CN116653920B - Automatic parking method and system based on 4D millimeter wave radar - Google Patents
Automatic parking method and system based on 4D millimeter wave radar Download PDFInfo
- Publication number
- CN116653920B CN116653920B CN202310415122.6A CN202310415122A CN116653920B CN 116653920 B CN116653920 B CN 116653920B CN 202310415122 A CN202310415122 A CN 202310415122A CN 116653920 B CN116653920 B CN 116653920B
- Authority
- CN
- China
- Prior art keywords
- parked
- area
- vehicle
- model
- parking
- 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000004873 anchoring Methods 0.000 claims abstract description 63
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 230000035945 sensitivity Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 10
- 230000001788 irregular Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- B60W2420/00—Indexing codes relating to the type of sensors based on the principle of their operation
- B60W2420/40—Photo, light or radio wave sensitive means, e.g. infrared sensors
- B60W2420/408—Radar; Laser, e.g. lidar
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Radar Systems Or Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of automatic parking, in particular to an automatic parking method and system based on a 4D millimeter wave radar, wherein the method comprises the following steps: acquiring an image of an area to be parked to obtain a target reference object; obtaining the reference distance between the tail end of the vehicle and each target reference object based on radar detection; constructing a to-be-parked area model; and determining the position of a parking preset area, and keeping the included angle theta 1 and the included angle theta 2 within a preset angle range until the tail end of the vehicle is parallel to the furthest anchoring boundary line, thus completing parking. According to the invention, under the condition that the parking area is an irregularly-shaped area, whether the area can be parked or not can be accurately judged, if so, the parking area is divided, automatic parking is completed, and the automatic parking efficiency is improved.
Description
Technical Field
The invention relates to the technical field of automatic parking, in particular to an automatic parking method and system based on a 4D millimeter wave radar.
Background
With the continuous development of intelligent traffic, the performance of the millimeter wave radar can be improved in all directions by the 4D millimeter wave radar. The ranging, speed measuring, angle measuring precision and resolution ratio of the 4D radar are greatly improved, the data precision is high, and the method is widely applied to the field of automatic parking.
The parking space of the large city is limited, the automobile is driven into a narrow space to be a common parking problem, a plurality of drivers feel very painful when selecting a plurality of temporary parking spaces, and the parking area or the area with an irregular shape facing the places without the marks for dividing the parking area can be manually judged to be parked, so that the scratch of the automobile is easy to cause or the parking is impossible, and whether the area to be parked can be used as a parking selection and automatic parking can not be accurately judged.
Disclosure of Invention
Aiming at the defects existing in the prior art, the invention provides an automatic parking method and system based on a 4D millimeter wave radar, which can effectively solve the problems of scratch and low parking efficiency of vehicles caused by inaccurate division of parking areas when the parking areas are irregularly shaped areas in the prior art.
In order to achieve the above purpose, the invention is realized by the following technical scheme:
The invention provides an automatic parking method based on a 4D millimeter wave radar, which comprises the following steps:
Acquiring an image of a region to be parked, inputting the image of the region to be parked into an image information feature extraction model to obtain image information features, obtaining a target reference object based on the image information features, and pre-training the image information feature extraction model based on the images of a plurality of parking regions;
Obtaining the reference distance between the tail end of the vehicle and each target reference object based on radar detection;
Determining the farthest anchoring boundary line and the nearest anchoring boundary line, and establishing a coordinate system based on the midpoint of the farthest anchoring boundary line as a coordinate origin, wherein a horizontal straight line of the to-be-parked area, which is farthest from the tail end of the vehicle, is the farthest anchoring boundary line, and a horizontal straight line of the to-be-parked area, which is closest to the tail end of the vehicle, is the nearest anchoring boundary line;
constructing a region model to be parked based on the reference distance, the nearest anchor boundary line and the coordinate system;
Generating a vehicle model based on the vehicle parameter data, judging whether the vehicle model is matched with the to-be-parked area model, and if so, enabling a to-be-parked area corresponding to the to-be-parked area model to be used for parking;
After confirming that the parking area to be parked corresponding to the area to be parked model can be used for parking, determining a parking preset area position according to the area to be parked model and the vehicle position;
after the driving vehicle reaches the position of the preset area, the radar detects the left end point of the tail end of the vehicle to the right end point of the farthest anchoring boundary line to generate a first set straight line, and an included angle theta 1 is formed between the first set straight line and the right side boundary of the to-be-parked area model;
the radar detects the left end point from the right end point of the tail end of the vehicle to the furthest anchoring boundary line to generate a second set straight line, and the second set straight line and the left side edge of the to-be-parked region model form an included angle theta 2;
The included angle theta 1 and the included angle theta 2 of the vehicle are kept within a preset angle range until the tail end of the vehicle is parallel to the farthest anchoring boundary line;
After the tail end of the vehicle is parallel to the furthest anchoring boundary line, the sum of the distances between the vehicle and the left and right side boundary lines of the to-be-parked area model is kept to be a fixed value until the vertical distance between the tail end of the vehicle and the furthest anchoring boundary line is within a preset threshold value interval, and the parking is completed.
