CN111845723A

CN111845723A - Full-automatic parking method and system

Info

Publication number: CN111845723A
Application number: CN202010776609.3A
Authority: CN
Inventors: 刘菲菲; 赵阳; 史钰玲; 周健; 张艳玲
Original assignee: Beijing Siwei Zhi Lian Technology Co ltd
Current assignee: Beijing Siwei Zhi Lian Technology Co ltd
Priority date: 2020-08-05
Filing date: 2020-08-05
Publication date: 2020-10-30

Abstract

The invention provides a full-automatic parking method and a system, which comprises the steps of obtaining space parking space data and obstacle data identified by an ultrasonic radar, and all-round parking space data and obstacle data identified by a fisheye camera; fusing based on a data fusion algorithm to obtain parking space fusion data and barrier space fusion data; planning by combining the parking space fusion data and the obstacle space fusion data according to a free trajectory planning algorithm to obtain a parking trajectory; and obtaining the moving position information of the current vehicle, and controlling the vehicle to run along the parking track in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm. In the scheme, the parking track is determined by calculating the parking space fusion data and the barrier space fusion data obtained by fusion by using a free track planning algorithm. And then controlling the vehicle to run along the parking track by using a high-precision positioning algorithm and a high-precision track tracking algorithm. The parking space recognition and automatic parking control accuracy is improved, and the method is suitable for more parking scenes.

Description

Full-automatic parking method and system

Technical Field

The invention relates to the technical field of automatic parking, in particular to a full-automatic parking method and system.

Background

At present, the existing vehicles usually use ultrasonic waves or fisheye cameras, and are combined with a full-automatic parking area controller to realize automatic parking.

In the prior art, the parking space is identified by the ultrasonic waves or the fisheye cameras respectively due to the identification limitations of the ultrasonic waves and the fisheye cameras, the identification precision is not high, and the track algorithm is intelligently applied to limited scenes. In the prior art, a PID control mode is generally adopted in vehicle parking control, and the control precision is low.

In summary, the existing automatic parking mode has the problems of few adaptive scenes, low recognition accuracy and low automatic parking control accuracy.

Disclosure of Invention

In view of this, embodiments of the present invention provide a full-automatic parking method and system, so as to solve the problems of few adaptive scenes, low recognition accuracy, and low automatic parking control accuracy in the existing automatic parking manner.

In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions:

the first aspect of the embodiment of the invention discloses a full-automatic parking method, which comprises the following steps:

acquiring spatial parking space data identified by the ultrasonic radar and all-round parking space data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain parking space fusion data;

acquiring obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain obstacle space fusion data;

according to a free trajectory planning algorithm, combining the parking space fusion data and the barrier space fusion data to carry out trajectory free planning to obtain a parking trajectory;

and acquiring the moving position information of the current vehicle in real time, and controlling the vehicle to run along the parking track in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm.

Optionally, the space parking space data that the ultrasonic radar discerned and the look around parking space data that the fisheye camera discerned are obtained, fuse based on the data fusion algorithm, obtain parking space fusion data, include:

acquiring space parking space characteristics identified based on the ultrasonic radar;

obtaining line parking space characteristics and parking space identification characteristics identified based on the fisheye camera;

and fusing the space parking space characteristic, the line parking space characteristic and the parking space identification characteristic based on a data fusion algorithm to obtain parking space fusion data.

Optionally, the obtaining obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain obstacle space fusion data includes:

obtaining the line parking space characteristics and the obstacle characteristics identified by the fisheye camera;

determining the obstacle feature identified by the ultrasonic radar according to the position information of the line-parking space feature;

and fusing the obstacle features identified by the fisheye camera and the obstacle features identified by the ultrasonic radar based on a data fusion algorithm to obtain obstacle space fusion data.

Optionally, the freely planning a trajectory by combining the parking space fusion data and the obstacle space fusion data according to a free trajectory planning algorithm to obtain a parking trajectory, including:

determining safe space data based on the parking space fusion data and the barrier space fusion data;

acquiring position information and vehicle operation parameters of a current vehicle;

and planning a free track according to the safe space data, the position information of the current vehicle and the vehicle running parameters to obtain a parking track.

Optionally, the obtaining of the moving position information of the current vehicle in real time, and controlling the vehicle to run along the parking trajectory in real time based on a high-precision positioning algorithm and a high-precision trajectory tracking algorithm includes:

controlling the vehicle to run along the parking track;

acquiring the moving position information of the current vehicle in real time, and determining whether the vehicle runs along the parking track based on a high-precision positioning algorithm and a high-precision track tracking algorithm;

and if the vehicle deviates from the parking track, adjusting the running track of the vehicle to enable the vehicle to run along the parking track.

Optionally, the method further includes:

detecting whether a moving object appears in a preset range of the vehicle in the process of controlling the vehicle to run along the parking track in real time based on the fisheye camera;

if the living body characteristic information of the moving object appears, acquiring the living body characteristic information of the moving object, carrying out living body detection on the living body characteristic information based on a living body detection algorithm, and outputting a detection result;

and/or the presence of a gas in the gas,

detecting whether a moving object appears in a blind area of the vehicle and/or at the rear part of the vehicle in the process of controlling the vehicle to run along the parking track or not in real time based on the ultrasonic radar;

and if so, generating and outputting safety prompt information.

A second aspect of the embodiment of the present invention discloses a full-automatic parking system, including:

the ultrasonic radar is arranged around the vehicle and used for acquiring space parking space data and obstacle data;

the fisheye cameras are arranged on the periphery of the vehicle and used for acquiring all-round parking space data and barrier data;

the controller is used for acquiring spatial parking space data identified by the ultrasonic radar and all-round parking space data identified by the fisheye camera, and fusing the spatial parking space data and all-round parking space data based on a data fusion algorithm to obtain parking space fusion data; acquiring obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain obstacle space fusion data; according to a free trajectory planning algorithm, combining the parking space fusion data and the barrier space fusion data to carry out trajectory free planning to obtain a parking trajectory; and acquiring the moving position information of the current vehicle in real time, and controlling the vehicle to run along the parking track in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm.

Optionally, the controller is configured to perform free trajectory planning by combining the parking space fusion data and the obstacle space fusion data according to a free trajectory planning algorithm to obtain a parking trajectory, and is specifically configured to:

determining safe space data based on the parking space fusion data and the barrier space fusion data; acquiring position information and vehicle operation parameters of a current vehicle; and planning a free track according to the safe space data, the position information of the current vehicle and the vehicle running parameters to obtain a parking track.

Optionally, the controller is configured to obtain the moving position information of the current vehicle in real time, and control the vehicle to run along the parking trajectory in real time based on a high-precision positioning algorithm and a high-precision trajectory tracking algorithm, and is specifically configured to:

controlling the vehicle to run along the parking track; acquiring the moving position information of the current vehicle in real time, and determining whether the vehicle runs along the parking track based on a high-precision positioning algorithm and a high-precision track tracking algorithm; and if the vehicle deviates from the parking track, adjusting the running track of the vehicle to enable the vehicle to run along the parking track.

