CN111437603B

CN111437603B - Method and system for generating map of electronic game

Info

Publication number: CN111437603B
Application number: CN202010238778.1A
Authority: CN
Inventors: 陈瑽; 寇京博; 任子敬; 覃哲林; 汪玉龙; 田吉亮; 庄涛; 杨凯允; 陈嘉伟; 殷宏亮; 张峰; 姚逸宁; 徐丹
Original assignee: Beijing Perfect Chijin Technology Co ltd
Current assignee: Beijing Perfect Chijin Technology Co ltd
Priority date: 2020-03-30
Filing date: 2020-03-30
Publication date: 2024-06-21
Anticipated expiration: 2040-03-30
Also published as: CN111437603A

Abstract

The invention provides a method and a system for generating a map of an electronic game, wherein the map comprises a common land block and a connecting land block. The method for generating the map of the electronic game comprises the steps of obtaining configuration information of the map to be generated, wherein the configuration information comprises the types and the numbers of common land parcels required by the generation of the map; according to the configuration information, randomly placing the number of common plots, and then acquiring the position information of each common plot in the number of common plots; determining a connection relationship between each of the number of common plots and other common plots; and connecting any two common plots having a connection relationship among the number of common plots using the connection plots to generate a map, wherein the connection plots include a level road plot and a stair plot.

Description

Method and system for generating map of electronic game

Technical Field

The present invention relates to computer game technology, and more particularly, to a method and system for generating a map for an electronic game.

Background

A game map is an important element that enables players to merge into an electronic game, which refers to a game area that is exclusive to a single or multiple users. In electronic games such as massively multiplayer online role-playing games (Massive Multiplayer Online Role-PLAYING GAME, MMORPG), players are often expected to be able to try more diverse game maps and more diverse game plays.

However, in the non-2D game, the complexity of the map is very high, and the simple stitching of the map cannot be performed as in the 2D game, but a plurality of factors such as the stitching degree of the map/model, the real-time light-shadow relation, the display effect or the operation efficiency need to be considered, so that when a large number of maps needed by the random map are manufactured, each map model needs to be redesigned and manufactured, which is time-consuming and labor-consuming, and brings great challenges to the artistic staff. In addition, how to improve game playability while randomizing play is also an important factor to consider when developing electronic games. In the current field, generation of random play is very blind, not taking into account the game experience of the player.

Methods and systems for generating maps for electronic games that comprehensively consider one or more of the above factors are lacking.

Disclosure of Invention

The invention provides a generation method of a map of an electronic game, wherein the map comprises a common land parcel and a connection land parcel, and the method comprises the steps of obtaining configuration information of the map to be generated, wherein the configuration information comprises the type and the number of the common land parcel required for generating the map; according to the configuration information, randomly placing the number of common plots, and then acquiring the position information of each common plot in the number of common plots; determining a connection relationship between each of the number of common plots and other common plots; and connecting any two common plots having the connection relationship among the number of common plots using a connection plot to generate the map, wherein the connection plot includes a level road plot and a stair plot.

The present invention also provides a system for generating a map of an electronic game, wherein the map comprises a common parcel and a connection parcel, and the system comprises means for obtaining configuration information of the map to be generated, wherein the configuration information comprises the type and number of the common parcel required for generating the map; means for randomly placing the number of ordinary plots according to the configuration information, and then acquiring position information of each of the number of ordinary plots; means for determining a connection relationship between each of the number of common plots and the other common plots; and means for connecting any two common plots of the number of common plots having the connection relationship using a connection plot to generate the map, wherein the connection plot includes a level road plot and a stair plot.

The present invention also provides a system comprising a processor and a memory having stored therein at least one instruction, at least one program, code set or instruction set that when loaded and executed by the processor is capable of carrying out the method described above.

The present invention also provides a computer readable medium having stored thereon at least one instruction, at least one program, code set or instruction set which when loaded and executed by a processor, enables the method described above to be carried out.

The method and the system for generating the map of the electronic game can generate the game map and/or the corresponding playing method under the condition that the display effect and the running efficiency are not affected and the randomness and the playability are ensured, so that the game experience of a player is improved. In particular, prior to implementing the method and system for generating a map of an electronic game of the present invention, corresponding art specifications and play have been designed by designers for different types of common plots in the map of the present invention. The method of the invention realizes random generation of the map by random of the position and the connection relation of the common land block of the preset type, and improves the manufacturing efficiency of the map of the electronic game. In addition, the method and system for generating a map of an electronic game of the present invention also correspond the randomly generated map to randomly generated play. The map generated above is related to a preset random playing method through the type of the common land block in the map, so that the consistency of the map and the playing method is considered while the diversity of the playing method is ensured, the blindness of the random playing method is avoided, and the game experience of a player in an electronic game is improved. Further, the method and system for generating a map of an electronic game of the present invention can also adjust the above play in real time through machine learning to provide a player with superior playability.

Drawings

FIG. 1 is a schematic diagram of the logical location of a general plot of land according to one embodiment of the present invention.

Fig. 2A and 2B are schematic diagrams of logic positions of gates of a normal block of 1×1 logic units and 2×2 logic units, respectively, according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of an art map and a logic map according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of an initial general plot of land according to one embodiment of the present invention.

FIG. 5 is a flow chart of generating a game map and play in accordance with one embodiment of the present invention.

