KR102234738B1 - Precise road map system for constructing digital map and road map through automation - Google Patents

Abstract

The present invention relates to a precise road map system, and more specifically, to an precise road map system that can build a figure map and a road map through automation that can detect and recognize objects by using a machine learning, comprising: an image sensor mounted on an upper part of the vehicle to capture image information; a laser sensor mounted on a top of a vehicle to detect distance information; a navigation sensor mounted on the top of the vehicle to provide navigation information; a data fusion unit that performs data fusion based on internal and external geometric models of each sensor; and a recognition unit that detects a candidate group in the data through machine learning and sequentially applies an additional machine learning model to the detected candidate group to recognize information indicated by a target object. The present invention relates to a map system with precision that can build numerical maps and road maps through an automated method that enables automatic detection and recognition of objects using machine learning technology.

Description

Map system with precision to build digital maps and road maps through an automated method {PRECISE ROAD MAP SYSTEM FOR CONSTRUCTING DIGITAL MAP AND ROAD MAP THROUGH AUTOMATION}

The present invention relates to a map system with precision, and more particularly, to a map system with precision capable of constructing a digital map and a road map through an automated method.

Conventional spatial modeling methods are generally performed by using data constructed manually through field surveys. This method has the disadvantage of not only incurring excessive labor costs, but also causing frequent errors in geographic information due to manual labor, and difficult to correct or update.

Recently, in order to solve this problem, geographic information data on buildings and road facilities is being constructed using surveying equipment such as a Mobile Mapping System (MMS).

However, the method using surveying equipment such as a mobile mapping system is also expensive due to the correction point measurement, and it is difficult to obtain accurate coordinate information due to distortion of the measured image, and furthermore, the accuracy of the information is lacking in urban areas. It causes many problems in the construction of the information database.

As described above, in the prior art, when creating a digital map or a precision road map, the 3D position coordinates are inversely calculated from the image sensor, or the 3D point cloud data is obtained directly from the laser sensor and the operator directly draws it on a computer or paper drawing. Because of the loss, it takes a lot of work time, and errors of workers and data errors act as causes of map errors.

Therefore, there is a situation in which the development of technologies for constructing new digital maps and road precision maps that can improve the problems of the prior art and map automatically and increase reliability is constantly required.

The matters described as background art above are only for improving understanding of the background of the present invention, and should not be taken as acknowledging that they correspond to the prior art already known to those of ordinary skill in the art.

The present invention is to solve the problems of the prior art described above, and a map system with precision capable of constructing a digital map and a road map through an automated method that enables automatic detection and recognition of objects using machine learning technology. Its purpose is to provide.

In addition, the present invention is an automated method that improves accuracy by fusion of data based on the internal and external geometric models of each sensor, and designing by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data are overlapped when designing the device. Another purpose is to provide a map system with precision that can build digital maps and road maps through this method.

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

The configuration of the present invention for achieving the above object includes: an image sensor mounted on an upper portion of a vehicle to capture image information; A laser sensor mounted on the upper part of the vehicle to detect distance information; A navigation sensor mounted on the upper part of the vehicle to provide navigation information; A data fusion unit that performs data fusion based on the internal and external geometric models of each sensor; And a recognition unit that detects a candidate group in the data through machine learning and sequentially applies an additional machine learning model to the detected candidate group to recognize information represented by a target object. It characterized in that it comprises a.

The geometric parameters of the image sensor in the map system with precision to build a digital map and a road map through the automated method according to the embodiment of the present invention are focal length, main point position, lens distortion parameter, sensor format size, sensor position And any one of an attitude value or a combination thereof, and the geometric parameter of the laser sensor is any one of an incidence angle of each laser, a distance scale, a distance direction offset, an axial offset, a position and an attitude value of the sensor, or a combination thereof, The geometric parameter of the navigation sensor is preferably any one of a scale in the axial direction, an offset in the axial direction, and a position and posture value of the sensor, or a combination thereof.

In the map system with precision to build a digital map and a road map through an automated method according to an embodiment of the present invention, a detachable part is coupled to the upper surface of the vehicle, a damping part is coupled to the upper part of the detachable part, and the insertion space of the attenuating part is It is preferable that a fluid buffer is inserted, an elevating unit is coupled to an upper portion of the fluid buffer, and an image sensor is coupled to an upper portion of the elevating portion.

In the map system with precision capable of constructing a digital map and a road map through an automated method according to an embodiment of the present invention, the detachable unit may include a detachable base that is coupled to an upper surface of a vehicle and has a detachable space formed therein; A detachable cover rotatably coupled to one side of the detachable space of the detachable base; And a detachable body that can be seated in the detachable space and detachable from the detachable base. Including, the detachable base, a pair of guide grooves formed in depressions on both sides along the longitudinal direction of the detachable space; An extension groove recessed in the rear end of the detachable space; A pair of detachment preventing portions formed at the rear of the upper end of the detachable space and protruding inward to prevent detachment of the detachable body; And a pair of position control grooves each recessed at the rear end of the pair of guide grooves. Including, the detachable body is formed protruding on both sides along the longitudinal direction of the detachable body, a pair of guide portions that are slidable in a pair of guide grooves; A detachable extension protruding from the rear end of the detachable body and capable of being accommodated in the extension groove; A protruding stopper protruding from the upper surface of the detachable body and having the same width as the width between the pair of detachment preventing portions; And a pair of position control units which are respectively recessed at the rear ends of the pair of guide units to insert a pair of position control units. It is preferable to include.

In the map system with precision to build a digital map and a road map through an automated method according to an embodiment of the present invention, the fluid buffer unit is coupled to the upper part of the damping unit and forms a space so that the fluid can be accommodated therein. Receiving part; A fluid lower plate coupled to an upper portion of the fluid receiving portion and having a lower plate through hole formed at one side thereof; A ring-shaped fluid rotating plate rotatably coupled to an upper portion of the lower fluid plate and having a rotation through hole formed on one side thereof and a hollow formed in a central portion thereof; A fluid upper plate coupled to an upper portion of the fluid rotating plate and having an upper plate through hole formed at one side thereof; A fluid supply unit coupled to an upper portion of the fluid upper plate and having a space formed therein to accommodate fluid; A fluid rotating unit coupled to an upper portion of the fluid lower plate and disposed in the hollow of the fluid rotating plate to rotate the fluid rotating plate; And a fluid connection unit connected between the fluid receiving unit and the fluid supply unit and having a one-way valve. Including, a ring-shaped lower plate communication groove in communication with the lower plate through hole is recessed on the upper surface of the fluid lower plate, and a ring-shaped rotary communication groove in communication with the rotary through hole is recessed on the upper surface of the fluid rotating plate, and the fluid As the rotating plate rotates, it is preferable that the fluid flowing from the fluid supply unit through the upper plate through hole sequentially passes through the rotary communication groove, the rotary through hole, the lower plate communication groove, and the lower plate through hole to be accommodated in the fluid receiving unit.