Optionally, the controller is further configured to detect, in real time, based on the fisheye camera, whether a moving object appears in a preset range of the vehicle in a process of controlling the vehicle to run along the parking track in real time; if the living body characteristic information of the moving object appears, acquiring the living body characteristic information of the moving object, carrying out living body detection on the living body characteristic information based on a living body detection algorithm, and outputting a detection result;

and/or the presence of a gas in the gas,

the controller is also used for detecting and controlling whether a moving object appears in a blind area of the vehicle and/or at the rear part of the vehicle in the process of the vehicle running along the parking track or not in real time based on the ultrasonic radar; and if so, generating and outputting safety prompt information.

Based on the full-automatic parking method and the system provided by the embodiment of the invention, the method comprises the following steps: acquiring spatial parking space data identified by an ultrasonic radar and all-round parking space data identified by a fisheye camera, and fusing based on a data fusion algorithm to obtain parking space fusion data; acquiring obstacle data identified by an ultrasonic radar and obstacle data identified by a fisheye camera, and fusing based on a data fusion algorithm to obtain obstacle space fusion data; according to a free trajectory planning algorithm, combining the parking space fusion data and the obstacle space fusion data to carry out trajectory free planning to obtain a parking trajectory; and acquiring the moving position information of the current vehicle in real time, and controlling the vehicle to run along the parking track in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm. In the embodiment of the invention, the ultrasonic radar and the fisheye cameras around the vehicle are arranged to look around the vehicle to determine the space parking space data, look around parking space data and obstacle data; and the spatial parking space data, the all-round parking space data and the barrier data are fused by using a data fusion algorithm so as to determine the parking space fusion data and the barrier space fusion data. And then, carrying out free trajectory planning on the parking space fusion data and the obstacle space fusion data by using a free trajectory planning algorithm to determine a parking trajectory. And finally, controlling the vehicle to run along the parking track in real time by using a high-precision positioning algorithm and a high-precision track tracking algorithm. By the aid of the method, the parking space identification precision and the automatic parking control precision can be improved in the automatic parking process of the vehicle, so that the full-automatic parking system is suitable for more parking scenes.

Drawings

In order to more clearly illustrate 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 described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a fully automatic parking system according to an embodiment of the present invention;

fig. 2 is an application architecture diagram of a fully automatic parking system according to an embodiment of the present invention;

FIG. 3 is a flow chart illustrating a fully automatic parking method according to an embodiment of the present invention;

FIG. 4 is a flow chart illustrating another method for fully automatic parking according to an embodiment of the present invention;

fig. 5 is a flowchart illustrating another fully automatic parking method according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The full-automatic parking system disclosed by the invention can be used for supporting the full-automatic parking operation of vehicles on indoor and outdoor vertical, parallel and diagonal parking spaces and various special-shaped parking spaces.

In the embodiment of the invention, the ultrasonic radar and the fisheye cameras around the vehicle are arranged to look around the vehicle to determine the space parking space data, look around parking space data and obstacle data; and the spatial parking space data, the all-round parking space data and the barrier data are fused by using a data fusion algorithm so as to determine the parking space fusion data and the barrier space fusion data. And then, carrying out free trajectory planning on the parking space fusion data and the obstacle space fusion data by using a free trajectory planning algorithm to determine a parking trajectory. And finally, controlling the vehicle to run along the parking track in real time by using a high-precision positioning algorithm and a high-precision track tracking algorithm. By the aid of the method, the parking space identification precision and the automatic parking control precision can be improved in the automatic parking process of the vehicle, so that the full-automatic parking system is suitable for more parking scenes.

Referring to fig. 1, a schematic structural diagram of a full-automatic parking system according to an embodiment of the present invention is shown, where the full-automatic parking system includes an ultrasonic radar 10, a fisheye camera 20, and a controller 30.

The ultrasonic radar 10 is arranged around the vehicle and used for acquiring space parking space data and obstacle data.

The number of the ultrasonic radars 10 is plural, and the plurality of ultrasonic radars 10 includes the ultrasonic radar 101, and the ultrasonic radars 102.

Wherein the value range of N is a positive integer greater than or equal to 1.

Specifically, each ultrasonic radar detects the surroundings of the vehicle to determine the recognized space parking space data and obstacle data.

It should be noted that the space and parking space data includes space and parking space characteristics.

The obstacle data includes objects that can obstruct the vehicle from parking, such as a parked vehicle in the line parking space and an ice cream cone parked in the line parking space.

The fisheye cameras 20 are arranged around the vehicle and used for acquiring all-round parking space data and obstacle data.

It should be noted that the number of the fisheye cameras 20 is plural, and the plural fisheye cameras 20 include fisheye cameras 201, and the fisheye cameras 202.

Wherein, the value range of M is a positive integer which is more than or equal to 1.

Specifically, each fisheye camera shoots the periphery of the vehicle, and the surrounding parking space data and the obstacle data are determined according to each shot picture.

It should be noted that the look-around parking space data includes a line parking space feature and a parking space identification feature.

This controller 30 is connected with ultrasonic radar 10 and fisheye camera 20 respectively, and is specific, and controller 30 is connected with ultrasonic radar 101 respectively, ultrasonic radar 102.

Note that the controller 30 may be a Chip on Chip (SOC) Chip.

The controller 30 is configured to obtain spatial parking space data identified by the ultrasonic radar 10 and all-round parking space data identified by the fisheye camera 20, and perform fusion based on a data fusion algorithm to obtain parking space fusion data; acquiring obstacle data identified by the ultrasonic radar 10 and obstacle data identified by the fisheye camera 20, and fusing based on a data fusion algorithm to obtain obstacle space fusion data; according to a free trajectory planning algorithm, combining the parking space fusion data and the obstacle space fusion data to carry out trajectory free planning to obtain a parking trajectory; and acquiring the moving position information of the current vehicle in real time, and controlling the prolonged parking track operation of the vehicle in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm.

It should be noted that the space and parking space data includes space and parking space characteristics. The around view parking space data comprises line parking space characteristics and parking space identification characteristics.

In a specific implementation, the controller 30 obtains spatial parking space data recognized by each of the ultrasonic radars 10 disposed on the vehicle and look-around parking space data recognized by each of the fisheye cameras 20. And fusing the space parking place data and the all-round parking place data by using a data fusion algorithm to obtain parking place fusion data. The obstacle data recognized by each ultrasonic radar 10 provided on the vehicle and the obstacle data recognized by each fisheye camera 20 provided on the vehicle are acquired at the same time. And fusing the obstacle data recognized by the ultrasonic radar 10 and the obstacle data recognized by the fisheye camera 20 by using a data fusion algorithm to obtain obstacle space fusion data. And then, freely planning the parking space fusion data and the obstacle space fusion data by using a free planning algorithm to obtain a parking track. And finally, acquiring the moving position of the current vehicle in real time through a Global Positioning System (GPS), determining the relative position between the vehicle and the parking space by using a high-precision positioning algorithm, and controlling the vehicle to park according to the parking track by using a high-precision track tracking algorithm so as to control the vehicle to run along the parking track in real time.