Fig. 6 is a flow chart of generating a map according to one embodiment of the invention.

FIG. 7 is a flow chart of real-time adjustment play in accordance with one embodiment of the present invention.

Detailed Description

The present disclosure will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are merely illustrative to enable those of ordinary skill in the art to better understand and thus practice the present invention and are not meant to limit the scope of the present invention in any way.

As used herein, the term "comprising" and variants thereof should be interpreted to mean "including but not limited to" open-ended terms. The term "based on" should be read as "based at least in part on". The terms "one embodiment" and "an embodiment" should be read as "at least one embodiment. The term "another embodiment" should be read as "at least one other embodiment.

As is well known in the art, a "2D game" expresses all game picture effects through a plane, and a "3D game" realizes three-dimensional coordinate axes in a virtual space through technical means, and a character controlled by a player can change effect positions on the three coordinate axes. The differences between 2D games and 3D games include: (1) The art resources in the 2D game are mainly graphic files of PNG or JPG, and the art resources in the 3D game are mainly models; (2) The 2D game cannot complete the viewing angle conversion, and the 3D game can complete the viewing angle conversion; and (3) art resources in 2D games can be used without the rendering of the engine, while art resources in 3D games must be used with the rendering of the engine.

As is well known in the art, a "pseudo 3D game" is also known as a "2.5D game" and refers to a game that uses or achieves some concept or effect in a 3D game, but does not fully meet the 3D game standard. For example, the viewing angle of some 2.5D games cannot be switched.

As is known in the art, "manhattan distance" refers to the L1-distance or city block distance, i.e., the sum of the projection distances generated by the line segment formed by two points on a fixed rectangular coordinate system of euclidean space to the axis. For example, on a plane, the manhattan distance d (i, j) = |x1-x2|+|y1-y2| between the i-point of the coordinate (x 1, y 1) and the j-point of the coordinate (x 2, y 2).

As is well known in the art, "euclidean distance" refers to the true distance between two points in m-dimensional space or the natural length of the vector (i.e., the distance of the point from the origin). The euclidean distance in two and three dimensions is the actual distance between two points. For example in a two-dimensional space, Wherein ρ is the Euclidean distance between the point (X2, y 2) and the point (X1, y 1), and |X| is the Euclidean distance from the point (X2, y 2) to the origin; in three-dimensional space,/>Where ρ is the Euclidean distance between the point (X2, y2, z 2) and the point (X1, y1, z 1), and |X| is the Euclidean distance of the point (X2, y2, z 2) to the origin.

As is well known in the art, a "minimum spanning tree" refers to a spanning tree for G in the graph theory field if a subgraph of the connected graph G is a tree that contains all vertices of G. A spanning tree is a very small connected subgraph of a connected graph that contains all vertices in the graph. The spanning tree of the graph is not unique. Different spanning trees can be obtained by traversing from different vertices. For example, common spanning tree algorithms include DFS spanning tree, BFS spanning tree, PRIM minimum spanning tree, and Kruskal minimum spanning tree algorithm. For the connected "weighted graph (connected network)" G, its spanning tree is also weighted. The sum of the weights of the edges of the spanning tree T is referred to as the weight of the tree, denoted W (T) = Σ _(u,v)∈TE W (u, v), where TE represents the set of edges of T and W (u, v) represents the weight of the edge (u, v). The smallest weighted spanning tree is the Minimum Spanning Tree (MST) of G.

As is well known in the art, "multiple linear regression" refers to regression that includes two or more independent variables. For example, all the variables including the dependent variables can be converted into standard components and then subjected to linear regression, and the obtained regression coefficients can reflect the importance of the corresponding independent variables. The regression equation at this time is called a standard regression equation, and the regression coefficient is called a standard regression coefficient, which is expressed as Z _y＝β₁Z·1+β₂Z·2+…+β_k Z.k. Since both are converted into standard components, there is no constant term a, and since the dependent variable should be averaged when the respective variables are averaged, and the average level corresponds exactly to standard component 0, the constant term is 0 when the variables at both ends of the equation are all 0. The multiple linear regression can estimate model parameters by using a least square method, and also needs to carry out statistical test on the model and the model parameters. The selection of the proper independent variable is one of the preconditions for correctly performing multiple regression prediction, and the selection of the independent variable of the multiple regression model can be solved by using a correlation matrix among the variables. The general least squares method (Ordinary Least Square, OLS) finds the best function by minimizing the sum of squares of the errors, solves the coefficient matrix by matrix operations: and adding an L2 norm into a cost function (cost function) to solve the overfitting problem of the linear function content. The L2 norm refers to the sum of squares and the reopened square of the absolute values of the vector elements: /(I)

As is well known in the art, "Mean Square Error (MSE)" is a measure reflecting the degree of difference between an estimated quantity and an estimated quantity. Let t be an estimate of the overall parameter θ determined from the subsamples, (θ -t) ², the mathematical expectation, called the mean square error of the estimate t. It is equal to σ ²+b², where σ ² and b are the variance and bias of t, respectively. In general, an index used for evaluating the quality standard of one point estimate is always a point estimate at a time when the sample size is constantDistance function from parameter truth value θ, the most common function is the square of distance, due to the estimator/>Having randomness, the function can be expected, i.e. mean square error

As is well known in the art, a "linear rectification function", also known as a modified linear unit, is an activation function commonly used in artificial neural networks, generally referring to a nonlinear function represented by a ramp function and its variants. In general terms, the linear rectification function refers to a ramp function in mathematics, i.e., f (x) =max (0, x). In the neural network, the linear rectification function is used as an activation function of the neuron, and a nonlinear output result of the neuron after linear transformation is defined. In other words, for an input vector x from the upper layer neural network into a neuron, a neuron using a linear rectification activation function will output max (0,w ^T x+b) to the next layer of neurons or as the output of the entire neural network (depending on where the neuron is located in the network structure).

