CN111516676B - Automatic parking method, system, automobile and computer readable storage medium - Google Patents
Automatic parking method, system, automobile and computer readable storage medium Download PDFInfo
- Publication number
- CN111516676B CN111516676B CN202010360392.8A CN202010360392A CN111516676B CN 111516676 B CN111516676 B CN 111516676B CN 202010360392 A CN202010360392 A CN 202010360392A CN 111516676 B CN111516676 B CN 111516676B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- obstacle
- parking
- target
- distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/06—Automatic manoeuvring for parking
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
- B60W40/105—Speed
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Traffic Control Systems (AREA)
Abstract
The scheme relates to an automatic parking method, an automatic parking system, an automobile and a computer readable storage medium, and can solve the problem that the automatic parking function cannot be realized when an obstacle exists. The method comprises the following steps: detecting whether an obstacle exists in an area of interest of the vehicle; if the obstacle exists, controlling the vehicle to brake and stop, and detecting the type of the obstacle; if the type of the obstacle is still a static obstacle after the vehicle stopping time reaches the first time, reducing the target parking speed of the vehicle, and controlling the vehicle to continue to run according to the parking path planned by the initial parking path at the reduced target parking speed; detecting whether a stationary obstacle is located in a parking path of the vehicle; if the vehicle is located at the target distance, reducing the target distance between the vehicle and the obstacle, and enabling the vehicle to run to a position where the distance between the vehicle and the obstacle is the target distance according to the reduced target parking vehicle speed; and controlling the vehicle to park according to the re-planned parking path by the adjusted target parking speed.
Description
Technical Field
The invention belongs to the field of automobile electric appliances, and particularly relates to an automatic parking method, an automatic parking system, an automobile and a computer readable storage medium.
Background
Under the rapid development of the automobile field, the intelligent assistant driving system is widely applied to various automobile models. An automatic parking assist system, which is one of the intelligent driving assist systems, also becomes a standard for high-class or high-end models of all models. With the widespread use of automated parking assist systems, the challenges they face from a complex and diverse parking environment are becoming more and more severe.
In the environment faced by parking, pedestrians are the most uncertain factors, and the randomness and uncertainty of the participation of the pedestrians in the traffic environment bring challenges to the safety guarantee of automatic parking. At present, a parking system mainly senses the surrounding environment based on an ultrasonic probe and a look-around camera, the problem of pedestrian recognition cannot be well solved due to the limitation of performance of ultrasonic waves, and pedestrians are mainly recognized through deep learning by the look-around camera.
Look around the camera and carry out pedestrian detection based on degree of depth study, under the condition in order to ensure the recognition rate, the misidentification is unavoidable, and even the probability of taking place is on the high side, and common situation is the square column misidentification with underground garage for the pedestrian.
When a pedestrian is found on or around a parking path, in consideration of uncertainty of pedestrian movement, measures for braking in advance are usually taken for ensuring safety, and if a pedestrian target still exists after braking, two strategies are mainly adopted:
the first strategy is: and (4) the pedestrian is considered to leave by self, the system is stopped to wait for the pedestrian to leave, and if the target continuously exists, the pedestrian exits overtime. If the pedestrian target is identified by mistake, the parking process is directly blocked and quit.
The second strategy is: taking the pedestrian as an obstacle, replanning the path and trying to get around the pedestrian. The path planning recognition rate is high, and if the path cannot be planned, the parking is directly quitted.
Therefore, a new processing strategy needs to be developed, which can adapt to the working conditions of most pedestrians, and eliminate the situation that parking cannot be completed due to false recognition under the condition of ensuring safety.
Disclosure of Invention
The invention provides an automatic parking method, an automatic parking system, an automobile and a computer readable storage medium, which can solve the problem that the automatic parking function cannot be realized when an obstacle exists.
The technical scheme of the invention is as follows:
the invention provides an automatic parking method, which is applied to an automatic parking system and comprises the following steps:
Step S101, after the automatic parking function is started and a parking path is planned, sequentially carrying out image recognition on each frame of image collected by a panoramic image system so as to detect whether an obstacle exists in an area of interest of a vehicle;
step S102, if the vehicle is braked and stopped, controlling the vehicle to brake and detecting the type of the obstacle within a first brake-keeping brake duration of the vehicle;
step S103, if the type of the obstacle is still a static obstacle after the vehicle brake-off time reaches the first time, reducing the target parking speed of the vehicle, and controlling the vehicle to continue to run according to the parking path planned in the step S101 at the reduced target parking speed; after step S103 is completed, step S104 is executed;
step S104, detecting whether the static obstacle is positioned in a parking path of the vehicle; if yes, go to step S105; otherwise, go to step S107;
step S105, reducing the target distance between the vehicle and the obstacle, and enabling the vehicle to drive to a position where the distance between the vehicle and the obstacle is the target distance according to the target parking vehicle speed reduced in the step S104;
step S106, when the vehicle runs to a position where the distance between the vehicle and the obstacle is the target distance, planning a parking path again according to the distance acquired by the ultrasonic distance sensor, and controlling the vehicle to park according to the re-planned parking path at the target parking speed reduced in the step S103;
And S107, adjusting the width of the region of interest to the width of the parking path, and adjusting and controlling the vehicle to park according to the parking path planned in the step S101 at the initial target parking speed.