It should be noted that the data fusion algorithm is to fuse different feature data in the same region identified by each sensor, so as to determine a more accurate parking space and an available safety space around the parking space by using the correlation of the different feature data and the complementarity of the information entropy.

The obstacle space fusion data refers to a space established by using a parking space obstacle, namely, a parking space which can be used for parking, so as to further determine a safe space for parking.

The free planning algorithm is used for planning a parking route according to the relationship between a vehicle and a parking space and by considering the motion characteristics of the vehicle.

The parking trajectory refers to an optimal path for the current vehicle to park.

The high-precision positioning algorithm is used for enabling the full-automatic parking system to determine the relative position between the vehicle and the parking space more accurately in the parking process.

The high-precision trajectory tracking algorithm is used for accurately controlling the vehicle to park according to a parking route when the system parks.

In the embodiment of the invention, the current vehicle position can be determined by a Beidou satellite navigation system or a Glonass positioning system besides a GPS.

Alternatively, the positioning system shown above may be provided in a navigation device inside a vehicle.

Optionally, based on the full-automatic parking system shown in fig. 1, the controller 30 is configured to obtain spatial parking space data identified by the ultrasonic radar and all-round parking space data identified by the fisheye camera, and perform fusion based on a data fusion algorithm to obtain parking space fusion data, and is specifically configured to: acquiring space parking space characteristics identified based on an ultrasonic radar; obtaining line parking space characteristics and parking space identification characteristics identified based on a fisheye camera; and fusing the space parking space characteristic, the line parking space characteristic and the parking space identification characteristic based on a data fusion algorithm to obtain parking space fusion data.

In the embodiment of the invention, the ultrasonic radar 10 uses the vehicle itself as an origin, and transmits ultrasonic waves to all directions around the vehicle through the ultrasonic transmitting device of the ultrasonic radar 10 to detect the parking spaces around the vehicle, when the ultrasonic waves are propagated in the air and touch the railings around the parking spaces, the reflected ultrasonic waves are returned to the ultrasonic radar 10 through the ultrasonic receiving device, the ultrasonic radar 10 records the time from the transmission of the ultrasonic waves to the return, and the distance between the railings around the parking spaces and the vehicle is calculated, so that the space parking space characteristics such as the size of each available parking space, the distance between the railing and the vehicle, the entrance direction and the like are determined.

The ultrasonic radar 10 incorporates an ultrasonic transmitter and an ultrasonic receiver.

The fisheye camera 20 looks around by taking itself as an origin to take pictures of all directions around the vehicle; and recognizing each photo, and determining the line parking space characteristics corresponding to the line segments around the parking space, and the parking space identification characteristics corresponding to the parking direction arrow and the disabled person identification.

In one embodiment, the controller 30 obtains spatial parking characteristics such as the size of each available parking space, the distance from the vehicle, and the entrance direction, which are identified based on the ultrasonic radar. The line parking space characteristics corresponding to the line segments around the parking space recognized by the fisheye camera 20 and the parking space identification characteristics corresponding to the parking direction arrow and the disabled person identification are obtained. The size of each available parking space recognized by the ultrasonic radar 10, the distance between the parking space and the vehicle, the entrance direction and other space parking space characteristics, the line parking space characteristics corresponding to the line segments around the parking space recognized by the fisheye camera 20, and the parking space identification characteristics corresponding to the parking direction arrow and the disabled person identification are fused to obtain parking space fusion data.

It should be noted that the spatial parking space characteristics refer to geometric characteristics such as the position, shape, size, etc. of the parking space in the space, and the spatial relationship with the adjacent parking spaces.

The line parking space characteristic refers to a line segment for distinguishing parking space positions. The sign identification feature is used to indicate the available objects for the parking space, the direction of parking, and the like.

The parking space fusion data comprise position information of available parking spaces and identification information in the parking spaces.

The size of the available parking space, the distance between the available parking space and the vehicle, line segments around the parking space and identification information in the parking space comprise a parking direction arrow and a disabled person identification.

In the embodiment of the invention, the ultrasonic radar and the fisheye cameras around the vehicle are arranged to look around the vehicle, so that the space parking space data and the look around parking space data are determined; and the data fusion algorithm is utilized to fuse the space parking spot data and the all-round parking spot data so as to determine the position information of the available parking spots and the identification information in the parking spots. So as to subsequently determine the obstacle space fusion data, thereby presetting a parking track. In the process of automatic parking of the vehicle, the accuracy of parking space identification is improved, so that the full-automatic parking system is suitable for more parking scenes.

Optionally, based on the full-automatic parking system shown in fig. 1, the controller 30 is configured to obtain obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and perform fusion based on a data fusion algorithm to obtain obstacle space fusion data, and is specifically configured to: obtaining the line parking space characteristics and the barrier characteristics identified by the fisheye camera; determining the obstacle characteristics identified by the ultrasonic radar according to the position information of the line-parking space characteristics; and fusing the obstacle features identified by the fisheye camera and the obstacle features identified by the ultrasonic radar based on a data fusion algorithm to obtain obstacle space fusion data.

In the embodiment of the invention, the line parking space characteristics refer to line segments around the parking space, and the obstacle characteristics further comprise a signboard, a railing of the parking space, vehicles around the vehicles and the like.

The fisheye camera 20 looks around by taking itself as an origin to take pictures of all directions around the vehicle; and recognizing each photo, and determining the line parking space characteristics corresponding to the line segments around the parking space, the signboard, the railing of the parking space, and the obstacle characteristics of the obstacles around the vehicle, such as the vehicle and the like.

In a specific implementation, the controller 30 obtains the line segment around the parking space, the signboard, the railing of the parking space, and the vehicle around the vehicle, which are recognized based on the fisheye camera 20. Ultrasonic waves are transmitted into line segments around the parking space through an ultrasonic transmitting device of the ultrasonic radar 10 so as to detect obstacles in the parking space except the line segments around the parking space. When the ultrasonic waves propagate in the air and collide with an obstacle to return to the ultrasonic radar 10, the ultrasonic radar 10 records the time from the transmission of the ultrasonic waves to the return, and calculates the distance between each direction of the obstacle and the vehicle, thereby determining the obstacle characteristics. And determining barrier space fusion data by using the relativity and the complementarity of the information entropy of the signboards, the railings of the parking spaces, the vehicles around the vehicles, the parked vehicles and the ice cream cylinders parked in the line parking spaces in the same area.

In the embodiment of the invention, the ultrasonic radar and the fisheye cameras around the vehicle are arranged to look around the vehicle, so that the characteristics of obstacles around the vehicle are determined; and fusing the obstacle features by using a data fusion algorithm to determine obstacle space fusion data. So as to plan the parking trajectory according to the free trajectory planning algorithm. In the process of automatic parking of the vehicle, the accuracy of parking space identification is improved, so that the full-automatic parking system is suitable for more parking scenes.