In an embodiment of the present invention, a "map" is made by arranging and splicing art resources according to a certain rule (e.g., position), also called an art map. In an embodiment of the present invention, art resources include, but are not limited to, maps and models, etc. In the embodiment of the present invention, the art map is represented as a building constituted by a plurality of rooms connected by hallways and/or stairs, etc., for example, the art standard thereof may be as follows: the minimum unit of a single room is 16 meters by 16 meters; the doors of the room are all at the edges of the room; the doors of the room are connected by using floating stone plates; the width of the door of the room is 6 meters, and the door is positioned at the center of the side length of 16 meters, namely, the left side and the right side are respectively free for 5 meters. Other art standards may also be used as desired in embodiments of the present invention. In the example of the present invention, prior to implementation of the scheme of the method and system for generating a map of an electronic game of the present invention, an art production is required in advance to support the above-described embodiments of the present invention. In the embodiment of the invention, the art making comprises making a splitting rule of an art map, namely splitting the map into a common land block and a connecting land block. In the embodiment of the invention, the common land block and the connecting land block are corresponding to art resources such as models, pictures and the like which are manufactured by art staff.

In embodiments of the present invention, a map, also referred to as a logical map, may also be defined using logical locations (logical coordinates). The map is made up of one or more base units in a two-dimensional plane. In the embodiment of the present invention, one basic unit (equivalent to one logical unit) may have an arbitrary shape and size, and preferably, one basic unit corresponds to a 16 m×16 m square of the art map. In the embodiment of the invention, two kinds of plots are shared in the logic map, namely a common plot and a connecting plot. In embodiments of the present invention, other plots may also be present in the logical map.

In the embodiment of the present invention, the general plot represents an area not connected by a path, which corresponds to a room in the art map. In the embodiment of the present invention, the size of a general plot is a multiple of one basic unit (i.e., is composed of N basic units, if 10×10 meters is one basic unit, one general plot may be 10×10×5, or may be 10×20×5, or 10×20×10, etc.). In the embodiment of the present invention, the basic unit corresponds to the minimum unit in the art map. In embodiments of the present invention, each common parcel may have one or more of walk-surface, trigger, and door elements. In an embodiment of the present invention, each common parcel necessarily has a walking surface and a door. In an embodiment of the invention, the connection block is configured to be able to connect common blocks through a door, e.g., the connection block connects common block a through door number 1 of common block a and door number 4 of common block B together.

In the embodiment of the invention, a certain angle of a common land block can be appointed as an origin (0, 0) of the common land block, and the common land block extends towards other directions, and positive directions and negative directions of x, y and z three axes are confirmed. The logical position of a common land parcel is defined in units of basic units, and one basic unit is moved in an axial forward direction, namely the next basic unit in the direction, and the logical position of the basic unit is the number +1 of the axis, and is conversely-1. For example, regardless of the height, the logical position is as shown in fig. 1 with the upper left corner of the ordinary plot as the origin position, the vertical direction as the y-axis, the downward direction as the positive direction, the lateral direction as the x-axis, and the rightward direction, with each point at the lower right of the origin representing one basic unit. Other forms of defining the base unit may also be used in embodiments of the invention.

In embodiments of the present invention, "walking surface" refers to an area of a common plot other than a trigger and a door; "trigger" refers to a monitoring and response mechanism for events that defines what events are monitored in a certain area in a common plot (e.g., entering, combat, moving, leaving, etc.), and when such events are monitored by the trigger, a predetermined response mechanism is triggered (e.g., generating an enemy, attaching an attribute to a player, triggering an organization, etc.); "door" refers to a access position that, when in an open position, can be moved from one normal parcel to the next through a connecting parcel. In an embodiment of the present invention, the "door" may correspond to a door, window, notch, etc. in an art map, as long as it can be used to indicate the location of a passageway from one general parcel to another. In an embodiment of the present invention, each area that may be logically a gate cannot be a gate for design reasons for each common parcel. It is necessary to define the position of the door that each general plot can open. In embodiments of the present invention that use logical positions to define the position of a gate, the defining direction is from the origin of the (0, 0) logical position of the normal plot. As fig. 2A and 2B show the logical positions of the gates of the ordinary plots of different embodiments, respectively, the ordinary plots of fig. 2A can open the gates from four directions, so that the gate positions are described as (0, -1), (-1, 0), (1, 0) and (0, 1) according to the previous definition; whereas the general plot of fig. 2B may have two doors in four directions each, the positions of the doors are described as (0, -1), (-1, 1), (1, 2) and (2, 0) in the case where only four of them are selected as the doors as needed.