Further, determining the parking predetermined area position according to the to-be-parked area model and the vehicle position includes the steps of:
Acquiring the right end point coordinate of the nearest anchoring boundary line and the left end point coordinate of the nearest anchoring boundary line according to the parking area model;
When the vehicle is positioned at the left side of the area to be parked, acquiring a central coordinate of the vehicle position, a right end point coordinate/> of the nearest anchoring boundary line, calculating a preset position angle value beta, and if the value of beta is within a first preset interval value, determining the vehicle position as the position of the area to be parked;
When the vehicle is positioned on the right side of the area to be parked, center coordinates of the vehicle position are acquired, left end point coordinates/> of the nearest anchor boundary line are calculated, and if the value of the predetermined position angle value gamma is within a second predetermined interval value, the vehicle position is determined to be the position of the area to be parked.
Further, constructing the model of the area to be parked based on the reference distance, the nearest anchor boundary line and the coordinate system includes the steps of:
Determining a horizontal distance F from the origin of coordinates to a target reference object on the left side of the origin of coordinates, and generating a left end point of the furthest anchoring boundary line;
Determining a horizontal distance F from the origin of coordinates to a target reference object on the right side of the origin of coordinates, and generating a right endpoint of the furthest anchoring boundary line;
And a rectangular area generated based on the left and right endpoints and the nearest anchor boundary line is the area model to be parked.
Further, the method for judging whether the vehicle model matches the to-be-parked region model includes the following steps:
acquiring the vehicle body length A and the vehicle body width B of a vehicle model, and the horizontal width M and the vertical length N of a region model to be parked;
Judging whether the vehicle model is matched with the to-be-parked area model by taking 0.85M as a threshold value for judging B, taking 0.9N as a threshold value for judging A, judging the numerical value magnitude relation between A and 0.9N and the numerical value magnitude relation between B and 0.85M;
If A is smaller than 0.9N and B is smaller than 0.85M, the vehicle model is matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model can be used for parking;
if A is greater than 0.9N and/or B is greater than 0.85M, the vehicle model is not matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model is not available for parking.
Further, the method further comprises:
If A is smaller than 0.9N and 2B is smaller than 0.85M, equally dividing the farthest boundary line into 2 sections, correspondingly generating 2 equal rectangular area models, wherein the 2 equal rectangular area models are matched with the vehicle model, and the to-be-parked area corresponding to the 2 equal rectangular area models can be used for parking;
If B is smaller than 0.85M, 2A is smaller than 0.9N, 2 bisectors parallel to the farthest boundary line and having a vertical distance of 0.5N are generated, wherein the to-be-parked area model is divided into 2 equal rectangular area models, the 2 equal rectangular area models are matched with the vehicle model, the to-be-parked area corresponding to the 2 equal rectangular area models can be used for parking, and the to-be-parked area corresponding to the rectangular area model close to the farthest boundary line is used as a priority parking area.