Optionally, based on the full-automatic parking system shown in fig. 1, according to a free trajectory planning algorithm, a trajectory free planning is performed by combining the parking space fusion data and the obstacle space fusion data, so as to obtain the controller 30 of the parking trajectory, which is specifically configured to: determining safe space data based on the parking space fusion data and the barrier space fusion data; acquiring position information and vehicle operation parameters of a current vehicle; and carrying out free track planning according to the safe space data, the current position information of the vehicle and the vehicle running parameters to obtain a parking track.

In a specific implementation, the controller 30 performs calculation according to the position information of the available parking space, the identification information in the parking space, and the obstacle space fusion data to obtain the safe space data for available parking. The position information of the current vehicle is acquired in real time through a GPS, and the speed of the current vehicle is acquired through a vehicle controller. And performing free trajectory planning on the safe space data of available parking, the position information of the current vehicle and the speed of the current vehicle by using a free planning algorithm to obtain a parking trajectory.

It should be noted that the safe space data for available parking refers to the size of the space in which the vehicle can operate when parking.

The position information of the vehicle refers to a position in space, such as a position that can be identified with coordinates.

In the embodiment of the invention, the parking space fusion data and the obstacle space fusion data are subjected to free trajectory planning by using a free trajectory planning algorithm, so that a parking trajectory is determined. In the process of automatic parking of the vehicle, the accuracy of parking space identification is improved, so that the full-automatic parking system is suitable for more parking scenes.

Optionally, based on the full-automatic parking system shown in fig. 1, the controller 30 is configured to obtain the moving position information of the current vehicle in real time, and control the vehicle to run along the parking trajectory in real time based on a high-precision positioning algorithm and a high-precision trajectory tracking algorithm, and is specifically configured to: controlling the vehicle to run along the parking track; the method comprises the steps of acquiring the moving position information of a current vehicle in real time, and determining whether the vehicle runs along a parking track based on a high-precision positioning algorithm and a high-precision track tracking algorithm; and if the vehicle deviates from the parking track, adjusting the running track of the vehicle to enable the vehicle to run along the parking track.

In a specific implementation, the controller 30 controls the vehicle to move along the parking trajectory at a preset parking speed. And acquiring the moving position of the current vehicle in real time through a GPS. And determining the relative position between the vehicle and the parking space by using a high-precision positioning algorithm, and controlling the vehicle to park according to the parking track by using a high-precision track tracking algorithm so as to determine the actual parking track of the vehicle. And when the actual parking track of the vehicle is not consistent with the parking track, indicating that the vehicle deviates from the parking track, and adjusting the actual parking track of the vehicle to enable the actual parking track of the vehicle to be consistent with the parking track planned by the free planning algorithm. When the actual parking track of the vehicle is consistent with the parking track, the situation that the vehicle deviates from the parking track is shown, and the vehicle is continuously controlled to continue to run along the parking track.

It should be noted that the preset parking speed is set according to a plurality of experiments, or may be set by a person skilled in the art according to experience, and the embodiment of the present invention is not limited thereto.

In the embodiment of the invention, whether the vehicle runs along the parking track is determined by utilizing a high-precision positioning algorithm and a high-precision track tracking algorithm. And controlling the vehicle to run along the parking track in real time when the vehicle is determined not to deviate from running along the parking track. By the method, the automatic parking control precision can be realized in the automatic parking process of the vehicle, so that the full-automatic parking system is suitable for more parking scenes.

Optionally, based on the full-automatic parking system shown in fig. 1, the controller 30 is further configured to detect, in real time, whether a moving object appears in a preset range of the vehicle in a process of controlling the vehicle to run along the parking track in real time based on the fisheye camera; and if so, acquiring the living body characteristic information of the moving object, carrying out living body detection on the living body characteristic information based on a living body detection algorithm, and outputting a detection result.

In a specific implementation process, in the process that the controller 30 controls the vehicle to run along the parking track, the fisheye camera 20 looks around the vehicle with the fisheye camera as an original point in real time to shoot pictures of all directions around the vehicle within a preset range; and identifying each photo, and acquiring the outline characteristics of the moving object by technologies such as illumination compensation, illumination removal and the like according to each photo of the moving object when the moving object is identified. And detecting the living body detection of the outline characteristic by using a living body detection algorithm to determine the moving track of the moving object in real time, and comparing the moving track of the moving object with the parking track of the vehicle, namely determining whether the comparison result is within a preset error to obtain a detection result. And when the moving object is not identified, the vehicle is continuously controlled to continue to run along the parking track.

The detection result is used to indicate whether or not there is a safety situation in parking in the current parking trajectory. If the detection result is that the comparison result is within the preset error, the parking is safe according to the current parking track; if the detection result is that the comparison result is not within the preset error, the parking is unsafe according to the current parking track, and a safety prompt is output to the driver.

Optionally, when determining that the vehicle is unsafe to park according to the current parking trajectory, the controller 30 outputs a control instruction to control the vehicle to stop parking; and when the detection result is determined to be that the comparison result is within the preset error, outputting a control command to control the vehicle to restart and continue parking.

In the embodiment of the invention, the vehicle is controlled to run along the parking track in real time by utilizing a high-precision positioning algorithm and a high-precision track tracking algorithm. The automatic parking control method has the advantages that in the automatic parking process of the vehicle and the automatic parking control precision, the full-automatic parking system is suitable for more parking scenes.

Optionally, based on the full-automatic parking system shown in fig. 1, the controller 30 is further configured to detect in real time whether a moving object is present in a blind area of the vehicle and/or at the rear of the vehicle during the process of the vehicle moving along the parking track based on the ultrasonic radar; and if so, generating and outputting safety prompt information.

In a specific implementation, in the process that the controller 30 controls the vehicle to move along the parking track, the ultrasonic radar 10 uses the vehicle itself as an origin, and transmits ultrasonic waves to the blind area and/or the rear portion of the vehicle through the ultrasonic transmitting device of the ultrasonic radar 10 to detect whether a moving object is present in the blind area and/or the rear portion of the vehicle, and when receiving the ultrasonic waves reflected by the ultrasonic receiving device, determines that a moving object is present in the blind area and/or the rear portion of the vehicle, and outputs a language prompt message and/or outputs a graphic prompt message to prompt the driver. And when the ultrasonic wave reflected by the ultrasonic wave receiving device is not received, controlling the vehicle to continue to run along the parking track.

The vehicle blind area refers to an included angle position which cannot be seen by the left and right side rearview mirrors and the middle rearview mirror.

It should be noted that both the blind zone detection alarm algorithm and the crossing detection alarm algorithm are used for judging the current safety condition according to the environment around the vehicle in the parking process so as to give a safety prompt to the client.

