KR102237927B1 - Prediction Method of Urban Land Use Change by Autonomous Vehicle - Google Patents

Abstract

In order to predict a change in land use in a city by an autonomous vehicle, the present invention selects any one of a plurality of scenarios after selecting a city to predict the change in land use, and the same according to the size of the city selected to apply the cellular automata. Divide into a plurality of cells having a size, divide the land use of the plurality of cells, extract the accessibility, land price and green area accessibility values according to the land use for each cell, and extract the neighboring effect value according to the land use for each cell In addition, different accessibility weights, land price weights, and green area accessibility weight values are set according to the land use, and the accessibility weight, land price weight, and green area in the conversion rule indicating the conversion possibility that land can be used for different purposes for each cell. The convertibility value is extracted for each land use by applying the accessibility weight value and the neighboring effect value, and the maximum value among the extracted convertibility values and the convertibility value of the current cell land use are compared to simulate the prediction of the change in land use of the cell. It relates to a method for predicting changes in urban land use by autonomous vehicles. According to the present invention, it is possible to predict changes in land use in a city when the autonomous vehicle becomes a major means of transportation by simulation based on scenarios divided according to the degree of automation of the autonomous vehicle. Accordingly, effective urban design and policy establishment are possible. It is possible.

Description

Prediction Method of Urban Land Use Change by Autonomous Vehicle}

The present invention relates to a land use change prediction method, and more specifically, to a method for predicting land use change in a city by an autonomous vehicle that predicts how the land use in a city will change when an autonomous vehicle is introduced.

If self-driving vehicles become the main means of transportation, people can engage in activities other than driving, such as reading, eating, and talking in the self-driving vehicle.

This means that people are no longer restricted from the act of driving, and it means that people's perception of travel time is different.

It can be expected that the perception of the changed travel time will affect people's choice of residence.

Existing studies predict that if autonomous vehicles are introduced, cities will become suburban, but no specific prediction results or models have been presented.

Therefore, there is a need for a model that can support the prediction of city change when autonomous vehicles become a major means of transportation.

An object of the present invention is to predict a change in land use as an autonomous vehicle is introduced, and to a method for predicting a change in land use in a city by an autonomous vehicle in which regulations and policies suitable for the change can be prepared.

In order to achieve the above object, the method for predicting land use change in a city by an autonomous vehicle according to an embodiment of the present invention is a city in which land use change is predicted in order to predict a change in land use in a city by an autonomous vehicle. After selecting, select any one of a plurality of scenarios, divide into a plurality of cells having the same size according to the size of a city selected to apply cellular automata, classify the land use of the plurality of cells, and use the land use for each cell. It extracts accessibility, land price, and green area accessibility values according to each cell, extracts neighboring effect values according to land use for each cell, sets different accessibility weights, land price weights, and green area accessibility weight values according to land use, and each cell The convertibility value is extracted for each land use by applying the accessibility weight, land price weight, green area accessibility weight value, and neighborhood effect value to a conversion rule indicating the conversion possibility that each land can be used for different purposes, and the extracted The maximum value of the convertibility values and the convertibility value of the current cell's land use are compared to simulate the prediction of the cell's land use change.

Here, the plurality of scenarios may be a scenario in which an autonomous vehicle having an automation level of 0 to 2 is common and an autonomous vehicle having an automation level of 3 to 5 is common.

In addition, the accessibility value is set in stages based on the distance between the cell and the main road and main traffic points,

The accessibility weight value may be set in stages using land use, the scenario, and the accessibility value.

And, the land price value is set step by step using an average value of land price data included in the cell,

The land price weight value may be set step by step using the land use, the scenario, and the land price value.

In addition, the green area accessibility value is set in stages based on the distance between the cell and the green area and the land use of the cell,

The green area accessibility weight may be set step by step using the land use, the scenario, and the green area accessibility value.