Further, in the parking process, the vehicle maintains the included angle 1 and the included angle 2 within the preset angle range, and the method comprises the following steps:
Acquiring a parking speed V of a vehicle, and acquiring a horizontal distance S from the middle point of the tail end of a vehicle model to the right side boundary of a region model to be parked;
correcting the values of the included angle theta 1 and the included angle theta 2 by using interval endpoint values alpha 1, alpha 2, alpha 3 and alpha 4 of a preset angle range;
If 2 > a 2, 1 < a 3, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
if 2 < a 1, 1 > a 4, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
Wherein E is a first constant correction coefficient, is a steering wheel sensitivity coefficient;
The values of the included angle theta 1 and the included angle theta 2 are corrected by controlling the steering wheel deflection angle .
The invention provides an automatic parking system based on a 4D millimeter wave radar, which comprises:
the information acquisition unit is used for acquiring an image of the parking area, and the reference distance between the tail end of the vehicle and each target reference object;
The modeling unit is used for determining the farthest anchoring boundary line and the nearest anchoring boundary line and constructing a region model to be parked and a vehicle model;
the judging unit is used for judging whether the vehicle model is matched with the to-be-parked area model or not;
The correcting unit is used for correcting the rotation angle of the steering wheel of the vehicle in the parking process;
The driving unit is used for driving the vehicle to reverse and park.
Compared with the known public technology, the technical scheme provided by the invention has the following beneficial effects:
the method comprises the steps of obtaining an image to be parked, obtaining target reference objects, obtaining reference distances between the tail end of a vehicle and each target reference object based on radar detection, and constructing a to-be-parked area model; after determining the position of the parking preset area, in the parking process, keeping the included angle theta 1 and the included angle theta 2 within a preset angle range until the tail end of the vehicle is parallel to the furthest anchoring boundary line, and completing parking. According to the invention, under the condition that the parking area is an irregularly-shaped area, whether the area can be parked or not can be accurately judged, if so, the parking area is divided, automatic parking is completed, and the automatic parking efficiency is improved.
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 evident that the drawings in the following description are only some embodiments of the present invention and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
FIG. 1 is a flow chart of an automatic parking method based on a 4D millimeter wave radar of the invention;
FIG. 2 is a schematic diagram of an automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 3 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 4 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 5 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 6 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 7 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
FIG. 8 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention;
Fig. 9 is a schematic diagram of another automatic parking method based on a 4D millimeter wave radar according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In some embodiments, fig. 1 is a flowchart of an automatic parking method based on a 4D millimeter wave radar according to the present invention, including the following steps:
Step S101: acquiring an image of a region to be parked, inputting the image of the region to be parked into an image information feature extraction model to obtain image information features, obtaining a target reference object based on the image information features, and pre-training the image information feature extraction model based on the images of a plurality of parking regions;
Step S102: obtaining the reference distance between the tail end of the vehicle and each target reference object based on radar detection;
step S103: determining the farthest anchoring boundary line and the nearest anchoring boundary line, and establishing a coordinate system based on the midpoint of the farthest anchoring boundary line as a coordinate origin, wherein a horizontal straight line of the to-be-parked area, which is farthest from the tail end of the vehicle, is the farthest anchoring boundary line, and a horizontal straight line of the to-be-parked area, which is closest to the tail end of the vehicle, is the nearest anchoring boundary line;
Step S104: constructing a to-be-parked area model based on the reference distance, the nearest anchoring boundary line and the coordinate system, generating a vehicle model based on vehicle parameter data, judging whether the vehicle model is matched with the to-be-parked area model, and if so, enabling a to-be-parked area corresponding to the to-be-parked area model to be used for parking;
After confirming that the parking area to be parked corresponding to the area to be parked model can be used for parking, determining a parking preset area position according to the area to be parked model and the vehicle position;
Step S105: after the driving vehicle reaches the position of the preset area, the radar detects the right end point from the left end point of the tail end of the vehicle to the farthest right end point of the anchoring boundary line to generate a first set straight line, the first set straight line forms an included angle theta 1 with the right end boundary of the to-be-parked area model, the radar detects the left end point from the right end point of the tail end of the vehicle to the farthest left end point of the anchoring boundary line to generate a second set straight line, and the second set straight line forms an included angle theta 2 with the left end boundary of the to-be-parked area model; keeping the included angle theta 1 and the included angle theta 2 within a preset angle range until the tail end of the vehicle is parallel to the furthest anchoring boundary line;
step S106: after the tail end of the vehicle is parallel to the furthest anchoring boundary line, the sum of the distances between the vehicle and the left and right side boundary lines of the to-be-parked area model is kept to be a fixed value until the vertical distance between the tail end of the vehicle and the furthest anchoring boundary line is within a preset threshold value interval, and the parking is completed.