Optionally, the controller 30 outputs a control instruction to control the vehicle to stop parking when it is determined that a moving object is present in the blind area and/or at the rear of the vehicle; and when the moving object in the blind area and/or the rear part of the vehicle is determined not to exist, outputting a control command to control the vehicle to restart and continue parking.

Optionally, the driver may also perform corresponding operations according to the language prompt information and/or the image-text prompt information.

In the embodiment of the invention, whether a moving object appears in a blind area of the vehicle and/or at the rear part of the vehicle is determined by utilizing a high-precision positioning algorithm and a high-precision track tracking algorithm, so that the vehicle can be controlled to run along a parking track in real time. Safety prompts may be output to the driver when it is determined that a moving object is present within the blind zone of the vehicle and/or at the rear of the vehicle. By the method, the precision of automatic parking control can be improved in the automatic parking process of the vehicle, so that the full-automatic parking system is suitable for more parking scenes.

Based on the fully-automatic parking system, the invention also shows a schematic structural diagram of the fully-automatic parking system, as shown in fig. 2.

The fully automatic parking system includes 4 fisheye cameras 20 and 12 ultrasonic radars 10. The controller 30 is an on-chip SOC30, and a main connector 2 for connecting the fisheye camera 20, the ultrasonic radar 10, and the SOC 30.

Specifically, the 4 fisheye cameras 20 include a fisheye camera 21, a fisheye camera 22, a fisheye camera 23, and a fisheye camera 24. The 12 ultrasonic radars 10 include an ultrasonic radar 11, an ultrasonic radar 12, an ultrasonic radar 13, an ultrasonic radar 14, an ultrasonic radar 15, an ultrasonic radar 16, an ultrasonic radar 17, an ultrasonic radar 18, an ultrasonic radar 19, an ultrasonic radar 110, an ultrasonic radar 111, and an ultrasonic radar 112.

In a specific implementation, the fisheye camera 21 is disposed at an arbitrary position right in front of the vehicle, the fisheye camera 22 is disposed at an arbitrary position on the right side of the vehicle, the fisheye camera 23 is disposed at an arbitrary position on the left side of the vehicle, and the fisheye camera 24 is disposed at an arbitrary position on the rear of the vehicle. The fisheye camera 21, the fisheye camera 22, the fisheye camera 23 and the fisheye camera 24 are respectively connected with the serial-parallel converter 1 through connectors Fakra.

The 12 ultrasonic radars are uniformly arranged at corresponding positions of the right front, the right side, the left side and the rear of the vehicle. The ultrasonic radar 11, the ultrasonic radar 12, the ultrasonic radar 13, the ultrasonic radar 14, the ultrasonic radar 15, the ultrasonic radar 16, the ultrasonic radar 17, the ultrasonic radar 18, the ultrasonic radar 19, the ultrasonic radar 110, the ultrasonic radar 111, and the ultrasonic radar 112 are connected to a serial-to-parallel converter (Deserializer1)190 through connectors, respectively.

Alternatively, the model number of the serial to parallel converter 190 may be MAX9286GTN/V + T.

Note that all the connectors are connected in series and connected to a Main connector (Main connector) 40.

Serial-to-parallel converter 1 is connected to on-chip SOC 30.

Alternatively, SOC30 may be model Telechipsc 8030.

And the SOC30 is used for acquiring spatial parking space data identified by each ultrasonic radar 10 arranged on the vehicle and all-round parking space data identified by each fisheye camera 20. And fusing the space parking place data and the all-round parking place data by using a data fusion algorithm to obtain parking place fusion data. The obstacle data recognized by each ultrasonic radar 10 provided on the vehicle and the obstacle data recognized by each fisheye camera 20 provided on the vehicle are acquired at the same time. And fusing the obstacle data recognized by the ultrasonic radar 10 and the obstacle data recognized by the fisheye camera 20 by using a data fusion algorithm to obtain obstacle space fusion data. And then, freely planning the parking space fusion data and the obstacle space fusion data by using a free planning algorithm to obtain a parking track. And finally, acquiring the moving position of the current vehicle in real time by using a high-precision positioning algorithm, determining the relative position between the vehicle and the parking space by using the high-precision positioning algorithm, and controlling the vehicle to park according to the parking track by using a high-precision track tracking algorithm so as to control the vehicle to run along the parking track in real time.

With continued reference to fig. 2, the fully automatic parking system further comprises: a decoder (Decode)50, a sequencer (Serializer)60, a Host (HU)70, an Ethernet (Ethernet)80, a first upgraded lan controller (CAN-FD)90 and a second upgraded lan controller (CAN-FD)100, a DC-DC chip (DC to DC)110, a power integrated circuit (PMIC)120, a non-volatile memory (SNOR)130, a flash controller (eMMC)140, a removable memory device (LPDDR4)150, a RESET module (RESET)160, a High-low Voltage module (High-low Voltage)170, and a battery test (battery test, B-DET) 180.

An input of decoder 50 is connected to an output of SOC30, an output of decoder 50 is connected to an input of sequencer 60, and an output of sequencer 60 is connected to host 70.

In a specific implementation, the decoder 50 is configured to decode the low-voltage differential signal sent by the SOC30, and send the decoded low-voltage differential signal to the serializer 60 through the MIPI-CSI of the mobile industry processor interface of the decoder 50, so that the serializer 60 processes the decoded low-voltage differential signal, and sends the processing result to the host 70 through the coaxial cable for display.

Alternatively, the decoder 50 may be of the type RAA278842 of Resa technologies RENeSAS, and the sequencer 60 may be of the type DS90UB 953-Q1.

Ethernet 80, first upgraded local area network controller 90, and second upgraded local area network controller 100 are communicatively coupled to SOC30 and main connector 40, respectively.

In a specific implementation, the ethernet 80 is connected to the SOC30 through a gigabit media independent interface RGMII, the ethernet 80 is connected to a connector, and the local area network 100BASE-T1 for operating at 100Mbps supports automotive applications through the main connector 40.

The first upgraded lan controller 90 is connected to the SOC30 via the transmit and receive CAN0 TX/RX interface of the CAN0 bus, and the first upgraded lan controller 90 is connected to the connector 40 via the positive or negative pole of the CAN0 bus for supporting automotive applications at flexible data rates.

The second upgraded lan controller 100 is connected to the SOC30 through the transmitting and receiving CAN1 TX/RX interface of CAN1, and the second upgraded lan controller 100 is connected to the connector 40 through the positive or negative pole of the CAN1 bus, which is dormant, for supporting automotive applications at flexible data rates.

Alternatively, the Ethernet 80 may be of the type RTL9000AN-VC available from Realtek, Swe, Inc., the first upgraded local area network controller 90 may be of the type TJA1042T, and the second upgraded local area network controller 100 may be of the type TJA 1043.

The DC-DC chip 110 is connected to the power IC 120, and the output terminal of the power IC 120 is connected via I₂The C bus is connected with the SOC30 and is used for providing electric energy for the full-automatic parking system.