In an embodiment of the invention, the connection block is configured to connect to a common block. In embodiments of the invention, the connection plots may include level road plots and stair plots. In an embodiment of the present invention, the level road plot corresponds to a corridor in the art map and the stair plot corresponds to a stair in the art map. In an embodiment of the invention, a stair parcel is a parcel having a difference in elevation from a normal parcel and a flat parcel. In this case, the z-coordinate needs to be introduced to represent the height, in addition to the x-and y-coordinates. In an embodiment of the present invention, to save resources, the logical position of the entrance of each stair plot is defined as altitude 0, the position of the exit is +1 above the entrance position, and-1 below the entrance position.

Fig. 3 shows a schematic diagram of an art map and a logical map according to an embodiment of the present invention. The right side of fig. 3 shows an art map to which art resources have been added, and the left side shows a logical map in the form of a two-dimensional plane. The art standard adopted by the map shown in fig. 3 is as follows: the minimum unit area of the map is 16 m×16 m, and the area of the map is 48 m×48 m, wherein the area of the platform a (room a) is 28 m×28 m, the area of the platform B (room B) is 14 m×14 m, and the area of the platform C (room C) is 28 m×14 m, and each room occupies a corresponding plurality of minimum units without exceeding the sum of the areas of the minimum units. In the two-dimensional map in fig. 3, the basic units refer to 9 areas represented by red lines, in which an a-land (a-ordinary land) occupies 2×2 basic units, a B-land (B-ordinary land) occupies 1×2 basic units, and a C-land (C-ordinary land) occupies 1×1 basic units.

In an embodiment of the present invention, there is also a special common plot, which is referred to as a starting common plot. Referring to fig. 4, a schematic diagram of a starting general plot of land is shown, according to one embodiment of the present invention. In an embodiment of the present invention, the starting common parcel is the first common parcel after the player enters the map, i.e., the start of the map. For example, the birth point of the player is set in the initial common land.

FIG. 5 illustrates a flow diagram for generating a game map and play according to one embodiment of the invention, wherein after triggering generation of the game map and/or play, configuration information for the map is first obtained, and then the map and play are generated based on the configuration information. In embodiments of the invention, the map and play are independent or non-independent elements of the electronic game. In embodiments of the present invention, reference to a "map" refers not only to a map of an electronic game, but also to play associated with the map. In an embodiment of the invention, the generation of the map is the generation of the map itself and the "play" associated with the map. In embodiments of the present invention, the methods of generating maps and playing methods of the present invention may be implemented independently.

Triggering generation of game maps and/or plays

In embodiments of the invention, the game map and/or play is triggered when one or more of the time, place, event, action, etc. requirements are met, and is in turn generated by the electronic game system, for example, starting to generate after the player triggers (e.g., clicks on) the game gate, preferably the electronic game system plays a segment of animation (e.g., a randomly stitched animation of the game map route) after the game map and/or play is triggered, and starts to generate the game map and/or play while playing the segment of animation.

Acquiring configuration information of map

In an embodiment of the present invention, the configuration information of the map includes, but is not limited to, the type and number (e.g., maximum or minimum) of common plots that may be generated in the map, the type, number and location of common plots that must be generated (e.g., appear in the middle or end of a path). In an embodiment of the present invention, the type of the ordinary parcel is configured to be able to determine the size of the ordinary parcel and the number and location of gates in the ordinary parcel. In an embodiment of the present invention, the type of the common parcel is configured to be able to determine an art resource corresponding to the common parcel. In an embodiment of the invention, the configuration information of the map further comprises the number and kind of play associated with the map, preferably the number and kind of play associated with a common parcel in the map, such as the number, kind, location, properties, generation scheme, etc. of enemies in the common parcel.

In an embodiment of the invention, the configuration information of the map includes one or more of the configuration information described above. In embodiments of the present invention, one or more of the configuration information of the map is random or preset (e.g., preset by a planner as needed). In an embodiment of the present invention, the configuration information of the map includes information for determining an art resource corresponding to a general block, which may be, for example, a type of the general block, or the like. In the embodiment of the invention, the common land parcels can be enabled to present corresponding artistic effects according to the types of the common land parcels. In an embodiment of the present invention, the configuration information further includes a type of the connection block. In the embodiment of the invention, the connecting land parcels can be enabled to present corresponding artistic effects according to the types of the connecting land parcels.

Generating a map from configuration information of the map

FIG. 6 illustrates a flow diagram for generating a map according to one embodiment of the invention.

Placing common plots at random locations in a map

In the embodiment of the invention, the common plots are placed in a way that the common plots are placed in sequence from the first common plot until the last common plot is also placed at a reasonable position in the map under the condition that the total of N common plots exist. In embodiments of the present invention, other methods of placing common plots may be used, such as randomly placing all common plots at once. In embodiments of the present invention, the placement of a common parcel is considered only for its location and not for its orientation. In embodiments of the present invention, not only the location but also the orientation of a common parcel need to be considered when it is placed. For example, in the case where the common parcel is not square and the door position is asymmetric, different orientations of the placed common parcel may result in different maps being generated.