Preferably, in step S101, the step of detecting whether an obstacle exists in the area of interest of the vehicle includes:
performing confidence cumulative calculation based on the recognition result of the predetermined frame image;
after the preset frame image recognition is carried out, if the confidence coefficient obtained by calculating the same target in each frame image recognition result reaches a set confidence value, whether an obstacle exists in the interested area of the vehicle is further detected according to whether the target is positioned in the interested area of the vehicle;
if the target is identified after image identification is carried out on one frame of image, carrying out confidence coefficient accumulation on a first numerical value; and if the target is not identified after the image identification is carried out on the frame of image, carrying out confidence deduction on a second numerical value.
Preferably, in step S102, the step of detecting the type of the obstacle specifically includes:
judging whether the vehicle is in a static state or not based on the four-wheel pulse signal of the vehicle;
if the position of the obstacle is in the preset position, calculating the position distance of the center point of the obstacle in the multi-frame two-frame images based on the continuous multi-frame images collected by the panoramic camera;
Calculating the instantaneous moving speed of the barrier according to the central point position distance;
performing Kalman filtering on the instantaneous moving speed to obtain the moving speed of the obstacle;
when the moving speed of the obstacle is lower than a preset speed value in a first time length, determining that the obstacle is a static obstacle;
otherwise, determining the obstacle as a movement obstacle.
Preferably, if the obstacle is detected to be a moving obstacle in step S102, the method further includes:
step S108, the parking path planning is carried out again; if a new parking path can be planned, controlling the vehicle to park according to the re-planned parking path at the initial target parking speed; if a new parking path cannot be planned, the process goes to step S109;
step S109, prolonging the vehicle brake time to a second time length when the first time length reaches, and simultaneously detecting whether the moving obstacle leaves the interested area of the vehicle or not during the vehicle prolonged brake time;
step S110, if the moving obstacle does not leave the interesting area of the vehicle when the brake-off time of the vehicle is prolonged to reach a second time length, the automatic parking function is quitted;
and step S111, if the vehicle is separated from the region of interest of the vehicle during the prolonged brake-off period, controlling the vehicle to park according to the parking path planned in the step S101 at the initial target parking speed.
Preferably, if it is detected in step S101 that no obstacle exists in the region of interest of the vehicle, the method further includes:
in step S112, the vehicle is controlled to park at the initial target parking vehicle speed according to the parking path planned in step S101.
Preferably, in step S101, the region of interest is: the vehicle driving direction is taken as a middle axis, and the longitudinal direction and the transverse direction respectively extend outwards for a certain distance and surround a formed area; the width of the region of interest in the transverse direction is greater than the width of the parking path; in the parking path planned in the step S101, a position where the vehicle needs to travel from the current position to a position where the distance between the vehicle and the obstacle is a first target distance is planned; after the target distance adjustment in step S105, the vehicle needs to travel from the current position to a position where the distance to the obstacle is the target distance; wherein the target distance in step S105 is smaller than the first target distance in step S101.
Preferably, the step of determining whether the vehicle is in a stationary state is:
judging whether the wheel pulse signals of 3 or more wheels of the vehicle are kept unchanged for a third time period at the same time;
if yes, the vehicle is determined to be in a static state.
An embodiment of the present invention further provides an automatic parking system, including:
the obstacle identification module is used for sequentially carrying out image identification on each frame of image collected by the all-around vision image system after the automatic parking function is started and a parking path is planned so as to detect whether an obstacle exists in an interested area of the vehicle;
the obstacle type detection module is used for controlling the vehicle to brake and stop if the obstacle type detection module exists, and detecting the type of the obstacle within a first brake and stop duration of the vehicle;
the target parking vehicle speed adjusting module is used for reducing the target parking vehicle speed of the vehicle and controlling the vehicle to continuously run according to the reduced target parking vehicle speed and the parking path planned in the obstacle identifying module if the type of the obstacle is a static obstacle;
the system comprises a static obstacle position detection module, a parking path detection module and a control module, wherein the static obstacle position detection module is used for detecting whether an obstacle is positioned in the parking path of the vehicle or not; if the position is in the parking control module, entering the vehicle parking control module; otherwise, entering a third vehicle parking control module;
the first vehicle parking control module is used for reducing the target distance between the vehicle and the obstacle and enabling the vehicle to run to a position where the distance between the vehicle and the obstacle is the target distance according to the reduced target parking vehicle speed in the target parking vehicle speed adjusting module;
The second vehicle parking control module is used for re-planning a parking path according to the distance acquired by the ultrasonic distance sensor when the vehicle runs to a position where the distance between the vehicle and the obstacle is the target distance, and controlling the vehicle to park according to the re-planned parking path at the reduced target parking speed in the target parking speed adjusting module;
and the third vehicle parking control module is used for adjusting the width of the region of interest to the width of the parking path and controlling the vehicle to park at the initial target parking speed according to the parking path planned in the step obstacle identification module.
The embodiment of the invention also provides an automobile which comprises the automatic parking system.
The embodiment of the invention also provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by the vehicle-mounted information processing module, the steps of the automatic parking method are realized.
The invention has the beneficial effects that:
the situation that a camera of the panoramic image system recognizes a static obstacle as a moving obstacle can be effectively processed, unnecessary exit is avoided on the premise that collision of the obstacle is prevented and safety is guaranteed, and the parking completion rate is improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The automatic parking system reasonably plans a parking path based on the information acquired by the ultrasonic probe and the look-around camera, and further realizes automatic parking of the vehicle. Based on the defects of the automatic parking function in the prior art, the embodiment of the application provides a method for processing the obstacle which influences the realization of the vehicle parking function in the automatic parking process of the vehicle. As shown in fig. 1, the automatic parking method specifically includes:
and S101, after the automatic parking function is started and a parking path is planned, sequentially carrying out image recognition on each frame of image acquired by the all-around vision image system so as to detect whether an obstacle exists in an interested area of the vehicle.
