AGV conflict determination method and system

Abstract

The invention discloses an AGV conflict determination method and system, and the method is applied to an AGV scheduling method, and comprises the steps: building a two-dimensional grid diagram, and defining each node in the two-dimensional grid diagram to correspond to a unique coordinate; acquiring a starting coordinate point and a target coordinate point of each AGV; in a single scheduling process, all AGVs only complete a moving task from a starting coordinate point to a target coordinate point once; calculating a common activity area of any two AGVs in the scheduling process; classifying the types of path conflict judgment according to the number of the advancing directions of the two AGVs; and judging that the conflict relation of the two AGVs is hostile, compatible or free based on the classification result and the common activity area of the two AGVs. According to the determination method provided by the invention, the AGV path conflict determination problem is solved through an elementary mathematical tool.

Description

Technical Field

[0001] The present invention relates to the field of automatic guided vehicle (AGV) path conflict detection, and in particular to a conflict determination method and system for AGVs. Background Art

[0002] In the process of automation in industrial production and logistics, the application of automated guided vehicles (AGVs) has become increasingly widespread, becoming a key device for improving transportation efficiency and intelligent level. In a transportation system with multi-AGV collaborative operations, planning an effective, safe, and fast travel path for each AGV has become the core issue for ensuring the efficient operation of the system. This issue is known as the multi-agent pathfinding problem (MAPF) in the academic and engineering fields, and its goal is to ensure that the paths of all AGVs from their respective starting points to their destinations do not collide with each other, while achieving the shortest travel time.

[0003] The MAPF problem has important applications in scenarios such as automated warehouses, intelligent transportation, and intelligent parking. In an automated warehouse, numerous AGVs need to quickly transport goods in a limited space, and reasonable path planning can improve the utilization rate of storage space and the turnover efficiency of goods; in the field of intelligent transportation, as a new type of transportation tool, the quality of path planning of AGVs directly affects traffic fluency and safety; in an intelligent parking system, AGVs need to accurately plan paths to achieve efficient vehicle parking and retrieval.

[0004] However, with the continuous increase in the number of AGVs and the growth of travel distances, the MAPF problem faces severe challenges. The search path space of this problem will increase exponentially. Even if it is restricted to study on a two-dimensional grid graph, it still belongs to an NP-hard problem. This means that as the problem scale expands, the computing resources and time required for solving will increase sharply, and traditional computing methods are difficult to find the optimal solution within a reasonable time, or even unable to find a feasible solution.

[0005] Among the numerous algorithms for solving the MAPF problem, the conflict - based search strategy (CBS) and its derivative algorithms play an important role. The CBS strategy greedily plans the shortest path that satisfies the constraints for each AGV and continuously resolves conflicts between AGVs in the solution space to achieve the purpose of overall path planning. However, in the existing technology, there is a lack of a complete method for judging and identifying the path conflict relationship between two AGVs, which has become the key bottleneck restricting the further optimization of the CBS algorithm and related path planning strategies. But in actual application scenarios, due to the lack of a method for accurately judging the AGV path conflict relationship, it may lead to frequent collisions, congestion, etc. during the driving process of AGVs, reducing the efficiency and stability of the transportation system. For example, in an automated warehouse, if the path conflict of AGVs cannot be accurately judged, multiple AGVs may drive towards the same area simultaneously, causing traffic jams and prolonging the cargo handling time; in an intelligent parking system, AGVs may not be able to find parking spaces in time due to path conflicts, affecting the user experience. This will not only increase the enterprise operation cost but also may affect the smooth progress of the entire production or service process.

[0006] In summary, the existing multi - AGV transportation system has significant deficiencies in path planning, especially in the link of AGV path conflict judgment, and urgently needs an efficient and accurate judgment method to solve these problems to promote the further development of related fields. Summary of the Invention

[0007] In view of the above - mentioned deficiencies in the current AGV path conflict determination technology, the present invention provides an AGV conflict determination method and system, which can fully determine the path conflict relationship between any two AGVs in a two - dimensional grid map.

[0008] To achieve the above - mentioned purpose, the first aspect of the present invention provides an AGV conflict determination method, including:

[0009] Establish a two - dimensional grid map and define that each node in the two - dimensional grid map corresponds to a unique coordinate;

[0010] Obtain the starting coordinate point and the target coordinate point of each AGV;

[0011] During a single scheduling process, all AGVs only complete one movement task from the starting coordinate point to the target coordinate point;

[0012] Calculate the common activity area of any two AGVs during this scheduling process;

[0013] Classify the type of path conflict determination of these two AGVs according to the number of their traveling directions;

[0014] Based on the classification result and the common activity area of two AGVs, determine the conflict relationship between the two AGVs as hostile, compatible or free.