The input terminal of the dc-dc chip 110 is connected to the main connector 40 for determining the battery power, and when the battery power is about to be exhausted, the battery power is displayed by the indicator light BATT built in the dc-dc chip 110 and the power ic 120.

Optionally, the model of the dc-dc chip 110 may be MPQ9840, and the model 120 of the power supply integrated circuit may be DA 9063L.

The non-volatile memory 130 is connected to the SOC30 through the secure data input output interface SDIO for storing secure data.

Optionally, the nonvolatile memory 130 is model MX25U1635FM1Q-13G from MACRONIX, Inc. of Wangchong electronics GmbH, and has a storage size of 16 Mb.

The flash controller 140 is connected to the SOC30 through the secure data input output interface SDIO for flashing the secure data.

Optionally, the flash controller 140 is model KLM8G1GESD-B04P4P019 from SAMSUNG, which stores a size of 8 gigabytes GB.

The mobile storage device 150 is connected with the SOC30 through an External memory interface External memory I/F, and is used for storing the security data.

Optionally, the mobile storage device 150 is model K4F6E3S4HM-GF CL, SAMSUNG, which stores 2 gigabytes GB in size.

The reset module 160 is connected to the SOC30 through a reset interface for resetting data in the nonvolatile memory 130, the flash controller 140, and the removable storage device 150.

The High-low Voltage module (High-low Voltage)170 is connected to the SOC30 through an AD interface, and is configured to adjust the Voltage.

The battery detection 180 is connected to the SOC30 through the B-det interface for detecting whether a battery powering the SOC30 is powered down.

Based on the above-mentioned full-automatic parking system disclosed in the embodiment of the present invention, the present invention also discloses a full-automatic parking method correspondingly, as shown in fig. 3, which is a flow diagram of the full-automatic parking method shown in the embodiment of the present invention, the full-automatic parking method includes:

step S301: and acquiring spatial parking space data identified by the ultrasonic radar and all-round parking space data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain parking space fusion data.

Optionally, each ultrasonic radar detects the surroundings of the vehicle to determine the identified space and parking space data. Each fisheye camera shoots the periphery of the vehicle to determine the data of the surrounding parking space.

In the process of implementing step S301, spatial parking space data identified by each ultrasonic radar and all-round parking space data identified by each fisheye camera provided on the vehicle are acquired. And fusing the space parking place data and the all-round parking place data by using a data fusion algorithm to obtain parking place fusion data.

Step S302: and acquiring obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and fusing based on a data fusion algorithm to obtain obstacle space fusion data.

In step S302, the obstacle data includes the parked vehicle in the line parking space, and items such as ice cream cones parked in the line parking space that can obstruct the parking of the vehicle.

Optionally, each ultrasonic radar detects the surroundings of the vehicle to determine the identified obstacle data; each fisheye camera detects the surroundings of the vehicle to determine the recognized obstacle data.

In the process of implementing step S302, obstacle data recognized by each ultrasonic radar provided on the vehicle and obstacle data recognized by each fisheye camera provided on the vehicle are acquired. And fusing the obstacle data identified by the ultrasonic radar and the obstacle data identified by the fisheye camera by using a data fusion algorithm to obtain obstacle space fusion data.

It should be noted that the obstacle space fusion data refers to a space created by using a parking space obstacle, that is, a parking space that can be used for parking, so as to further determine a safe space for parking.

It should be noted that the execution sequence of step S301 and step S302 is not limited to the above, and may be executed in parallel, or step S302 may be executed first and then step S301 is executed, which is not limited in this embodiment of the present invention.

Step S303: and according to a free trajectory planning algorithm, combining the parking space fusion data and the obstacle space fusion data to carry out free trajectory planning to obtain a parking trajectory.

In the process of implementing step S303, a free planning algorithm is used to perform trajectory free planning on the parking space fusion data and the obstacle space fusion data, so as to obtain a parking trajectory.

It should be noted that the free planning algorithm is to plan a parking route according to the relationship between the vehicle and the parking space and considering the motion characteristics of the vehicle.

Step S304: and acquiring the moving position information of the current vehicle in real time, and controlling the vehicle to run along the parking track in real time based on a high-precision positioning algorithm and a high-precision track tracking algorithm.

In the process of implementing step S304, a Global Positioning System (GPS) is used to obtain a current moving position of the vehicle in real time, and a high-precision positioning algorithm is used to determine a relative position between the vehicle and the parking space, and a high-precision trajectory tracking algorithm is used to control the vehicle to park according to a parking trajectory, so as to control the vehicle to run along the parking trajectory in real time.

It should be noted that, in addition to determining the moving position of the current vehicle by using the GPS, the position of the current vehicle may be determined by using a positioning system such as a beidou satellite navigation system or GLONASS.

It should be noted that the high-precision positioning algorithm is used for enabling the fully automatic parking system to determine the relative position between the vehicle and the parking space more accurately during parking.

Based on the above-mentioned full-automatic parking method shown in the embodiment of the present invention, in the step S301, the spatial parking space data identified by the ultrasonic radar and the all-round parking space data identified by the fisheye camera are obtained, and are fused based on the data fusion algorithm to obtain the parking space fusion data, which includes the following steps:

step S11: and acquiring the space parking space characteristics identified based on the ultrasonic radar.

Optionally, the ultrasonic radar uses the vehicle itself as an origin, and transmits ultrasonic waves to all directions around the vehicle through an ultrasonic transmitting device of the ultrasonic radar so as to detect a parking space around the vehicle, when the ultrasonic waves are propagated in the air and touch the railings around the parking space, the reflected ultrasonic waves are returned to the ultrasonic radar through an ultrasonic receiving device, the ultrasonic radar records the time from the transmission of the ultrasonic waves to the return, and the distance between the railings around the parking space and the vehicle is calculated, so that the size of each available parking space, the distance between the parking space and the vehicle, and the space parking space characteristics such as the entrance direction are determined.

In the process of implementing step S11, the spatial parking space characteristics such as the size of each available parking space, the distance from the vehicle, and the entrance direction, which are identified based on the ultrasonic radar, are obtained.

Step S12: the line parking space characteristic and the parking space identification characteristic identified based on the fisheye camera are obtained.

Optionally, the fisheye camera looks around by taking the fisheye camera as an origin to shoot pictures in all directions around the vehicle; and recognizing each photo, and determining the line parking space characteristics corresponding to the line segments around the parking space, and the parking space identification characteristics corresponding to the parking direction arrow and the disabled person identification.

In the process of specifically implementing the step S12, a line parking space feature corresponding to a line segment around the parking space recognized based on the fisheye camera, and a parking space identification feature corresponding to a parking direction arrow and a disabled person identification are obtained.

It should be noted that the line parking space feature refers to a line segment for distinguishing parking space positions. The sign identification feature is used to indicate the available objects for the parking space, the direction of parking, and the like.