The automatic parking function is started by a user, after the automatic parking function is started, the automatic parking system automatically searches for a parking space and plans an optimal parking path, wherein the automatic parking function and the automatic parking system plan the optimal parking path both belong to mature technical means in the prior art, and the automatic parking function and the optimal parking path can be realized by adopting the existing means in the embodiment. In this optimal parking path, a specific steering wheel angle and a specific gear when the vehicle is moved from the current position to the next position are planned. Meanwhile, after the automatic parking function is started, the default initial target running speed of the vehicle is 2 km/h.
In the automatic parking process, the cameras in the panoramic image system, which are consistent with the target gear direction, start the pedestrian target detection function (namely, the front-view camera in the panoramic image system is started in the D gear, the rear-view camera in the panoramic image system is started in the R gear, and the detection function is closed in the P gear or the N gear). The automatic parking system processes the image picture at the speed of not less than 20fps, performs object attribute identification of external features such as the head, the shoulders, the trunk and the legs of the pedestrian on the basis of the deep convolutional neural network, and judges whether the object is an obstacle or not according to the identification result for more than 2 times. When the obstacle is detected and identified according to the image acquisition result of the camera, the detection information is sent to the automatic parking system, and the automatic parking system starts to carry out confidence cumulative calculation on the obstacle.
Considering the influence of external environments such as illumination, background and the like, the missing detection rate of about 6% exists in obstacle identification, so that the obstacles cannot be processed according to real-time input information processed by each frame of image, and the obstacles cannot be judged in a confidence accumulation mode. The confidence cumulative calculation mode is that if the camera identifies one frame again, 30% confidence is accumulated; if the camera output result does not continuously identify the target, the confidence coefficient of each frame is reduced by 10%. The confidence upper limit value is 100%, and when the confidence is 0, the tracking of the obstacle is cancelled.
After the automatic parking function of the panoramic image system is started, the cameras at all positions in the system start to acquire images or images, and the acquired images are transmitted to the automatic parking function in real time. When the automatic parking function is used for image recognition, image features are extracted, and the type of the obstacle is recognized according to some inherent attributes of different obstacles, such as pedestrians, animals or other non-biological obstacles. The automatic parking system can output an image recognition result once when finishing the recognition of one frame of image, and the image recognition result comprises the result of whether the obstacle exists in the frame of image. For the same camera, the automatic parking system can perform image recognition according to a plurality of continuously acquired images to accurately confirm whether the obstacle exists. Specifically, confidence accumulation is performed on the image recognition results of continuous multi-frame images acquired by a camera, and whether an obstacle exists in the region of interest of the vehicle is judged according to the confidence accumulation results of the continuous multi-frame images. Specifically, the steps specifically include: performing confidence cumulative calculation based on the recognition result of the predetermined frame image; after the preset frame image recognition is carried out, if the confidence coefficient obtained by calculating the same target in each frame image recognition result reaches a set confidence value (90%), whether an obstacle exists in the region of interest of the vehicle is further detected according to whether the target is located in the region of interest of the vehicle, and if the target is located in the region of interest of the vehicle, it is determined that the obstacle exists in the region of interest of the vehicle, and the object is the obstacle; if the target is identified after image identification is carried out on one frame of image, carrying out confidence degree accumulation on a first numerical value (30%); and if the target is not identified after the image identification is carried out on the frame of image, carrying out confidence deduction on a second numerical value (10%). For example, after the first frame image is subjected to image recognition, the confidence of the object a is increased by 30% in 5 frames of images continuously acquired by the camera, if the second frame image continues to recognize the object a, the confidence is accumulated again by 30%, and if one of the 5 frames of images does not recognize the object a, the confidence is reduced by 10%. After image recognition is performed on the continuous 5 frames of images, a confidence accumulation result can be obtained. And if the confidence accumulated result exceeds the set confidence value, determining whether the object A is further detected to be positioned in the region of interest of the vehicle, and if so, determining that the object A is an obstacle existing in the region of interest of the vehicle. In practice, of course, there may be a plurality of objects in one frame of image, or there may be no object, and in a continuous multi-frame image, the same object in each frame of image is identified based on image feature comparison.
In addition, in step 101, with the driving direction of the vehicle as a central axis, the longitudinal direction and the transverse direction respectively extend outwards for a certain distance and then surround the formed area; this region is a region having a longitudinal direction of 2m and a transverse direction of 3 m. Wherein the transverse width (3 m) of the region of interest is greater than the transverse width (2 m) of the parking path.
In step S101, if it is determined that no obstacle exists in the region of interest of the vehicle, the automatic parking system performs no processing, that is, executes step S112, and controls the vehicle to park at the initial target parking speed (2 km/h) according to the parking route planned in step S101.
And when the confidence coefficient reaches 90% or more, the target is considered to be valid and the step S102 is locked, if the target exists, the vehicle is controlled to carry out braking action, and the type of the obstacle is detected within the first time (3S) of the vehicle keeping braking.
The purpose of controlling the vehicle to brake is to cause the vehicle to collide with an obstacle to cause damage to the vehicle or the human body. Meanwhile, the event that the vehicle remains braked and stopped continues for a first duration (i.e., 3S), during which the automatic parking system recognizes whether the obstacle is a moving obstacle or a stationary obstacle by means of the images collected by the panoramic imaging system.
