KR101679699B1 - Specification ananyzation system for producing lidar sensor and method for optimizing sensor specification using the same - Google Patents

Abstract

The present invention relates to a specification analysis system for producing a LIDAR sensor and a method for optimizing the specification using the system. The system analyzes the sensor specification defined for the LIDAR sensor based on the camera modeling, A specification analyzer for extracting design specifications that are close to each other, an optimization specification calculator for re-calculating the optimum sensor specifications based on the design specifications extracted by the specification analyzer, the optimum sensor specifications calculated through the optimization specification calculator, And a final calculation unit for finally extracting a design specification that minimizes an error between the design specification and the design specification.

Description

TECHNICAL FIELD [0001] The present invention relates to a specification analyzing system for producing a lidar sensor, and a specification optimization method using the same. Technical Field [0001]

Embodiments of the present invention relate to a specification analysis system for producing a lidar sensor and a specification optimization method using the same.

Laser distance measurement sensors are often referred to as LIDAR (Light Detection and Ranging) sensors. Usually, the distance measurement applied to the Raidasensor is made by transmitting the laser from the emitter and receiving the laser reflected from the receiver, and inversing the time difference between the two.

The Raidasensor has a scanning type and a flash type. In the scanning type, the rotating body exists in the sensor, and the emitter and the receiver are synchronized with the rotating body, so that the distance can be measured at various angles. The flash type is the same as the scanning type and the distance measuring method, but there is a structural difference. That is, there is no rotating body different from the scanning type.

Referring to FIG. 1, a general flash-type Lada sensor structure is shown. When the laser is irradiated to a specific area (Illuminated Area) 3 through the reflector 2 in the laser emitter 1, the receiver 5 receives the laser reflected in the specific area 3 and returns.

As described above, since the flash type Lidar sensor (hereinafter referred to as a flash Lidar sensor) is mounted on the receiver 5 side of the lens 4, a field of view (FOV) Can be easily changed. In addition, since there is no rotating body, it is possible to guarantee vibration resistance when the vehicle is mounted, and time synchronization between data can be ensured because the sensor values are received simultaneously as photographing.

However, in spite of these advantages, the conventional flash ray sensor has not become popular at present. That's because of the high price of $ 150,000 per sensor. As demand increases in the future, prices may decline. However, in order to increase demand, it will take a lot of time for the car or parts maker to adopt Flash Lidar sensor and develop related applications. In addition, in order to mount a flash Lida sensor on an intelligent vehicle, the sensor specifications required for vehicle applications (Forward Collision Warning, Blind Spot Detection, Pedestrian Detection, etc.) must be satisfied.

The sensor specifications are determined by a number of design factors of the flash Lida sensor. However, it is not easy to derive the specifications of the design parameters (hereinafter referred to as the design specifications) satisfying the sensor specifications, which causes a problem of increasing the development cost and time required for the development of the sensor.

Korean Registered Patent No. 10-1417431 (Three-dimensional spatial information generation system using LIDAR sensor, registered on Apr. 1, 2014) Korean Patent Registration No. 10-0817612 (Registered on Mar. 31, 2008, Laser vision system capable of changing measurement range and measurement accuracy)

An embodiment of the present invention provides a specification analysis system and a specification optimization method using the specification analysis system that can reduce the cost and time required for the sensor production by deriving the optimum design specification satisfying the sensor specification required for the Lidar sensor do.

The problems to be solved by the present invention are not limited to the above-mentioned problems, and other matters not mentioned can be clearly understood by those skilled in the art from the following description.

The specification analyzing system for manufacturing a Lidar sensor according to an embodiment of the present invention analyzes a sensor specification defined for a LIDAR sensor based on a camera modeling to determine a design specification close to the defined sensor specification A specification analyzing unit for extracting; An optimization specification calculator for re-calculating an optimal sensor specification based on the design specification extracted by the specification analyzer; And a final calculation unit for finally extracting a design specification for minimizing an error between the optimum sensor specification calculated through the optimization specification unit and the defined sensor specification.

Also, in the specification analysis system, the defined sensor specification includes a field of view (FOV) and a maximum detection range (MDR) that can be detected by a sensor, A design specification close to the defined sensor specification can be extracted using the following equation defined based on the camera modeling for the view area (FOV) and the maximum detection distance (MDR).