In summary, the invention judges whether the vehicle model matches the to-be-parked area model under the condition that the parked area is an irregularly-shaped area, if so, the to-be-parked area corresponding to the to-be-parked area model can be used for parking, automatic parking is completed, the condition that the vehicle is scratched or cannot be parked due to manual judgment is avoided, and the parking safety and the parking efficiency can be improved.
The method of the present invention will be described with reference to the schematic diagrams of the 4D millimeter wave radar-based automatic parking method of the present invention shown in fig. 2 to 9.
As shown in fig. 2, in the case that the vehicle is located on the left side of the area to be parked, the 4D millimeter wave radar and the vehicle-mounted looking-around image device located at the tail end of the vehicle collect information of the area to be parked, in step S101, the vehicle-mounted looking-around image device may acquire an image of the area to be parked in real time, input the image of the area to be parked into the image information feature extraction model to obtain image information features, and obtain a target reference object based on the image information features, where the image information feature extraction model is obtained by pre-training based on images of a plurality of parking areas, and the target reference object may be used to represent position information of an object in the parking area. Based on radar detection, obtaining the reference distance between the tail end of the vehicle and each target reference object, it is worth noting that the radar detection equipment can determine the maximum area in a to-be-parked area according to the target reference object, determine the farthest anchoring boundary line and the nearest anchoring boundary line, and establish a coordinate system based on the midpoint of the farthest anchoring boundary line as a coordinate origin, wherein the horizontal straight line farthest from the vertical distance of the tail end of the vehicle in the to-be-parked area is the farthest anchoring boundary line, the horizontal straight line closest to the vertical distance of the tail end of the vehicle in the to-be-parked area is the nearest anchoring boundary line, determine the horizontal distance F from the coordinate origin to the target reference object on the left side of the coordinate origin, and generate the left end point of the farthest anchoring boundary line; and determining the horizontal distance F from the coordinate origin to the target reference object on the right side of the coordinate origin, generating the right end point of the furthest anchoring boundary line, taking the rectangular area generated based on the left and right end points and the closest anchoring boundary line as a to-be-parked area model, and taking the shadow area shown in fig. 2 as the to-be-parked area. As shown in fig. 3, in the case where the vehicle is located on the right side of the area to be parked, the hatched area in fig. 3 is the area to be parked.
In some embodiments, a method of determining whether a vehicle model matches a model of an area to be parked includes the steps of:
Acquiring the vehicle body length A and the vehicle body width B of a vehicle model, and the horizontal width M and the vertical length N of a region model to be parked; judging whether the vehicle model is matched with the to-be-parked area model by taking 0.85M as a threshold value for judging B, taking 0.9N as a threshold value for judging A, judging the numerical value magnitude relation between A and 0.9N and the numerical value magnitude relation between B and 0.85M; if A is smaller than 0.9N and B is smaller than 0.85M, the vehicle model is matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model can be used for parking; if A is greater than 0.9N and/or B is greater than 0.85M, the vehicle model is not matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model is not available for parking.
In conclusion, according to the parameter values of the vehicle model and the parameter values of the to-be-parked area model, the parking area is quantized, whether the area with the irregular shape can be used for parking can be accurately judged, and if yes, parking is automatically completed; if not, the method can give the driver timely feedback, warn the driver not to drive in, help the driver save time, avoid the occurrence of vehicle scratch caused by human judgment, and correspondingly improve the parking efficiency.
In some embodiments, referring to the schematic diagrams of the 4D millimeter wave radar-based automatic parking method shown in fig. 4 and 5, the coordinate axis is forward on the right side of the origin.
Acquiring the right end point coordinate of the nearest anchoring boundary line and the left end point coordinate of the nearest anchoring boundary line according to the parking area model;
When the vehicle is located at the left side of the area to be parked, the center coordinate of the vehicle position is acquired based on the coordinate system, the right end point coordinate/> of the nearest anchor boundary line is calculated, and the predetermined position angle value is calculated, wherein , if the value is within the first preset interval value, the vehicle position is determined to be the position of the predetermined area to be parked, and as shown in fig. 4, the dotted area corresponding to P is the position of the predetermined area to be parked.
When the vehicle is located on the right side of the area to be parked, the center coordinates of the vehicle position are acquired based on the coordinate system, the left end point coordinates/> of the nearest anchor boundary line, the value of the predetermined position angle value gamma is calculated, , and if the value of gamma is within the second preset interval value, the vehicle position is determined to be the position of the predetermined area to be parked, and as shown in fig. 5, the dotted area corresponding to P is the position of the predetermined area to be parked.
In some embodiments, in step S105, referring to fig. 6, during the parking process, the vehicle maintains the included angle 1 and the included angle 2 within the preset angle range, including the following steps:
Acquiring a parking speed V of a vehicle, and acquiring a horizontal distance S from the middle point of the tail end of a vehicle model to the right side boundary of a region model to be parked;
The values of the included angles theta 1 and theta 2 are corrected according to the interval end point values alpha 1, alpha 2, alpha 3 and alpha 4 of the preset angle range, and it is worth to be explained that the alpha 1, the alpha 2, the alpha 3 and the alpha 4 are correspondingly generated based on the data of different regional models to be parked and the vehicle models.
If 2 > a 2, 1 < a 3, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
If 2 < a 1, 1 > a 4, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
Wherein E is a first constant correction coefficient, is a steering wheel sensitivity coefficient;
If theta 2 is larger than alpha 2 and theta 1 is smaller than alpha 3, the vehicle driving unit controls the steering wheel to deflect leftwards by an angle to correct; if 2 < a1, 1 > a 4, the vehicle driving unit controls the steering wheel to deflect right by an angle/> to correct, and corrects the values of the included angle 1 and the included angle 2 by controlling the deflection angle/> of the steering wheel, and keeps the included angle 1 and the included angle 2 within a preset angle range until the tail end of the vehicle is parallel to the farthest anchoring boundary line, as shown in fig. 7, the tail end of the vehicle is parallel to the farthest anchoring boundary line, and the sum of the left and right boundary distances between the vehicle and the to-be-parked area model is kept to be a fixed value, the fixed value is generated based on the to-be-parked area model, different to-be-parked area model parameters are different, and the fixed value is also different until the vertical distance between the tail end of the vehicle and the farthest anchoring boundary line is within a preset threshold interval T, which is worth explaining that the preset threshold interval T can be determined according to the vehicle length a and the vertical length N of the to-be-parked area model, and parking is completed.
In some embodiments, the area to be parked is if there are two cases:
first case: as shown in fig. 8, if a is smaller than 0.9n and 2b is smaller than 0.85M, the furthest boundary line is equally divided into 2 segments, and 2 equal rectangular area models are correspondingly generated, the 2 equal rectangular area models are matched with the vehicle model, the area to be parked corresponding to the 2 equal rectangular area models can be used for parking, and any area to be parked can be selected for automatic parking.
Second case: as shown in fig. 9, if B is smaller than 0.85M, 2A is smaller than 0.9N, 2 bisectors parallel to the furthest boundary line and having a vertical distance of 0.5N are generated, wherein the to-be-parked area model is divided into 2 equal rectangular area models, then the 2 equal rectangular area models each match the vehicle model, the to-be-parked area corresponding to the 2 equal rectangular area models is available for parking, and the to-be-parked area corresponding to the rectangular area model close to the furthest boundary line is taken as the priority parking area. When the vehicle is automatically parked, firstly taking a to-be-parked area corresponding to a rectangular area model close to the nearest boundary line as a reference, keeping the included angle theta 1 and the included angle theta 2 within a preset angle range until the tail end of the vehicle is parallel to a bisector, keeping the sum of the boundary distances between the left side and the right side of the vehicle and the to-be-parked area model as a fixed value, and finishing parking until the vertical distance between the tail end of the vehicle and the farthest anchoring boundary line is within a preset threshold value interval.
In summary, under the two conditions, the vehicles can divide the parking area more accurately, the lower parking space can be saved, the number of parked vehicles is increased under the condition of limited parking space, and the utilization rate of the parking area is improved.
The invention also provides an automatic parking system based on the 4D millimeter wave radar, which comprises an information acquisition unit, a modeling unit, a judging unit, a correction unit and a driving unit;
The information acquisition unit is used for acquiring an image of the parking area, and the reference distance between the tail end of the vehicle and each target reference object; the modeling unit is used for determining the farthest anchoring boundary line and the nearest anchoring boundary line and constructing a region model to be parked and a vehicle model; the judging unit is used for judging whether the vehicle model is matched with the to-be-parked area model or not; the correcting unit is used for correcting the rotation angle of the steering wheel of the vehicle in the parking process; the driving unit is used for driving the vehicle to reverse and park.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; these modifications or substitutions do not depart from the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. An automatic parking method based on a 4D millimeter wave radar is characterized by comprising the following steps:
Acquiring an image of a region to be parked, inputting the image of the region to be parked into an image information feature extraction model to obtain image information features, obtaining a target reference object based on the image information features, and pre-training the image information feature extraction model based on the images of a plurality of parking regions;
Obtaining the reference distance between the tail end of the vehicle and each target reference object based on radar detection;
Determining the farthest anchoring boundary line and the nearest anchoring boundary line, and establishing a coordinate system based on the midpoint of the farthest anchoring boundary line as a coordinate origin, wherein a horizontal straight line of the to-be-parked area, which is farthest from the tail end of the vehicle, is the farthest anchoring boundary line, and a horizontal straight line of the to-be-parked area, which is closest to the tail end of the vehicle, is the nearest anchoring boundary line;
constructing a region model to be parked based on the reference distance, the nearest anchor boundary line and the coordinate system;
Generating a vehicle model based on the vehicle parameter data, judging whether the vehicle model is matched with the to-be-parked area model, and if so, enabling a to-be-parked area corresponding to the to-be-parked area model to be used for parking;
After confirming that the parking area to be parked corresponding to the area to be parked model can be used for parking, determining a parking preset area position according to the area to be parked model and the vehicle position;
after the driving vehicle reaches the position of the preset area, the radar detects the left end point of the tail end of the vehicle to the right end point of the farthest anchoring boundary line to generate a first set straight line, and an included angle theta 1 is formed between the first set straight line and the right side boundary of the to-be-parked area model;
the radar detects the left end point from the right end point of the tail end of the vehicle to the furthest anchoring boundary line to generate a second set straight line, and the second set straight line and the left side edge of the to-be-parked region model form an included angle theta 2;
keeping the included angle theta 1 and the included angle theta 2 within a preset angle range until the tail end of the vehicle is parallel to the furthest anchoring boundary line;
after the tail end of the vehicle is parallel to the furthest anchoring boundary line, keeping the sum of the distances between the vehicle and the left and right side boundary lines of the to-be-parked area model to be a fixed value until the vertical distance between the tail end of the vehicle and the furthest anchoring boundary line is within a preset threshold value interval, and completing parking;
The vehicle keeping the included angle theta 1 and the included angle theta 2 within the preset angle range comprises the following steps:
Acquiring a parking speed V of a vehicle, and acquiring a horizontal distance S from the middle point of the tail end of a vehicle model to the right side boundary of a region model to be parked;
correcting the values of the included angle theta 1 and the included angle theta 2 by using interval endpoint values alpha 1, alpha 2, alpha 3 and alpha 4 of a preset angle range;
If 2 > a 2, 1 < a 3, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
If 2 < a 1, 1 > a 4, correlating to generate a deflection angle , where the deflection angle/> is obtained by the formula:
wherein E is a first constant correction coefficient, is a steering wheel sensitivity coefficient;
The values of the included angle theta 1 and the included angle theta 2 are corrected by controlling the steering wheel deflection angle .
2. The 4D millimeter wave radar-based automatic parking method according to claim 1, wherein determining the parking predetermined area position from the to-be-parked area model and the vehicle position comprises the steps of:
Acquiring the right end point coordinate of the nearest anchoring boundary line and the left end point coordinate of the nearest anchoring boundary line according to the parking area model;
When the vehicle is positioned at the left side of the area to be parked, acquiring a central coordinate of the vehicle position, a right end point coordinate/> of the nearest anchoring boundary line, calculating a preset position angle value beta, and if the value of beta is within a first preset interval value, determining the vehicle position as the position of the area to be parked;
When the vehicle is positioned on the right side of the area to be parked, center coordinates of the vehicle position are acquired, left end point coordinates/> of the nearest anchor boundary line are calculated, and if the value of the predetermined position angle value gamma is within a second predetermined interval value, the vehicle position is determined to be the position of the area to be parked.
3. The 4D millimeter wave radar-based automatic parking method according to claim 2, wherein constructing the region model to be parked based on the reference distance, the nearest anchor boundary line, and the coordinate system comprises the steps of:
Determining a horizontal distance F from the origin of coordinates to a target reference object on the left side of the origin of coordinates, and generating a left end point of the furthest anchoring boundary line;
Determining a horizontal distance F from the origin of coordinates to a target reference object on the right side of the origin of coordinates, and generating a right endpoint of the furthest anchoring boundary line;
And a rectangular area generated based on the left and right endpoints and the nearest anchor boundary line is the area model to be parked.
4. The 4D millimeter wave radar-based automatic parking method according to claim 3, wherein the method of judging whether the vehicle model matches the model of the area to be parked comprises the steps of:
acquiring the vehicle body length A and the vehicle body width B of a vehicle model, and the horizontal width M and the vertical length N of a region model to be parked;
Judging whether the vehicle model is matched with the to-be-parked area model by taking 0.85M as a threshold value for judging B, taking 0.9N as a threshold value for judging A, judging the numerical value magnitude relation between A and 0.9N and the numerical value magnitude relation between B and 0.85M;
If A is smaller than 0.9N and B is smaller than 0.85M, the vehicle model is matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model can be used for parking;
if A is greater than 0.9N and/or B is greater than 0.85M, the vehicle model is not matched with the to-be-parked area model, and the to-be-parked area corresponding to the to-be-parked area model is not available for parking.
5. The 4D millimeter wave radar-based automatic parking method according to claim 4, further comprising:
If A is smaller than 0.9N and 2B is smaller than 0.85M, equally dividing the farthest boundary line into 2 sections, correspondingly generating 2 equal rectangular area models, wherein the 2 equal rectangular area models are matched with the vehicle model, and the to-be-parked area corresponding to the 2 equal rectangular area models can be used for parking;
If B is smaller than 0.85M, 2A is smaller than 0.9N, 2 bisectors parallel to the farthest boundary line and having a vertical distance of 0.5N are generated, wherein the to-be-parked area model is divided into 2 equal rectangular area models, the 2 equal rectangular area models are matched with the vehicle model, the to-be-parked area corresponding to the 2 equal rectangular area models can be used for parking, and the to-be-parked area corresponding to the rectangular area model close to the farthest boundary line is used as a priority parking area.
6. An automatic parking system applied to the 4D millimeter wave radar-based automatic parking method of claim 5, comprising:
the information acquisition unit is used for acquiring an image of the parking area, and the reference distance between the tail end of the vehicle and each target reference object;
The modeling unit is used for determining the farthest anchoring boundary line and the nearest anchoring boundary line and constructing a region model to be parked and a vehicle model;
the judging unit is used for judging whether the vehicle model is matched with the to-be-parked area model or not;
The correcting unit is used for correcting the rotation angle of the steering wheel of the vehicle in the parking process;
The driving unit is used for driving the vehicle to reverse and park.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310415122.6A CN116653920B (en) | 2023-04-18 | 2023-04-18 | Automatic parking method and system based on 4D millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310415122.6A CN116653920B (en) | 2023-04-18 | 2023-04-18 | Automatic parking method and system based on 4D millimeter wave radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116653920A CN116653920A (en) | 2023-08-29 |
| CN116653920B true CN116653920B (en) | 2024-04-16 |
Family
ID=87723120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310415122.6A Active CN116653920B (en) | 2023-04-18 | 2023-04-18 | Automatic parking method and system based on 4D millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116653920B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110444044A (en) * | 2019-08-27 | 2019-11-12 | 纵目科技（上海）股份有限公司 | Vehicle pose detection system, terminal and storage medium based on ultrasonic sensor |
| WO2020168769A1 (en) * | 2019-02-19 | 2020-08-27 | 惠州市德赛西威智能交通技术研究院有限公司 | Vertical parking method |
| CN113320474A (en) * | 2021-07-08 | 2021-08-31 | 长沙立中汽车设计开发股份有限公司 | Automatic parking method and device based on panoramic image and human-computer interaction |
| CN114523959A (en) * | 2022-04-08 | 2022-05-24 | 南昌智能新能源汽车研究院 | Automatic parking trajectory planning method based on vehicle kinematics model |
| CN115629386A (en) * | 2022-12-21 | 2023-01-20 | 广州森弘信息科技有限公司 | High-precision positioning system and method for automatic parking |
-
2023
- 2023-04-18 CN CN202310415122.6A patent/CN116653920B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020168769A1 (en) * | 2019-02-19 | 2020-08-27 | 惠州市德赛西威智能交通技术研究院有限公司 | Vertical parking method |
| CN110444044A (en) * | 2019-08-27 | 2019-11-12 | 纵目科技（上海）股份有限公司 | Vehicle pose detection system, terminal and storage medium based on ultrasonic sensor |
| CN113320474A (en) * | 2021-07-08 | 2021-08-31 | 长沙立中汽车设计开发股份有限公司 | Automatic parking method and device based on panoramic image and human-computer interaction |
| CN114523959A (en) * | 2022-04-08 | 2022-05-24 | 南昌智能新能源汽车研究院 | Automatic parking trajectory planning method based on vehicle kinematics model |
| CN115629386A (en) * | 2022-12-21 | 2023-01-20 | 广州森弘信息科技有限公司 | High-precision positioning system and method for automatic parking |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116653920A (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1890264B1 (en) | Vehicle, image processing system, image processing method, image processing program, image processing system configuration method, and server | |
| CN109522804B (en) | Road edge identification method and system | |
| US10023198B2 (en) | Method for determining a relative gradient of a roadway | |
| CN101750049B (en) | Monocular vision vehicle distance measuring method based on road and vehicle information | |
| US20140063252A1 (en) | Method for calibrating an image capture device | |
| US20100118140A1 (en) | Parking assisting apparatus and method | |
| CN108470142B (en) | Lane positioning method based on inverse perspective projection and lane distance constraint | |
| CN112009462B (en) | Forward automatic parking method and device | |
| CN103909930A (en) | Method for auxiliary control of traveling along with vehicle ahead | |
| JP6020729B2 (en) | Vehicle position / posture angle estimation apparatus and vehicle position / posture angle estimation method | |
| CN110927762B (en) | Positioning correction method, device and system | |
| LU502288B1 (en) | Method and system for detecting position relation between vehicle and lane line, and storage medium | |
| CN111267862B (en) | Method and system for constructing virtual lane line depending on following target | |
| CN112166336A (en) | Method and device for calibrating pitching installation angle of millimeter wave radar, vehicle control system and vehicle | |
| CN109827516B (en) | Method for measuring distance through wheel | |
| CN111325187B (en) | Lane position identification method and device | |
| US8031908B2 (en) | Object recognizing apparatus including profile shape determining section | |
| CN116653920B (en) | Automatic parking method and system based on 4D millimeter wave radar | |
| KR101424636B1 (en) | Automatic parking system for vehicle | |
| CN111231946B (en) | Low-sight-distance vehicle safe driving control method | |
| JP3930309B2 (en) | Vehicle driving support device | |
| CN115222779A (en) | Vehicle cut-in detection method and device and storage medium | |
| CN112061118B (en) | Vehicle collision prediction method and device | |
| CN112070839A (en) | Method and equipment for positioning and ranging rear vehicle transversely and longitudinally | |
| CN114820819B (en) | Expressway automatic driving method and system |
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 |