest to the interior of the parking space. Based on the parking direction selected by the user, two entry points are generated on the corresponding side of the virtual vehicle, and the entry points are associated with the first entrance reference point and the second entrance reference point respectively. Based on the vehicle's steering angle range, generate an entry curve from the entry point to the first and second entry reference points; If the curvature of the parking curve is within the vehicle's turning angle range, it means that the corresponding parking curve can meet the vehicle's parking requirements, and the generated parking curve is recorded as the automatic parking route. If the curvature of the entry curve is not within the vehicle's steering angle range, a standard entry curve is generated based on the first and second entry reference points, and an optimal entry point is set on the standard entry curve. It should be noted that the standard entry curve and the optimal entry point are preset by the system and depend on the vehicle's basic information. A reference line is generated based on the optimal entry point, parallel to the line connecting the first and second entry reference points. A trajectory line passing through the entry point is generated based on the vehicle's current location and its orientation at that location; the trajectory line is a straight line. Obtain the intersection point of the trajectory line and the standard inbound curve, and obtain the relative positional relationship between the intersection point and the optimal inbound point; If the intersection point is not between the optimal parking point and the entrance reference point, then the intersection point is marked as the parking point, and a parking curve from the parking point to the first entrance reference point and the second entrance reference point is generated. The trajectory line and the generated parking curve are marked as automatic parking routes. If the intersection point is between the optimal parking point and the entrance reference point, adjust the vehicle's orientation so that the intersection point of the trajectory line and the standard parking curve coincides with the optimal parking point. The vehicle's route during the adjustment process, the trajectory line generated after the adjustment, and the parking curve between the optimal parking point and the first and second entrance reference points are marked as the automatic parking route. The total length of the automatic parking routes generated for each parking space is sorted from low to high, and the parking space with the highest length in the sort is selected as the target parking space. The car is then parked according to the automatic parking route corresponding to the target parking space.

[0023] In another embodiment of the present invention, when there are no available parking spaces and only parking spaces to be evaluated, the edge area of ​​the safe parking space corresponding to the parking space is redefined based on the suspected obstacle area in the edge area of ​​the safe parking space. At the same time, a new safe parking space of the same size as the safe parking space is generated based on the redefined edge area, and the generated new safe parking space is evaluated to determine whether there is a suspected obstacle area in the safe parking space. If there is, it exists only in the edge area of ​​the safe parking space, and the above process is repeated. If no such space exists, a parking space is automatically drawn based on the newly generated safe parking space, and a corresponding automatic parking route is generated based on the automatically drawn parking space.

[0024] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications or equivalent substitutions made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An intelligent collision avoidance method for automatic parking of vehicles, characterized in that, Includes the following steps: The system acquires environmental image information of the vehicle's location based on the automatic parking command, constructs a simulated parking area based on the acquired environmental image information, and identifies the parking spaces contained within the environmental image information. By acquiring vehicle sensor data, information about obstacles around the vehicle's location can be obtained. The system assesses the parking feasibility of each parking space identified within the simulated parking area to determine if a parking space is available. If a parking space is available, it generates an automatic parking route from the vehicle's location to that space.

2. The intelligent collision avoidance method for automatic parking of vehicles according to claim 1, characterized in that, The process of acquiring environmental image information about the vehicle's location based on automatic parking instructions includes: The user selects an automatic parking command based on the vehicle's location. The automatic parking command includes a first command, a second command, a third command, and a fourth command. Based on the automatic parking command selected by the user, images of the vehicle in the corresponding parking direction are acquired to obtain the corresponding environmental image information.

3. The intelligent collision avoidance method for automatic parking of vehicles according to claim 2, characterized in that, The process of constructing a simulated parking area space based on the acquired environmental image information and identifying the parking spaces contained within the environmental image information includes: Construct a three-dimensional spatial coordinate system based on the vehicle's location; The obtained environmental image information is rasterized, and the rasterized environmental image information is mapped to the corresponding position in the three-dimensional spatial coordinate system; The rasterized environmental image information is converted to grayscale to obtain the corresponding grayscale image. Feature extraction is performed on the grayscale image to obtain the edge features of the parking space, and a simulated parking space is generated based on the corresponding position of the obtained edge features in the three-dimensional spatial coordinate system, thereby completing the construction of the simulated parking area space.