(Where pixel size is the size of one pixel of the receiver, Resolution is the number of pixels lying on the horizontal and vertical sides of the light receiving unit, Focal Length is the distance between the lens and the light receiving unit, The minimum number of points that the sensor recognizes the object to be detected.)

The design specification includes a pixel size, a resolution, a focal length, an object size, and a minimum number of points to recognize a detection object.

In the specification analysis system, the final calculation unit may calculate an error by weighting the defined sensor specification.

In this case, the specification analysis system may further include a weight adjusting unit for adjusting the weight of the sensor specification to be optimized first among the defined sensor specifications.

Further, in the specification analyzing system, the final calculating unit may calculate the final design specification by the following equation (3) that minimizes the error between the optimum sensor specification and the defined sensor specification.

,

는 정의된 센서 사양의 x,y 방향 뷰 영역(FOV),

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 x,y 방향 뷰 영역(FOV),

,

,

는 정의된 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

,

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

는 각 엘리먼트(element)에 대한 가중치임.)(here,

,

Is the x, y direction view area (FOV) of the defined sensor specification,

,

(X, y) view area (FOV) of the optimum sensor specification calculated through the optimization specification calculator,

,

,

(MDR) of the defined sensor specifications for vehicles, motorcycles, and pedestrians,

,

,

(MDR) of the optimum sensor specification for the vehicle, the motorcycle, and the pedestrian calculated through the optimization specification calculator,

Is the weight for each element.)

Meanwhile, a specification optimization method for manufacturing a LIDAR sensor according to an exemplary embodiment of the present invention is a method in a system for analyzing specifications for manufacturing a LIDAR sensor, Analyzing the sensor specification defined on the basis of the camera modeling and extracting a design specification close to the defined sensor specification; Shipment of the optimal sensor specification based on the design specification extracted by the system; And a step of calculating an error between the optimal sensor specification of the system and the defined sensor specification and finally extracting the design specification that minimizes the error.

In addition, in the step of extracting the design specification close to the defined sensor specification, the defined sensor specification may include a field of view (FOV), a maximum detection range (MDR The design specification can be extracted using the following Equation 1 and Equation 2 defined on the basis of the camera modeling with respect to the view area FOV and the maximum detection distance MDR.

[Equation 1]

&Quot; (2) "

(Where pixel size is the size of one pixel of the receiver, Resolution is the number of pixels lying on the horizontal and vertical sides of the light receiving unit, Focal Length is the distance between the lens and the light receiving unit, The minimum number of points that the sensor recognizes the object to be detected.)

The step of finally extracting the design specification with which the error is minimized may further comprise the step of calculating the final design specification by minimizing the error between the optimum sensor specification calculated through the optimization specification calculating section and the defined sensor specification, Can be calculated.

&Quot; (3) "

,

는 정의된 센서 사양의 x,y 방향 뷰 영역(FOV),

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 x,y 방향 뷰 영역(FOV),

,

,

는 정의된 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

,

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

는 각 엘리먼트(element)에 대한 가중치임.) (here,

,

Is the x, y direction view area (FOV) of the defined sensor specification,

,

(X, y) view area (FOV) of the optimum sensor specification calculated through the optimization specification calculator,

,

,

(MDR) of the defined sensor specifications for vehicles, motorcycles, and pedestrians,

,

,

(MDR) of the optimum sensor specification for the vehicle, the motorcycle, and the pedestrian calculated through the optimization specification calculator,

Is the weight for each element.)

The step of finally extracting the design specification that minimizes the error may weight the sensor specification defined above and adjust the weight of the sensor specification to be optimized first among the defined sensor specifications.

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

According to one embodiment of the present invention, it is possible to easily produce a sensor by deriving an optimal design specification that satisfies the sensor specifications required when applying the Lidar sensor to an intelligent vehicle, a defense, or an aerospace application.

In addition, since the required specifications for securing the distance vary depending on the detection object of a vehicle, a motorcycle, a pedestrian, etc., the present invention can derive the optimum design specification by adjusting the weight according to the purpose of use, .