[0015] In some embodiments according to the first aspect of the present invention, each AGV moves from the starting coordinate point to the target coordinate point according to the Manhattan distance and cannot stop before reaching the target coordinate point. When the AGV reaches the target coordinate point, it stops at this point until all AGVs reach their respective target coordinate points.

[0016] In some embodiments according to the first aspect of the present invention, calculating the common activity area of any two AGVs in the current scheduling process includes: determining the area formed by the starting coordinate points and target coordinate points of the two AGVs on the two-dimensional grid map, and the common activity area of the two AGVs is the intersection of the activity areas of the two AGVs on the two-dimensional grid map.

[0017] In some embodiments according to the first aspect of the present invention, classifying the type of path conflict determination according to the number of moving directions of the two AGVs includes: calculating the sum of the absolute values of the moving direction values of the two AGVs along the X-axis and Y-axis on the two-dimensional grid map, and the moving direction values include 1, -1, and 0, which respectively represent that the AGV moves along the positive direction of the axis, the negative direction of the axis, and no axial movement.

[0018] In some embodiments according to the first aspect of the present invention, determining the conflict relationship between the two AGVs as hostile, compatible or free based on the classification result and the common activity area of the two AGVs includes: according to the states of the two AGVs, dividing the determination process into a first stage and a second stage. The first stage is the time period when the two AGVs move together, and the second stage is the time period when one AGV reaches the target coordinate point and the other AGV continues to move until it reaches the target coordinate point.

[0019] In some embodiments according to the first aspect of the present invention, determining the conflict relationship between the two AGVs as hostile, compatible or free based on the classification result and the common activity area of the two AGVs includes: if the first stage or the second stage is a hostile relationship, then determine the overall as a hostile relationship; if neither the first stage nor the second stage is a hostile relationship and there is a compatible relationship, then determine the overall as a compatible relationship; if both the first stage and the second stage are free relationships, then determine the overall as a free relationship.

[0020] In some embodiments according to the first aspect of the present invention, determining the conflict relationship between the two AGVs as hostile, compatible or free based on the classification result and the common activity area of the two AGVs includes: if the two AGVs have no common activity area, then determine the conflict relationship between the two AGVs as free; if the two AGVs have a common activity area, then make a further determination according to the number of moving directions of the two AGVs.

[0021] In some embodiments according to the first aspect of the present invention, the further determination based on the number of travel directions of two AGVs includes: when the number of travel directions of two AGVs is 2, the determination rule is:

[0022] When the travel directions of two AGVs are the same, the determination process includes:

[0023] In the first stage, the two AGVs are in a free relationship;

[0024] In the second stage, if one AGV reaches its target coordinate point first and the target coordinate point blocks the only path of the other AGV, it is a hostile relationship; otherwise, it is a free relationship;

[0025] When the travel directions of two AGVs are opposite, it is a hostile relationship;

[0026] When the travel directions of two AGVs are perpendicular, the determination process includes:

[0027] In the first stage, if the two AGVs enter the only intersection point at the same time, it is a hostile relationship; otherwise, it is a free relationship;

[0028] In the second stage, if one AGV reaches its target coordinate point first and the target coordinate point blocks the only path of the other AGV, it is a hostile relationship; otherwise, it is a free relationship.

[0029] In some embodiments according to the first aspect of the present invention, for the AGV conflict determination method according to claim 5, the further determination based on the number of travel directions of two AGVs includes: when the number of travel directions of two AGVs is 3, the determination rule is:

[0030] First scenario, when the first AGV moves along the positive X-axis and positive Y-axis while the second AGV only moves along the positive X-axis, the determination process includes:

[0031] In the first stage, when the sum of the coordinate values of the starting coordinate points of the two AGVs is not equal, it is a free relationship; when the sum of the coordinate values of the starting coordinate points of the two AGVs is equal and the activity area of one AGV penetrates the activity area of the other AGV, it is a hostile relationship; otherwise, it is a compatible relationship;

[0032] In the second stage, when the second AGV reaches the target coordinate point first, if the target coordinate point of the second AGV is in the common activity area of the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the first AGV and that of the second AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are compatible; otherwise, they are free. When the first AGV reaches the target coordinate point first, if the target coordinate point of the first AGV is in the common activity area of the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the second AGV and that of the first AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are in a hostile relationship; otherwise, they are in a free relationship.