In a map system with precision to build a digital map and a road map through an automated method according to an embodiment of the present invention, the lifting part is coupled to the upper part of the fluid buffer part, and a space formed so as to accommodate the fluid therein. case; An elevating rod having a lower end disposed inside the elevating case and capable of elevating up and down; A ring-shaped vertical copper plate inserted into the lifting rod and disposed between the outer surface of the lifting rod and the inner surface of the lifting case and movable up and down; An upper stopper protruding from the side of the elevating rod and disposed at the top with respect to the vertical copper plate; A lower stopper protruding from the side of the elevating rod and disposed at a lower portion with respect to the vertical copper plate; An upper elastic portion disposed between the lower surface of the upper stopper and the upper surface of the vertical copper plate; A lower elastic portion disposed between the upper surface of the lower stopper and the lower surface of the vertical copper plate; A rod through hole through which a fluid can pass through the inside of the lifting rod; And a hydraulic pump connected to the lifting case to apply pressure to the fluid inside the lifting case. Including, wherein the rod through hole, a first transverse through hole that passes through the lifting rod in a transverse direction and is disposed under the upper stopper; A second transverse through hole penetrating the elevating rod in a transverse direction, disposed above the lower stopper, and disposed relatively lower than the first transverse through hole; And a vertical through hole penetrating the lifting rod in a longitudinal direction and connecting between the first transverse through hole and the second transverse through hole. It is preferable to include.

The present invention having the above configuration automatically detects and recognizes objects through machine learning, and automatically maps social infrastructure facilities and target structures through convergence technology between image sensors, laser sensors, and navigation sensors to improve work efficiency. And there is an effect of increasing the reliability.

In addition, the present invention performs data fusion based on the internal and external geometric models of each sensor, and when designing the device, it is possible to increase accuracy by designing by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data are overlapped. There is an effect.

In addition, the present invention detects a candidate group in data through machine learning based on the shape, color, and texture of the target object, and recognizes information represented by the target object by sequentially applying an additional machine learning model to the detected candidate group. Classification has the effect of improving accuracy.

Further, the present invention has the effect of remarkably increasing the stability when acquiring information of each sensor using the damping unit and the fluid buffer unit, adjusting the height of each sensor by using the elevating unit, and being able to freely detach and detach according to the situation.

1 is a block diagram showing each configuration of a map system with precision capable of constructing a digital map and a road map through an automated method according to an embodiment of the present invention.
2 is a block diagram showing each configuration of a map system with precision capable of constructing a digital map and a road map through an automated method according to a second embodiment of the present invention.
3 is a block diagram showing each configuration of a map system with precision capable of constructing a digital map and a road map through an automated method according to a third embodiment of the present invention.
4 is a view for explaining a process of data fusion based on an internal geometric model and an external geometric model according to an embodiment of the present invention.
5 is a view for explaining the concept of an MMS geometric model according to an embodiment of the present invention.
6 is a block diagram showing a process of applying a machine learning model according to an embodiment of the present invention.
7A and 7B are diagrams showing an example of training data and fusion data according to an embodiment of the present invention.
8A and 8B are diagrams showing an example of a data preprocessing process according to an embodiment of the present invention.
9 is a diagram showing an example of a process for detecting a candidate group according to an embodiment of the present invention.
10 is a view showing an example of an object recognition process according to an embodiment of the present invention.
11 is a diagram showing an example of a mapping process through a fusion technology according to an embodiment of the present invention.
12 to 14 are flow charts showing a method of operating a map system with precision according to the first, second, and third embodiments of the present invention.
15 is an exemplary view showing an image sensor, a laser sensor, and a navigation sensor mounted on an upper portion of a vehicle according to an embodiment of the present invention.
Figure 16 is an exploded perspective view showing the appearance of a detachable portion according to an embodiment of the present invention.
17 is a view showing a cross-sectional view of a detachable base according to an embodiment of the present invention.
18 is a view showing the overall appearance of a damping unit according to an embodiment of the present invention.
19 is a view showing a longitudinal section of the attenuation unit according to an embodiment of the present invention.
20 is a view showing an exploded view of each configuration of the fluid buffer unit according to an embodiment of the present invention.
21 is a view showing a cross-sectional view of a lifting unit according to an embodiment of the present invention.

Hereinafter, based on the accompanying drawings, the present invention will be described in detail so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein.

In order to clearly describe the present invention, parts irrelevant to the description have been omitted, and the same reference numerals are attached to the same or similar components throughout the specification.

In addition, terms or words used in this specification and claims should not be construed as being limited to their usual or dictionary meanings, and the inventors appropriately explain the concept of terms in order to explain their own invention in the best way. Based on the principle that it can be defined, it should be interpreted as a meaning and concept consistent with the technical idea of the present invention.

1 is a configuration diagram showing each configuration of a map system with precision capable of constructing a digital map and a road map through an automated method according to an embodiment of the present invention, and FIG. 2 is a second embodiment of the present invention. It is a configuration diagram showing each configuration of a map system with precision capable of constructing a digital map and a road map through the automation method according to the present invention, and FIG. 3 is a digital map and a road map through the automation method according to the third embodiment of the present invention. It is a configuration diagram showing each configuration of the map system with the precision that can be constructed.

As shown, the map system with precision according to the present invention automatically detects and recognizes objects and maps automatically through a fusion technology between the image sensor 10, the laser sensor 20, and the navigation sensor 30. For.

To this end, data fusion is performed based on the internal and external geometric models of each sensor, and when the device is designed, the image sensor data and the laser sensor data are designed to overlap each other by reflecting the viewing angle of each sensor to increase accuracy. .