1 is a structural view of a flash type Lada sensor.
FIG. 2 is a configuration diagram of a specification analyzing system for manufacturing a Lidar sensor according to an embodiment of the present invention.
3 is a view for explaining a camera model in relation to a Lidar sensor according to an embodiment of the present invention.
4 and 5 are conceptual diagrams for explaining the mathematical relationship between the sensor specifications and the design specifications using the camera model.
FIG. 6 is a flowchart illustrating a method for optimizing specifications for manufacturing a Lidar sensor according to an embodiment of the present invention. Referring to FIG.
Figs. 7 to 10 show examples in which an optimum design specification is calculated according to various sensor specifications.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and / or features of the present invention, and how to accomplish them, will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. It should be understood, however, that the invention is not limited to the disclosed embodiments, but is capable of many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, To fully disclose the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

For reference, the 'sensor specification' referred to in the present invention refers to an output specification required when the Lidar sensor is applied to an automobile, an aerospace, or the like. The present embodiment includes, but is not limited to, a view area detected by a sensor, a maximum detection distance, and the like.

In addition, the 'design specification' referred to in the following description of the present invention is a specification that internal parts must have in order to output a required sensor specification, and it is a specification required at the design stage.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a configuration diagram of a specification analyzing system for manufacturing a Lidar sensor according to an embodiment of the present invention.

The specification analysis system according to an embodiment of the present invention includes a modeling unit 11, a specification analysis unit 12, an optimization specification calculation unit 13, a final calculation unit 14, and a weight adjustment unit 15.

The modeling unit 11 models the Lidar sensor to derive an optimal design specification. The modeling unit 11 of the present embodiment can model the Lada sensor based on the camera modeling by applying characteristics similar to those of the Lada sensor and the camera. Modeling can reveal mathematical relationships between sensor specifications and design specifications.

For example, referring to FIG. 3, as in the case where the camera receives light reflected from an object through a lens (CCD or CMOS), the Lidar sensor also detects a laser reflected from the object, (FPA or ROIC). Here, the light and the lens of the camera can correspond to the laser and the lens of the Lidar sensor, respectively. Camera modeling can be applied to this structural similarity.

Camera modeling can be applied to a pin-hole camera model without distortion even for long exposure.

The specification analyzing unit 12 determines whether or not the sensor requirement specification (sensor specification) defined for the Lidar sensor is related to the relational expression (that is, the mathematical relation formula between the sensor specification and the design specification) based on the camera modeling in the modeling unit 11 . As a result of analysis, the specification analyzer 12 extracts the design specifications that the sensor internal components should have in order to output the defined sensor specifications close to the defined sensor specifications.

In this case, the defined sensor specification refers to the output specification required when designing the Lidar sensor for automobile, defense, ship, and aerospace, and it is a field of view (FOV) that can be detected by the sensor, A maximum detection range (MDR), and the like.

The view area FOV is divided into an FOV (horizontal FOV) in the x direction and a FOV (vertical FOV) in the y direction as shown in FIG. 4, and can be defined as shown in Equation 1 based on the camera modeling .

As shown in FIG. 5, the maximum detection range (MDR) is divided into a vehicle maximum detection distance, a bicycle maximum detection distance, a pedestrian maximum detection distance, Can be defined as Equation (2) based on camera modering.

In this case, the pixel size, resolution, focal length, object size, minimum points, and the like are used for the internal lens and the light receiving unit This is a design specification to have.

pixel size is the size of a pixel of a receiver, Resolution is the number of pixels lying on the horizontal and vertical sides of the light receiving unit, Focal Length is the distance between the lens and the light receiving unit, Object Size is the size of the object to be detected, It is the minimum number of points that recognize the object.

The optimization specification calculator 13 re-calculates the optimum sensor specification based on the design specification extracted by the specification analyzer 12. [ At this time, the optimum sensor specification can be calculated by using Equation 1 and Equation 2 previously defined in the specification analyzing unit 12. However, since the design specification extracted by the specification analyzing unit 12 is a value optimized for the sensor specification, the sensor specification output by substituting it can be regarded as an optimized value. Therefore, it is called the optimum sensor specification.

The final calculation unit 14 calculates an error between the optimum sensor specification calculated through the optimization specification calculation unit 13 and the predefined sensor specification, and finally extracts the design specification that minimizes the calculated error.

The final calculation unit 14 can be calculated using the following equation (3).

)을 기준값으로 설정하고, 설정한 기 정의된 센서 사양(예컨대,

)과 최적화 사양 산출부를 통해 산출된 최적의 센서 사양(예컨대,

)과의 오차를 계산하는 함수이다. fcost is a predefined sensor specification (eg,

) Is set as a reference value, and the predetermined sensor specification (for example,

) And the optimum sensor specification (for example,

).

