KR101196084B1 - System for optimal path planning of mobile object with genetic algorithm - Google Patents

Abstract

An optimal path generation system for a moving object using a genetic algorithm is disclosed. An optimal path generation system for a moving object includes: an information input unit for providing real-time road situation information indicating an average moving speed for each road in a moving area and information on an urban area including location information of roads and buildings located in a moving area; And a path generation unit generating a moving path in the moving area through a genetic algorithm using the urban area information and the average moving speed.

Description

SYSTEM FOR OPTIMAL PATH PLANNING OF MOBILE OBJECT WITH GENETIC ALGORITHM}

Embodiments of the present invention are directed to a path generation system for generating an optimal moving path of a moving object in a downtown area by using a genetic algorithm capable of searching various areas in consideration of a complex urban area situation.

In recent years, a map service of various environments has been required as the hardware performance has been improved and advanced, and a 3D map service displaying a moving path of a moving object through a 3D map has been provided.

Meanwhile, in producing a 3D image related to a video or a game, a moving path of a character may be generated by applying a predetermined path finding algorithm based on a starting point and a target point designated for the character. The pathfinding algorithm refers to an algorithm that allows enemy characters or AI units that are not directly controlled by the user to go to a specific destination by avoiding obstacles by themselves. For example, in a game program such as StarCraft, it is also possible to find a way for friendly units to avoid the blocked terrain. Such a path finding algorithm is implemented by various methods, and recently, a path finding algorithm called 'A * algorithm' is applied to a video or a game as an efficient path finding algorithm.

Such a path finding algorithm is utilized for a method of generating a proper movement path of a character in a situation where complex 3D computer graphics is used. However, it is necessary to generate the optimum moving route by using complex area information as in the situation in the 3D map. In the case of a wide area, the path generation time is delayed due to the large amount of area information.

In the complex and extensive urban area, the optimal path generation system for moving objects is provided to consider the location of buildings and roads for each area and to generate the optimal moving path of moving objects by using the real-time information of the area including the road. .

By using goodness-of-fit evaluation considering various factors that are characteristic of genetic algorithms, and selecting algorithms, crossing, and searching for mutant regions in a wide range of regions, algorithm computation time can be minimized in generating optimal moving paths of moving objects in urban areas. An optimal path generation system for a moving object is provided.

An information input unit for providing real-time road situation information indicating an average moving speed for each road in the moving area and the downtown area information including the location information of roads and buildings located in the moving area; And a path generation unit for generating a moving path in the moving area through a genetic algorithm using the city center information and the average moving speed.

According to one side, the information input unit, a downtown area information acquisition module for obtaining location information of roads and buildings in a moving area in association with a map providing server for maintaining and managing map-related data; And a real-time road situation information acquisition module for obtaining an average moving speed of moving objects on each road in a moving area by interworking with a road information providing server that maintains and manages road condition information for each road.

According to another aspect, the route generation unit may include: an area information database for maintaining city center information and an average moving speed provided from the information input unit; And an initial path group including a plurality of paths that can be moved within the moving area by using the information of the urban area and the average moving speed, and then evaluating the fitness for each path through the fitness evaluation of the genetic algorithm and then initial path group according to the fitness. It includes a path generation module that selects one path among the paths.

According to another aspect, the path generation module selects two paths from the initial path group in a tournament method that compares the fitness of two paths by two paths, and selects one path having a high degree of fitness, and the paths meet each other for the two paths. After changing the two paths by exchanging sections from one intersection point to another intersection point, the fitness of two paths can be re-evaluated to finally select a path having a high degree of fitness as a moving path.

Genetic algorithm's initial path group generation, suitability evaluation, selection, using building location information and road information in urban areas with complex building locations and irregular roads, and real-time average moving speeds due to congestion of moving objects in urban areas. The hybridization and mutation methods generate the optimal movement path of the moving object. Therefore, it is possible to ensure the optimum travel time and distance from the starting point to the target point through the fitness evaluation considering various factors.

Since urban areas for route generation have a wide variety of considerations, various movement paths can be explored by applying selection, breeding, and mutation methods based on genetic algorithms to utilize various information in a wide range of regions. By reducing unnecessary memory consumption, the computational time of the algorithm can be effectively reduced.

1 is a block diagram illustrating an internal configuration of a system for generating an optimal path of a moving object using a genetic algorithm according to an embodiment of the present invention.
2 is a flowchart illustrating a process of generating a path of a mobile object using a genetic algorithm according to an embodiment of the present invention.
3 is a diagram for describing a process of selecting a movement path through tournament selection of a genetic algorithm according to an embodiment of the present invention.
4 is a diagram for explaining a process of crossing a moving path through two-point crossing of a genetic algorithm according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating an internal configuration of a system for generating an optimal path of a moving object using a genetic algorithm according to an embodiment of the present invention.