이며, 상기

는 K 토지 용도의 이웃효과이고, 상기

는 c거리 내의 토지 용도 L의 가중치 매개변수이며, 상기

는 관성효과일 수 있다.And, the functional formula of the neighboring effect value,

Is, above

Is the neighborhood effect of K land use, and above

Is the weighting parameter of the land use L within distance c, above

May be an inertial effect.

The c can be set by Perry's neighborhood unit theory and main road coverage.

는, 상기 c 범위 내에 동일한 용도의 셀을 포함하면 상기

은 1 값을 가지고, 동일한 용도의 셀을 포함하지 않으면 상기

은 0 값을 가질 수 있다.Also, above

Is, if the cell of the same purpose is included in the range c, the

Has a value of 1, and if it does not contain cells for the same purpose, the above

Can have a value of 0.

이며, 상기 K는 셀의 토지용도이고, 상기

는 토지 용도 K로의 전환가능성이며,

는 토지용도가 K로 변할 접근성 가중치 값이며,

는 토지용도가 K로 변할 토지가격 가중치 값이고,

는 토지용도가 K로 변할 녹지접근성 가중치 값이며,

는 토지용도가 K인 셀의 이웃효과 값일 수 있다.The convertibility value function formula is,

And K is the land use of the cell, and

Is the possibility of converting to land use K,

Is the accessibility weight value at which the land use will change to K,

Is the land price weight value at which the land use will change to K,

Is the green area accessibility weight value at which the land use will change to K,

May be the neighboring effect value of a cell whose land use is K.

,

,

및

는 0이상 1이하의 값을 가질 수 있다.And, above

,

,

And

Can have a value of 0 or more and 1 or less.

According to the method for predicting changes in land use in a city by an autonomous vehicle of the present invention,

First, it is possible to predict changes in land use in cities when autonomous vehicles become the main means of transportation by simulation based on scenarios divided according to the degree of automation of autonomous vehicles, thereby enabling effective urban design and policy establishment.

1 is an exemplary diagram showing main route data.
2 is an exemplary diagram showing data of major traffic points.
3 is an exemplary view showing green area data.
4 is an exemplary diagram showing the distribution of land use in the metropolitan area.
5 is an exemplary diagram showing the distribution of accessibility in the metropolitan area.
6 is an exemplary diagram showing the distribution of land prices in the metropolitan area.
7 is an exemplary diagram showing the distribution of green area accessibility in the metropolitan area.
8 is an algorithm showing a process of predicting a change in land use of a cell.
9 is an exemplary diagram showing the distribution of land use in the metropolitan area according to a scenario.
10 is an exemplary diagram showing the distribution of land use in the metropolitan area according to a scenario.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. In this case, it should be noted that the same components in the accompanying drawings are indicated by the same reference numerals as possible. In addition, detailed descriptions of known functions and configurations that may obscure the subject matter of the present invention will be omitted. For the same reason, some components in the accompanying drawings are exaggerated, omitted, or schematically illustrated.

1 is an exemplary diagram showing main route data, FIG. 2 is an exemplary diagram showing data of major traffic points, FIG. 3 is an exemplary diagram showing green area data, and FIG. 4 is an exemplary diagram showing land use distribution in the metropolitan area, 5 is an exemplary diagram showing the distribution of accessibility in the metropolitan area.

And, FIG. 6 is an exemplary diagram showing the distribution of land prices in the metropolitan area, FIG. 7 is an exemplary diagram showing the distribution of green area accessibility in the metropolitan area, and FIG. 8 is an algorithm showing a process of predicting a change in land use of a cell, and FIG. 9 is It is an exemplary diagram showing the distribution of land use in the metropolitan area according to the scenario, and FIG. 10 is an exemplary view showing the distribution of land use in the metropolitan area according to the scenario.

The method for predicting changes in land use in a city by an autonomous vehicle according to the present invention,

After selecting a city to predict land use change, select any one of multiple scenarios,

Divide into a plurality of cells having the same size according to the size of the city selected to apply the cellular automata,

The land uses of the plurality of cells are identified.

Here, the plurality of the above scenarios are a scenario in which an autonomous vehicle having an automation level of 0 to 2 is common and an autonomous vehicle having an automation level of 3 to 5 is common.