,

는 정의된 센서 사양의 x,y 방향 뷰 영역(FOV),

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 x,y 방향 뷰 영역(FOV),

,

,

는 정의된 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

,

,

는 최적화 사양 산출부를 통해 산출된 최적의 센서 사양의 차량, 이륜차, 보행자에 대한 최대검출거리(MDR),

는 각 엘리먼트(element)에 대한 가중치를 나타낸다. Also,

,

Is the x, y direction view area (FOV) of the defined sensor specification,

,

(X, y) view area (FOV) of the optimum sensor specification calculated through the optimization specification calculator,

,

,

(MDR) of the defined sensor specifications for vehicles, motorcycles, and pedestrians,

,

,

(MDR) of the optimum sensor specification for the vehicle, the motorcycle, and the pedestrian calculated through the optimization specification calculator,

Represents a weight for each element.

At this time, the weight adjusting unit 15 assigns weights to the sensor specifications defined by the final calculating unit 14, and in particular, by weighting the sensor specifications to be optimized first among the defined sensor specifications, You can choose to perform the optimization with.

It is assumed that the area of the sensor to be manufactured is determined when the optimization is performed. Since the increase in the size of the sensor means an increase in manufacturing cost, optimization of the sensor specification is performed in a given area.

The operation process in the specification analysis system configured as above will be described in detail with reference to FIG.

First, in step S10, the specification analysis system 10 performs modeling on the Lada sensor through the modeling unit 11. [ In other words, based on the camera modeling, we define the mathematical relationship between the sensor specifications and the design specifications of the LIDAR sensor.

Thereafter, in step S11, the specification analysis unit 12 of the system 10 analyzes the sensor requirement specification (sensor specification) defined for the Lada sensor based on the mathematical relationship defined through the modeling unit 11 do. As a result of the analysis, it is possible to extract the optimum design specification close to the defined sensor specification.

For example, a defined sensor specification may include a field of view (FOV), a maximum detection range (MDR), etc. that can be detected by the sensor.

The view area and the maximum detection distance can be expressed by equations (1) and (2) described above, and the design specification can be extracted through this.

The view area FOV and the maximum detection distance MDR through the specification analyzer 12 are influenced simultaneously by changes in the focal length and a plurality of maximum detection distances MDR are given depending on the object It is not easy to find the optimum design specification. Therefore, the following optimization process is further performed.

That is, in step S12, the optimization specification calculator 13 of the system 10 re-calculates the optimum sensor specification based on the design specification extracted through the specification analyzer 12. [

In the next step S13, the final calculation unit 14 of the system 10 sets the predefined sensor specification as a reference value and calculates the difference (error) between the re-calculated optimal sensor specification and the calculated error Obtain design specifications that can be minimized.

At this time, if there is a sensor specification to be optimized first among the sensor specifications, for example, if it is requested to secure the distance for the vehicle detection more preferentially, in the next step S14, the weight adjusting section 15 of the system 10 The weights can be adjusted for the sensor specification of the maximum detection distance (MDR). Therefore, it is possible to selectively perform the optimization based on the sensor specifications.

FIGS. 7 to 10 show examples in which such an optimization method is applied to various sensor specifications.

In the embodiments of FIGS. 7 to 10, the object size is generally divided into a vehicle, a motorcycle, and a pedestrian, and defines a constant value for each object size. It is assumed that the minimum number of detection points is 3, the sensor size is 1024 pixels, and one pixel size is 100 占 퐉 占 100 占 퐉. The size of the vehicle is assumed to be 5 m (Height) × 2 m (Width) × 1.5 m (Length), the size of the motorcycle is 2 m × 0.8 m × 1.2 m, and the pedestrian size is 0.5 m × 0.5 m × 1.8 m.

As shown in FIG. 7 to FIG. 10, as a result of optimization for various sensor specifications, the optimized design specifications are derived as shown in (b), the error is displayed for each sensor specification, And shows the difference in optimized sensor specifications.

7 and 8, the absolute value sum of each error with respect to the view area FOV and the maximum detection distance is smaller in the embodiment of FIG. 8 as compared with the embodiment of FIG. That is, it can be seen that the sensor specification capable of producing a given sensor size of 1024 pixels can be optimized to the design specification according to the embodiment of FIG. This allows the correct sensor specification to be established.

On the other hand, in order to satisfy the sensor specification according to the embodiment of FIG. 7, the sensor size must be enlarged, which may lead to a cost increase.

Next, referring to the embodiment of FIG. 9, this is an optimization method for the remote RR sensor specification.