[0033] In the second scenario, when the first AGV moves along the positive X-axis and positive Y-axis while the second AGV only moves along the negative X-axis, the determination process includes:

[0034] In the first stage, when there is a coordinate point in the common activity area of the two AGVs and the Manhattan distances from this coordinate point to the two AGVs are equal, the two AGVs are in a compatible relationship. When there are two adjacent coordinate points on the X-axis in the common activity area of the two AGVs, and the sum of the X coordinate value and Y coordinate value of the starting coordinate point of the first AGV plus the difference between the X coordinate value and Y coordinate value of the starting coordinate point of the second AGV is equal to the sum of the X coordinate values of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0035] In the second stage, the determination method is the same as that in the second stage of the first scenario.

[0036] In the third scenario, when one AGV moves along the positive X-axis and positive Y-axis while the other AGV only moves along the positive Y-axis, the determination method is the same as that in the first scenario.

[0037] In the fourth scenario, when one AGV moves along the positive X-axis and positive Y-axis while the other AGV only moves along the negative Y-axis, the determination method is the same as that in the second scenario.

[0038] In some embodiments according to the first aspect of the present invention, the further determination based on the number of traveling directions of the two AGVs includes: If the number of traveling directions of the two AGVs is 4, the determination rule is:

[0039] In the first scenario, when both AGVs move along the positive X-axis and positive Y-axis, the determination process includes:

[0040] In the first stage, when the sum of the coordinate values of the starting coordinate points of the two AGVs is not equal, the two AGVs are in a free relationship; when the sum of the coordinate values of the starting coordinate points of the two AGVs is equal and the active areas of the two AGVs penetrate each other, the two AGVs are in a hostile relationship; otherwise, they are in a compatible relationship.

[0041] In the second stage, when the target coordinate point of the first-arriving AGV is located in the common active area of the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the later-arriving AGV and the Manhattan distance from the starting coordinate point to the target coordinate point of the first-arriving AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are in a compatible relationship; otherwise, they are in a free relationship.

[0042] Second scenario, when the first AGV moves along the positive X-axis and positive Y-axis while the second AGV moves along the positive X-axis and negative Y-axis, the determination process includes:

[0043] In the first stage, when there is a coordinate point in the common active area of the two AGVs such that the Manhattan distances from the starting coordinate points of the first AGV and the second AGV to the coordinate point are equal, the two AGVs are compatible; when there are two adjacent coordinate points on the Y-axis in the common active area of the two AGVs and the sum of the X coordinate value and Y coordinate value of the starting coordinate point of the first AGV plus the difference between the Y coordinate value and X coordinate value of the starting coordinate point of the second AGV is equal to the sum of the Y coordinate values of the two adjacent coordinate points, the two AGVs are compatible; otherwise, the two AGVs are free.

[0044] In the second stage, the determination method is the same as the determination method in the second stage of the first scenario.

[0045] Third scenario, when the first AGV moves along the positive X-axis and positive Y-axis while the second AGV moves along the negative X-axis and positive Y-axis, the determination method is the same as the determination method of the second scenario.

[0046] Fourth scenario, when the first AGV moves along the positive X-axis and positive Y-axis while the second AGV moves along the negative X-axis and negative Y-axis, the determination process includes:

[0047] In the first stage, when there is a coordinate point in the common active area of the two AGVs and the Manhattan distances from the coordinate point to the starting coordinate points of the two AGVs are both equal, the two AGVs are in a compatible relationship; when there are two adjacent coordinate points in the common active area of the two AGVs and the sum of the coordinate values of the starting coordinate point of the first AGV and the coordinate values of the starting coordinate point of the second AGV is equal to the sum of the coordinate values of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0048] In the second stage, the determination method is the same as that in the second stage of the first scenario.

[0049] In some embodiments according to the first aspect of the present invention, the adversarial relationship between the two AGVs includes point conflict and edge conflict. The point conflict means that two AGVs reach the same position at the same time, and the edge conflict means that two AGVs exchange their positions at adjacent times.

[0050] To achieve the above object, the second aspect of the present invention provides an AGV conflict determination system, including:

[0051] A network construction module, configured to establish a two-dimensional grid map and define that each node in the two-dimensional grid map corresponds to a unique coordinate;

[0052] A coordinate point acquisition template, configured to acquire the starting coordinate point and the target coordinate point of each AGV;

[0053] A task scheduling module, configured to execute the scheduling task of moving each AGV from the starting coordinate point to the target coordinate point;

[0054] A conflict type classification module, configured to classify the conflict types of two AGVs based on the number of traveling directions of the two AGVs;

[0055] A conflict determination module, configured to determine the conflict relationship between two AGVs based on the common activity area and conflict types of the two AGVs.