Self-driving vehicles are classified into 6 levels, ranging from level 0 to level 5, depending on the degree of control and ability to cope with the situation (SAE, 2014).

Autonomous vehicles at level 0 are vehicles that have no automation function, and autonomous vehicles at level 1 can automatically control the speed or direction, and autonomous vehicles at level 2 can control the speed and direction at the same time within the designated area. Can be controlled.

In addition, autonomous vehicles of level 3 can autonomously drive within a given area, but driver intervention is required in case of inclement weather such as heavy snow conditions, and autonomous vehicles of level 4 are capable of self-driving within a designated area. No driver intervention is required, and the 5-level autonomous vehicle is a fully autonomous vehicle that does not require driver intervention in all situations.

Here, the division of land use of the cell is determined by the total area of the building according to the use in the cell, and the use of the cell representing the cell is the land use with the largest area among the buildings for each use included in the cell.

Land uses include residential, commercial, social, industrial, agricultural, vacant, public and green areas.

And, for each cell, the accessibility, land price, and green area accessibility values according to the land use are extracted,

For each cell, the neighboring effect value according to the land use is extracted,

For each cell, different accessibility weights, land price weights, and green area accessibility weights are set according to land use.

For each cell, the convertibility value is extracted for each land use by applying the accessibility weight, land price weight, green area accessibility weight value, and neighborhood effect value to a conversion rule indicating the conversion possibility that land can be used for different purposes,

The maximum value of the extracted convertibility values and the convertibility value of the current cell land use are compared to simulate a change prediction of the land use of the cell.

Here, the accessibility weights, land price weights, and green area accessibility weights are set as the distribution ratios for each land use in a cell-divided area in the previously published paper and current Korean spatial information.

The accessibility value is set step by step based on the distance between the cell and major roads and major traffic points.

The accessibility weight value is set step by step according to the distance using land use, scenario and accessibility values.

Data on major roads uses OpenStreetMap, and is displayed as shown in FIG. 1.

And, the distance to the main road is set as a straight line distance from the center of the cell to the nearest main road.

The data on the main intersection point uses the intelligent transport system standard node data (National Transport Information Center), and is shown as shown in FIG. 2.

The accessibility value is set by the distance to the nearest major road that autonomous vehicles can drive and the distance to the nearest traffic point where the autonomous public transport system is to be built using these data.

The land price value is set as the average value of the price of the land included in the cell based on data notified to the National Geographic Information Institute, and is set step by step according to the land price.

The land price weight value is set differently by using the land use, scenario and land price value.

The green area accessibility value is set step by step based on the distance between the cell and the green area and the land use of the cell.

The green area accessibility weight value is set differently according to the distance based on the distance between the cell and the green area using land use, scenario and green area accessibility values.

Green area data is limited to Level 1 and Level 2 areas in the ecological characteristic map based on the national environment information network system, and is shown as shown in FIG. 3.

이다.The neighborhood effect is calculated using land use, and the functional equation is

to be.

In each cell, ranges of 200m, 400m, 600m, and 1000m are formed, and the land use ratio of the cells that overlap the ranges is used.

는 K 토지용도의 이웃효과이고,

는 c거리 내의 토지용도 L의 가중치 매개변수이며,

는 관성효과를 나타낸다.here,

Is the neighborhood effect of K land use,

Is the weight parameter of land use L within distance c,

Represents the inertial effect.

c is determined by Perry's neighborhood unit theory and main road coverage,

L represents the current land use of the cell, and the use is any one of residential, commercial, industrial, agricultural, public, green and vacant land.

값은 c 거리에 따라 미리 설정된다.here,

The value is preset according to the c distance.

는 측정되는 셀이 c 범위 내에 동일한 용도의 셀을 포함하면

은 1 값을 가지고, 동일한 용도의 셀을 포함하지 않으면

은 0 값을 가진다.

Is, if the cell to be measured contains cells of the same purpose within the range c

Has a value of 1, and if it does not contain cells of the same purpose

Has a value of 0.