Step S13: and fusing the space parking space characteristic, the line parking space characteristic and the parking space identification characteristic based on a data fusion algorithm to obtain parking space fusion data.

In the process of implementing step S13 specifically, spatial parking space characteristics such as the size of each available parking space recognized by the ultrasonic radar, the distance to the vehicle, the entrance direction, and the like, line parking space characteristics corresponding to line segments around the parking space recognized by the fisheye camera, and parking space identification characteristics corresponding to the parking direction arrow and the disabled person identification are fused to obtain parking space fusion data.

It should be noted that the parking space fusion data includes position information of available parking spaces and identification information in the parking spaces.

In the embodiment of the invention, the data fusion algorithm fuses different feature data in the same region identified by each sensor, so that more accurate parking spaces and available safety spaces around the parking spaces are determined by utilizing the correlation of the different feature data and the complementarity of the information entropy.

It should be noted that the data fusion algorithm may be a high-precision perceptual data fusion algorithm.

That is, the size of each available parking space recognized by the ultrasonic radar, the distance from the vehicle, the vehicle entrance direction, the line segment around the parking space recognized by the fisheye camera, the parking direction arrow and the feature data of the disabled person identification in the same area are determined. And determining the position information of the available parking space and the identification information in the parking space by utilizing the size of the available parking space in the same area, the distance between the available parking space and the vehicle, the entrance direction, the line segments around the parking space, the correlation between the arrow of the parking direction and the identification of the disabled person and the complementarity of the information entropy.

In the embodiment of the invention, the ultrasonic radar and the fisheye cameras around the vehicle are arranged to look around the vehicle, and the space parking space data and the look around parking space data are determined; and the data fusion algorithm is utilized to fuse the space parking spot data and the all-round parking spot data so as to determine the position information of the available parking spots and the identification information in the parking spots. So as to subsequently determine the obstacle space fusion data, thereby presetting a parking track. In the process of automatic parking of the vehicle, the accuracy of parking space identification is improved, so that the full-automatic parking system is suitable for more parking scenes.

Based on the above-mentioned full-automatic parking method shown in the embodiment of the present invention, in the step S302, the obstacle data identified by the ultrasonic radar and the obstacle data identified by the fisheye camera are obtained, and are fused based on the data fusion algorithm to obtain the obstacle space fusion data, the method includes the following steps:

step S21: and acquiring the line parking space characteristic and the obstacle characteristic identified by the fisheye camera.

In step S21, the line parking space features refer to line segments around the parking space, and the obstacle features further include a signboard, a rail of the parking space, vehicles around the vehicle, and the like.

Optionally, the fisheye camera looks around by taking the fisheye camera as an origin to shoot pictures in all directions around the vehicle; and recognizing each photo, and determining the line parking space characteristics corresponding to the line segments around the parking space, the signboard, the railing of the parking space, and the obstacle characteristics of the obstacles around the vehicle, such as the vehicle and the like.

In the process of specifically implementing the step S21, a line segment around the parking space recognized based on the fisheye camera, a presence signboard, a rail of the parking space, and a vehicle around the vehicle are acquired.

Step S22: and determining the obstacle characteristics identified by the ultrasonic radar according to the position information of the line-parking space characteristics.

In the process of specifically implementing step S22, based on the line parking space characteristics corresponding to the line segments around the parking space identified by the fisheye camera, the ultrasonic wave transmitting device of the ultrasonic radar transmits ultrasonic waves to the line parking space characteristics to detect obstacles in the parking space, except the line segments around the parking space. When the ultrasonic waves are transmitted in the air and collide with an obstacle to return to the ultrasonic radar, the ultrasonic radar records the time from the transmission of the ultrasonic waves to the return, and calculates the distance between each direction of the obstacle and the vehicle, thereby determining the characteristics of the obstacle.

Step S23: and fusing the obstacle features identified by the fisheye camera and the obstacle features identified by the ultrasonic radar based on a data fusion algorithm to obtain obstacle space fusion data.

In the process of implementing step S23, the obstacle space fusion data is determined by using the correlations and information entropy complementarity of the signboards, the rails of the parking spaces, the vehicles around the vehicle, the parked vehicles, and the ice cream cone parked in the line parking space in the same area.

Based on the above-mentioned full-automatic parking method shown in the embodiment of the present invention, in the step S303, according to the free trajectory planning algorithm, the trajectory free planning is performed in combination with the parking space fusion data and the obstacle space fusion data, so as to obtain the parking trajectory, which includes the following steps:

step S31: and determining safety space data based on the parking space fusion data and the obstacle space fusion data.

In the process of specifically implementing the step S31, calculation is performed according to the position information of the available parking space, the in-parking-space identification information, and the obstacle space fusion data, so as to obtain the safe space data for available parking.

Step S32: and acquiring the position information and the vehicle operation parameters of the current vehicle.

In the process of step S32, the vehicle operation parameters include at least the vehicle speed of the current vehicle.

In the process of implementing step S32, the position information of the current vehicle is acquired in real time by the GPS, and the speed of the current vehicle is acquired by the vehicle controller.

It should be noted that the position information of the vehicle refers to a position in space, such as a position that can be identified by coordinates.

Step S33: and carrying out free track planning according to the safe space data, the current position information of the vehicle and the vehicle running parameters to obtain a parking track.

In the process of implementing step S33 specifically, a free-path planning is performed on the safe space data of available parking, the position information of the current vehicle, and the vehicle speed of the current vehicle by using a free-path planning algorithm, so as to obtain a parking path.

Based on the above-mentioned full-automatic parking method shown in the embodiment of the present invention, in the process of executing step S304 to obtain the moving position information of the current vehicle in real time and controlling the vehicle to run along the parking trajectory in real time based on the high-precision positioning algorithm and the high-precision trajectory tracking algorithm, the method includes the following steps:

step S41: and controlling the vehicle to run along the parking track.

In a specific implementation process, the SOC controls the vehicle to move along a parking track according to a preset parking speed.

Step S42: and acquiring the moving position of the current vehicle in real time.

In the process of implementing step S42, the current moving position of the vehicle may be acquired in real time by GPS.

Step S43: and determining whether the vehicle runs along the parking track based on the high-precision positioning algorithm and the high-precision track tracking algorithm, if not, executing the step S44, and if so, continuously controlling the vehicle to run along the parking track.

In the process of specifically implementing the step S43, determining the relative position between the vehicle and the parking space by using a high-precision positioning algorithm, and controlling the vehicle to park according to the parking track by using a high-precision track tracking algorithm, thereby determining the actual parking track of the vehicle, and determining whether the actual parking track of the vehicle is consistent with the parking track determined in the step S33, if not, indicating that the vehicle deviates from the parking track, and executing a step S44, if so, indicating that the vehicle does not deviate from the parking track, and continuing to control the vehicle to run along the parking track.

Step S44: and adjusting the running track of the vehicle to enable the vehicle to run along the parking track.

In the process of implementing step S44, the actual parking trajectory of the vehicle is adjusted so that the actual parking trajectory of the vehicle coincides with the parking trajectory planned by the free planning algorithm.