Specifically, the manner of judging the type of the obstacle is specifically as follows: judging whether the vehicle is in a static state or not based on four-wheel pulse signals of the vehicle;
if the position of the obstacle is in the center point position distance, calculating the center point position distance of the obstacle in the two frames of images based on the continuous multi-frame images collected by the panoramic camera;
calculating the instantaneous moving speed of the barrier according to the central point position distance;
performing Kalman filtering on the instantaneous moving speed to obtain the moving speed of the obstacle;
determining the obstacle to be a static obstacle when the moving speed of the obstacle is lower than a preset speed value (0.1 m/s) within a first time period (0.3 s);
otherwise, determining the obstacle as a movement obstacle.
Since the vehicle moves at a low speed and the movement speed of the obstacle may be slow in the parking process, the accuracy of judging whether the obstacle target moves is low when the vehicle moves, and whether the pedestrian target is stationary is judged after the vehicle is stationary. In this embodiment, the four-wheel pulse signals are used to determine that the vehicle is stationary, and when the wheel pulse signals of 3 or more wheels are kept unchanged for 300ms at the same time, the vehicle is determined to be stationary.
Furthermore, after the vehicle is stationary, the center point of the identified obstacle is set as the obstacle coordinate, and the instantaneous movement speed of the obstacle is calculated according to the position change of the center point of 2 continuous frames of image processing. And considering the detection error, performing Kalman filtering processing on the instantaneous movement speed of the obstacle, eliminating the movement speed error caused by the detection error, and obtaining the movement speed of the obstacle. And then, carrying out time delay judgment on the moving speed of the obstacle, and determining the obstacle to be a static obstacle when the moving speed value of the obstacle lasting for more than 300ms is less than 0.1 m/s. Otherwise, determining the obstacle as a movement obstacle.
Meanwhile, the vehicle is kept still for 3s for delay confirmation, namely, the time for the real moving obstacle (such as an animal, a vehicle or a pedestrian) to leave is given, and whether the obstacle is in a continuous static state is confirmed.
And step S103, if the type of the obstacle is still a static obstacle after the vehicle brake-off time reaches the first time, reducing the target parking speed (1km/h) of the vehicle, and controlling the vehicle to continuously run according to the parking path planned in the step S101 at the reduced target parking speed (1 km/h).
After the vehicle is stationary for 3 seconds, the pedestrian target is still in a stationary state, and the vehicle is allowed to restart because the pedestrian target needs to be closer to the current target obstacle in order to finish parking or eliminate the possibility of mistaken identification of the obstacle. The parking control speed is reduced to 1km/h, and compared with the parking speed of a whole vehicle of 2km/h, the detection times of ultrasonic waves or a camera can be doubled within the same moving distance, so that the effectiveness of ultrasonic detection when the vehicle approaches a target is improved; meanwhile, the braking distance in emergency can be greatly reduced by reducing the vehicle speed.
After step S103 is completed, step S104 is executed to detect whether the stationary obstacle is located in the parking path of the vehicle, and if so, step S105 is executed.
And step S105, adjusting the target running distance (50cm), and enabling the vehicle to run for the target running distance from the current position according to the target parking speed (1km/h) adjusted to be small in step S104.
Specifically, in step S105, the stationary obstacle positioning coordinates of the locked position are compared with the parking planned path, and the target travel distance is updated when the stationary obstacle is on the parking planned path. The specific value of the target travel distance is set to be the distance between the host vehicle and the target minus 50 cm. After the target running distance is adjusted, the vehicle can continue to run to a position 30-50cm away from the target. That is, for the vehicle, if the vehicle travels along the parking route in step S101, the distance traveled by the vehicle at the current steering wheel angle becomes short, and the distance between the stationary obstacle when the vehicle stops at the current steering wheel angle is between 80cm and 100 cm.
In the parking path planned in the step S101, a position where the vehicle needs to travel from the current position to a position where the distance to the obstacle is a first target distance (80 cm-100 cm) is planned; after the target driving distance adjustment in step S105, the vehicle needs to drive from the current position to a position where the distance to the obstacle is the target distance (50 cm); wherein the target distance (30-50 cm) in step S105 is smaller than the first target distance (80-100 cm) in step S101.
And S106, after the vehicle runs the target running distance from the current position, planning a parking path again according to the distance acquired by the ultrasonic distance sensor, and controlling the vehicle to park according to the re-planned parking path at the target parking speed (1km/h) reduced in the step S103.
At the moment, the camera on the vehicle is a fisheye camera, the installation position of the camera is lower, a complete target image cannot be shot, or the distortion of the target image at the edge of the fisheye camera is too large, the automatic parking system does not identify a pedestrian target through image processing any more, but more accurate object distance information can be acquired by means of ultrasonic close-range detection, and the path is updated by taking the detected object as a parking space boundary.
If the stationary obstacle is not located in the parking path of the vehicle in the step S104, the step S107 is executed, the width (3m) of the region of interest is adjusted to the width (2m) of the parking path, and the vehicle is controlled to park according to the parking path planned in the step S101 at the initial target parking speed (2 km/h). That is, the coverage area of the region of interest is narrowed, only the region on the parking planned path is set as the dangerous region, the target parking speed of the vehicle adjusted by the detection of the obstacle is returned to the initial target parking speed, and the vehicle is continuously parked according to the parking path planned path other than the one in S101.