In an embodiment of the present invention, the number of common plots that need to be generated in common in the map is first determined from the configuration information of the map obtained above. In an embodiment of the invention, the size of the map is related to the number of common plots, e.g., proportional. In an embodiment of the invention, the size of the map is represented by one or more base units, for example by N base units, where N is an integer greater than 0. The basic cell defined herein is a logical cell size regardless of the specifications of the art fabrication (e.g., 16 meters by 16 meters). In the embodiment of the present invention, the definition of the basic unit of the map is the same as the definition of the basic unit of the above ordinary land parcel. In an embodiment of the present invention in which the number of ordinary plots to be generated is M, the range of the map size may be determined by the value of M, for example, such that the size of the map as a square is between 5.5×log (M) to 6.5×log (M) to the power of the square, where the base of the Log function is 2. For example, when there are a total of 1000 common plots, 5.5×log (1000) is approximately equal to 55, the size of the last calculated map is 55 to the square, 3025 base units. In an embodiment of the present invention, rounding or rounding the value calculated by the above formula is the size of the map. The method can solve the problem that the size of the map is linearly increased due to the increase of the number of the common plots when the random map is formed by limiting the size of the map, so that the connecting path between the common plots becomes very long. If the number of ordinary plots is taken as X and the size of the map is taken as a Y construction function, the calculation method of the invention enables the increase amount of Y to be converged rather than linearly increased when X is increased. In embodiments of the present invention, the linear increase in map size due to the increase in the number of ordinary plots may also be limited by other methods. Other formulas or calculations may also be used to define the size of the map in embodiments of the invention.

In an embodiment of the present invention, if a random placement location for a common parcel is not available (i.e., there are other common parcel within the placement location or beyond a defined map boundary, referred to herein as a "placement failure"), then a location is re-randomized. In an embodiment of the invention, the number of re-randomizations is limited (there is a threshold), for example 20 times, and if no reasonable location has been determined within the threshold, the map needs to be increased in size and regenerated. In the embodiment of the present invention, the length and width of the map are each +1 basic units (logical units). In an embodiment of the present invention, the threshold refers to the number of single ordinary parcel placement failures or the cumulative number of current all ordinary parcel placement failures. In embodiments of the present invention, regenerating a map refers to repositioning all ordinary plots from the first ordinary plot or just repositioning the current ordinary plot.

In an embodiment of the present invention, if a starting common parcel is present in the map, the placement is started from the starting common parcel. In an embodiment of the present invention, if there is no starting common plot, one is randomly selected as the starting common plot. In the embodiment of the invention, the map configuration information may have a common land block which is necessarily generated, and when all the common land blocks are placed, the probability of using the common land block which is necessarily generated when the common land block is placed is: (all ordinary plots-ordinary plots generated)/number of necessary generations, the smaller the number the greater the probability of occurrence, the 1 indicates that this ordinary plot must be used this time.

Determining connection relation between common plots

In the embodiment of the invention, the connection relation between the ordinary plots is determined by taking Euclidean distance between the ordinary plots as a weight to acquire the spanning tree. In embodiments of the present invention, the connection relationship between ordinary plots may also be determined by other methods.

In the embodiment of the invention, after each common land parcels are reasonably placed in the map, the Euclidean distance between the common land parcels and other common land parcels is calculated from the initial common land parcels, and connection with the smallest distance is selected. Through the operation of this step, the connection relationship between all the ordinary plots is determined. In the embodiment of the invention, the Euclidean distance between two common plots can be calculated by setting two corresponding points between the two common plots according to the requirement. In the embodiment of the present invention, the point of the general plot for calculating the euclidean distance may be a center point thereof or any point of up, down, left, and right, preferably, a center point calculation is used.

In an embodiment of the present invention, the "in" and "out" of this common plot are considered in addition to the minimum Euclidean distance when obtaining the spanning tree. In an embodiment of the present invention, "incomes" are defined as: the number of edges connected to this node; "out-degree" is defined as: the number of edges from this nodule to the other nodules. In an embodiment of the invention, the definition of the nodules is the same as the definition of vertices in the production tree above. In an embodiment of the present invention, the number of edges of a nodule refers to the number of edges of a spanning tree. In the embodiment of the invention, the entering degree and the exiting degree of the common land parcel have the following requirements: the sum of the entrance degree and the exit degree of each common land is smaller than the number of the doors of the common land; the initial common land block only has the outgoing degree and the incoming degree; common plots as endpoints have only an in-degree and no out-degree. In the embodiment of the invention, after all the conditions are met, the Euclidean distance is selected as the weight value, and the connection is performed by the edge with the minimum weight value in the spanning tree. In an embodiment of the invention, each node in the spanning tree refers to a corresponding common plot.

Connecting common plots using connecting plots

In an embodiment of the present invention, a connecting block is used to connect the respective doors of two common blocks. In an embodiment of the invention, connections are made to connect plots according to Manhattan distances between gates of ordinary plots by traversing the spanning tree. In embodiments of the present invention, the common land parcels may also be connected using the connection land parcels by other methods.

In an embodiment of the present invention, path connection is made based on the spanning tree obtained above starting from the original ordinary plot. In an embodiment of the present invention, a first ordinary block (i.e., a target ordinary block) connected to a starting ordinary block (a starting ordinary block) is first selected, all unused gates in the starting ordinary block are traversed, two gates having the smallest Manhattan distance between the gate or gates and all unused gates of the target ordinary block are calculated, and then the two gates are connected. In the embodiment of the invention, after the starting common land parcels and the target common land parcels are connected, the target common land parcels are used as new starting common land parcels to be continuously connected. In the embodiment of the invention, if the doors selected for connection in the target common plot are already used in other paths, the doors selected for connection in the departure common plot are replaced first, one door with the minimum Manhattan distance is selected from the remaining doors in the departure common plot for connection, and if the doors cannot be connected, the remaining doors in the target common plot are replaced. In an embodiment of the present invention, if all doors in the departure common parcel and the destination common parcel are traversed without completing the path connection, the above step of "place common parcel at random position in the map" is returned, and the entire map is regenerated.

