KR102614256B1 - 3D vector coordinate system for autonomous driving - Google Patents

Abstract

A three-dimensional vector coordinate system for autonomous driving according to an embodiment includes an autonomous driving unit mounted on a vehicle and connected to a central processing unit; A roadside installation unit installed at the roadside and in communication with the central processing unit; and a central processing unit in communication with the autonomous driving unit and the roadside installation unit, wherein the autonomous driving unit includes a vehicle sensing unit that senses the surroundings of the vehicle, a speed control unit that autonomously controls the speed of the vehicle, and It includes a steering control unit that autonomously controls steering.

Description

3D vector coordinate system for autonomous driving}

The present invention relates to an autonomous driving coordinate system, and more specifically, to a three-dimensional vector coordinate system for autonomous driving.

The content described in this section simply provides background information for this embodiment and does not constitute prior art.

A typical car is driven by the driver's steering, braking, etc. operations, while an autonomous vehicle's steering and braking operations are performed without the driver's intervention. With the development of sensors and information and communication technologies, self-driving cars, which have recently appeared, are expected to be commercialized in the near future.

Meanwhile, prior to the commercialization of self-driving cars, ordinary cars driven by drivers are being equipped with cutting-edge driving assistance devices that can replace the driver's eyes and ears for the driver's convenience. For example, vehicles are equipped with various sensors such as ultrasonic sensors, image sensors, radar sensors, LiDAR sensors, etc., so when an object comes close to the vehicle or the vehicle approaches an object while the vehicle is driving or parking. Configurations that warn drivers of this are becoming common.

In particular, recognition judgment technology is being applied, such as being able to recognize lanes marked on the road through the camera sensor mounted on the vehicle, and recognizing and judging moving objects through the fusion of the camera sensor and LiDAR sensor.

In this way, an autonomous driving control system that can control driving based on information recognized from various cognitive means mounted on the vehicle is being applied to self-driving cars.

0001) Republic of Korea Intellectual Property Office Public Patent Publication No. 10-2019-0000843 0002)Korea Intellectual Property Office Registered Patent Publication No. 10-0904767 0003) Republic of Korea Patent Office Publication No. 10-2017-0077332

Accordingly, the present invention was created to eliminate the above-mentioned problems, and includes an autonomous driving unit mounted on a vehicle and connected to a central processing unit; A roadside installation unit installed at the roadside and in communication with the central processing unit; and a central processing unit in communication with the autonomous driving unit and the roadside installation unit, wherein the autonomous driving unit includes a vehicle sensing unit that senses the surroundings of the vehicle, a speed control unit that autonomously controls the speed of the vehicle, and It was completed as a technical task with a focus on a three-dimensional vector coordinate system for autonomous driving, including a steering control unit that autonomously controls steering.

The problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

A three-dimensional vector coordinate system for autonomous driving according to an embodiment to solve the above problem includes an autonomous driving unit mounted on a vehicle and connected to a central processing unit; A roadside installation unit installed at the roadside and in communication with the central processing unit; and a central processing unit in communication with the autonomous driving unit and the roadside installation unit, wherein the autonomous driving unit includes a vehicle sensing unit that senses the surroundings of the vehicle, a speed control unit that autonomously controls the speed of the vehicle, and It includes a steering control unit that autonomously controls steering.

The autonomous driving unit may further include a vehicle communication unit that communicates wirelessly with the central processing unit.

The autonomous driving unit may further include a vehicle database unit that stores the vehicle registration number and vehicle model information, and provides the registration number and model information to the central processing unit through the vehicle communication unit.

Specific details of other embodiments are included in the detailed description and drawings.

According to the present invention described above, an autonomous driving unit mounted on a vehicle and connected to a central processing unit; A roadside installation unit installed at the roadside and in communication with the central processing unit; and a central processing unit in communication with the autonomous driving unit and the roadside installation unit, wherein the autonomous driving unit includes a vehicle sensing unit that senses the surroundings of the vehicle, a speed control unit that autonomously controls the speed of the vehicle, and By providing a three-dimensional vector coordinate system for autonomous driving that includes a steering control unit that autonomously controls steering, not only the movement of immediately adjacent vehicles, but also the movement of vehicles located front-to-front, back-to-back, side-to-side, etc. (speed value, position value, Steering values) can also be easily recognized through 3D mesh coordinates, which has the advantage of being excellent for accident prevention.

Effects according to the embodiments are not limited to the contents exemplified above, and further various effects are included in the present specification.

Figure 1 is a diagram showing the configuration of a 3D vector coordinate system for autonomous driving according to an embodiment.
Figure 2 is a diagram showing the configuration of the roadside installation part of Figure 1.
FIG. 3 is a diagram showing the configuration of the central processing unit of FIG. 1.
FIG. 4 is a diagram showing the configurations of the autonomous driving unit of FIG. 1.
Figure 5 is a diagram showing the relationship between the vehicle sensing unit, speed control unit, steering control unit, roadside sensing unit, and database unit.
Figure 6 is a diagram showing the relationship between the database unit and the vehicle model determination unit.
Figure 7 is a diagram showing the relationship between vehicle sensing data, cadastral map data, lidar data, and vision recognition data and the coordinate determination unit for each vehicle part.
Figure 8 is a diagram showing the relationship between the coordinate determination unit for each vehicle part, the derived data communication unit, and adjacent vehicles.
Figure 9 is a diagram showing the configuration of derived data.
Figure 10 is a diagram showing the relationship between the vector analysis unit, the coordinate determination unit for each vehicle part, and the database unit.
Figure 11 is a diagram showing the relationship between the vehicle database unit, vehicle communication unit, and other vehicles.
Figure 12 is a diagram showing the configuration of relative vehicle operation information.

Hereinafter, the configuration of the present invention and its operations and effects will be described collectively with reference to the accompanying drawings.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete, and are provided by those skilled in the art in the technical field to which the present invention pertains. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. In addition, the same reference numerals refer to the same elements throughout the specification.

Figure 1 is a diagram showing the configuration of a 3D vector coordinate system for autonomous driving according to an embodiment. Figure 2 is a diagram showing the configuration of the roadside installation part of Figure 1. FIG. 3 is a diagram showing the configuration of the central processing unit of FIG. 1. FIG. 4 is a diagram showing the configurations of the autonomous driving unit of FIG. 1.

Referring to FIGS. 1 to 4, a three-dimensional vector coordinate system 1 for autonomous driving according to an embodiment includes an autonomous driving unit 10 mounted on a vehicle and connected to the central processing unit 50, and installed on the roadside. It may include a roadside installation unit 30 in communication with the central processing unit 50, the autonomous driving unit 10, and a central processing unit 50 in communication with the roadside installation unit 30.

The autonomous driving unit 10 includes a vehicle sensing unit 11 that senses the surroundings of the vehicle, a speed control unit 12 that autonomously controls the speed of the vehicle, and a steering control unit 13 that autonomously controls the steering of the vehicle. may include.

The vehicle sensing unit 11 of the autonomous driving unit 10 may include sensors such as cameras, radars, and ultrasonic waves attached to the vehicle. The vehicle sensing unit 11 can determine the situation by sensing the surroundings of the vehicle through the sensor. The vehicle sensing unit 11 may further include a linear communication device such as a beacon that determines the location of the vehicle through sensors such as the radar and ultrasonic waves.

The autonomous driving unit 10 may further include an electronic control unit (ECU) that determines the sensed surrounding information of the vehicle. After determining the surrounding information of the vehicle sensed by the electronic control unit (ECU), the electronic control unit (ECU) transmits control signals to the speed control unit 12 and the steering control unit 13, respectively, to control the vehicle's speed and braking. , and steering can be controlled.

The autonomous driving unit 10 may further include a GPS coordinate recognition unit. The GPS coordinate recognition unit may be in communication with a communication unit that provides GPS signals.

The autonomous driving unit 10 may further include a vehicle communication unit 14 and a vehicle database unit 15.

Meanwhile, the autonomous driving unit of a conventional vehicle can only consider the surrounding environment sensed by the sensor. In other words, only the surrounding environment in front, rear, and sides of the vehicle could be considered, and the surrounding environment not sensed by the sensor could not be considered.

Furthermore, the sensor has a predetermined viewing angle, but since the autonomous driving system of existing vehicles relies on the sensor having a limited viewing angle, there is a problem in that it is difficult to accurately determine the surrounding environment.

Furthermore, because the sensor depends on the relative scalar amount (e.g., speed information) between the vehicle mounted on the vehicle and the adjacent surrounding environment, direction information and/or the sum of direction information and speed information (e.g., speed information) is recognized. There was a problem that it was difficult to do.

Furthermore, the GPS signal of the autonomous driving unit of an existing vehicle is a signal composed of two-dimensional coordinates, and the GPS coordinate recognition unit may not recognize the GPS signal in tunnels, building forests, indoors, etc. Because the GPS signal is a signal composed of two-dimensional coordinates, it may be difficult to recognize vertical coordinates.

In addition, the linear communication device of the autonomous driving unit of existing vehicles may have high inaccuracy in the real environment due to various communication obstacles.

In order to eliminate the above-mentioned problems, the specific configuration of the present invention will be described.

The roadside installation unit 30 may be installed on the roadside. For example, the roadside installation unit 30 may be installed on a streetlight located on the roadside. The roadside installation units 30 are provided in plural numbers, and the plurality of roadside installation units 30 may be installed for each streetlight located on the roadside at predetermined intervals.

The roadside installation unit 30 may include a roadside sensing unit 31 and a roadside communication unit 35.

The roadside sensing unit 31 may be a sensor such as a radar or camera installed on the roadside. LiDAR data of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle, is generated through the radar sensor of the roadside sensing unit 31, and the radar data of the vehicles passing the roadside at a predetermined time is generated through the camera sensor of the roadside sensing unit 31. Vision recognition data of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle, can be generated.

The LIDAR data may include three-dimensional coordinate information related to vehicles passing the roadside at a predetermined time. The vision recognition data may include three-dimensional coordinate information related to vehicles passing the roadside at a predetermined time.

The roadside sensing unit 31 may acquire three-dimensional coordinate information related to vehicles passing the roadside at the predetermined time based on the lidar data and the vision recognition data.

The LIDAR data may obtain partial information and/or all information about vehicles passing the roadside at the predetermined time.

The vision recognition data may obtain partial information and/or all information about vehicles passing the roadside at the predetermined time.

As described above, the roadside installation unit 30 is installed on the roadside and is installed at each streetlight spaced apart at a predetermined interval, so the roadside sensing unit 31 of the roadside installation unit 30 is also spaced apart at a predetermined interval. and can be installed.

The roadside sensing unit 31 can sense by covering a predetermined range. For example, the predetermined range may be set in consideration of the distance between the roadside sensing unit 31 and the adjacent roadside sensing unit 31. For example, if the distance between adjacent roadside sensing units 31 is 100m, the predetermined range covered by the roadside sensing unit 31 is set to the center of a circle with the roadside sensing unit 31 having a radius of 50m. It can cover up to However, the above example is not limited to this, and the coverage ranges of adjacent roadside sensing units 31 may overlap.

The roadside communication unit 35 may receive the LIDAR data and the vision recognition data generated by the roadside sensing unit 31 and provide them to the database unit 54.

That is, the roadside communication unit 35 may provide the database unit 54 with three-dimensional coordinate information related to vehicles passing along the roadside obtained by the roadside sensing unit 31.

The central processing unit 50 may include a vehicle model determination unit 51, a coordinate determination unit for each vehicle part 52, a derived data communication unit 53, a database unit 54, and a vector analysis unit 55.

The vehicle model determination unit 51 may determine models of vehicles passing along the roadside at a predetermined time, including the vehicle and vehicles adjacent to the vehicle.

The vehicle part coordinate determination unit 52 may determine the coordinates of each part of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle. The coordinate determination unit 52 for each vehicle part determines coordinate information (3D absolute coordinate information of individual vehicles) between the vehicle and adjacent vehicles (front, rear, side, front, rear, rear, side, etc.) of the vehicle. can be created. In addition, the coordinate determination unit 52 for each vehicle part can generate 3-dimensional coordinates of the road based on the cadastral map data (2-dimensional coordinates) and the roadside sensing unit 31, and based on the actual measurement reference point, which will be described later.

Figure 5 is a diagram showing the relationship between the vehicle sensing unit, speed control unit, steering control unit, roadside sensing unit, and database unit.

Referring to FIG. 5, vehicle sensing data from the vehicle sensing unit 11, speed data from the speed control unit 12, and steering data from the steering control unit 13 are each transmitted to the database unit 54 through the vehicle communication unit 14. can be provided. In addition to the vehicle sensing data, speed data, and steering data, predicted driving path data of the vehicle may also be provided to the database unit 54 through the vehicle communication unit 14.

Additionally, the lidar data and vision recognition data of the roadside sensing unit 31 may be provided to the database unit 54 through the roadside communication unit 35, respectively.

Additionally, vehicle registration data stored in the vehicle database unit 15 may also be provided to the database unit 54 through the vehicle communication unit 14.

Figure 6 is a diagram showing the relationship between the database unit and the vehicle model determination unit.

Referring to FIG. 6, the database unit 54 may store vehicle sensing data provided from the vehicle sensing unit 11, lidar data provided from the roadside sensing unit 31, vision recognition data, and standard shape data for each vehicle. there is. In addition, the database unit 54 contains cadastral map data, vehicle registration data provided from the vehicle database unit 15, speed data provided from the speed control unit 12, steering data provided from the steering control unit 13, and the predicted driving path data. can be saved.

The vehicle sensing data, lidar data, vision recognition data, standard shape data for each vehicle, and vehicle registration data stored in the vehicle database unit 15 are to be provided to the vehicle model determination unit 51, as shown in FIG. 6. You can. The vehicle model determination unit 51 determines the vehicle and adjacent vehicles (front, rear, side, front) through the provided vehicle sensing data, lidar data, vision recognition data, standard shape data for each vehicle, and vehicle registration data. You can determine the vehicle model (front, rear, side, etc.). The vehicle model determination unit 51 determines the vehicle and adjacent vehicles (front, rear, side, front, rear, side, etc.) of the vehicle through the vehicle sensing data, the lidar data, and the vision recognition data. etc.) can play a role in determining the vehicle model.

However, if the vehicle model determined by the vehicle model determination unit 51 is not accurate, the vehicle model determination unit 51 may apply the default model to the vehicle and adjacent vehicles of the vehicle. For example, when the discrimination degree (or accuracy) between the vehicle and adjacent vehicles of the vehicle by the vehicle model determination unit 51 is less than about 70%, the vehicle model determination unit 51 determines the vehicle registration data. As a basis, the default model extracted based only on the vehicle-specific standard shape data of the corresponding vehicle may be determined as the vehicle model of the corresponding vehicle. However, the above figure of about 70% is not limited, and of course can be designed to vary depending on the situation.

The vehicle model determined by the vehicle model determination unit 51 may be provided in the form of a 3D shape mesh.

Once the vehicle model of each vehicle is determined, an ID may be assigned to each vehicle.

Meanwhile, the vehicle model determination unit 51 additionally includes real-time vehicle types (or on-site vehicle types) such as each vehicle as well as cargo loaded on the vehicle (e.g., loaded containers, loaded goods, etc.). Thus, the vehicle model can be determined.

Figure 7 is a diagram showing the relationship between vehicle sensing data, cadastral map data, lidar data, and vision recognition data and the coordinate determination unit for each vehicle part.

Referring to FIG. 7, vehicle sensing data, cadastral map data, lidar data, and vision recognition data stored in the database unit 54 may be provided to the coordinate determination unit 52 for each vehicle part. The cadastral map data may be a constant value (or absolute value). That is, the cadastral map data may be data representing a road surface, a structure on the road surface, and/or an obstacle on the road surface, regardless of the relationship with adjacent vehicles. The cadastral map data may be two-dimensional coordinate data.

The coordinate determination unit 52 for each vehicle part determines the vehicle and adjacent vehicles (front, rear, side, front, rear, side) through the provided vehicle sensing data, cadastral map data, lidar data, and vision recognition data. Coordinates (hereinafter referred to as derived data) for each vehicle part (side, etc.) can be determined. The vehicle part-specific coordinate determination unit 52 may determine the vehicle part-specific coordinates by referring to the vehicle model determined by the vehicle model determination unit 51 described above in FIG. 6.

The coordinates for each vehicle part may be not only two-dimensional coordinates on a plane, but also three-dimensional coordinates including height. The actual measurement reference point of the coordinates for each vehicle part may be a paying sensor or facility installed in a lane or on the roadside. That is, based on the actual measurement reference point, the coordinates for each vehicle part can be determined by considering the angle and distance between the vehicle parts. After the coordinates for each vehicle part are determined, they can be corrected at any time.

To be more specific, the vehicle part coordinate determination unit 52 may determine the vehicle part coordinates of an individual vehicle based on the individual vehicle model determined by the vehicle model determination unit 51 and the cadastral map data.

As described above, the coordinate determination unit 52 for each vehicle part additionally generates 3-dimensional coordinates of the road based on the cadastral map data (2-dimensional coordinates) and the roadside sensing unit 31, and based on the actual measurement reference point to be described later. can do. The vehicle part coordinate determination unit 52 additionally generates three-dimensional coordinates (mesh form) of the road, so there is no need to newly calculate the coordinate values of the vehicle each time the vehicle passes. In other words, if the coordinate values of the vehicle are newly calculated each time the vehicle passes, the possibility of recognition errors may occur depending on various situations. As described above, the coordinate determination unit 52 for each vehicle part determines the coordinates of the road. By generating additional 3D coordinates, there is an advantage in that the possibility of recognition errors can be prevented in advance depending on various situations. Furthermore, the vehicle part coordinate determination unit 52 has the advantage of being able to express the vehicle, etc. more quickly and accurately by additionally generating three-dimensional coordinates (mesh form) of the road.

The three-dimensional mesh-shaped coordinates are a three-dimensional straight mesh of the space itself within the predetermined range covered by the roadside sensing unit 31 based on the cadastral map data (two-dimensional coordinates) and the roadside sensing unit 31. , the intersection of the three-dimensional straight meshes can be set to the default value.

In some embodiments, the curvature of the road (or road surface) or the presence of fallen objects on the road can be detected in real time by the roadside sensing unit 31 and updated in real time in the three-dimensional mesh coordinates.

As in the present embodiment, by generating the three-dimensional mesh coordinates, even if only the transmitted angle of the radar and the reflection distance from the roadside sensing unit 31 to the reflector (e.g., vehicle or load, etc.) are known, There is an advantage in that the position of the reflector can be accurately determined.

Data derived from the vehicle part coordinate determination unit 52 (vehicle part coordinates of individual vehicles and 3D coordinates of the road) are hereinafter referred to as derived data (ID).

The vector analysis unit 55 determines a vector value between the vehicle and adjacent vehicles of the vehicle based on the vehicle part coordinates of each vehicle, the speed data, and the steering data among the derived data (ID). You can. The vector value is an absolute coordinate value (3D) between the vehicle and adjacent vehicles of the vehicle, the relative speed value between the vehicle and adjacent vehicles of the vehicle, and the proximity of the vehicle to the vehicle. It may include relative direction values (3D) between vehicles.

The vector analysis unit 55 determines the vector value between the vehicle and adjacent vehicles of the vehicle, and can determine the vector value based on the coordinates of each representative part of the vehicle. The vector analysis unit 55 may substitute the determined vector value of the coordinates for each representative part of the vehicle into the three-dimensional coordinates of the road among the derived data (ID) derived from the vehicle part coordinate determination unit 52.

Figure 8 is a diagram showing the relationship between the coordinate determination unit for each vehicle part, the derived data communication unit, and adjacent vehicles.

Referring to FIG. 8, derived data (ID) derived from the vehicle part-specific coordinate determination unit 52 may be provided to the derived data communication unit 53. The derived data (ID) provided to the derived data communication unit 53 may be provided to adjacent vehicles (front, rear, side, front, rear, side, etc.) of the vehicle.

When providing derived data (ID) to adjacent vehicles (front, rear, side, front, rear, side, etc.) of the vehicle, it is stored in the vehicle database unit 15 of each vehicle and is stored in the vehicle communication unit 14. ), based on the vehicle registration data stored in the database unit 54, can be provided to adjacent vehicles (front, rear, side, front, rear, side, etc.) of the vehicle.

Meanwhile, derived data (ID) can be provided through the derived data communication unit 53 not only to adjacent vehicles of the vehicle, but also to the vehicle.

In FIG. 8, it is explained that the derived data (ID) itself can be provided to adjacent vehicles through the derived data communication unit 53, but it is not limited thereto, and the coordinates for each representative part of the vehicles determined by the vector analysis unit 55 The vector value of can be provided to adjacent vehicles through the derived data communication unit 53 by substituting the derived vector data for the three-dimensional coordinates of the road among the derived data (ID) derived from the vehicle part coordinate determination unit 52. The derived vector data may be provided in the form of coordinates in the three-dimensional mesh format described above. In some embodiments, the derived vector data may further include navigation destination information as well as steering data and speed data (including brake data, acceleration data, etc.) of the vehicle.

The derived vector data may be generated for each ID assigned to each vehicle by determining the vehicle model of each vehicle.

The derived vector data for each ID may be provided to adjacent vehicles through the derived data communication unit 53 along with road situation information (eg, abnormal flow information, traffic light information, etc.). In addition, the derived vector data can be communicated in two directions, so of course, the derived vector data generated from the adjacent vehicles can also be provided to the corresponding vehicle. As a result, not only real-time driving information, but also expected traffic flow and expected lane changes can be easily known, which has the advantage of distributing autonomous driving traffic and promoting defensive driving.

Figure 9 is a diagram showing the configuration of derived data.

Referring to FIG. 9, derived data (ID) may be provided in the form of numerical data and image data. The numerical data may be coordinate values for each vehicle part of the vehicle and vehicles adjacent to the vehicle, and 3D coordinate values of a road. Additionally, the numerical data may include accurate coordinates regarding obstacles (taking height, etc. into account) around the vehicle. The image data may be an accurate image of the vehicle, adjacent vehicles of the vehicle, and obstacles (considering height, etc.) around the vehicle.

Figure 10 is a diagram showing the relationship between the vector analysis unit, the coordinate determination unit for each vehicle part, and the database unit.

Referring to FIG. 10, the vector analysis unit 55 may receive derived data (ID) from the vehicle part coordinate determination unit 52, and may receive speed data and steering data from the database unit 54.

The vector analysis unit 55 determines a vector value between the vehicle and adjacent vehicles of the vehicle based on the vehicle part coordinates of each vehicle, the speed data, and the steering data among the derived data (ID). You can. The vector value is an absolute coordinate value (3D) between the vehicle and adjacent vehicles of the vehicle, the relative speed value between the vehicle and adjacent vehicles of the vehicle, and the proximity of the vehicle to the vehicle. It may include relative direction values (3D) between vehicles.

According to this embodiment, the advantage is that there is no need to calculate the distance between the sensor and the vehicle or the distance between vehicles each time by deriving absolute coordinate values (3D) between adjacent vehicles of the vehicle on the coordinates of a 3D mesh. There is.

Figure 11 is a diagram showing the relationship between the vehicle database unit, vehicle communication unit, and other vehicles. Figure 12 is a diagram showing the configuration of relative vehicle operation information.

Referring to Figures 11 and 12, the operation information of the other vehicle (or adjacent vehicle) may be provided to the vehicle database unit 15 of the corresponding vehicle through the vehicle communication unit 14. The other vehicle operation information may be a vector value between the other vehicle and adjacent vehicles of the other vehicle determined by the vector analysis unit 55 of the other vehicle.

The present invention described above has been described with reference to an embodiment shown in the drawings, but this is merely illustrative, and various modifications and other equivalent embodiments can be made by those skilled in the art. should be clarified. Therefore, the true technical protection scope of the present invention should be interpreted in accordance with the attached claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

1: 3D vector coordinate system for autonomous driving
10: Autonomous Driving Department
30: Roadside installation section
50: Central processing unit

Claims (3)

an autonomous driving unit (10) mounted on the vehicle and connected to the central processing unit (50);
A roadside installation unit (30) installed at the roadside and in communication with the central processing unit (50); and
It includes the central processing unit 50 in communication with the autonomous driving unit 10 and the roadside installation unit 30,
The autonomous driving unit 10 includes a vehicle sensing unit 11 that senses the surroundings of the vehicle,
A speed control unit 12 that autonomously controls the speed of the vehicle, and
Includes a steering control unit 13 that autonomously controls steering of the vehicle,
The autonomous driving unit 10 further includes a vehicle communication unit 14 that communicates wirelessly with the central processing unit 50,
The autonomous driving unit 10 stores the vehicle registration number and vehicle model information, and sends vehicle registration data including the registration number and model information to the central processing unit 50 through the vehicle communication unit 14. It further includes a vehicle database unit 15 that provides,
The autonomous driving unit 10 further includes an electronic control unit (ECU) that determines the sensed surrounding information of the vehicle,
After determining the surrounding information of the vehicle sensed by the electronic control unit (ECU), the electronic control unit (ECU) transmits control signals to the speed control unit 12 and the steering control unit 13, respectively, to determine the vehicle's speed, Controls braking and steering,
The autonomous driving unit 10 further includes a GPS coordinate recognition unit,
The GPS coordinate recognition unit is in communication with a communication unit that provides GPS signals,
The roadside installation units 30 are provided in plural numbers, and the plurality of roadside installation units 30 are installed at each streetlight located on the roadside at predetermined intervals,
The roadside installation unit 30 includes a roadside sensing unit 31 and a roadside communication unit 35,
The roadside sensing unit 31 includes a sensor, and the sensor includes a radar sensor and a camera sensor,
LiDAR data of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle, is generated through the radar sensor of the roadside sensing unit 31, and the camera of the roadside sensing unit 31 Generating vision recognition data of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle, through a sensor,
The LIDAR data includes three-dimensional coordinate information related to vehicles passing the roadside at a predetermined time, and the vision recognition data includes three-dimensional coordinate information related to vehicles passing the roadside at a predetermined time,
The roadside sensing unit 31 acquires three-dimensional coordinate information related to vehicles passing the roadside at the predetermined time based on the lidar data and the vision recognition data,
The LIDAR data acquires partial or total information about vehicles passing by the roadside at the predetermined time,
The vision recognition data acquires partial or total information about vehicles passing the roadside at the predetermined time,
Since the roadside installation unit 30 is installed on the roadside and installed at each streetlight spaced apart at a predetermined interval, the roadside sensing unit 31 of the roadside installation unit 30 is also installed at a predetermined interval,
The roadside sensing unit 31 covers a predetermined range and senses,
The predetermined range is set in consideration of the distance between the roadside sensing unit 31 and the adjacent roadside sensing unit 31,
The roadside communication unit 35 receives the LIDAR data and the vision recognition data generated by the roadside sensing unit 31 and provides them to the database unit 54,
Further comprising a central processing unit 50,
The central processing unit 50 includes a vehicle model determination unit 51, a coordinate determination unit for each vehicle part 52, a derived data communication unit 53, a database unit 54, and a vector analysis unit 55,
The vehicle model determination unit 51 determines models of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle,
The vehicle part coordinate determination unit 52 determines the coordinates of each part of vehicles passing the roadside at a predetermined time, including the vehicle and adjacent vehicles of the vehicle,
The coordinate determination unit 52 for each vehicle part determines coordinate information (3D absolute coordinate information of individual vehicles) between the vehicle and adjacent vehicles (front, rear, side, front, rear, rear, side) of the vehicle. create,
The coordinate determination unit 52 for each vehicle part generates three-dimensional coordinates of the road based on cadastral map data (two-dimensional coordinates) and the actual measurement reference point,
The vehicle registration data stored in the vehicle database unit 15 is provided to the database unit 54 through the vehicle communication unit 14,
The database unit 54 stores vehicle sensing data provided from the vehicle sensing unit 11, the lidar data provided from the roadside sensing unit 31, the vision recognition data, and standard shape data for each vehicle,
The database unit 54 includes the cadastral map data, the vehicle registration data provided from the vehicle database unit 15, speed data provided from the speed control unit 12, steering data provided from the steering control unit 13, and predicted driving. Route data is saved,
The vehicle sensing data, the lidar data, the vision recognition data, the standard shape data for each vehicle, and the vehicle registration data stored in the vehicle database unit 15 are provided to the vehicle model determination unit 51,
The vehicle model determination unit 51 determines the vehicle and adjacent vehicles (front) of the vehicle through the provided vehicle sensing data, the lidar data, the vision recognition data, the standard shape data for each vehicle, and the vehicle registration data. , rear, side, front-front, rear-rear, side-to-side) to determine the vehicle model,
The vehicle model determination unit 51 determines the vehicle and adjacent vehicles (front, rear, side, front-front, rear-rear, side-side) of the vehicle through the vehicle sensing data, the lidar data, and the vision recognition data. ) determine the vehicle model,
If the vehicle model determined by the vehicle model determination unit 51 is not accurate, the vehicle model determination unit 51 applies a default model to the vehicle and adjacent vehicles of the vehicle,
If the vehicle model determination unit 51 determines that the differentiation between the vehicle and adjacent vehicles of the vehicle is less than 70%, the vehicle model determination unit 51 determines each vehicle of the vehicle based on the vehicle registration data. Determine the default model extracted based only on standard shape data as the vehicle model of the vehicle,
The vehicle model determined by the vehicle model determination unit 51 is provided in the form of a 3D shape mesh,
When the vehicle model of each vehicle is determined, an ID is assigned to each vehicle,
The vehicle model determination unit 51 determines the vehicle model by additionally including the real-time vehicle type of each vehicle as well as the load loaded on the vehicle,
The vehicle sensing data, the cadastral map data, the lidar data, and the vision recognition data stored in the database unit 54 are provided to the vehicle part-specific coordinate determination unit 52,
The cadastral map data may be a constant value, the cadastral map data is data representing a road surface, a structure on the road surface, and an obstacle on the road regardless of the relationship with adjacent vehicles, and the cadastral map data is two-dimensional coordinate data,
The coordinate determination unit 52 for each vehicle part determines the vehicle and adjacent vehicles (front, rear, side, front, front) through the vehicle sensing data, cadastral map data, lidar data, and vision recognition data. , rear-rear, side-to-side) determine the coordinates (hereinafter derived data) for each vehicle part,
The vehicle part-specific coordinate determination unit 52 determines the vehicle part-specific coordinates with reference to the vehicle model determined by the vehicle model determination unit 51,
The coordinates for each vehicle part are not only two-dimensional coordinates on a plane, but also three-dimensional coordinates including height,
The actual measurement reference point of the coordinates for each vehicle part is a paying sensor or facility installed in the lane or roadside,
Based on the actual measurement reference point, determine the coordinates for each vehicle part by considering the angle and distance between the vehicle parts,
After the coordinates for each vehicle part are determined, they can be corrected at any time,
The vehicle part coordinate determination unit 52 determines the vehicle part coordinates of each individual vehicle based on the individual vehicle model determined by the vehicle model determination unit 51 and the cadastral map data,
The coordinate determination unit 52 for each vehicle part additionally generates three-dimensional coordinates of the road based on the actual measurement reference point, based on the cadastral map data and the roadside sensing unit 31, and the three-dimensional coordinates are in a mesh form. ,
The three-dimensional coordinates are based on the cadastral map data and the roadside sensing unit 31, and the space itself within the predetermined range covered by the roadside sensing unit 31 is composed of three-dimensional straight meshes, and the three-dimensional straight line The intersection of the meshes is set to default,
The vector analysis unit 55 determines a vector value between the vehicle and adjacent vehicles of the vehicle based on the vehicle part coordinates of each vehicle among the IDs, the speed data, and the steering data,
The vector value is an absolute coordinate value (3D) between the vehicle and adjacent vehicles of the vehicle, the relative speed value between the vehicle and adjacent vehicles of the vehicle, and the proximity of the vehicle to the vehicle. Contains relative direction values (3D) between vehicles,
The vector analysis unit 55 determines the vector value between the vehicle and adjacent vehicles of the vehicle, and determines the vector value based on the coordinates of each representative part of the vehicle, and the vector analysis unit 55 determines the vector value between the vehicle and adjacent vehicles. Substituting the vector value of the coordinates for each representative part of the vehicle into the three-dimensional coordinates of the road among the IDs derived from the vehicle part coordinate determination unit 52,
The ID derived from the vehicle part coordinate determination unit 52 may be provided to the derived data communication unit 53. The ID provided to the derived data communication unit 53 is provided to adjacent vehicles (front, rear, side, front, rear, side) of the vehicle,
When providing the ID to adjacent vehicles (front, rear, side, front, rear, rear, and side) of the vehicle, it is stored in the vehicle database unit 15 of each vehicle and is sent to the vehicle communication unit 14. Based on the vehicle registration data stored in the database unit 54, it can be provided to adjacent vehicles (front, rear, side, front, rear, and side) of the vehicle.
The ID is provided through the derived data communication unit 53 not only to adjacent vehicles of the vehicle, but also to the vehicle,
The vector value of the coordinates for each representative part of the vehicle determined by the vector analysis unit 55 is the derived vector data substituted for the three-dimensional coordinates of the road among the IDs derived from the vehicle part coordinate determination unit 52. Provided to adjacent vehicles through the communication unit 53,
The derived vector data is provided in the form of a mesh of the three-dimensional coordinates,
The derived vector data is generated for each ID assigned to each vehicle by determining the vehicle model of each vehicle,
The derived vector data for each ID is provided to adjacent vehicles through the derived data communication unit 53 together with road situation information, and the derived vector data can be communicated in two directions, so that the derived vector data generated from the adjacent vehicles Data is also provided to the vehicle,
The ID is provided in the form of numerical data and image data, the numerical data are the coordinates of each vehicle part of the vehicle and adjacent vehicles of the vehicle, and the three-dimensional coordinates of the road, and the numerical data are the three-dimensional coordinates of the road around the vehicle. Contains accurate coordinates of obstacles, and the image data is an accurate image of the vehicle, adjacent vehicles of the vehicle, and obstacles around the vehicle,
The vector analysis unit 55 receives the ID from the vehicle part coordinate determination unit 52, the speed data, and the steering data from the database unit 54,
The vector analysis unit 55 determines a vector value between the vehicle and adjacent vehicles of the vehicle based on the vehicle part coordinates, the speed data, and the steering data of each vehicle among the IDs,
The vector value is an absolute coordinate value (3D) between the vehicle and adjacent vehicles of the vehicle, a relative speed value between the vehicle and adjacent vehicles of the vehicle, and the proximity of the vehicle to the vehicle. A 3D vector coordinate system for autonomous driving that includes relative direction values (3D) between vehicles.
delete delete