As shown in fig. 1, for the method in this embodiment, if it is detected in step S102 that the obstacle is a moving obstacle, the method further includes:
step S108, the parking path is planned again; if a new parking path can be planned, controlling the vehicle to park according to the re-planned parking path at the initial target parking speed (2 km/h); if a new parking path cannot be planned, the process goes to step S109;
step S109, prolonging the vehicle brake time to a second time length when the first time length reaches, and simultaneously detecting whether the moving obstacle leaves the interested area of the vehicle or not during the vehicle prolonged brake time;
step S110, if the moving obstacle does not leave the interesting area of the vehicle when the brake-off time of the vehicle is prolonged to reach a second time length, the automatic parking function is quitted;
and step S111, if the vehicle is separated from the region of interest of the vehicle during the prolonged brake-off period, controlling the vehicle to park according to the parking path planned in the step S101 at the initial target parking speed (2 km/h).
For moving obstacles, the design aim is to replan a parking path for parking if the obstacle can be avoided, and to stop parking if the path cannot be replanned.
The method can effectively process the condition that the camera of the panoramic image system identifies the static barrier as the moving barrier, avoids unnecessary exit on the premise of preventing the collision of the barrier and ensuring safety, and improves the parking completion rate.
The system is simple and effective to realize, and excessive and complex modification and change of the system are not needed.
An embodiment of the present invention further provides an automatic parking system, including:
the obstacle identification module is used for sequentially carrying out image identification on each frame of image collected by the all-around vision image system after the automatic parking function is started and a parking path is planned so as to detect whether an obstacle exists in an interested area of the vehicle;
the obstacle type detection module is used for controlling the vehicle to brake and stop if the obstacle type detection module exists, and detecting the type of the obstacle within a first time period of vehicle brake and stop;
the target parking vehicle speed adjusting module is used for reducing the target parking vehicle speed of the vehicle and controlling the vehicle to continuously run according to the reduced target parking vehicle speed and the parking path planned in the obstacle identifying module if the type of the obstacle is a static obstacle;
the system comprises a static obstacle position detection module, a parking path detection module and a control module, wherein the static obstacle position detection module is used for detecting whether an obstacle is positioned in the parking path of the vehicle or not; if the position is in the parking control module, entering the vehicle parking control module; otherwise, entering a third vehicle parking control module;
The first vehicle parking control module is used for reducing the target distance between the vehicle and the obstacle and enabling the vehicle to run to a position where the distance between the vehicle and the obstacle is the target distance according to the reduced target parking vehicle speed in the target parking vehicle speed adjusting module;
the second vehicle parking control module is used for planning a parking path again according to the distance acquired by the ultrasonic distance sensor when the vehicle runs to the position where the distance between the vehicle and the obstacle is the target distance, and controlling the vehicle to park according to the re-planned parking path at the reduced target parking speed in the target parking speed adjusting module;
and the third vehicle parking control module is used for adjusting the width of the region of interest to the width of the parking path and controlling the vehicle to park at the initial target parking speed according to the parking path planned in the step obstacle identification module.
It should be understood that the system in the embodiment of the present invention further includes modules and units corresponding to other steps in the above method, which are not described herein again. The system of the invention has the same technical effects as the method.
The embodiment of the invention also provides an automobile which comprises the automatic parking system.
The embodiment of the invention also provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by the vehicle-mounted information processing module, the steps of the automatic parking method are realized. The embodiments described above describe only some of the one or more embodiments of the present invention, but those skilled in the art will recognize that the present invention can be embodied in many other forms without departing from the spirit and scope thereof. Accordingly, the present examples and embodiments are to be considered as illustrative and not restrictive, and various modifications and substitutions may be made therein without departing from the spirit and scope of the present invention as defined by the appended claims.
Claims (10)
1. An automatic parking method is applied to an automatic parking system, and is characterized by comprising the following steps:
step S101, after the automatic parking function is started and a parking path is planned, sequentially carrying out image recognition on each frame of image collected by a panoramic image system so as to detect whether an obstacle exists in an area of interest of a vehicle;
step S102, if the obstacle exists, controlling the vehicle to perform brake-off action, and detecting the type of the obstacle within a first time period of vehicle brake-off keeping;
Step S103, if the type of the obstacle is still a static obstacle after the vehicle brake-off time reaches the first time, reducing the target parking speed of the vehicle, and controlling the vehicle to continue to run according to the parking path planned in the step S101 at the reduced target parking speed; after step S103 is completed, step S104 is executed;
step S104, detecting whether the static obstacle is positioned in a parking path of the vehicle; if yes, go to step S105; otherwise, go to step S107;
step S105, reducing the target distance between the vehicle and the obstacle, and enabling the vehicle to drive to a position where the distance between the vehicle and the obstacle is the target distance according to the target parking vehicle speed reduced in the step S104;
step S106, when the vehicle runs to a position where the distance between the vehicle and the obstacle is the target distance, planning a parking path again according to the distance acquired by the ultrasonic distance sensor, and controlling the vehicle to park according to the re-planned parking path at the target parking speed reduced in the step S103;
and S107, adjusting the width of the region of interest to the width of the parking path, and adjusting and controlling the vehicle to park according to the parking path planned in the step S101 at the initial target parking speed.
2. The method according to claim 1, wherein the step of detecting whether an obstacle exists in the area of interest of the vehicle in step S101 comprises:
performing confidence cumulative calculation based on the recognition result of the predetermined frame image;
after the preset frame image recognition is carried out, if the confidence coefficient obtained by calculating the same target in each frame image recognition result reaches a set confidence value, whether an obstacle exists in the interested area of the vehicle is further detected according to whether the target is positioned in the interested area of the vehicle;
if the target is identified after image identification is carried out on one frame of image, carrying out confidence coefficient accumulation on a first numerical value; and if the target is not identified after the image identification is carried out on the frame of image, carrying out confidence deduction on a second numerical value.