In an embodiment of the invention, after a complete path is generated, the path is recorded in a path list. In an embodiment of the invention, a level road plot and a stair plot are generated from the path using the inlet relative position and the outlet relative position, wherein:

portal relative position = position of previous common parcel-position of current common parcel;

exit relative position = position of the next ordinary plot-position of the current ordinary plot.

In the above embodiments of the present invention, the positions of the former, latter and current general plots refer to the positions of the doors for connection.

In an embodiment of the invention it is possible to determine which level road or stair parcels to use based on the relative positions of the entrance and exit, e.g. based on the difference in height between the two doors. In embodiments of the present invention, when there is no difference in height between the two doors, either a level road plot or a stair plot may be used instead. In an embodiment of the present invention, if a stair plot is used between two common plots, the height of the stair plot needs to be added to the spanning tree, ensuring that the nodes generated after the spanning tree are consistent in height. This is done because all the ordinary plots are generated after the gate and the stair plots, so the ordinary plots are at the same height on the logic map, if the path generated at this time determines to pass the stair plots, the heights of all the ordinary plots and the road plots and the stair plots after that need to be increased, and thus the heights are kept consistent.

Alternatively, in the embodiment of the present invention, corresponding art resources are attached to generate an art map according to the logical positions of the common land parcel and the connection land parcel determined above and the connection relationship (path) between each other. In the embodiment of the invention, the additional art resource refers to making the common land block present the art effect brought by the art resource. In the embodiment of the invention, the common land parcels and the connecting land parcels are already added with corresponding art resources. In an embodiment of the present invention, the art resources corresponding to the connection block are random or preset.

Generating playing method according to configuration information of map

In an embodiment of the invention, the map has play corresponding thereto. In an embodiment of the invention, each common parcel in the map has play corresponding to it. In an embodiment of the present invention, play corresponding to each common plot is randomly generated or preset by a planner. In the above-mentioned design of playing methods by the planner, the planner may design 1 to N (N is an integer greater than 1) sets of different playing methods for each common land according to the type, size, etc. of each common land, and each set of playing methods may include, but is not limited to, the number of enemies, the number or type of enemies per wave, the enemies configuration, the organization configuration, the appearance conditions, time, etc. In embodiments of the invention, there is a many-to-many relationship between play and plots. In the embodiment of the invention, 1 to N sets of playing methods generated by the above design are all first playing methods, and each of the 1 to N sets of playing methods is called a first sub-playing method.

In the embodiment of the invention, the common land parcels and the play list which can be used by the common land parcels can be output in the form of an array Tile and synchronized to the server from the front end. For example: { id, x, y, z, w, h, play = { }, { id, x, y, z, w, h, play = { }, a list of { id, x, y, z, w, h, play = { }, play list of random or planned common land parcels available in the map is included in play (id, x, y, z, w, h are each various attributes of the common land parcels). The play (play) may be zero or more. For example, a common land parcel a has play= { a, b, c, d … j } for a total of 10 plays, a play being configured by a planner: a= { { id=01.hp=11, att=12, pos= { x=1, y=1, z=10 } … … }, { id=02, hp=12, att=14, pos= { x=11, y=12, z=12 } } } … … }. In the embodiment of the invention, each time a player enters a common land block, one play method can be randomly selected from the play method list corresponding to the common land block for the player to play.

Machine learning configuration information for maps

In an embodiment of the present invention, in order to ensure playability of an electronic game while the play is random, configuration information of a map is specially designed by a planner. In an embodiment of the present invention, the configuration information designed by the planner may be machine-learned for subsequent prediction by the electronic game system of generating random configuration information based on certain specific information without degrading the playability of the electronic game.

In embodiments of the present invention, machine learning is performed using a variety of methods known in the art, preferably using a multiple linear regression algorithm. In the embodiment of the present invention, machine learning uses [ player information, environment information ] as the input vector X, and [ configuration information of map ] as the output vector Y. In embodiments of the present invention, player information includes, but is not limited to, battle, life, attack, defense, skill level, occupation, and the like; the context information includes, but is not limited to, map type, map difficulty, whether time is targeted, how much time is, whether points are targeted, how much points are, etc.; the configuration information of the map is defined as above, including, but not limited to, length, width, height of the map, maximum value of land (room), kind and number of ordinary land, play corresponding to ordinary land. In embodiments of the present invention, some or all of the above environmental information is random or designed by the planner. Other input and output vectors may also be used to train the machine learning model in embodiments of the present invention.

In the embodiment of the present invention, first, a planner configures an appropriate amount (for example, about 100 pieces) of data for machine learning according to the data items in the input vector. In embodiments of the invention, the training set, the test set, and the validation set may be set using a ratio of 6:2:2. Because of the predictability of the data, as long as the configuration of the boundary data is noted, the input data is regularized, and a linear w parameter vector is calculated by directly using a multiple linear regression algorithm, and then the vector can be given to a server for direct calculation.

Real-time adjusting playing method

Referring back to fig. 5, in some embodiments of the invention, the play that has been generated above may also be adjusted in real time. In the embodiment of the invention, the applicable playing method can be dynamically changed on the same map, and the map is considered not to be changed but is adjusted or updated. In an embodiment of the present invention, the original play is referred to as a first play and the adjusted play is referred to as a second play, relative to the adjusted play.