The optimal path generation system of a moving object according to an embodiment divides the urban area at regular intervals, inputs whether the moving object is movable for each region, and inputs average moving speed information of the real-time road. These data are then used to construct chromosome generation through initial routing. In the generated chromosome generation, the fitness evaluation according to the mathematical formula can be performed, the selection, crossing, and mutagenesis methods can be applied to change the migration route configured in the chromosome generation and generate the optimal migration path through the stop condition.

As illustrated in FIG. 1, an optimal path generation system of a moving object according to an embodiment includes an information input unit 100, a path generation unit 200, and a viewer generation unit 300.

Information Input (100)

The information input unit 100 obtains the information on the moving area of the city center of the moving object in response to the information request in real time and provides the obtained path information to the path generation unit 200 for path generation. The information input unit 100 collects in real time road information of a region separated by a predetermined interval in an urban area, location information of various buildings, and information obtained from an average moving speed of moving objects on a road in real time. The collected information is provided to the path generation unit 200. An internal configuration of the information input unit 100 for performing the above-described functions includes an information management module 110, a downtown area information acquisition module 120, and a real-time road situation information acquisition module 130.

The downtown area information acquisition module 120 periodically acquires downtown area information including road information and location information of buildings in association with a map providing server (not shown) that maintains and manages map-related data.

The real-time road situation information acquisition module 130 interlocks with a road information providing server (not shown) that maintains and manages real-time road situation information in real time, including an average moving speed of moving objects on each road in an urban area in real time. Obtain information.

The information management module 110 generates a path by using the urban area information acquired by the urban area information acquisition module 120 and the real-time road situation information obtained by the real-time road situation information acquisition module 130 when a movement route request is generated from the outside. Provided to the unit 200, and updates the downtown area information and the real-time road situation information periodically or in real time.

Route Generator (200)

The route generation unit 200 performs a function of generating an optimal route using a genetic algorithm by utilizing the urban area information and the real-time road situation information on the urban area provided from the information input unit 100. The internal configuration of the path generation unit 200 for performing the above function includes a local information database 210 and a path generation module 220.

The area information database 210 receives the city area information and real-time road situation information (hereinafter, collectively referred to as 'city information') for the city area from the information input unit 100 and updates the city information while updating the city information. Keep it. In this case, the updated downtown information is transmitted to the path generation module 220.

The path generation module 220 receives the downtown information from the local information database 210 and generates an optimal moving path of the moving object by utilizing the downtown information through the genetic algorithm application module 222.

The process of generating the optimum path in the path generation module 220 is as follows.

2 is a flowchart illustrating a process of generating a path of a mobile object using a genetic algorithm according to an embodiment of the present invention.

In step S2, the route generation module 220 receives city information from the local information database 210. The route generation module 220 divides the downtown area where the moving object moves to at regular intervals and inputs the downtown information received from the regional information database 210 into each divided area.

In step S4, the route generation module 220 divides the urban areas at regular intervals, allocates coordinates of the divided regions in the situation where the downtown information is input to each region, and proceeds to the initial path group generation process of the genetic algorithm. The initial path group generation process means that a random path is generated from a starting point to a target point at a starting point and a target point of a moving object to form a chromosome group consisting of n paths.

The fitness of each route is evaluated in a randomly generated route group from the starting point to the target point. The fitness evaluation function may be defined as in Equation 1. Here, | P _i -P _{i + 1} | denotes a moving distance of the path generated for the moving object from the i th to i + 1 th point. In addition, S denotes whether or not the corresponding area can be moved by a numerical value. If the numerical value is 1, the area becomes a movable area. And, α represents the average moving speed of the moving objects in the region and can derive the T _{p (i)? P (i + 1)} value through these values. That is, T _{p (i)? P (i + 1)} is the average moving time of the moving object from the i th to i + 1 th. Through T _{p (i)? P (i + 1)} , it is possible to derive the travel time for each section in the travel path. Finally, F represents the total travel time from the starting point of the travel route to the target point.

In step S6, the path generation module 220 performs fitness evaluation of the respective paths in the initial path group and the modified moving path group through Equation (1).

In step S8, the path generation module 220 indicates the suitability of the moving path as a numerical value and selects a relatively excellent moving path through the evaluation of the fitness of the paths in the moving path group. If the total travel distance and total travel time of the selected travel path satisfy the stop condition, the optimal travel path is selected. If not, the travel path is changed by selecting, crossing, and mutating the genetic algorithm. In other words, if a relatively good movement route is selected according to the fitness evaluation, but it is not satisfied under the stop condition, the area search and the route are changed through genetic algorithm selection, crossover, and mutation process.