The present invention detects a candidate group in data through machine learning based on the shape, color, and texture of the target object, and sequentially applies an additional machine learning model to the detected candidate group to recognize and classify the information represented by the target object. Includes composition.

An embodiment of the present invention includes a configuration in which machine learning is sequentially performed after data fusion is performed before machine learning.

Another embodiment of the present invention includes a configuration in which a target object is detected/recognized by applying machine learning to each data, and then maps by inputting information from the navigation sensor 30.

In another embodiment of the present invention, the laser sensor data is converted to absolute coordinates by inputting information from the navigation sensor 30 to the laser sensor data, and machine learning is applied to each of the absolute coordinated laser sensor data and image sensor data to detect objects and It includes a configuration for mapping by additionally applying laser sensor data converted to absolute coordinates to the object recognized by the recognition and image sensor.

First, a detailed description of the configuration of the map system with precision according to the first embodiment of the present invention is as follows.

As shown in FIG. 1, data fusion is performed before machine learning, and then machine learning is sequentially performed. An image sensor 10 for photographing image information, a laser sensor 20 for sensing distance information, and positioning information A navigation sensor (30) that provides navigation information such as the speed, position, and posture of a moving object, a data fusion unit (40) that performs data fusion based on the internal and external geometric models of each sensor, and the shape of the object object , Color, texture, etc., detects candidate groups in the data through machine learning, and sequentially applies additional machine learning models to the detected candidate groups to recognize and classify the information represented by target objects, and digital map or road precision map data It includes a recognition unit 50 to output.

Here, the image sensor data and the laser sensor data are designed to overlap each other by reflecting the viewing angle of each sensor.

In addition, the geometric parameter of the image sensor 10 may be any one of a focal length, a main point position, a lens distortion parameter, a sensor format size, a sensor position, and a posture value, or a combination thereof.

In addition, the geometric parameter of the laser sensor 20 may be any one of an incidence angle of each laser, a distance scale, a distance direction offset, an axial offset, a position and an attitude value of the sensor, or a combination thereof.

In addition, the geometric parameter of the navigation sensor 30 may be any one of a scale in an axial direction, an offset in an axial direction, a position and posture value of the sensor, or a combination thereof.

The map system with precision according to the second embodiment of the present invention detects/recognizes an object object by applying machine learning to each data, as shown in FIG. 2, and then inputs and maps information from the navigation sensor 30, An image sensor 10 for photographing image information, a laser sensor 20 for sensing distance information, and a navigation sensor 30 for providing navigation information such as positioning information and speed, position and posture of a moving object, and an image sensor 10 ), based on the shape, color, and texture of the target object, detects the candidate group in the data through machine learning, and sequentially applies additional machine learning models to the detected candidate group to recognize the information represented by the target object. Receives data from the first recognition unit 50a to classify and the laser sensor 20 to detect candidate groups in the data through machine learning based on the shape, color, and texture of the target object, and sequentially When object data of the second recognition unit 50b, which recognizes and classifies the information represented by the target object by applying an additional machine learning model, and the object data of the first recognition unit 50a and the second recognition unit 50b are input, the navigation sensor information It includes an absolute coordinate conversion unit 60 for outputting digital map or road precision map data by inputting and performing absolute coordinates.

Here, the image sensor data and the laser sensor data are designed to overlap each other by reflecting the viewing angle of each sensor.

In addition, the geometric parameter of the image sensor 10 may be any one of a focal length, a main point position, a lens distortion parameter, a sensor format size, a sensor position, and a posture value, or a combination thereof.

In addition, the geometric parameter of the laser sensor 20 may be any one of an incidence angle of each laser, a distance scale, a distance direction offset, an axial offset, a position and an attitude value of the sensor, or a combination thereof.

In addition, the geometric parameter of the navigation sensor 30 may be any one of a scale in an axial direction, an offset in an axial direction, a position and posture value of the sensor, or a combination thereof.

The map system with precision according to the third embodiment of the present invention includes an image sensor 10 for photographing image information, a laser sensor 20 for sensing distance information, and positioning information and speed and position of a moving object, as shown in FIG. 3. And a navigation sensor 30 that provides navigation information such as a posture, and a first absolute coordinate unit that converts the data of the laser sensor 20 into absolute coordinates by inputting the information of the navigation sensor 30 to the data of the laser sensor 20 (80a), a first recognition unit 70a for detecting and recognizing an object by applying machine learning to each of the laser sensor 20 data and image sensor 10 data in absolute coordinates, and the laser sensor 20 in absolute coordinates. ) A second recognition unit 70b that detects a candidate group in the data through machine learning, and sequentially applies an additional machine learning model to the detected candidate group to recognize and classify the information represented by the target object, and the first recognition unit ( Includes a second absolute coordinates unit 80b that performs absolute coordinates by inputting navigation sensor information to the object data of 70a), and laser sensor 20 data that is absolute coordinates on the object recognized from the image sensor 10 By applying additionally, the data for mapping is output.

Here, the image sensor data and the laser sensor data are designed to overlap each other by reflecting the viewing angle of each sensor.

In addition, the geometric parameter of the image sensor 10 may be any one of a focal length, a main point position, a lens distortion parameter, a sensor format size, a sensor position, and a posture value, or a combination thereof.

In addition, the geometric parameter of the laser sensor 20 may be any one of an incidence angle of each laser, a distance scale, a distance direction offset, an axial offset, a position and an attitude value of the sensor, or a combination thereof.

In addition, the geometric parameter of the navigation sensor 30 may be any one of a scale in an axial direction, an offset in an axial direction, a position and posture value of the sensor, or a combination thereof.

As described above, the system for constructing a map with precision according to the present invention includes an image using image-laser sensor data fusion and machine learning, an object object detection and recognition automation configuration from a laser sensor.

The data fusion is based on the internal and external geometric models of each sensor, and is designed by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data are overlapped when designing the device.

4 is a view for explaining a process of data fusion based on an internal geometric model and an external geometric model according to an embodiment of the present invention, and FIG. 5 is a diagram illustrating the concept of an MMS geometric model according to an embodiment of the present invention. It is a drawing for explanation.