3. The method according to claim 1, wherein in step S102, the step of detecting the type of the obstacle is specifically:
judging whether the vehicle is in a static state or not based on the four-wheel pulse signal of the vehicle;
if the position of the obstacle is in the preset position, calculating the position distance of the center point of the obstacle in the multi-frame two-frame images based on the continuous multi-frame images collected by the panoramic camera;
Calculating the instantaneous moving speed of the barrier according to the central point position distance;
performing Kalman filtering on the instantaneous moving speed to obtain the moving speed of the obstacle;
when the moving speed of the obstacle is lower than a preset speed value in a first time length, determining that the obstacle is a static obstacle;
otherwise, determining the obstacle as a movement obstacle.
4. The method according to claim 1, wherein if the obstacle is detected to be a moving obstacle in step S102, the method further comprises:
step S108, the parking path planning is carried out again; if a new parking path can be planned, controlling the vehicle to park according to the re-planned parking path at the initial target parking speed; if a new parking path cannot be planned, the process goes to step S109;
step S109, prolonging the vehicle brake time to a second time length when the first time length reaches, and simultaneously detecting whether the moving obstacle leaves the interested area of the vehicle or not during the vehicle prolonged brake time;
step S110, if the moving obstacle does not leave the interesting area of the vehicle when the brake-off time of the vehicle is prolonged to reach a second time length, the automatic parking function is quitted;
And step S111, if the vehicle is separated from the region of interest of the vehicle during the prolonged brake-off period, controlling the vehicle to park according to the parking path planned in the step S101 at the initial target parking speed.
5. The method according to claim 1, wherein if no obstacle is detected in the region of interest of the vehicle in step S101, the method further comprises:
in step S112, the vehicle is controlled to park at the initial target parking vehicle speed according to the parking path planned in step S101.
6. The method according to claim 1, wherein in step S101, the regions of interest are: the vehicle driving direction is taken as a central axis, and the longitudinal direction and the transverse direction respectively extend outwards for a certain distance and then surround to form an area; the width of the region of interest in the transverse direction is larger than the width of the parking path in the transverse direction; in the parking path planned in the step S101, a position where the vehicle needs to travel from the current position to a position where the distance between the vehicle and the obstacle is a first target distance is planned; after the target distance adjustment in step S105, the vehicle needs to travel from the current position to a position where the distance to the obstacle is the target distance; wherein the target distance in step S105 is smaller than the first target distance in step S101.
7. The method of claim 3, wherein the step of determining whether the vehicle is stationary is:
judging whether the wheel pulse signals of 3 or more wheels of the vehicle are kept unchanged for a third time period at the same time;
if yes, the vehicle is determined to be in a static state.
8. An automatic parking system, comprising:
the obstacle identification module is used for sequentially carrying out image identification on each frame of image collected by the all-around vision image system after the automatic parking function is started and a parking path is planned so as to detect whether an obstacle exists in an interested area of the vehicle;
the obstacle type detection module is used for controlling the vehicle to brake and stop if the obstacle type detection module exists, and detecting the type of the obstacle within a first brake and stop duration of the vehicle;
the target parking vehicle speed adjusting module is used for reducing the target parking vehicle speed of the vehicle and controlling the vehicle to continuously run according to the reduced target parking vehicle speed and the parking path planned in the obstacle identifying module if the type of the obstacle is a static obstacle;
the system comprises a static obstacle position detection module, a parking path detection module and a parking path detection module, wherein the static obstacle position detection module is used for detecting whether an obstacle is positioned in the parking path of a vehicle; if the position is in the parking control module, entering the vehicle parking control module; otherwise, entering a third vehicle parking control module;
The first vehicle parking control module is used for reducing the target distance between the vehicle and the obstacle, so that the vehicle can drive to a position where the distance between the vehicle and the obstacle is the target distance according to the reduced target parking vehicle speed in the target parking vehicle speed adjusting module;
the second vehicle parking control module is used for planning a parking path again according to the distance acquired by the ultrasonic distance sensor when the vehicle runs to the position where the distance between the vehicle and the obstacle is the target distance, and controlling the vehicle to park according to the re-planned parking path at the reduced target parking speed in the target parking speed adjusting module;
and the third vehicle parking control module is used for adjusting the width of the region of interest to the width of the parking path and controlling the vehicle to park at the initial target parking speed according to the parking path planned in the step obstacle identification module.