이다.The functional formula for convertibility is

to be.

는 토지용도 K로의 전환가능성이다.K is the cell's land use,

Is the possibility of converting to land use K.

는 셀의 토지용도가 K로 변할 접근성 가중치이며,

는 셀의 토지용도가 K로 변할 토지가격 가중치이고,

는 토지용도가 K로 변할 녹지접근성 가중치이며,

는 토지용도가 K인 셀의 이웃효과 값이다.

Is the accessibility weight at which the cell's land use will change to K,

Is the land price weight at which the cell's land use will change to K,

Is the green area accessibility weight at which the land use will change to K,

Is the neighboring effect value of a cell whose land use is K.

,

,

및

는 0이상 1이하의 값을 가진다.here,

,

,

And

Has a value of 0 or more and 1 or less.

It will be described in detail with an example.

Select a city in the Seoul metropolitan area to predict land use change, and select one of the two scenarios described above.

The area of Seoul's metropolitan area is about 11,700 square kilometers and has a population of 26 million.

If the Seoul metropolitan area is divided into 200m×200m cells to apply the cellular automata, the Seoul metropolitan area is divided into 279,742 cells.

And, as shown in Figure 4, 279,742 cells are divided according to the current land use,

Here, the use of the cell may be classified into residential, commercial, industrial, agricultural, public, green and empty land.

For each cell, the accessibility value according to the land use, the land price value, and the green area accessibility value are extracted, and the neighboring effect value according to the land use is extracted.

And, using the accessibility value, the land price value, and the green area accessibility value, steps are divided to set the accessibility weight value, the land price weight value, and the green area accessibility weight value as shown in FIGS. 5 to 7.

Here, the accessibility weight value, the land price weight value, and the green area accessibility weight value are set differently for residential, commercial, industrial, agricultural, public, green and vacant lands.

Thereafter, by applying the accessibility weight, land price weight, green area accessibility weight, and neighbor effect values to the conversion rule for each cell, a conversion possibility value that can be used for different purposes of land for each cell is extracted.

Here, all convertible values of one cell for residential, commercial, industrial and agricultural use are extracted.

The use of public and green cells is fixed and the use does not change, and vacant land can be changed to residential, agricultural, industrial and commercial, but residential, agricultural, industrial and commercial cannot be changed to vacant land.

Therefore, if the current use of the cell is residential, agricultural, industrial and commercial, all convertibility values for residential, commercial, industrial and agricultural use are calculated; for public and green cells, the convertibility value is not calculated. If so, calculate the convertibility values for residential, commercial, industrial and agricultural use.

In addition, the maximum value of convertibility values for uses other than the current land use of each cell is extracted, and the change in land use can be predicted by comparing the convertibility value of the current land use with the extracted maximum value.

Here, the change in land use is predicted by comparing the convertibility value of the current use of any one cell with the maximum value of the convertibility value of the other uses excluding the current use.

8 is an algorithm showing a process for predicting a change in land use.

As shown in FIG. 8, for each cell, convertibility values of land uses other than the current land use are extracted, and a maximum value is extracted among them.

In addition, by comparing the convertibility value of the current land use with the extracted maximum value, if the convertibility value of the current land use is higher, the current land use is maintained, and if the extracted maximum value is higher, the change in land use can be predicted. have.

For example, all convertibility values for residential, commercial, industrial, agricultural, and vacant land of one cell are extracted.

If the cell's current use is in agriculture, the maximum of the convertible values for residential, commercial and industrial excluding agriculture is extracted.

If the extracted maximum value is a commercial convertibility value, the agricultural convertibility value and the commercial convertibility value are compared, and if the agricultural use is larger, the current cell retains its existing use, and if the commercial use is larger, it can be expected to change to commercial use.

Here, it is assumed that cells can be changed into residential, industrial, commercial, and agricultural cells, and that the use of public and green areas does not change.

In addition, vacant land can be converted into residential, commercial, industrial and agricultural cells, but residential, commercial, industrial and agricultural cells cannot be converted into vacant land.

Therefore, if the current use of the cell is residential, agricultural, industrial and commercial, all convertibility values for residential, commercial, industrial and agricultural use are calculated, and for public and green cells, the convertibility value is not calculated. If so, calculate the convertibility values for residential, commercial, industrial and agricultural use.

9 shows simulation results for a scenario in which an autonomous vehicle having 0 to 2 levels is common.

10 shows simulation results for a scenario in which an autonomous vehicle having 3 to 5 levels is common.

Therefore, it is possible to predict changes in land use by analyzing the results derived according to the scenario.

Although described with reference to Examples, those skilled in the art can understand that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention described in the following claims. There will be.

Claims (10)

To predict changes in urban land use by autonomous vehicles,
After selecting a city to predict land use change, select any one of multiple scenarios,
Divide into a plurality of cells having the same size according to the size of the city selected to apply the cellular automata,
Classify the land use of the plurality of cells,
For each cell, the accessibility value, land price value, and green area accessibility value according to land use are extracted,
For each cell, the neighboring effect value according to the land use is extracted,
Different accessibility weights, land price weights, and green area accessibility weights are set according to land use,
For each cell, a conversion possibility value for each land use is extracted by applying the accessibility weight, land price weight, green area accessibility weight value, and neighborhood effect value to a conversion rule indicating the conversion possibility that land can be used for different purposes,
A method for predicting land use change in an urban area by means of an autonomous vehicle that simulates predicting a change in land use of a cell by comparing a maximum value among the extracted convertibility values with a convertibility value of a current cell's land use. The method of claim 1,
A plurality of the above scenarios,
A method for predicting changes in land use in a city by an autonomous vehicle, characterized in that a scenario in which an autonomous vehicle having an automation level of 0 to 2 levels becomes common and an autonomous vehicle having an automation level of 3 to 5 levels is universal . The method of claim 2,
The accessibility value is,
It is set in stages based on the distance between the cell and major roads and major traffic points,
The accessibility weight value is,
A method for predicting changes in land use in a city by an autonomous vehicle, characterized in that it is set step by step using the land use, the scenario, and the accessibility value. The method of claim 3,
The land price value is,
It is set step by step using the average value of the land price data included in the cell,
The land price weight value is,
A method for predicting changes in land use in a city by an autonomous vehicle, characterized in that it is set step by step using the land use, the scenario, and the land price value. The method of claim 4,
The green area accessibility value is,
It is set in stages based on the distance between the cell and the green area and the land use of the cell,
The green area accessibility weight is,
A method for predicting changes in land use in a city by an autonomous vehicle, characterized in that it is set step by step using the land use, a scenario, and the green area accessibility value. The method of claim 5,
The functional formula of the neighboring effect value is,

Is,
remind

Is the neighborhood effect of K land use, and above

Is the weighting parameter of the land use L within distance c, above

A method for predicting changes in land use in cities by autonomous vehicles, characterized in that is an inertial effect. The method of claim 6,
C is,
A method for predicting changes in land use in cities by autonomous vehicles, characterized in that it is set by Perry's neighborhood unit theory and main road coverage. The method of claim 7,
remind

Is,
If the cell of the same purpose is included in the range c, the

Has a value of 1,
If it does not contain cells of the same purpose, the above

A method for predicting changes in land use in a city by an autonomous vehicle, characterized in that has a value of 0. The method of claim 8,
The convertibility value function formula is,

Is,
K is the land use of the cell, and

Is the possibility of converting to land use K,

Is the accessibility weight value at which the land use will change to K,

Is the land price weight value at which the land use will change to K,

Is the green area accessibility weight value at which the land use will change to K,

Is a method for predicting changes in land use in cities by autonomous vehicles, characterized in that the neighboring effect value of a cell whose land use is K. The method of claim 9,
remind

,

,

And

A method for predicting changes in land use in an urban area by an autonomous vehicle, characterized in that it has a value of 0 or more and 1 or less.

Images