Machine learning is performed in two stages: object object detection and object object recognition, and in the object object detection step, candidate groups in the data are detected through machine learning based on the shape, color, and texture of the object object, and the detected candidate groups By sequentially applying additional machine learning models to the object, the information represented by the target object is recognized and classified.

As shown in Figure 4, by modeling the internal geometry of the sensor of the image sensor 10, modeling the internal geometry of the sensor of the laser sensor 20, and applying a defined image sensor geometry model and a defined laser sensor geometry model. Data fusion.

The internal geometric model of the laser sensor 20 may be defined as follows.

In addition, the internal geometric model of the image sensor 10 may be defined as follows.

And for data fusion by applying the defined image sensor 10 geometric model and the defined laser sensor 20 geometric model, the external geometric model of the laser sensor 20 is defined as in Equation 3, and the image sensor ( 10) The external geometric model may be defined as in Equation 4.

As shown in FIG. 5, the present invention includes a configuration in which a target point observed from a sensor is converted into a standardized absolute coordinate system with a geometric model of a mathematically defined MMS and mapped.

Here, the absolute coordinate system may be defined as follows.

In order to define the position of an arbitrary point in space, there must be a coordinate system as a reference. And the coordinate system for defining the position in the three-dimensional space is composed of the origin of the coordinate axis and three orthogonal coordinate axes.

The coordinate system can be freely set according to the purpose. If it is set large, it can be divided into an absolute coordinate system, which is a fixed coordinate system in which the position of the origin or the direction of the coordinate axis, never changes, and a variable coordinate system, which is a coordinate system that can move and rotate.

The present invention uses an absolute coordinate system that is a standard for shape modeling or numerical analysis, and converts an observed target point into a standardized absolute coordinate system and maps it. A detailed description of the process of applying the machine learning model in the precision guidance system according to the present invention is as follows.

6 is a block diagram showing a process of applying a machine learning model according to an embodiment of the present invention, and FIGS. 7A and 7B are views showing an example of training data and fusion data according to an embodiment of the present invention. 8A and 8B are diagrams showing an example of a data preprocessing process according to an embodiment of the present invention, FIG. 9 is a diagram showing an example of a candidate group detection process according to an embodiment of the present invention, and FIG. A diagram showing an example of a process of recognizing an object according to an embodiment of the present invention, and FIG. 11 is a diagram showing an example of a mapping process through a fusion technology according to an embodiment of the present invention.

Pre-processing the training data as shown in FIG. 6 in order to detect the candidate group in the data through machine learning and sequentially apply additional machine learning models to the detected candidate group to recognize and classify the information represented by the target object; and A candidate group detection step and an object recognition step are performed.

The pre-processing of the training data may include a process of adjusting an image brightness value as shown in FIG. 8A and a process of removing unnecessary data from the laser sensor data as shown in FIG. 8B.

In addition, the step of detecting the candidate group may include a process of vectorizing brightness, color information, and shape information to detect the candidate group.

As shown in FIG. 9, when a candidate group is detected, an object recognition process is performed by applying a machine learning algorithm.

As shown in FIG. 10, when the object is recognized and classified by performing pre-processing of training data, detecting a candidate group, and recognizing an object, mapping is performed through a data fusion technology.

In addition, a detailed description of the method of operating the map system according to the present invention is as follows.

12 to 14 are flow charts showing a method of operating a map system with precision according to the first, second, and third embodiments of the present invention.

As shown in FIG. 12, the method of operating the map system with precision according to the first embodiment of the present invention is designed by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data are overlapped (S601), and Data fusion is performed based on the sensor's internal and external geometric models (S602).

Subsequently, a candidate group is detected in the data through machine learning based on the shape, color, and texture of the object object (S603).

Then, additional machine learning models are sequentially applied to the detected candidate group to recognize and classify information represented by the target object (S604).

Then, the digital map or the road precision map data is output and mapped (S605).

As shown in Fig. 13, the method of operating the map system with precision according to the second embodiment of the present invention is designed by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data are overlapped (S701), and the image A candidate group in the data is detected through machine learning based on the shape, color, and texture of the target object of the sensor and the laser sensor (S702).

Subsequently, information represented by a target object is recognized and classified by sequentially applying an additional machine learning model to the detected candidate group (S703).

Then, by inputting the navigation sensor information, absolute coordinates are performed (S704).

Subsequently, the digital map or the detailed road map data is output and mapped (S705).

As shown in FIG. 14, the method of operating the map system with precision according to the third embodiment of the present invention is designed by reflecting the viewing angle of each sensor so that the image sensor data and the laser sensor data overlap (S801), and the laser By inputting navigation sensor information to the sensor data, absolute coordinates of the laser sensor data are performed (S802).

Subsequently, machine learning is applied to each of the absolute coordinated laser sensor data and image sensor data to detect a candidate group in the data based on the shape, color, and texture of the object object (S803).

Then, additional machine learning models are sequentially applied to the detected candidate group to recognize and classify information represented by the target object (S804).

Subsequently, an absolute coordinated laser sensor data is additionally applied to the object recognized by the image sensor to map (S805).

FIG. 15 is a diagram illustrating an exemplary image sensor, a laser sensor, and a navigation sensor mounted on an upper portion of a vehicle according to an embodiment of the present invention, and FIG. 16 is a view of a detachable part according to an embodiment of the present invention. It is an exploded perspective view showing, and Figure 17 is a view showing a cross-sectional view of the detachable base according to an embodiment of the present invention.

As shown, the image sensor 10 and the laser sensor 20 are mounted on the upper part of the vehicle 100, respectively, the image sensor 10 is used to photograph image information, and the laser sensor 20 is a distance It is used to detect information.

At this time, the image sensor 10 is mounted on the upper portion of the vehicle 100 via the detachable unit 200, the damping unit 300, the fluid buffer unit 400, and the elevating unit 500 to prevent external shock or vibration. At the same time as this attenuation, the height can be adjusted.

The detachable part 200 is coupled to the upper part of the vehicle 100, the attenuating part 300 is coupled to the upper part of the detachable part 200, and the fluid buffer part 400 is inserted in the insertion space 311 of the attenuating part 300. ) Is mounted, the elevating unit 500 is coupled to the upper portion of the fluid buffer unit 400, and the image sensor 10 is coupled to the upper portion of the elevating unit 500.