9. An automobile characterized by comprising the automatic parking system according to claim 8.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which, when being executed by an onboard information processing module, implements the steps of the automatic parking method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010360392.8A CN111516676B (en) | 2020-04-30 | 2020-04-30 | Automatic parking method, system, automobile and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010360392.8A CN111516676B (en) | 2020-04-30 | 2020-04-30 | Automatic parking method, system, automobile and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111516676A CN111516676A (en) | 2020-08-11 |
| CN111516676B true CN111516676B (en) | 2022-06-07 |
Family
ID=71905514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010360392.8A Active CN111516676B (en) | 2020-04-30 | 2020-04-30 | Automatic parking method, system, automobile and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111516676B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112277935B (en) * | 2020-10-30 | 2022-03-11 | 广州小鹏自动驾驶科技有限公司 | Automatic parking method and device |
| JP2022087977A (en) * | 2020-12-02 | 2022-06-14 | フォルシアクラリオン・エレクトロニクス株式会社 | Parking support device and parking support method |
| CN112801024B (en) * | 2021-02-09 | 2023-08-29 | 广州小鹏自动驾驶科技有限公司 | Detection information processing method and device |
| CN113071480B (en) * | 2021-04-30 | 2022-06-03 | 重庆长安汽车股份有限公司 | Automatic parking obstacle detection method, parking method and system and vehicle |
| CN113511191B (en) * | 2021-05-12 | 2022-06-07 | 重庆长安汽车股份有限公司 | Vertical parking completion judgment system, method, vehicle and storage medium |
| CN113200040B (en) * | 2021-06-17 | 2022-08-16 | 广州小鹏自动驾驶科技有限公司 | Automatic parking method and device |
| CN114454874B (en) * | 2022-02-21 | 2023-06-23 | 岚图汽车科技有限公司 | A method and system for automatic parking to prevent sudden braking |
| CN114995421B (en) * | 2022-05-31 | 2024-06-18 | 重庆长安汽车股份有限公司 | Automatic driving obstacle avoidance method, device, electronic equipment, storage medium and program product |
| CN115158295A (en) * | 2022-06-29 | 2022-10-11 | 重庆长安汽车股份有限公司 | Parking control method and device for vehicle |
| CN115158297A (en) * | 2022-08-12 | 2022-10-11 | 浙江吉利控股集团有限公司 | A kind of automatic parking method, device and electronic equipment |
| CN116279502A (en) * | 2023-03-20 | 2023-06-23 | 浙江极氪智能科技有限公司 | Parking space identification method, device, equipment and storage medium |
| CN117261877B (en) * | 2023-08-18 | 2024-05-14 | 广州优保爱驾科技有限公司 | Self-correction image acquisition system and method based on vehicle appearance change |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH075921A (en) * | 1993-06-15 | 1995-01-10 | Nippon Yusoki Co Ltd | Traveling stop controller for unmanned work vehicle |
| JPH07210245A (en) * | 1994-01-14 | 1995-08-11 | Sony Corp | Transfer control method |
| JP2002229645A (en) * | 2001-01-31 | 2002-08-16 | Shin Kobe Electric Mach Co Ltd | Autonomous vehicle control system |
| JP2004221871A (en) * | 2003-01-14 | 2004-08-05 | Auto Network Gijutsu Kenkyusho:Kk | Vehicle periphery monitoring device |
| WO2006064544A1 (en) * | 2004-12-14 | 2006-06-22 | Hitachi, Ltd. | Automatic garage system |
| GB0621182D0 (en) * | 2006-10-25 | 2006-12-06 | Li Shih Hsiung | Car reversal radar that automatically modifies the sensor scanning range and method of the same |
| JP2006343309A (en) * | 2005-05-09 | 2006-12-21 | Nippon Soken Inc | Obstacle detector |
| JP2008207732A (en) * | 2007-02-27 | 2008-09-11 | Denso Corp | Drive assisting device |
| CN101412401A (en) * | 2007-10-17 | 2009-04-22 | 罗伯特·博世有限公司 | Control device and method for assisting parking |
| JP2011156955A (en) * | 2010-01-29 | 2011-08-18 | Equos Research Co Ltd | Travel control device and travel control method |
| WO2012095717A2 (en) * | 2011-01-12 | 2012-07-19 | Toyota Jidosha Kabushiki Kaisha | Travel support apparatus |
| CN102652091A (en) * | 2009-12-12 | 2012-08-29 | 大众汽车有限公司 | Method and device for steering a vehicle toward an object during a parking operation |
| DE102011012541A1 (en) * | 2011-02-26 | 2012-08-30 | Conti Temic Microelectronic Gmbh | Method for performing longitudinal control of vehicle, involves using driver input and additional environment information to perform longitudinal control of vehicle when vehicle distance to obstacle is less than preset distance |
| JP2012176748A (en) * | 2011-02-02 | 2012-09-13 | Nissan Motor Co Ltd | Parking assisting system |
| WO2013072132A1 (en) * | 2011-11-15 | 2013-05-23 | Robert Bosch Gmbh | Method and driver assistance system for the detection of a vehicle environment |
| CN104401260A (en) * | 2014-10-21 | 2015-03-11 | 奇瑞汽车股份有限公司 | Automatic parking system |
| CN106471554A (en) * | 2014-06-10 | 2017-03-01 | 株式会社电装 | Driving support device |
| CN106985815A (en) * | 2017-03-14 | 2017-07-28 | 重庆长安汽车股份有限公司 | Remote control parking system and method |
| CN107226088A (en) * | 2016-03-25 | 2017-10-03 | 松下电器（美国）知识产权公司 | Controller, driving control method and program |
| CN108928341A (en) * | 2017-05-23 | 2018-12-04 | 株式会社万都 | Intelligent parking auxiliary system and its control method |
| CN110356394A (en) * | 2019-07-31 | 2019-10-22 | 重庆长安汽车股份有限公司 | A kind of vehicle actively avoids the method, apparatus and automobile of barrier |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10809078B2 (en) * | 2018-04-05 | 2020-10-20 | Symbol Technologies, Llc | Method, system and apparatus for dynamic path generation |
-
2020