The error for the view area (FOV) was small, but the error for the maximum detection distance of the motorcycle was large. If distance is required for the detection of the motorcycle, weights can be adjusted. However, as other features are affected, the final choice must be made for the purpose of the sensor.

Next, referring to the embodiment of FIG. 10, it is a method for optimizing the sensor specifications for a near-infrared sensor.

Since the error is considerably large for each sensor specification, it can be concluded that the given sensor size can not satisfy the required conditions. After all, it is necessary to make the sensor size increase.

As described above, the present invention can reduce the cost and development time required for manufacturing the sensor by deriving the optimum design specification that satisfies the required specification from the given requirement for the Lidar sensor.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the scope of the appended claims and equivalents thereof.

10: Specification Analysis System
11: Modeling unit
12: Specification analysis section
13: Optimization specification calculating section
14:
15:

Claims (8)

A specification analyzing unit for analyzing the sensor specifications defined for the LIDAR sensor based on the relationship established based on the camera modeling and extracting the design specifications to be provided by the sensor components in order to output the defined sensor specifications, ;
An optimization specification calculator for re-calculating the sensor specification based on the design specification extracted by the specification analysis unit; And
And a final calculation unit for finally extracting a design specification that minimizes an error between the sensor specification recalculated through the optimization specification unit and the sensor specification,
The final calculation unit calculates,
And the final design specification is calculated by the following equation (3) for minimizing an error between the sensor specification re-calculated through the optimization specification unit and the sensor specification defined above.
&Quot; (3) "

(here,

,

Is the x, y direction view area (FOV) of the defined sensor specification,

,

(X, y) view area (FOV) of the optimum sensor specification calculated through the optimization specification calculator,

,

,

(MDR) of the defined sensor specifications for vehicles, motorcycles, and pedestrians,

,

,

(MDR) of the optimum sensor specification for the vehicle, the motorcycle, and the pedestrian calculated through the optimization specification calculator,

Is the weight for each element.) The method according to claim 1,
The defined sensor specification includes a field of view (FOV) and a maximum detection range (MDR) that can be detected by the sensor,
The specification analyzer
The design specification is extracted using the following Equation 1 and Equation 2 defined on the basis of the camera modeling with respect to the view area FOV and the maximum detection distance MDR. .
[Equation 1]

&Quot; (2) "

(Where pixel size is the size of one pixel of the receiver, Resolution is the number of pixels lying on the horizontal and vertical sides of the light receiving unit, Focal Length is the distance between the lens and the light receiving unit, The minimum number of points that the sensor recognizes the object to be detected.) 3. The method according to claim 1 or 2,
The design specification includes a pixel size, a resolution, a focal length, an object size, and a minimum number of points for recognizing an object to be detected. Specification Analysis System for Lidar Sensor Fabrication. The method according to claim 1,
The final calculator
And the error is calculated by giving a weight to the sensor specification defined above. The method according to claim 1 or 4,
A weight adjustment unit for adjusting and applying a weight to a sensor specification to be optimized first among the sensor specifications defined above,
Wherein the system further comprises a controller for controlling the system. delete As a method in a system for analyzing specifications for LIDAR sensor fabrication,
The system analyzes the sensor specifications defined for the LIDAR sensor based on the relationship established based on the camera modeling and extracts the design specifications that the sensor components should have in order to output the defined sensor specifications step;
Re-calculating the sensor specification based on the extracted design specification; And
The system calculates an error between the re-calculated sensor specification and the defined sensor specification, and finally extracts the design specification with the minimum error,
Finally, the step of finally extracting the design specification,
Wherein the system calculates a final design specification according to the following formula (3) that minimizes an error between the re-calculated sensor specification and the defined sensor specification.
&Quot; (3) "

(here,

,

Is the x, y direction view area (FOV) of the defined sensor specification,

,

(X, y) view area (FOV) of the optimum sensor specification calculated through the optimization specification calculator,

,

,

(MDR) of the defined sensor specifications for vehicles, motorcycles, and pedestrians,

,

,

(MDR) of the optimum sensor specification for the vehicle, the motorcycle, and the pedestrian calculated through the optimization specification calculator,

Is the weight for each element.) 8. The method of claim 7,
Finally, the step of finally extracting the design specification,
Wherein weighting is applied to the sensor specifications defined above, and the weights can be adjusted with respect to sensor specifications to be preferentially optimized among the sensor specifications defined above.

Images