The detachable part 200 is coupled to the upper surface of the vehicle 100 and has a detachable base 210 having a detachable space 211 formed therein, a detachable cover 220 that is rotatably coupled to one side of the detachable space of the detachable base, and It can be seated in the detachable space and includes a detachable body 230 that is detachable from the detachable base.

The detachable base 210 is generally formed in a rectangular parallelepiped shape, and a detachable space 211 is formed so that the detachable body 230 can be accommodated therein. The detachable space 211 has a cross section of a long rectangular shape in the front-rear direction.

A pair of guide grooves 212 are recessed in the left and right sides of the detachable space 211 along the longitudinal direction (front and rear direction). Along the guide groove 212, the detachable body 230 is slidable back and forth.

At the rear end of the detachable space 211, an extension groove 213 is recessed toward the rear. At the rear of the upper end of the detachable space 211, a pair of detachment preventing portions 214 are formed to protrude toward the inner direction of the detachable space 211. That is, when the detachable space 211 is viewed from the top, the detachable space 211 has a relatively wide width and has a relatively narrow width with the front portion to which the detachable cover 220 is rotated, and the detachment preventing part 214 is It can be divided into a protruding rear part.

A pair of position regulation grooves 215 are recessed at the rear end of the pair of guide grooves 212. In addition, a pair of cover fixing grooves 216 are recessed in the upper central portion of the detachable space 211. In the cover fixing groove 216, a cover fixing portion 221 protruding from the side of the detachable cover 220 is inserted, and accordingly, the detachable cover 220 may be fixed by closing the upper end of the detachable space 211. .

The detachable body 230 has a size that can be inserted into the detachable space 211 of the detachable base 210, and can slide back and forth within the detachable space 211. A pair of guide portions 231 are formed to protrude along the longitudinal direction (front and rear direction) on both left and right sides of the detachable body 230. The guide part 231 is inserted into the guide groove 212 and can slide back and forth.

At the rear end of the detachable body 230, a detachable extension 232 is formed to protrude toward the rear. The detachable extension 232 may be inserted into the extension groove 213. As the detachable extension part 232 is inserted into the extension groove 213, the detachable body 230 may be fixed without moving in the vertical direction.

On the upper surface of the detachable body 230, a protruding stopper 233 protrudes in the form of a rectangular parallelepiped. The protruding stopper 233 has the same width as the width between the pair of separation prevention parts 214 and the same length as the length of the separation prevention part 214. That is, the protruding stopper 233 is formed in a shape that fits perfectly into the space between the pair of detachment prevention parts 214, and accordingly, when the detachable cover 220 closes the upper end of the detachable space 211, the detachable body The 230 may be fixed without being moved back and forth, left and right.

A pair of position regulation portions 234 are protruding from the rear end of the pair of guide portions 231. A pair of position control units 234 may be inserted into the position control groove 215, thereby limiting the movement of the detachable body 230 in the front and rear directions.

When the user slides the detachable body 230 to the rear end of the guide groove 212 while inserting the detachable body 230 into the detachable space 211, the position control unit 234 is inserted into the position control groove 215 (click It is possible to clearly recognize the completion of the movement of the detachable body 230 through the feeling), and at the same time, it is possible to limit the movement in the front and rear directions to some extent.

Looking at the assembly process of the detachable body 230, first, the user inserts the detachable body 230 into the front portion of the detachable space 211 while the detachable cover 220 is open. Then, while the pair of guide portions 231 are inserted into the guide groove 212, the detachable body 230 is slid to the rear portion of the detachable space 211.

The detachable body 230 is completely moved to the rear part of the detachable space 211 so that a pair of position control parts 234 are inserted into the position control groove 215, and the detachable extension part 232 is inserted into the extension groove 213 Then, the detachable cover 220 is closed to close the front portion of the detachable space 211. Accordingly, the detachable body 230 is seated inside the detachable base 210 so that movement in the front and rear, left and right and up and down directions is limited and can be firmly fixed.

On the contrary, when the detachable body 230 is to be separated from the detachable base 210, the detachable cover 220 is opened so that the front part of the detachable space 211 is opened, and the detachable body 230 is placed in front of the detachable space 211. After sliding to the part, it is enough to separate it through the front part of the detachable space 211.

In this way, the present invention can be fixed to the detachable base 210 or separated from the detachable base 210 with only a simple configuration consisting of the detachable base 210, the detachable body 230 and the detachable cover 220. Therefore, there is an advantage in that the image sensor can be freely separated and stored from the vehicle 100 when the image sensor 10 is not used in consideration of surrounding conditions.

18 is a view showing the overall appearance of the attenuation unit according to an embodiment of the present invention, Figure 19 is a view showing a longitudinal section of the attenuation unit according to an embodiment of the present invention.

As shown, the attenuation unit 300, a cylindrical damping case 310 having an insertion space 311 formed in the center, a plurality of damping holes 312 perforated in the transverse direction on the side of the damping case, a plurality of It includes a damping elastic part 320 that can be selectively inserted into each of the damping holes of the attenuation case and a damping locking part 330 coupled to the side of the damping case to prevent the damping elastic part from being separated from the damping hole.

In addition, a rail 313 is installed in the transverse direction on the inner upper end of the damping case 310, and a plurality of elastic rings 314 are slidably coupled to the rail 313. The plurality of elastic rings 314 is made of an elastic material, and may be folded or unfolded according to an applied impact.

The plurality of elastic rings 314 slide along the rail 313 when the fluid buffer unit 400 inserted into the insertion space 311 is strongly shaken back and forth, left and right, and the external impact applied to the fluid buffer unit 400 by colliding with each other. It plays a role in alleviating.

A plurality of attenuation holes 312 are formed in four directions of the attenuation case 310 by forming a set of four attenuation holes. Four attenuation holes 312 forming a set are stacked up and down, and the attenuation locking part 330 is coupled in the longitudinal direction to cover the attenuation hole 312. The damping locking part 330 may be coupled with a conventional bolt or the like.

The damping elastic part 320 includes a rod-shaped damping insertion part 321 that can be inserted into the damping hole 312, a damping spring 322 having an elasticity and a damping spring having one end coupled to the damping insertion part. It includes a damping cover 323 coupled to the other end.