In step S10, the path generation module 220 selects the movement path through the selection process of the genetic algorithm. In genetic algorithms, the selection process is the process of selecting some pathways from a population of migration paths, either stoichiometrically or effec- tively for mating and mutation processes. Here, the method proceeds to a method of selecting a moving path having a relatively good suitability in order to select the optimum moving path. The movement route is a method of selecting a relatively efficient route, and a tournament selection method is selected among the selection methods. The tournament selection method is to select one excellent moving path by comparing two paths in a path group consisting of n moving paths. Through the tournament selection process shown in FIG. 3, two paths may be finally selected by comparing n movement paths.

In step S12, the path generation module 220 selects a moving path through a crossing process of the genetic algorithm. In the genetic algorithm, the breeding process is applied to the two paths selected through the selection process, such as uniform breeding, one-point crossing, two-point crossing, multi-point crossing, and so on. How to proceed. In the two-point hybridization process shown in FIG. 4, each path is divided according to two intersections in two moving paths, and the mating is completed by exchanging partial paths from the first intersection point to the second intersection point.

In step S14, the path generation module 220 re-evaluates the fitness of the two paths changed through the crossing process, selects one migration path having excellent fitness in the two paths, and proceeds with the mutation process. The mutation process is applied to one path that has been altered through crosses, and randomly generates random paths to the target points by selecting random points from the movement path. If a path that is repeated and a path that is repeated in the randomly generated process is generated, the movement path is generated by removing it.

The path generation module 220 transmits the optimal movement path generated through the gene algorithm application module 222 to the viewer generator 300.

viewer Generator (300)

The viewer generating unit 300 serves to provide a user with a moving path situation in a map of a downtown area through an optimum moving path of the moving object received from the path generating unit 200. The internal configuration of the viewer generator 300 for performing the above-described functions includes a path management module 310 and an optimal movement path viewer module 320.

The route management module 310 updates the corresponding region in the downtown area through the optimal movement route received from the route generation module 220 and transmits the corresponding region to the optimal movement route viewer module 320.

The optimal movement route viewer module 320 displays the optimal movement route of the moving object on the map of the downtown area according to the user's request.

As described above, according to embodiments of the present invention, by using a genetic algorithm considering various information such as building location, road information, and traffic congestion in a downtown area, it is possible to search for a moving path of a moving object and generate an optimal moving path. have. In the path generation process through genetic algorithm, the initial path group is generated by using the urban area data divided into a certain area, and the fitness of each generated path is evaluated. Depending on the suitability of the pathways, selection, crossover, and mutation methods can be used to explore the various regions and modify the pathways to create optimal pathways based on stationary conditions.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

100: information input unit
110: information management module
120: urban area information acquisition module
130: real time road situation information acquisition module
200: path generation unit
210: Places database
220: path generation module
222: gene algorithm application module

Claims (5)

In the optimal path generation system for generating a path to the moving area of the moving object,
An information input unit for providing urban area information including location information of roads and buildings located in the moving area and real-time road situation information indicating an average moving speed of each road in the moving area; And
A route generation unit for generating a movement route in the moving area through a genetic algorithm using the city area information and the average moving speed.
Including,
The path-
A regional information database for maintaining the downtown area information and the average moving speed provided by the information input unit; And
After generating an initial path group including a plurality of paths that can be moved within the moving area by using the urban area information and the average moving speed, the fitness is evaluated for each path through the fitness evaluation of the genetic algorithm. Path generation module for selecting one path from the initial path group according to
/ RTI >
The path generation module,
Selecting two paths from the initial path group in a tournament manner in which one of the plurality of paths is compared and the one of the plurality of paths is selected.
The two paths are changed through a two-point crossing method in which sections from one intersection point to another intersection point are exchanged with respect to the two paths.
Reassessing the goodness of fit of the two changed paths to select a path having the high goodness of fit as the movement path
Optimal path generation system of a moving object, characterized in that. The method of claim 1,
The information input unit,
An urban area information acquisition module for obtaining location information of roads and buildings in the moving area by interworking with a map providing server for maintaining and managing map related data; And
Real-time road situation information acquisition module for acquiring an average moving speed of moving objects on each road in the moving area by linking with a road information providing server that maintains and manages road situation information for each road
Optimal path generation system of a moving object comprising a. delete delete The method of claim 1,
The path generation module,
Dividing the moving area into intervals of a predetermined interval to generate the initial route group by combining the intervals, and then evaluating the fitness using the movement time of each interval.
Optimal path generation system of a moving object, characterized in that.

Images