4. The intelligent collision avoidance method for automatic parking of vehicles according to claim 3, characterized in that, The process of obtaining obstacle information around the vehicle's location by acquiring vehicle sensor data includes: Data acquisition terminals are installed at various locations on the vehicle to acquire corresponding vehicle sensor data. Based on the vehicle sensor data obtained from each data acquisition terminal, the location and distance of each obstacle are determined, and corresponding simulated obstacles are generated at their corresponding positions in the three-dimensional spatial coordinate system. The coordinates of each simulated obstacle in the three-dimensional coordinate system are labeled, and the coordinates of all simulated obstacles in the three-dimensional coordinate system are summarized to obtain the obstacle information at the vehicle's location.

5. The intelligent collision avoidance method for automatic parking of vehicles according to claim 4, characterized in that, The process of assessing the parking feasibility of each identified parking space within the simulated parking area to determine the existence of available parking spaces includes: Based on the vehicle's basic information, a virtual vehicle is generated at the vehicle's current location, and a corresponding safe parking space is generated at the location of each simulated parking space. The coordinate range of the safe parking space is then obtained. The safe parking space is divided into peripheral and core areas; The coordinate range of each safe parking space is compared with the coordinate information of all simulated obstacles in the obstacle information. If the coordinate information in the obstacle information is within the coordinate range of any safe parking space, then the coordinate information within the coordinate range of the safe parking space is marked as overlapping coordinates. If the coordinates of the obstacle are not within a safe parking space, then the corresponding parking space is a parking space available. By summing up the overlapping coordinates within the same safe parking space, the suspected obstacle area corresponding to that safe parking space can be obtained. If there are any suspected obstruction areas within the core area of ​​the safe parking space, it means that the corresponding parking space is not a parking space. If the suspected obstacle area exists only in the edge area, the corresponding parking space will be marked as a parking space to be evaluated.

6. The intelligent collision avoidance method for automatic parking of vehicles according to claim 5, characterized in that, The vehicle's basic information includes its length, width, height, steering angle range, and orientation at its current location.

7. The intelligent collision avoidance method for automatic parking of vehicles according to claim 6, characterized in that, The process of generating an automated parking route from the vehicle's location to an available parking space includes: When there is a parking space available, mark the two ends of the entrance to the available parking space, and generate a first entrance reference point and a second entrance reference point on the two ends and the side closest to the interior of the parking space. Based on the parking direction selected by the user, two parking points are generated on the corresponding side of the virtual vehicle; Based on the vehicle's steering angle range, generate an entry curve from the entry point to the first and second entry reference points; If the curvature of the parking curve is within the vehicle's turning angle range, it means that the corresponding parking curve can meet the vehicle's parking requirements, and the generated parking curve is recorded as the automatic parking route.

8. The intelligent collision avoidance method for automatic parking of a vehicle according to claim 7, characterized in that, If the curvature of the entry curve is not within the vehicle's steering angle range, a standard entry curve is generated based on the first and second entry reference points, and the optimal entry point is set on the standard entry curve. A reference line is generated based on the optimal entry point, parallel to the line connecting the first and second entry reference points. Generate a trajectory line passing through the entry point based on the vehicle's current location and orientation at that location; Obtain the intersection of the trajectory line and the standard parking curve. If the intersection point is not between the optimal parking point and the entrance reference point, mark the intersection point as the parking point. Then generate the parking curve from the parking point to the first entrance reference point and the second entrance reference point, and mark the trajectory line and the generated parking curve as the automatic parking route. If the intersection point is between the optimal parking point and the entrance reference point, adjust the vehicle's orientation so that the intersection point of the trajectory line and the standard parking curve coincides with the optimal parking point. The vehicle's route during the adjustment process, the trajectory line generated after the adjustment, and the parking curve between the optimal parking point and the first and second entrance reference points are marked as the automatic parking route. The total length of the automatic parking routes generated for each parking space is sorted from low to high, and the parking space with the highest length in the sort is selected as the target parking space. The car is then parked according to the automatic parking route corresponding to the target parking space.

9. The intelligent collision avoidance method for automatic parking of vehicles according to claim 8, characterized in that, When there are only parking spaces to be evaluated, the edge area of ​​the safe parking space corresponding to the parking space is redefined based on the suspected obstacle area in the edge area of ​​the safe parking space. At the same time, a new safe parking space of the same size as the original safe parking space is generated based on the redefined edge area. The generated new safe parking space is evaluated to determine whether there is a suspected obstacle area in the safe parking space. If there is, it exists only in the edge area of ​​the safe parking space, and the above process is repeated. If no such space exists, a parking space is automatically drawn based on the newly generated safe parking space, and a corresponding automatic parking route is generated based on the automatically drawn parking space.

Images