The damping elastic part 320 may be selectively inserted into the damping hole 312 or separated from the damping hole 312. That is, the user may insert the damping elastic part 320 only into a desired damping hole among the 16 damping holes 312, and accordingly, the overall buffering force of the damping part 300 may be adjusted.

The attenuation insert 321 is made of a magnetic material such as metal, and a permanent magnet 315 is coupled to the inner end of the attenuation hole 312, so when the damping elastic part 320 is inserted into the attenuation hole 312, it is attenuated. The insertion part 321 may be naturally located at the inner end of the damping hole 312.

The attenuation cover 323 is formed to have the same size as the attenuation hole 312, and attenuated as the attenuation locking part 330 coupled to the attenuation case 310 in the longitudinal direction blocks the attenuation cover 323 The elastic part 320 is not separated from the damping hole 312.

As described above, in the present invention, when the fluid buffer unit 400, which will be described later, is inserted into the insertion space 311, the damping unit 300 surrounds the fluid buffer unit 400 from the front, rear, left and right, so that the fluid is buffered from external shock or vibration. The unit 400 and the image sensor 10 coupled thereto may be protected.

In addition, in the present invention, the damping elastic part 320 can be inserted only into a desired damping hole among the plurality of damping holes 312 in consideration of the surrounding situation or the weight of each component, so that the user can freely adjust the buffering force according to the situation. There is an advantage that there is.

20 is a view showing an exploded state of each configuration of a fluid buffer unit according to an embodiment of the present invention.

As shown, the fluid buffer unit 400 according to the present invention is inserted into the insertion space 311 of the damping unit 300, and the fluid supply unit 410, the fluid upper plate 420, the fluid rotating plate 430, the fluid It includes a rotating part 440, a fluid lower plate 450, a fluid receiving part 460, and a fluid connection part 470.

The fluid supply unit 410 is formed in a cylindrical shape with an empty inside so that fluid can be accommodated therein, and a fluid supply hole 411 is perforated to supply a fluid to the lower surface.

The upper surface of the fluid supply unit 410 is made of a material having elasticity such as rubber, and accordingly, when the lifting unit 500 coupled to the upper portion of the fluid supply unit 410 is shaken up and down, the internal fluid is supplied by the pressure. It exits from the hole 411.

The fluid upper plate 420 is disposed under the fluid supply unit 410, and an upper plate through hole 421 is vertically perforated on one side thereof. The fluid upper plate 420 is formed in a disk shape, and the upper plate through hole 421 is formed at a position corresponding to the fluid supply hole 411.

The fluid rotating plate 430 is disposed under the fluid upper plate 420, and a rotation through hole 431 is vertically perforated on one side thereof. A hollow 433 is formed in the central portion of the fluid rotating plate 430 so that the fluid rotating plate is formed in a ring shape as a whole.

A ring-shaped rotation communication groove 432 communicating with the rotation through hole 431 is recessed on the upper surface of the fluid rotating plate 430. Since the fluid rotating plate 430 is rotatable, the fluid transferred from the upper plate through hole 421 to the lower side is transferred directly down through the rotation through hole 431 or through the rotation communication groove 432. ) Can be passed down and then passed down.

The fluid lower plate 450 is disposed under the fluid rotating plate 430, and a lower plate through hole 451 is vertically perforated on one side thereof. The lower fluid plate 450 is formed in a disk shape, and a lower plate communication groove 452 in the form of a ring communicating with the lower plate through hole 451 is recessed on the upper surface of the lower fluid plate 450.

Since the fluid rotating plate 430 is rotatable, the fluid delivered to the lower side through the rotation through hole 431 is transferred directly below through the lower plate through hole 451, or through the lower plate communication groove 452. 451) and then down.

The fluid rotating part 440 is disposed inside the hollow 433 of the fluid rotating plate 430 and coupled to the upper surface of the fluid lower plate 450 to rotate the fluid rotating plate 430. The fluid rotating part 440 is composed of a fluid rotating motor 441 and a fluid rotating gear 442, and the diameter of the fluid rotating gear 442 is formed equal to the inner diameter of the hollow 433.

The fluid rotation motor 441 may be operated by being electrically connected to a control unit (not shown) of the ground measuring device 300, and the fluid rotation gear 442 rotates as the fluid rotation motor 441 operates. The rotating plate 430 may rotate based on the fluid rotating part 440.

The fluid receiving part 460 is disposed under the fluid lower plate 450 and is formed in a cylindrical shape with an empty inside so that the fluid can be accommodated therein. The fluid receiving hole 461 is perforated so that the fluid may be accommodated in the upper surface of the fluid receiving part 460.

The fluid connection part 470 connects the fluid receiving part 460 and the fluid supply part 410. That is, the fluid supplied from the fluid supply unit 410 and transferred downward may be accommodated in the fluid receiving unit 460 and then transferred to the fluid supply unit 410 again through the fluid connection unit 470.

A one-way valve 471 is coupled to the central portion of the fluid connection part 470. In the one-way valve 471, the lifting unit 500 and the image sensor 10 coupled to the upper portion of the fluid supply unit 410 are shaken by an external shock, so that the fluid pressure inside the fluid supply unit 410 increases and the fluid moves downward. When transmitted, it is not transmitted through the fluid connection part 470, but can be transmitted through the fluid upper plate 420, the fluid rotating plate 430, and the fluid lower plate 450.

In other words, as shown by the dotted line in the drawing, the fluid is transferred from top to bottom through the fluid supply unit 410, the fluid upper plate 420, the fluid rotating plate 430, the fluid lower plate 450, and the fluid receiving unit 460.

As described above, in the present invention, since the fluid rotating plate 430 can be rotated by the fluid rotating unit 440, the moving distance of the fluid transferred from the fluid supply unit 410 to the fluid receiving unit 460 can be varied depending on the situation. .

That is, as shown by way of example in the drawing, when the rotation through hole 431 of the fluid rotating plate 430 is rotated to form 180 degrees with the upper plate through hole 421 and the lower plate through hole 451 (longest distance), The fluid supplied from the fluid supply unit 410 sequentially passes through the upper plate through hole 421, the rotary communication groove 432, the rotary through hole 431, the lower plate communication groove 452, and the lower plate through hole 451. It is accommodated in the receiving portion 460.

Although not shown, when the rotation through hole 431 of the fluid rotating plate 430 is disposed on the same axis (shortest distance) to form 0 degrees with the upper plate through hole 421 and the lower plate through hole 451, the fluid supply unit ( The fluid supplied from 410 is sequentially passed through the upper plate through hole 421, the rotation through hole 431, and the lower plate through hole 451 to be received in the fluid receiving part 460.

If the rotation through hole 431 of the fluid rotating plate 430 is rotated to achieve more than 0 degrees and less than 180 degrees with the upper plate through hole 421 and the lower plate through hole 451 (intermediate distance), the moving distance of the fluid is also It is variable to fit.

The frequency applied by external shock or vibration is inversely proportional to the moving distance of the fluid. That is, in order to control high-frequency vibration, the moving distance of the fluid must be set short, and in order to control the low-frequency vibration, the moving distance of the fluid must be set long.

As described above, in the present invention, since the moving distance of the fluid can be varied by rotating the fluid rotating plate 430, the best buffering effect can be provided in consideration of the surrounding environment or applied vibration.

21 is a view showing a cross-sectional view of a lifting unit according to an embodiment of the present invention.

As shown, the lifting part 500 according to the present invention includes a lifting case 510, a lifting rod 520, an upright copper plate 530, an upper stopper 521, a lower stopper 522, and an upper elastic part 540 ), a lower elastic part 541, a rod through hole 550, and a hydraulic pump 560.

The elevating case 510 is coupled to an upper portion of the fluid buffer unit 400 and a space is formed so as to accommodate a fluid therein. The lifting case 510 is formed in a cylindrical shape with an empty inside, and a fluid is accommodated therein.

The elevating rod 520 has a lower end disposed inside the elevating case 510 and can be elevated vertically. The lifting rod 520 is formed in a cylindrical shape with a full inside, and an image sensor 10 is coupled to the upper end of the lifting rod 520.

The vertical copper plate 530 is inserted into the lifting rod 520 and disposed between the outer surface of the lifting rod 520 and the inner surface of the lifting case 510 and formed in a ring shape that can move up and down. The inner diameter of the upright copper plate 530 is the same as the outer diameter of the lifting rod 520, and the outer diameter of the upright copper plate 530 is the same as the inner diameter of the lifting case 510.

The upper stopper 521 is formed to protrude from the side of the lifting rod 520 and is disposed at the top with respect to the upright copper plate 530. The lower stopper 522 is formed to protrude from the side of the lifting rod 520 and is disposed at the lower part with respect to the upright copper plate 530. That is, between the upper stopper 521 and the lower stopper 522, the upright copper plate 530 is coupled so as to be movable up and down.

The upper elastic part 540 is disposed between the lower surface of the upper stopper 521 and the upper surface of the upright copper plate 530, and the lower elastic part 541 is formed between the upper surface of the lower stopper 522 and the upper surface of the upright copper plate 530. It is placed between the lower surfaces. As shown, the vertical copper plate 530 is normally spaced apart from the upper stopper 521 and the lower stopper 522 by a predetermined interval by the elastic force of the upper elastic portion 540 and the lower elastic portion 541.

The rod through hole 550 penetrates the inside of the lifting rod 520 to allow a fluid to pass therethrough. The rod through hole 550 passes through the lifting rod 520 in the transverse direction and penetrates the first transverse through hole 551 and the lifting rod 520 disposed under the upper stopper 521 in the transverse direction, and It is disposed above the lower stopper 522 and penetrates the second transverse through hole 552 and the elevating rod 520 disposed relatively lower than the first transverse through hole 551 in the longitudinal direction, It includes a longitudinal through-hole 553 connecting between the 551 and the second transverse through-hole 552.

In other words, the first transverse hole 551 is disposed in a space spaced apart between the upper stopper 521 and the upright copper plate 530 in normal times, and in the spaced apart between the lower stopper 522 and the upright copper plate 530 Since the second transverse hole 552 is disposed, the fluid inside the lifting case 510 can move freely up and down.

Meanwhile, the hydraulic pump 560 may be connected to the upper inlet 511 and the lower inlet 512 formed on one side of the lifting case 510 to apply pressure to the fluid inside the lifting case 510. The hydraulic pump 560 may be operated by being electrically connected to a control unit (not shown) or the like.

When the fluid pressure in the lower part of the lifting case 510 is higher than the fluid pressure in the upper part of the lifting case 510 through the lower inlet 512, the upper elastic part 540 ) Overcomes the elastic force and moves upward to close the first transverse hole 551. In this state, when the upright copper plate 530 moves further upward, the upper surface of the upright copper plate 530 comes into contact with the lower surface of the upper stopper 521, and the lifting rod 520 rises as a whole.

On the contrary, when the fluid pressure in the upper part of the lifting case 510 is higher than the fluid pressure in the lower part of the lifting case 510 based on the upright copper plate 530 through the upper inlet 511, the upright copper plate 530 becomes the lower elastic part. It overcomes the elastic force of 541 and moves downward to close the second transverse hole 552. In this state, when the moving plate 530 moves further downward, the lower surface of the moving plate 530 comes into contact with the upper surface of the lower stopper 522, and the lifting rod 520 descends as a whole.

When the height of the lifting rod 520 is determined, the hydraulic pump 560 uses the upper inlet 511 and the lower inlet 512 with respect to the vertical copper plate 530, and the fluid pressure and the lifting case on the upper part of the lifting case 510 (510) The lower fluid pressure can be maintained equally, and the upright copper plate 530 is positioned at a predetermined distance between the upper stopper 521 and the lower stopper 522.

At this time, since the first transverse hole 551 and the second transverse hole 552 are both open, the fluid inside the lifting case 510 can be freely moved, and a buffering effect to some extent is obtained by the movement of the fluid. I can.

The present invention described above is not limited by the above-described embodiments and the accompanying drawings, and it is common in the technical field to which the present invention pertains that various substitutions, modifications and changes can be made without departing from the technical spirit of the present invention. It will be obvious to those who have the knowledge of.

10: image sensor 20: laser sensor 30: navigation sensor
40: data fusion unit 50: recognition unit 60: absolute coordinates unit
100: vehicle 200: detachable unit 210: detachable base
211: detachable space 212: guide groove 213: extension groove
214: departure prevention part 215: position regulation groove 216: cover fixing groove
220: detachable cover 221: cover fixing unit 230: detachable body
231: guide part 232: detachable extension 233: protruding stopper
234: position regulation unit 300: attenuation unit 310: attenuation case
311: insertion space 312: damping hole 313: rail
314: elastic ring 315: permanent magnet 320: damping elastic part
321: attenuation insert 322: attenuation spring 323: attenuation cover
330: damping locking part 400: fluid buffering part 410: fluid supplying part
411: fluid supply hole 420: fluid upper plate 421: upper plate through hole
430: fluid rotating plate 431: rotation through hole 432: rotation communication groove
433: hollow 440: fluid rotating part 441: fluid rotating motor
442: fluid rotation gear 450: fluid lower plate 451: lower plate through hole
452: lower plate communication groove 460: fluid receiving part 461: fluid receiving hole
470: fluid connection part 471: one-way valve 500: elevating part
510: lifting case 511: upper inlet 512: lower inlet
520: lifting rod 521: upper stopper 522: lower stopper
530: Shanghai copper plate 540: upper elastic portion 541: lower elastic portion
550: rod through hole 551: first through hole 552: second through hole
553: vertical through hole 560: hydraulic pump

Claims (1)

An image sensor mounted on the upper part of the vehicle to capture image information; A laser sensor mounted on the upper part of the vehicle to detect distance information; A navigation sensor mounted on the upper part of the vehicle to provide navigation information; A data fusion unit that performs data fusion based on the internal and external geometric models of each sensor; And a recognition unit that detects a candidate group in the data through machine learning and sequentially applies an additional machine learning model to the detected candidate group to recognize information represented by a target object. Including,
The geometric parameter of the image sensor is any one or a combination of focal length, main point position, lens distortion parameter, sensor format size, sensor position and posture value, and the geometric parameter of the laser sensor is the incident angle of each laser, distance scale , Distance direction offset, axial offset, sensor position and attitude value, or a combination thereof, and the geometric parameter of the navigation sensor is any one of axial scale, axial offset, sensor position and attitude value Or a combination thereof,
The detachable part is coupled to the upper surface of the vehicle, the damping part is coupled to the upper part of the detachable part, the fluid buffer is inserted into the insertion space of the damping part, the lifting part is coupled to the upper part of the fluid buffer, and the image sensor is coupled to the upper part of the lifting part,
The detachable part,
A detachable base coupled to the upper surface of the vehicle and having a detachable space formed therein; A detachable cover rotatably coupled to one side of the detachable space of the detachable base; And a detachable body that can be seated in the detachable space and detachable from the detachable base. Including,
The detachable base,
A pair of guide grooves recessed on both sides along the longitudinal direction of the detachable space; An extension groove recessed in the rear end of the detachable space; A pair of detachment preventing portions formed at the rear of the upper end of the detachable space and protruding inward to prevent detachment of the detachable body; And a pair of position control grooves each recessed at the rear end of the pair of guide grooves. Including,
The detachable body,
A pair of guide portions protruding on both sides along the longitudinal direction of the detachable body and slidable in a pair of guide grooves; A detachable extension protruding from the rear end of the detachable body and capable of being accommodated in the extension groove; A protruding stopper protruding from the upper surface of the detachable body and having the same width as the width between the pair of detachment preventing portions; And a pair of position control units which are respectively recessed at the rear ends of the pair of guide units to insert a pair of position control units. Including,
The fluid buffer unit,
A fluid receiving unit coupled to the upper portion of the damping unit and having a space formed therein so as to accommodate fluid therein; A fluid lower plate coupled to an upper portion of the fluid receiving portion and having a lower plate through hole formed at one side thereof; A ring-shaped fluid rotating plate rotatably coupled to an upper portion of the lower fluid plate and having a rotation through hole formed on one side thereof and a hollow formed in a central portion thereof; A fluid upper plate coupled to an upper portion of the fluid rotating plate and having an upper plate through hole formed at one side thereof; A fluid supply unit coupled to an upper portion of the fluid upper plate and having a space formed therein to accommodate fluid; A fluid rotating unit coupled to an upper portion of the fluid lower plate and disposed in the hollow of the fluid rotating plate to rotate the fluid rotating plate; And a fluid connection unit connected between the fluid receiving unit and the fluid supply unit and having a one-way valve. Including,
A ring-shaped lower plate communication groove in communication with the lower plate through hole is recessed on the upper surface of the fluid lower plate, and a ring-shaped rotary communication groove in communication with the rotary through hole is recessed on the upper surface of the fluid rotating plate,
As the fluid rotating plate rotates, the fluid introduced through the upper plate through hole from the fluid supply unit can be received in the fluid receiving unit by sequentially passing through the rotary communication groove, the rotary through hole, the lower plate communication groove and the lower plate through hole,
The elevating part,
An elevating case coupled to an upper portion of the fluid buffer and having a space formed therein to accommodate fluid; An elevating rod having a lower end disposed inside the elevating case and capable of elevating up and down; A ring-shaped vertical copper plate inserted into the lifting rod and disposed between the outer surface of the lifting rod and the inner surface of the lifting case and movable up and down; An upper stopper protruding from the side of the elevating rod and disposed at the top with respect to the upright copper plate; A lower stopper protruding from the side of the elevating rod and disposed at a lower portion with respect to the vertical copper plate; An upper elastic portion disposed between the lower surface of the upper stopper and the upper surface of the vertical copper plate; A lower elastic portion disposed between the upper surface of the lower stopper and the lower surface of the vertical copper plate; A rod through hole through which a fluid can pass through the inside of the lifting rod; And a hydraulic pump connected to the lifting case to apply pressure to the fluid inside the lifting case. Including,
The rod through hole may include a first transverse through hole that passes through the elevating rod in a transverse direction and is disposed under an upper stopper; A second transverse through hole passing through the lifting rod in the transverse direction and disposed above the lower stopper and disposed at a relatively lower portion than the first transverse through hole; And a vertical through hole penetrating the lifting rod in a longitudinal direction and connecting between the first transverse through hole and the second transverse through hole. A map system with precision capable of constructing a digital map and a road map through an automated method comprising a.

Images