In the embodiment of the invention, the method for adjusting the playing method in real time uses a multiple linear regression algorithm as a basis, simultaneously uses a deep learning model, and adds some nonlinear transformation, so that the generated playing method is more abundant in change. In embodiments of the present invention, machine learning may also be performed using a variety of methods consistent with the art.

In an embodiment of the present invention, the machine learning model is built based on the input vector [ player information, environment information, play information ] and the output vector [ enemy information ]. In embodiments of the present invention, player information includes, but is not limited to, battle, life, attack, defense, skill level, occupation, and the like; environmental information including, but not limited to, current points, time remaining, level difficulty, rating score, etc.; the play information includes, but is not limited to, the current common parcel ID, the enemy total wave number, the enemy current wave number, the enemy total number, the current enemy rank, the enemy type, and the like. In embodiments of the present invention, enemy information includes, but is not limited to, enemy statement, attack, defense, type, whether to generate and evaluate scores, and the like.

In the embodiment of the invention, the input layer is a vector composed of N groups of player information, environment information and play information; the output layer corresponds to a current enemy adjusting scheme, which is a series of numerical adjusting coefficients including basic properties, number, position offset, time offset of appearance, equipment information, skill information and the like of the enemy. The machine learning model of the present invention also includes hidden layers (two to three hidden layer designs): first layer activate Function uses RELU. The second layer activate Function uses relu/sigmoid. In an embodiment of the invention, the machine learning model also involves COST Function, where the COST Function of the output layer takes MSE (mean square error), and for whether this is generated, a cross entropy Function, i.e. COST function= [ MSE, MSE, MSE, MSE, CEL, MSE ], is used. Since the learned data is mostly linear, to prevent the over-fitting problem, L2 parameters are added to the MSE. Algorithms for such multiple linear regression, loss functions, prevention of overfitting, etc., are known in the art. In embodiments of the present invention, specific designs are also contemplated, for example, where the evaluation score is a full map accumulated data, the data is initially manually filled in and tested by the planner, and when such an output is relatively difficult to input, a positive score is output, and conversely a negative score is generated, and whether to generate such an enemy may be determined based on the difficulty parameter. More specifically, the evaluation score is the number of attributes and states of the current player in a battle (such as a one-wave enemy), the difficulty parameter is the low difficulty, medium difficulty or high difficulty of the map, for example, a 5-level player encounters 5 10-level enemies on a high-difficulty map for the first time, the score of the battle is positive 3, the score of the low-difficulty player is 10, and thus the number or attributes of the generated enemies can be reduced in the next battle due to the excessively high score, and the parameter is a special parameter in the training set and possibly affects the attributes of the generated enemies.

In an embodiment of the invention, the machine learning process is as follows: the generated data is prepared by a planner (for example, a set of simple and visual linear tables (excels)) and is stored as learning data by checking the input of each common land block according to a developed special editor in the game process, adjusting the output according to the feeling of the planner and testing. For example, training may begin when there are around 1000 pieces of data, and then the model is saved.

FIG. 7 illustrates a flow chart of real-time adjustment play in accordance with one embodiment of the present invention. In the embodiment of the invention, the method for adjusting the playing method in real time can be implemented between different common plots of the map, and can also be implemented between enemies of different waves of a certain common plot. In the embodiment of the invention, the first playing method obtained after the player enters the common land block is completely randomly generated, can be randomly generated in the play corresponding to the common land block by the method, and can be randomly generated by a machine learning model for adjusting the playing method in real time. In embodiments of the present invention, the randomly generated play may be determined entirely when the map is generated, i.e., when the player enters the first common parcel, the play in all common parcel is determined. In the above case, however, the various conditions and states may have changed when the player enters the next normal land, and the play generated by the initial random or machine learning may no longer be appropriate, requiring adjustment based on the current conditions and states. In addition, various conditions and states may be different in different wave-times of a single common land parcel, where adjustments to already generated play are also required, e.g., where the configuration and attributes of enemies in the wave-times that have not yet occurred may be adjusted to match the player. In an embodiment of the invention, the generated second play may also be fed back to the machine learning model for further updating.

The methods and apparatus of embodiments of the present invention may be implemented as pure software modules (e.g., software programs written in Java language), as pure hardware modules (e.g., application specific ASIC chips or FPGA chips), or as modules combining software and hardware (e.g., firmware systems with fixed code stored). In an embodiment of the invention, the method of the invention is implemented using Lua, c++, python coordination.

Another aspect of the invention is a computer readable medium having stored thereon computer readable instructions which, when executed, may implement the methods of the embodiments of the invention.

Those of ordinary skill in the art will appreciate that the foregoing description is by way of example only, and is not intended to limit the invention so described. The present invention is also capable of numerous modifications and variations. Any modifications or variations which come within the spirit and scope of the invention are desired to be protected.

Claims

1. A method of making a map of an electronic game, wherein the map comprises a common plot and a connected plot, and the method comprises:

Acquiring configuration information of the map to be manufactured, wherein the configuration information comprises the types and the numbers of the common land parcels required for manufacturing the map;

Placing the number of common plots in a random position according to the configuration information, and then acquiring the position information of each common plot in the number of common plots, wherein before the number of common plots are placed in the random position, the size of the map is determined according to the number of common plots, and when one common plot in the number of common plots overlaps with other common plots or exceeds a boundary defined by the size of the map after being placed, the common plots which are not placed currently are replaced randomly;

Determining a connection relationship between each of the number of common plots and other common plots; and

Connecting any two common plots having the connection relationship among the number of common plots using a connection plot, wherein the connection plot includes a level road plot and a stair plot, and the connecting operation includes:

Connecting two doors with minimum Manhattan distance between unused doors of the departure common land parcels and the target common land parcels, and continuously connecting the target common land parcels as a new departure common land parcels;

If the doors selected for connection in the target common land parcels are used in other paths, replacing the doors selected for connection in the departure common land parcels, selecting one door with the minimum Manhattan distance from the remaining doors in the departure common land parcels for connection, and if none of the remaining doors in the departure common land parcels can be connected, replacing the remaining doors in the target common land parcels; and

And if all doors in the departure common land parcels and the target common land parcels cannot be connected, carrying out random placement operation again.

2. The method of claim 1, wherein the type of the common parcel is configured to enable determination of the size of the common parcel and the number and location information of gates in the common parcel.

3. The method as recited in claim 1, further comprising: after the number of common plots is randomly placed, orientation information of each common plot of the number of common plots is obtained.

4. A method according to claim 3, wherein the position information, orientation information and connection relation are represented by logical coordinates.

5. The method of claim 1, wherein the size of the map is represented by one or more base units.

6. The method of claim 1, wherein the random placement operation is a multiple of random placements and only one common parcel is randomly placed at a time.

7. The method of claim 1, wherein the map size is increased when the number of cumulative placement failures for the number of common plots or when the number of placement failures for one of the number of common plots exceeds a threshold; and

Wherein each of the number of common plots is replaced randomly or a common plot that fails to be placed currently is replaced randomly after the size of the map is increased.

8. The method of claim 1, wherein determining the connection relationship comprises obtaining a spanning tree with a minimum euclidean distance between each of the number of common plots and the other common plots as a weight.

9. The method of claim 8, wherein the operation of determining the connection relationship further comprises obtaining the spanning tree based on an ingress and egress of each of the number of common plots respectively as a plurality of nodes of the spanning tree.

10. The method of claim 8, wherein the method further comprises:

If a first common parcel of the number of common parcels is connected to a second common parcel by the stair parcel, adjusting a height of the second common parcel and connection parcel that are located further back in the spanning tree than the second common parcel, wherein the height is a height in the spanning tree.

11. The method according to any of claims 1-10, wherein the configuration information further comprises play information, and the method further comprises:

and determining a playing method corresponding to the map as the playing method of the map according to the configuration information.

12. The method of claim 11, wherein determining the play comprises determining a first play corresponding to each of the number of common plots in the map as the play of the common plot based on a type of each of the number of common plots.

13. The method of claim 12, wherein the first play comprises a plurality of first sub-plays, and wherein the operation of determining the first play comprises randomly selecting one of the plurality of first sub-plays as a play of the common parcel.

14. The method of claim 12, wherein determining the play comprises adjusting the play of the common plot from a first play to a second play, and wherein the second play is different from the first play.

15. The method of claim 14, wherein the second play is determined by a deep learning model.

16. The method of claim 15, wherein the deep learning model utilizes at least one of a multiple linear regression algorithm, a loss function, and a prevent over-fitting algorithm; and

The deep learning model is trained by taking real-time player information, environment information and playing information as input vectors and enemy information as output vectors.

17. The method of claim 14, wherein the act of adjusting the first play to the second play is performed when a player enters a new common parcel in the map; or (b)

When player information changes, an operation of adjusting the first play to the second play is performed.

18. The method according to any one of claims 12-17, wherein after determining the play of the common parcel, determining the play of the map according to the connection relation.

19. The method of claim 1, wherein the configuration information is pre-designed by a planner or generated by a machine learning model.

20. The method of claim 19, wherein when using a machine learning model, the machine learning model utilizes a multiple linear regression algorithm; and

The machine learning model is trained with player information and environment information as input vectors and configuration information of the map as output vectors.

21. A system for making a map of an electronic game, wherein the map comprises a common parcel and a connection parcel, and the system comprises:

means for obtaining configuration information of the map to be made, wherein the configuration information includes a type and a number of the ordinary plots required for making the map;

Means for placing the number of ordinary plots in a random position according to the configuration information, and then acquiring position information of each of the number of ordinary plots, wherein the size of the map is determined according to the number of ordinary plots before the number of ordinary plots is placed in the random position, and when one of the number of ordinary plots overlaps with other ordinary plots or exceeds a boundary defined by the size of the map after placement, the ordinary plots that are currently placed in failure are replaced randomly;

means for determining a connection relationship between each of the number of common plots and the other common plots; and

Means for connecting any two common plots of the number of common plots having the connection relationship using a connection plot, wherein the connection plot comprises a level road plot and a stair plot, and the connecting operation comprises:

22. A system comprising a processor and a memory having stored therein at least one instruction, at least one program, code set or instruction set that when loaded and executed by the processor is capable of implementing a method according to any of claims 1-20.

23. A computer readable medium having stored thereon at least one instruction, at least one program, code set or instruction set which when loaded and executed by a processor, enables the method according to any one of claims 1-20.