Based on the above-mentioned full-automatic parking method illustrated in fig. 3 according to the embodiment of the present invention, referring to fig. 4 in conjunction with fig. 3, after the step S304 is executed to control the vehicle to run along the parking trajectory, the full-automatic parking method includes:

step S401: and detecting whether a moving object appears in a preset range of the vehicle in the process of controlling the vehicle to run along the parking track in real time based on the fisheye camera, if so, executing the step S402, and if not, continuously controlling the vehicle to run along the parking track.

In the process of specifically implementing the step S401, in the process of controlling the vehicle to run along the parking track, whether a moving object appears in the preset range of the vehicle is detected in real time by using the fisheye camera, if so, the step S402 is executed, and if not, the vehicle is continuously controlled to run along the parking track.

In the embodiment of the invention, in the process of controlling the vehicle to run along the parking track by the SOC, the fisheye camera looks around by taking the fisheye camera as an original point in real time to shoot pictures in all directions around the vehicle within a preset range; and recognizing each photo, judging whether each photo is recognized to have a moving object, if so, executing step S402, and if not, continuously controlling the vehicle to continuously run along the parking track.

Step S402: and acquiring the living body characteristic information of the moving object, carrying out living body detection on the living body characteristic information based on a living body detection algorithm, and outputting a detection result.

In the process of specifically implementing step S402, according to each photo of the moving object, the contour feature of the moving object is obtained through technologies such as illumination compensation and illumination removal, a living body detection algorithm is used to detect a living body for the contour feature, so as to determine the moving trajectory of the moving object in real time, and the moving trajectory of the moving object is compared with the parking trajectory of the vehicle, that is, whether the comparison result is within a preset error is determined, so as to obtain the detection result.

Optionally, when it is determined that parking according to the current parking track is unsafe, parking is stopped, and when it is determined that the detection result is that the comparison result is within the preset error, the parking is restarted.

In the embodiment of the invention, the living body detection algorithm is used for judging the current safety condition according to the environment around the vehicle in the parking process so as to give a safety prompt to a client.

It should be noted that the moving object refers to an object currently in motion, such as: a moving person, or a moving vehicle.

Based on the above-mentioned full-automatic parking method illustrated in fig. 3 according to the embodiment of the present invention, referring to fig. 5 in conjunction with fig. 3, after the step S304 is executed to control the vehicle to run along the parking trajectory, the full-automatic parking method includes:

step S501: and detecting whether a moving object appears in a blind area of the vehicle and/or at the rear part of the vehicle in the process of controlling the vehicle to run along the parking track in real time based on the ultrasonic radar, if so, executing the step S502, and if not, continuously controlling the vehicle to run along the parking track.

In the process of specifically implementing the step S501, in the process of controlling the vehicle to run along the parking track, whether a moving object appears in a blind area of the vehicle and/or in the rear of the vehicle is detected in real time based on the ultrasonic radar, if so, the step S502 is executed, and if not, the vehicle is continuously controlled to run along the parking track.

In the embodiment of the present invention, in the process of controlling the vehicle to run along the parking track by the SOC, the ultrasonic radar transmits ultrasonic waves into the blind area and/or the rear portion of the vehicle by using the vehicle itself as the origin through the ultrasonic transmitting device of the ultrasonic radar to detect whether a moving object is present in the blind area and/or the rear portion of the vehicle, and when receiving the ultrasonic waves reflected by the ultrasonic receiving device, determines that a moving object is present in the blind area and/or the rear portion of the vehicle, and performs step S502; and when the ultrasonic wave reflected by the ultrasonic wave receiving device is not received, controlling the vehicle to continue to run along the parking track.

Optionally, in the process of controlling the vehicle to run along the parking track, in addition to determining whether a moving object appears in the blind area and/or the rear portion of the vehicle by using the ultrasonic radar, it may also be determined whether a moving object appears in the blind area and/or the rear portion of the vehicle in other manners, which is specifically implemented as follows:

in the process of controlling the vehicle to run along the parking track, each picture of the moving object identified by the fisheye camera is obtained, whether the moving object occurs in the blind area of the vehicle is determined by using a blind area detection alarm algorithm based on the pictures of the blind area of the vehicle identified by the fisheye camera, whether the moving object occurs in the rear area of the vehicle is determined by using a crossing detection alarm algorithm based on the pictures of the rear part of the vehicle identified by the fisheye camera, if at least one of the pictures determines that the moving object occurs, the step S502 is executed, and if none of the pictures of the rear part of the vehicle determines that the moving object occurs, the vehicle is continuously controlled to.

Step S502: and generating and outputting safety prompt information.

In step S502, the safety prompt message includes a text prompt message and/or a language prompt message.

In the process of implementing step S502 specifically, the language prompt information is output to prompt the driver of the vehicle in language, and/or the image-text prompt information is output to prompt the driver of the vehicle.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for fully automatic parking, the method comprising:

2. The method according to claim 1, wherein the acquiring of the spatial parking space data identified by the ultrasonic radar and the all-round parking space data identified by the fisheye camera is performed based on a data fusion algorithm to obtain parking space fusion data, and the method comprises:

3. The method according to claim 1, wherein the obtaining obstacle data identified by the ultrasonic radar and obstacle data identified by the fisheye camera, and fusing the obstacle data based on a data fusion algorithm to obtain obstacle space fusion data comprises:

4. The method of claim 1, wherein the freely planning a trajectory by combining the parking space fusion data and the obstacle space fusion data according to a free trajectory planning algorithm to obtain a parking trajectory comprises:

5. The method according to claim 1, wherein the obtaining of the moving position information of the current vehicle in real time and the controlling of the vehicle along the parking trajectory in real time based on a high-precision positioning algorithm and a high-precision trajectory tracking algorithm comprise:

controlling the vehicle to run along the parking track;

6. The method of any one of claims 1 to 5, further comprising:

and/or the presence of a gas in the gas,

and if so, generating and outputting safety prompt information.

7. A fully automatic parking system, comprising:

8. The system according to claim 7, wherein the controller is configured to perform free trajectory planning by combining the parking space fusion data and the obstacle space fusion data according to a free trajectory planning algorithm to obtain a parking trajectory, and is specifically configured to:

9. The system according to claim 7, wherein the controller for acquiring the moving position information of the current vehicle in real time and controlling the vehicle to run along the parking trajectory in real time based on a high-precision positioning algorithm and a high-precision trajectory tracking algorithm is specifically configured to:

10. The system according to any one of claims 7 to 9, wherein the controller is further configured to detect in real time whether a moving object occurs within a preset range of the vehicle during the real-time control of the vehicle along the parking trajectory based on the fisheye camera; if the living body characteristic information of the moving object appears, acquiring the living body characteristic information of the moving object, carrying out living body detection on the living body characteristic information based on a living body detection algorithm, and outputting a detection result;

and/or the presence of a gas in the gas,