- 2020-04-30 CN CN202010360392.8A patent/CN111516676B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH075921A (en) * | 1993-06-15 | 1995-01-10 | Nippon Yusoki Co Ltd | Traveling stop controller for unmanned work vehicle |
| JPH07210245A (en) * | 1994-01-14 | 1995-08-11 | Sony Corp | Transfer control method |
| JP2002229645A (en) * | 2001-01-31 | 2002-08-16 | Shin Kobe Electric Mach Co Ltd | Autonomous vehicle control system |
| JP2004221871A (en) * | 2003-01-14 | 2004-08-05 | Auto Network Gijutsu Kenkyusho:Kk | Vehicle periphery monitoring device |
| WO2006064544A1 (en) * | 2004-12-14 | 2006-06-22 | Hitachi, Ltd. | Automatic garage system |
| JP2006343309A (en) * | 2005-05-09 | 2006-12-21 | Nippon Soken Inc | Obstacle detector |
| GB0621182D0 (en) * | 2006-10-25 | 2006-12-06 | Li Shih Hsiung | Car reversal radar that automatically modifies the sensor scanning range and method of the same |
| JP2008207732A (en) * | 2007-02-27 | 2008-09-11 | Denso Corp | Drive assisting device |
| CN101412401A (en) * | 2007-10-17 | 2009-04-22 | 罗伯特·博世有限公司 | Control device and method for assisting parking |
| CN102652091A (en) * | 2009-12-12 | 2012-08-29 | 大众汽车有限公司 | Method and device for steering a vehicle toward an object during a parking operation |
| JP2011156955A (en) * | 2010-01-29 | 2011-08-18 | Equos Research Co Ltd | Travel control device and travel control method |
| WO2012095717A2 (en) * | 2011-01-12 | 2012-07-19 | Toyota Jidosha Kabushiki Kaisha | Travel support apparatus |
| JP2012176748A (en) * | 2011-02-02 | 2012-09-13 | Nissan Motor Co Ltd | Parking assisting system |
| DE102011012541A1 (en) * | 2011-02-26 | 2012-08-30 | Conti Temic Microelectronic Gmbh | Method for performing longitudinal control of vehicle, involves using driver input and additional environment information to perform longitudinal control of vehicle when vehicle distance to obstacle is less than preset distance |
| WO2013072132A1 (en) * | 2011-11-15 | 2013-05-23 | Robert Bosch Gmbh | Method and driver assistance system for the detection of a vehicle environment |
| CN106471554A (en) * | 2014-06-10 | 2017-03-01 | 株式会社电装 | Driving support device |
| CN104401260A (en) * | 2014-10-21 | 2015-03-11 | 奇瑞汽车股份有限公司 | Automatic parking system |
| CN107226088A (en) * | 2016-03-25 | 2017-10-03 | 松下电器（美国）知识产权公司 | Controller, driving control method and program |
| CN106985815A (en) * | 2017-03-14 | 2017-07-28 | 重庆长安汽车股份有限公司 | Remote control parking system and method |
| CN108928341A (en) * | 2017-05-23 | 2018-12-04 | 株式会社万都 | Intelligent parking auxiliary system and its control method |
| CN110356394A (en) * | 2019-07-31 | 2019-10-22 | 重庆长安汽车股份有限公司 | A kind of vehicle actively avoids the method, apparatus and automobile of barrier |
Non-Patent Citations (4)
| Title |
|---|
| Design and Implementation of Autonomous Vehicle Valet Parking System;Kyoung-Wook Min;《Proceedings of the 16th International IEEE Annual Conference on Intelligent Transportation Systems》;20131009;全文 * |
| 一种基于A*算法的空间多自由度机械臂路径规划方法;宗成星等;《合肥工业大学学报(自然科学版)》;20170228(第02期);全文 * |
| 乡村道路环境下农业机器人导航避障算法研究;刘琼;《中国博士学位论文全文数据库 (信息科技辑)》;20180215;全文 * |
| 基于超声波雷达的泊车位类型检测;付鹏等;《汽车实用技术》;20200415(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111516676A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111516676B (en) | Automatic parking method, system, automobile and computer readable storage medium | |
| CN106945660B (en) | A kind of automated parking system | |
| CN110103967B (en) | Automatic lane changing method for vehicle, vehicle control system and vehicle | |
| CN113492851B (en) | Vehicle control device, vehicle control method, and computer program for vehicle control | |
| CN112154455B (en) | Data processing method, equipment and movable platform | |
| CN110481526B (en) | A smart car sensor blind spot pedestrian detection and active collision avoidance method | |
| CN106470886B (en) | Method for establishing the ambient enviroment model of means of transport | |
| CN108928343A (en) | A kind of panorama fusion automated parking system and method | |
| US20190232953A1 (en) | Parking control method and parking control device | |
| CN109733392B (en) | Obstacle avoidance method and device | |
| CN110968086B (en) | Vehicle arrival control method and system | |
| US12073722B2 (en) | Parking lot identification system and method | |
| CN114274952B (en) | Autonomous parking method and system for vertical parking space, storage medium and electronic equipment | |
| CN112537294B (en) | Automatic parking control method and electronic equipment | |
| CN114312760B (en) | Auxiliary parking method with road parking spaces, electronic equipment and automobile | |
| CN111231965A (en) | Method and device for adjusting vehicle control mode and unmanned vehicle | |
| CN112985425A (en) | Vehicle positioning method, device and system based on heterogeneous sensing data fusion | |
| JP2018063476A (en) | Apparatus, method and computer program for driving support | |
| CN113650607B (en) | Low-speed scene automatic driving method, system and automobile | |
| CN114495066A (en) | Method for assisting backing | |
| CN113830080A (en) | Interaction system for automobile parking based on human-computer interaction | |
| US11884265B1 (en) | Parking assistance method and parking assistance device | |
| CN116022126A (en) | Autonomous learning parking method, electronic device and storage medium | |
| CN118545081A (en) | Lane departure warning method and system | |
| CN211044536U (en) | Automatic line patrol and parking system based on tag positioning and recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |