CN112216142A - Vehicle vision positioning system based on specific scene - Google Patents
Vehicle vision positioning system based on specific scene Download PDFInfo
- Publication number
- CN112216142A CN112216142A CN202011088251.1A CN202011088251A CN112216142A CN 112216142 A CN112216142 A CN 112216142A CN 202011088251 A CN202011088251 A CN 202011088251A CN 112216142 A CN112216142 A CN 112216142A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- parking space
- pose
- target parking
- position information
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/042—Detecting movement of traffic to be counted or controlled using inductive or magnetic detectors
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096805—Systems involving transmission of navigation instructions to the vehicle where the transmitted instructions are used to compute a route
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of vehicle positioning systems, and discloses a vehicle vision positioning system based on a specific scene, which comprises: the system comprises a parking lot, a vehicle detection unit, a scene database, an image acquisition unit and a server side, wherein the server side is provided with a target parking space configuration module, a specific identification module and a pose calculation module; the server side acquires image information to be compared from the image acquisition unit and determines corresponding reference image information; the pose calculation module calculates three-dimensional position information and relative pose information of the vehicle through a PNP pose algorithm, calculates position information and orientation information of the vehicle relative to a geodetic coordinate system according to the relative pose information of the vehicle, generates actual position information according to the position information and the orientation information, and sends the actual position information to the vehicle end.
Description
Technical Field
The invention relates to the technical field of vehicle positioning systems, in particular to a vehicle vision positioning system based on a specific scene.
Background
A Self-driving vehicle (Self-driving car), also called as an unmanned vehicle, a computer-driven vehicle, or a wheeled mobile robot, is an intelligent vehicle that realizes unmanned driving through a computer system. With the popularization of the autonomous vehicle, the autonomous vehicle may be used as a taxi or a public transportation, and when a passenger uses the autonomous vehicle, the passenger needs to input a destination, and the autonomous vehicle generates a driving route based on a current position and the destination and drives according to the generated driving route.
When a car drives into a parking lot, because the environment of the parking lot is complex, a lot of time is spent for automatically searching for a parking space after the car drives into the parking lot, and under certain scenes in the field of automatic driving, the surrounding environment needs to be correct and timely responded; in order to enable the automobile to automatically drive into the target parking space from the entrance of the parking lot, not only the coordinates of the automobile under a physical coordinate system need to be accurately known, but also the self pose needs to be known.
At present, methods for positioning an automobile mainly determine the position of the automobile on a map through a GPS signal and a radar, but the methods can only roughly determine position information and cannot accurately obtain the pose of the automobile.
Disclosure of Invention
The invention aims to provide a vehicle vision positioning system based on a specific scene, which has the advantages of being capable of accurately positioning the coordinate position and the body posture of an automobile and facilitating accurate parking and warehousing of the automobile.
In order to achieve the above purpose, the basic scheme of the invention is as follows: a scene-specific based vehicle vision localization system, comprising:
the image acquisition unit is arranged in the vehicle and used for acquiring real-time images of scene areas and producing image information to be compared;
the vehicle detection unit is arranged at an entrance of the parking lot and is used for detecting the condition of a vehicle running in the area where the entrance is located;
defining a plurality of scene areas corresponding to a parking lot, wherein each scene area corresponds to a specific marker, and the distance between the adjacent specific markers is smaller than a first preset distance so that each image information to be compared, which is acquired by an image acquisition unit, at least comprises two specific markers at different positions;
the vehicle vision positioning system comprises a scene database, wherein the scene database is configured with reference image information, and the reference image information comprises a characteristic map area corresponding to a specific marker;
the server end is connected with the vehicle end; the server side is provided with a target parking space configuration module, the target parking space configuration module is connected to the vehicle detection unit, and when the vehicle detection unit detects that a vehicle enters an entrance of a parking lot, the target parking space configuration module configures a target parking space corresponding to the vehicle; the target parking space configuration module is configured with a vehicle control strategy, the vehicle control strategy comprises a plurality of vehicle control instructions, and each vehicle control instruction takes a target parking space as an index and controls a vehicle to drive to the target parking space;
the server side is also provided with a specific identification module, the specific identification module is connected with the image acquisition unit and the scene database, the specific identification module extracts feature points in the image information to be compared and generates a feature region according to the feature points, a specific marker closest to the feature region is called from the scene database according to the feature region, and reference image information corresponding to the specific marker is determined according to the specific marker;
the server side is also provided with a pose calculation module, the pose calculation module is provided with a pose calculation strategy, the pose calculation strategy is provided with a PNP (plug-and-play) pose algorithm, the pose calculation strategy is used for solving three-dimensional position information and relative pose information of the vehicle through the PNP pose algorithm according to specific markers in the image information to be compared, the relative pose information comprises a pitch angle, a yaw angle and a roll angle of the vehicle, the position information and the orientation information of the vehicle relative to a geodetic coordinate system are calculated according to the relative pose information of the vehicle, actual position information is generated according to the position information and the orientation information, and the actual position information is sent to the vehicle side;
the vehicle control instruction comprises a plurality of initial position information and target position information, and the target position information of each vehicle control instruction corresponds to the initial position information of the next vehicle control instruction;
the vehicle control strategy collects images to be compared in real time and uploads the images to the server, the specific identification module determines reference image information, the pose calculation module calculates the reference image information to obtain actual position information of the vehicle, a vehicle execution command is generated according to target position information corresponding to the vehicle control command to control the vehicle to move to a position corresponding to the target position information, and the vehicle control command is sequentially executed until the vehicle moves to the target parking space.
Further, the target parking space configuration module configures a target parking space for the vehicle according to vehicle parking information of a parking lot, position information of the vehicle and pre-stored map information of the parking lot; the target parking space configuration module is configured with a target parking space configuration strategy, and the target parking space configuration strategy is configured with a target parking space configuration step;
a first target parking space configuration step: the target parking space configuration strategy calculates the distance between each empty parking space and the position information of the vehicle, and the empty parking space with the shortest distance is preconfigured to the vehicle;
step two, target parking space configuration: the target parking space configuration strategy calculates a parking driving route from the vehicle to an empty parking space allocated to the vehicle according to the map information of the parking lot;
step three, target parking space configuration: and the target parking space configuration strategy sends the parking driving route to each vehicle control instruction, and the vehicle control instruction controls the vehicle to drive to the target parking space.
The vehicle body data acquisition unit is arranged in the vehicle and connected to the server side, and is used for acquiring the relative distance between the vehicle and the specific marker in the current scene area, the vehicle travel distance and the wheel torque.
Further, the PNP pose algorithm is configured with preset weight parameters, and the preset weight parameters are obtained by establishing a weight relationship according to the relative distance between the vehicle and the specific marker in the current scene area, the vehicle travel distance, the wheel torque and the pixel occupied by the specific marker in the image information to be compared.
Further, the pose calculation module is configured with a pose calculation correction strategy, and the pose calculation correction strategy obtains a correction weight parameter according to the deviation between the actual position information of the vehicle under the current vehicle control instruction and the initial position information of the vehicle under the next vehicle control instruction;
when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is larger than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; and when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is smaller than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
Further, when the PNP pose algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is larger than the vehicle travel distance acquired by the vehicle body data acquisition unit, the preset weight parameter is reduced, a correction weight parameter is generated, and the PNP pose correction algorithm is generated according to the correction weight parameter; and when the PNP pose algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is smaller than the vehicle travel distance acquired by the vehicle body data acquisition unit, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
Further, when the wheel torque of the vehicle in the current scene area is calculated and obtained by the PNP pose correction algorithm and is larger than the wheel torque obtained by the vehicle body data obtaining unit, the preset weight parameter is reduced, a correction weight parameter is generated, and the PNP pose correction algorithm is generated according to the correction weight parameter; and when the wheel torque of the vehicle in the current scene area is smaller than the wheel torque acquired by the vehicle body data acquisition unit through calculation of the PNP pose correction algorithm, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
Furthermore, the scene database also comprises a plurality of obstacle blocks and obstacle types corresponding to the obstacle blocks, when the server side acquires the image information to be compared, the specific identification module can extract the obstacle blocks in the image information to be compared and send the corresponding obstacle types to the target parking space configuration module, and the target parking space configuration module resets the vehicle driving route.
Further, the vehicle control strategy comprises a vehicle speed control command and is configured with a safe vehicle speed threshold value, and when the current vehicle speed of the vehicle is greater than the safe vehicle speed threshold value, the vehicle control strategy command generates a vehicle deceleration control command and sends the vehicle deceleration control command to a brake control unit at the vehicle end.
Further, the vehicle detection unit comprises an entrance ground induction coil and a controller, the entrance ground induction coil is buried under the ground surface of the entrance and used for detecting whether a vehicle passes through, and the controller is used for processing the acquired data and controlling an entrance barrier gate.
Compared with the prior art, the scheme has the beneficial effects that:
1. when the vehicle drives in the area of the entrance, the target parking space configuration module pre-configures the empty parking space to the vehicle; and calculating a parking driving route from the vehicle to the empty parking space allocated to the vehicle according to the map information of the parking lot, sending the parking driving route to each vehicle control instruction, generating a vehicle execution command by the target position information corresponding to the vehicle control instruction so as to control the vehicle to move to the position corresponding to the target position information, and sequentially executing the vehicle control instruction until the vehicle moves to the target parking space.
2. The pose calculation strategy calculates three-dimensional position information and relative pose information of the vehicle through a PNP (plug-and-play) pose algorithm according to specific markers of image information to be compared in a scene, the relative pose information comprises a pitch angle, a yaw angle and a roll angle of the vehicle, the position information and the orientation information of the vehicle relative to a geodetic coordinate system are calculated according to the relative pose information of the vehicle, actual position information is generated according to the position information and the orientation information, and the actual position information is sent to a vehicle end.
3. The vehicle sequentially passes through the first scene area, the second scene area and the third scene area according to the parking driving route until the vehicle reaches the target parking space; before the vehicle reaches the target parking space, in the process that the vehicle passes through each scene area, the camera in the vehicle compares the real-time image information to be compared with the reference image information, and calculates the real-time three-dimensional position information and the relative posture information of the vehicle, so that the vehicle can be accurately controlled to drive into the target parking space.
Drawings
FIG. 1 is a system architecture diagram of the present invention.
Reference numerals in the drawings of the specification include: the system comprises a parking lot 1, a vehicle end 2, an image acquisition unit 3, a vehicle detection unit 4, a vehicle body data acquisition unit 5, a scene database 6, a target parking space configuration module 7, a server end 8, a specific identification module 9 and a pose calculation module 10.
Detailed Description
The invention will be described in further detail by means of specific embodiments with reference to the accompanying drawings:
example (b):
a scene-specific based vehicle vision positioning system, as shown in fig. 1, comprising:
the image acquisition unit 3 is a camera arranged in the vehicle and is used for acquiring a real-time image of a scene area and producing image information to be compared;
the vehicle body data acquisition unit 5 is arranged in the vehicle and connected to the server end 8, and the vehicle body data acquisition unit 5 is used for acquiring the relative distance between the vehicle and a specific marker, the vehicle travel distance and the wheel torque of the vehicle in the current scene area;
the vehicle detection unit 4 comprises an entrance ground induction coil and a controller, the entrance ground induction coil is buried under the ground surface of an entrance and used for detecting whether a vehicle passes through, and the controller is used for processing the acquired data and controlling an entrance barrier gate;
defining a plurality of scene areas corresponding to a parking lot 1, wherein each scene area corresponds to a specific marker, and the distance between adjacent specific markers is smaller than a first preset distance so that each image information to be compared, acquired by an image acquisition unit 3, at least comprises two specific markers at different positions;
the vehicle vision positioning system comprises a scene database 6, the scene database 6 is configured with reference image information, and the reference image information comprises a characteristic map area corresponding to a specific marker;
the server end 8 is connected to the vehicle end 2; the server end 8 is configured with a target parking space configuration module 7, the target parking space configuration module 7 is connected to the vehicle detection unit 4, and when the vehicle detection unit 4 detects that the vehicle enters the entrance of the parking lot 1, the target parking space configuration module 7 configures a target parking space corresponding to the vehicle; the target parking space configuration module 7 configures a target parking space for the vehicle according to the vehicle parking information of the parking lot 1, the position information of the vehicle and the pre-stored map information of the parking lot 1; a target parking space configuration strategy is configured in the target parking space configuration module 7, and the target parking space configuration strategy is configured with a target parking space configuration step;
a first target parking space configuration step: the target parking space configuration strategy calculates the distance between each empty parking space and the position information of the vehicle, and the empty parking space with the shortest distance is pre-configured to the vehicle;
step two, target parking space configuration: the target parking space configuration strategy calculates a parking driving route from the vehicle to an empty parking space allocated to the vehicle according to the map information of the parking lot 1;
step three, target parking space configuration: and the target parking space configuration strategy sends the parking driving route to each vehicle control instruction, and the vehicle control instruction controls the vehicle to drive to the target parking space.
The target parking space configuration module 7 is configured with a vehicle control strategy, the vehicle control strategy comprises a plurality of vehicle control instructions, and each vehicle control instruction takes a target parking space as an index and controls a vehicle to drive to the target parking space; the vehicle control instruction comprises a plurality of initial position information and target position information, and the target position information of each vehicle control instruction corresponds to the initial position information of the next vehicle control instruction;
the vehicle control strategy collects images to be compared in real time and uploads the images to the server, the reference image information is determined through the specific identification module 9, the actual position information of the vehicle is obtained through calculation of the pose calculation module 10 according to the reference image information, a vehicle execution command is generated according to the target position information corresponding to the vehicle control command so as to control the vehicle to move to the position corresponding to the target position information, and the vehicle control command is sequentially executed until the vehicle moves to the target parking space;
the vehicle control strategy comprises a vehicle speed control command and is configured with a safe vehicle speed threshold value, and when the current vehicle speed of the vehicle is greater than the safe vehicle speed threshold value, the vehicle control strategy command generates a vehicle deceleration control command and sends the vehicle deceleration control command to a brake control unit of the vehicle end 2.
The server 8 is further configured with a specific identification module 9, the specific identification module 9 is connected to the image acquisition unit 3 and the scene database 6, the specific identification module 9 extracts feature points in the image information to be compared and generates a feature region according to the feature points, and calls a specific marker closest to the feature region from the scene database 6 according to the feature region, and determines reference image information corresponding to the specific marker according to the specific marker.
The server end 8 is further provided with a pose calculation module 10, the pose calculation module 10 is provided with a pose calculation strategy, the pose calculation strategy is provided with a PNP pose algorithm, the pose calculation strategy is used for calculating three-dimensional position information and relative pose information of the vehicle according to specific markers in the image information to be compared and the PNP pose algorithm, the relative pose information comprises a pitch angle, a yaw angle and a roll angle of the vehicle, the position information and orientation information of the vehicle relative to a ground coordinate system are calculated according to the relative pose information of the vehicle, actual position information is generated according to the position information and the orientation information, and the actual position information is sent to the vehicle end 2.
The PNP pose algorithm is configured with preset weight parameters, and the preset weight parameters are obtained by establishing a weight relationship according to the relative distance between the vehicle and the specific marker in the current scene area, the vehicle travel distance, the wheel torque and the pixel occupied by the specific marker in the image information to be compared.
The pose calculation module 10 is configured with a pose calculation and correction strategy, and the pose calculation and correction strategy obtains a correction weight parameter according to the deviation between the actual position information of the vehicle under the current vehicle control instruction and the initial position information of the vehicle under the next vehicle control instruction;
when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is larger than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit 5, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is smaller than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit 5, the preset weight parameter is reduced, a correction weight parameter is generated, and the PNP pose correction algorithm is generated according to the correction weight parameter.
When the PNP pose correction algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is larger than the vehicle travel distance acquired by the vehicle body data acquisition unit 5, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; and when the PNP pose algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is smaller than the vehicle travel distance acquired by the vehicle body data acquisition unit 5, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
When the wheel torque of the vehicle in the current scene area is larger than the wheel torque acquired by the vehicle body data acquisition unit 5 through calculation of the PNP pose correction algorithm, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; when the wheel torque of the vehicle in the current scene area is smaller than the wheel torque acquired by the vehicle body data acquisition unit 5 through calculation of the PNP pose correction algorithm, the preset weight parameters are increased, correction weight parameters are generated, and the PNP pose correction algorithm is generated according to the correction weight parameters.
The scene database 6 further comprises a plurality of obstacle blocks and obstacle types corresponding to the obstacle blocks, when the server 8 obtains the image information to be compared, the specific identification module 9 can extract the obstacle blocks in the image information to be compared and send the corresponding obstacle types to the target parking space configuration module 7, and the target parking space configuration module 7 resets a vehicle driving route.
The specific implementation mode of the scheme is as follows:
a vehicle drives at an entrance of a parking lot 1, an image of a scene area at the entrance is acquired by an image acquisition unit 3 in the vehicle, and a corresponding map model of the parking lot 1 is called from a server end 8; meanwhile, when the ground induction coil at the entrance of the parking lot 1 detects that the vehicle runs in the area of the entrance, the barrier at the entrance of the parking lot 1 is automatically opened, and the vehicle establishes initial position coordinates (X0, Y0) at the entrance of the parking lot 1; synchronously, when a vehicle enters the parking lot 1, the server 8 receives a signal that the vehicle enters the parking lot 1, and the target parking space configuration strategy of the target parking space configuration module 7 calculates the distance between each empty parking space and the position information of the vehicle, and pre-configures the empty parking space with the shortest distance to the vehicle.
After the target parking space configuration module 7 configures the target parking space for the vehicle, a plurality of continuous scene areas are generated between the vehicle located at the entrance of the parking lot 1 and the target parking space, meanwhile, a vehicle control strategy generates a plurality of vehicle control instructions according to the plurality of continuous scene areas, each vehicle control instruction correspondingly controls the vehicle to run between adjacent scene areas, target position information of each vehicle control instruction corresponds to initial position information of the next vehicle control instruction, and each vehicle control instruction takes the target parking space as an index and controls the vehicle to run to the target parking space.
The first vehicle control command controls the vehicle to drive from the vehicle at the entrance of the parking lot 1 to the preset target position (X1, Y1) of the first scene area, the vehicle reaching the target position can shoot the image of the first scene area and generate first scene comparison image information, the camera sends the first scene comparison image information to the server, the specific identification module 9 extracts the feature points in the first scene comparison image information, generates the feature areas according to the feature points, and calls the specific markers closest to the feature areas from the scene database 6 according to the feature areas,
the pose calculation strategy calculates three-dimensional position information and relative pose information of the vehicle through a PNP pose algorithm according to the specific markers of the image information to be compared in the first scene, calculates position information and orientation information of the vehicle relative to a geodetic coordinate system according to the relative pose information of the vehicle, generates actual position information according to the position information and the orientation information, and sends the actual position information to the vehicle end 2 to obtain the coordinate position (X1 ', Y1') of the actual position information of the vehicle end 2.
And the pose calculation correction strategy obtains correction weight parameters according to the deviation of the coordinate positions (X1 ', Y1') of the actual position information of the vehicle end 2 and the preset target positions (X1, Y1) of the first scene area, and generates a PNP pose first correction algorithm according to the correction weight parameters.
The second vehicle control command controls the vehicle to drive from the actual position of the first scene area to the preset target position (X2, Y2) of the first scene area, the camera of the vehicle collects the image of the second scene area and generates second scene comparison image information, the camera sends the second scene comparison image information to the server, the pose calculation strategy obtains the three-dimensional position information and the relative pose information of the vehicle through a PNP pose first correction algorithm according to the specific marker of the second scene comparison image information and generates actual position information, the actual position information is sent to the vehicle end 2, the coordinate position of the obtained actual position information of the vehicle end 2 is (X2 ', Y2'), the corrected PNP pose first correction algorithm can more accurately reflect the deviation between the actual position and the target position of the vehicle, and the vehicle continuously drives into each subsequent scene area, and the PNP pose algorithm is continuously corrected, so that the vehicle can smoothly run to a target parking space under the guidance of a plurality of vehicle control instructions.
When the vehicle runs to the target parking space, the vehicle control instruction ends the control, and the target parking space configuration module 7 updates the vehicle parking information of the parking lot 1, so that the target parking space can be accurately configured for the subsequent vehicles entering the parking lot 1.
The foregoing is merely an example of the present invention and common general knowledge of known specific structures and features of the embodiments is not described herein in any greater detail. It should be noted that, for those skilled in the art, without departing from the structure of the present invention, several changes and modifications can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicability of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.
Claims (10)
1. Vehicle vision positioning system based on specific scene, its characterized in that:
the method comprises the following steps:
the image acquisition unit is arranged in the vehicle and used for acquiring real-time images of scene areas and producing image information to be compared;
the vehicle detection unit is arranged at an entrance of the parking lot and is used for detecting the condition of a vehicle running in the area where the entrance is located;
defining a plurality of scene areas corresponding to a parking lot, wherein each scene area corresponds to a specific marker, and the distance between the adjacent specific markers is smaller than a first preset distance so that each image information to be compared, which is acquired by an image acquisition unit, at least comprises two specific markers at different positions;
the vehicle vision positioning system comprises a scene database, wherein the scene database is configured with reference image information, and the reference image information comprises a characteristic map area corresponding to a specific marker;
the server end is connected with the vehicle end; the server side is provided with a target parking space configuration module, the target parking space configuration module is connected to the vehicle detection unit, and when the vehicle detection unit detects that a vehicle enters an entrance of a parking lot, the target parking space configuration module configures a target parking space corresponding to the vehicle; the target parking space configuration module is configured with a vehicle control strategy, the vehicle control strategy comprises a plurality of vehicle control instructions, and each vehicle control instruction takes a target parking space as an index and controls a vehicle to drive to the target parking space;
the server side is also provided with a specific identification module, the specific identification module is connected with the image acquisition unit and the scene database, the specific identification module extracts feature points in the image information to be compared and generates a feature region according to the feature points, a specific marker closest to the feature region is called from the scene database according to the feature region, and reference image information corresponding to the specific marker is determined according to the specific marker;
the server side is also provided with a pose calculation module, the pose calculation module is provided with a pose calculation strategy, the pose calculation strategy is provided with a PNP (plug-and-play) pose algorithm, the pose calculation strategy is used for solving three-dimensional position information and relative pose information of the vehicle through the PNP pose algorithm according to specific markers in the image information to be compared, the relative pose information comprises a pitch angle, a yaw angle and a roll angle of the vehicle, the position information and the orientation information of the vehicle relative to a geodetic coordinate system are calculated according to the relative pose information of the vehicle, actual position information is generated according to the position information and the orientation information, and the actual position information is sent to the vehicle side;
the vehicle control instruction comprises a plurality of initial position information and target position information, and the target position information of each vehicle control instruction corresponds to the initial position information of the next vehicle control instruction;
the vehicle control strategy collects images to be compared in real time and uploads the images to the server, the specific identification module determines reference image information, the pose calculation module calculates the reference image information to obtain actual position information of the vehicle, a vehicle execution command is generated according to target position information corresponding to the vehicle control command to control the vehicle to move to a position corresponding to the target position information, and the vehicle control command is sequentially executed until the vehicle moves to the target parking space.
2. The scene-specific vehicle visual positioning system of claim 1, wherein: the target parking space configuration module configures a target parking space for the vehicle according to vehicle parking information of a parking lot, position information of the vehicle and pre-stored map information of the parking lot; the target parking space configuration module is configured with a target parking space configuration strategy, and the target parking space configuration strategy is configured with a target parking space configuration step;
a first target parking space configuration step: the target parking space configuration strategy calculates the distance between each empty parking space and the position information of the vehicle, and the empty parking space with the shortest distance is preconfigured to the vehicle;
step two, target parking space configuration: the target parking space configuration strategy calculates a parking driving route from the vehicle to an empty parking space allocated to the vehicle according to the map information of the parking lot;
step three, target parking space configuration: and the target parking space configuration strategy sends the parking driving route to each vehicle control instruction, and the vehicle control instruction controls the vehicle to drive to the target parking space.
3. The scene-specific vehicle visual positioning system of claim 1, wherein: the vehicle body data acquisition unit is arranged in the vehicle and connected to the server side, and the vehicle body data acquisition unit is used for acquiring the relative distance between the vehicle and the specific marker in the current scene area, the vehicle travel distance and the wheel torque.
4. The scene-specific vehicle visual positioning system of claim 3, wherein: the PNP pose algorithm is configured with preset weight parameters, and the preset weight parameters are obtained by establishing a weight relationship according to the relative distance between the vehicle and the specific marker in the current scene area, the vehicle travel distance, the wheel torque and the pixel occupied by the specific marker in the image information to be compared.
5. The scene-specific vehicle visual positioning system of claim 4, wherein: the pose calculation module is configured with a pose calculation correction strategy, and the pose calculation correction strategy obtains a correction weight parameter according to the deviation between the actual position information of the vehicle under the current vehicle control instruction and the initial position information of the vehicle under the next vehicle control instruction;
when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is larger than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; and when the PNP pose correction algorithm calculates that the relative distance between the vehicle and the specific marker in the current scene area is smaller than the relative distance between the vehicle and the specific marker acquired by the vehicle body data acquisition unit, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
6. The scene-specific vehicle visual positioning system of claim 5, wherein: when the PNP pose algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is larger than the vehicle travel distance acquired by the vehicle body data acquisition unit, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; and when the PNP pose algorithm calculates that the vehicle travel distance of the vehicle in the current scene area is smaller than the vehicle travel distance acquired by the vehicle body data acquisition unit, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
7. The scene-specific vehicle visual positioning system of claim 5, wherein: when the wheel torque of the vehicle in the current scene area is larger than the wheel torque acquired by the vehicle body data acquisition unit through calculation of the PNP pose correction algorithm, reducing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter; and when the wheel torque of the vehicle in the current scene area is smaller than the wheel torque acquired by the vehicle body data acquisition unit through calculation of the PNP pose correction algorithm, increasing the preset weight parameter and generating a correction weight parameter, and generating the PNP pose correction algorithm according to the correction weight parameter.
8. The scene-specific vehicle visual positioning system of claim 1, wherein: the scene database also comprises a plurality of obstacle blocks and obstacle types corresponding to the obstacle blocks, when the server side obtains the image information to be compared, the specific identification module can extract the obstacle blocks in the image information to be compared and send the corresponding obstacle types to the target parking space configuration module, and the target parking space configuration module resets the vehicle driving route.
9. The scene-specific vehicle visual positioning system of claim 1, wherein: the vehicle control strategy comprises a vehicle speed control instruction and is configured with a safe vehicle speed threshold, and when the current vehicle speed of the vehicle is greater than the safe vehicle speed threshold, the vehicle control strategy instruction generates a vehicle deceleration control instruction and sends the vehicle deceleration control instruction to a brake control unit at the vehicle end.
10. The scene-specific vehicle visual positioning system of claim 1, wherein: the vehicle detection unit comprises an entrance ground induction coil and a controller, wherein the entrance ground induction coil is buried under the ground surface of an entrance and used for detecting whether a vehicle passes through, and the controller is used for processing the acquired data and controlling an entrance barrier gate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011088251.1A CN112216142B (en) | 2020-10-13 | 2020-10-13 | Vehicle vision positioning system based on specific scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011088251.1A CN112216142B (en) | 2020-10-13 | 2020-10-13 | Vehicle vision positioning system based on specific scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112216142A true CN112216142A (en) | 2021-01-12 |
| CN112216142B CN112216142B (en) | 2021-07-13 |
Family
ID=74053644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011088251.1A Active CN112216142B (en) | 2020-10-13 | 2020-10-13 | Vehicle vision positioning system based on specific scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112216142B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113605763A (en) * | 2021-07-29 | 2021-11-05 | 中汽创智科技有限公司 | Stereo garage auxiliary positioning method and device for AVP |
| CN113928320A (en) * | 2021-10-13 | 2022-01-14 | 中科云谷科技有限公司 | Reversing guide method and device for engineering vehicle and processor |
| CN114267200A (en) * | 2021-12-28 | 2022-04-01 | 广东伟邦科技股份有限公司 | Vehicle management method based on visual identification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441813A (en) * | 2008-12-19 | 2009-05-27 | 北京中星微电子有限公司 | Parking ground management system based on image processing |
| CN106710293A (en) * | 2016-10-31 | 2017-05-24 | 中原智慧城市设计研究院有限公司 | Dynamic and intelligent vehicle guidance method for underground parking lot |
| CN106781688A (en) * | 2017-03-28 | 2017-05-31 | 重庆大学 | Pilotless automobile Entrance guides system and method |
| CN106952495A (en) * | 2017-03-06 | 2017-07-14 | 深圳市贝尔信智能***有限公司 | A kind of parking lot multifunctional management system of intelligent building |
| CN109949609A (en) * | 2019-04-30 | 2019-06-28 | 广州小鹏汽车科技有限公司 | A kind of positioning correction method and system, vehicle of vehicle |
-
2020
- 2020-10-13 CN CN202011088251.1A patent/CN112216142B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441813A (en) * | 2008-12-19 | 2009-05-27 | 北京中星微电子有限公司 | Parking ground management system based on image processing |
| CN106710293A (en) * | 2016-10-31 | 2017-05-24 | 中原智慧城市设计研究院有限公司 | Dynamic and intelligent vehicle guidance method for underground parking lot |
| CN106952495A (en) * | 2017-03-06 | 2017-07-14 | 深圳市贝尔信智能***有限公司 | A kind of parking lot multifunctional management system of intelligent building |
| CN106781688A (en) * | 2017-03-28 | 2017-05-31 | 重庆大学 | Pilotless automobile Entrance guides system and method |
| CN109949609A (en) * | 2019-04-30 | 2019-06-28 | 广州小鹏汽车科技有限公司 | A kind of positioning correction method and system, vehicle of vehicle |
Non-Patent Citations (1)
| Title |
|---|
| 吴晓蝶 等: "一种车辆入库自动循环停车***的方案设计", 《电脑知识与技术》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113605763A (en) * | 2021-07-29 | 2021-11-05 | 中汽创智科技有限公司 | Stereo garage auxiliary positioning method and device for AVP |
| CN113928320A (en) * | 2021-10-13 | 2022-01-14 | 中科云谷科技有限公司 | Reversing guide method and device for engineering vehicle and processor |
| CN114267200A (en) * | 2021-12-28 | 2022-04-01 | 广东伟邦科技股份有限公司 | Vehicle management method based on visual identification |
| CN114267200B (en) * | 2021-12-28 | 2023-11-21 | 广东伟邦科技股份有限公司 | Vehicle management method based on visual recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112216142B (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112216142B (en) | Vehicle vision positioning system based on specific scene | |
| JP6930600B2 (en) | Vehicle position estimation device and vehicle control device | |
| US11731665B2 (en) | Driving control method and driving control device | |
| CN107933560B (en) | Vehicle control device | |
| US10095239B1 (en) | Autonomous vehicle paletization system | |
| CN109866759B (en) | Automatic parking method | |
| US11584388B2 (en) | Driving control method and driving control apparatus | |
| US11099561B1 (en) | Control of an autonomous vehicle in unmapped regions | |
| EP3696788B1 (en) | Driving control method and driving control apparatus | |
| US10787172B2 (en) | Driving assistance device and driving assistance method | |
| US20190031193A1 (en) | Travel control device for vehicle | |
| US11294376B2 (en) | Moving body control device | |
| CN113791621B (en) | Automatic steering tractor and airplane docking method and system | |
| JP2017165309A (en) | Travel control device for vehicle | |
| US11623494B1 (en) | Sensor calibration and verification using induced motion | |
| CN111273673A (en) | Automatic driving following method and system of unmanned vehicle and unmanned vehicle | |
| US11120277B2 (en) | Apparatus and method for recognizing road shapes | |
| CN110576849A (en) | Vehicle travel control device | |
| WO2021153176A1 (en) | Autonomous movement device, autonomous movement control method, and program | |
| US20190180117A1 (en) | Roadside object recognition apparatus | |
| CN116312014B (en) | Intelligent navigation method for parking lot | |
| CN115050205B (en) | Map generation device and position recognition device | |
| US11858534B2 (en) | Automated driving system | |
| CN115050203B (en) | Map generation device and vehicle position recognition device | |
| US20230406299A1 (en) | Moving body control system and moving body control method |
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 |