[0056] Advantages of the implementation of the present invention: First, the present invention takes the starting coordinate point and the target coordinate point of the AGV as inputs and uses elementary mathematical tools to determine the conflict type, reducing the computing resource requirements and improving the real-time performance of the system. Second, by enhancing the algorithm performance, it clarifies the priority and direction of conflict resolution for the CBS solution strategy of the MAPF problem and can find a globally optimal solution faster during the high-conflict path planning process. Third, by providing an accurate conflict determination basis, it can plan a reasonable path for the AGV in advance, reducing the waiting and avoidance time of the AGV and thus improving the cargo handling efficiency and the utilization rate of the storage space. Finally, the present invention is applicable to multi-AGV systems of different scales and complexities, has strong versatility and can meet the diverse scenario requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0058] Figure 1Schematic diagram of the process of an AGV conflict determination method according to the present invention. Specific embodiments

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] Before further elaborating on the present invention, the nouns and terms involved in the embodiments of the present invention are described. The nouns and terms involved in the embodiments of the present invention are applicable to the following explanations:

[0061] AGV: Automated Guided Vehicle is a mobile robot that can automatically travel along a predetermined route and carry goods without manual driving, and is commonly used in scenarios such as warehouses, factories, hospitals, ports, etc.

[0062] Manhattan distance: The Manhattan distance is a metric for measuring the distance between two points in a geometric space. It is named after the Manhattan borough of New York City because the street layout in this area is similar to a grid, and the shortest path between two points is usually to walk along these vertical and horizontal streets. In a two-dimensional plane, assuming we have two points P 1 (x 1 , y 1 ) and P 2 (x 2 , y 2 ), then the formula for calculating their Manhattan distance d is d = |x 1 - x 2 | + |y 1 - y 2 |, where |x 1 - x 2 | represents the distance between the two points on the x-axis, and |y 1 - y 2 | identifies the distance between the two points on the y-axis; for two points in a multi-dimensional space, such as for two points P 1 (x 11 , x 12 ,... x 1n ) and P 2 (x 21 , x 22 ,... x 2n ), their Manhattan distance is the sum of the absolute values of the differences in each dimension, and the mathematical formula is:

[0063]

[0064] Figure 1 Disclosed is a method for determining AGV conflicts in an embodiment of the present invention. This conflict determination method is applied to an AGV scheduling method and includes the following processes:

[0065] Step S1: Establish a two-dimensional grid map and define that each node in the two-dimensional grid map corresponds to a unique coordinate.

[0066] In an embodiment of the present invention, first, a two-dimensional grid map based on Manhattan distance needs to be created. In this two-dimensional grid map, each grid (node) is connected to adjacent grids in four directions: up, down, left, and right, and the distance between them is 1. This connection method uses Manhattan distance. At the same time, the coordinate values of each node are all integers.

[0067] Step S2: Obtain the starting coordinate point and the target coordinate point of each AGV.

[0068] In the implementation of the present invention, to avoid conflicts between the starting coordinate points and target coordinate points of AGVs and simplify the discussion of the problem, the starting coordinate points and target coordinate points of each AGV are all different.

[0069] Step S3: Set that in a single scheduling process, all AGVs only complete one movement task from the starting coordinate point to the target coordinate point.

[0070] Here, each AGV moves from its starting point to its target coordinate point according to the Manhattan distance, and the Manhattan distance has been explained above. According to the rules of Manhattan distance, from the decision-making space of the AGV, there are a total of 5 actions of the AGV on the map at each moment, that is , so the optional actions of each AGV before reaching the target coordinate point are affected by the starting and ending points. Specifically, if the target coordinate point is in the lower right of the starting coordinate point, according to the limitation of Manhattan distance, the optional actions of the AGV at each step can only be to the right or down, and cannot be to the left or up. For example, if the starting point of an AGV is (0,0) and the ending point is (5,6), since it must take the shortest path, before reaching the ending point, the AGV needs to execute 5 times , 6 times ↑ actions. After reaching the ending point, the AGV stays at the ending point, so the only action is stop.

[0071] To define the premise for discussing the AGV conflict determination method, in this embodiment, each AGV cannot stay at the coordinate point before reaching its target coordinate point, that is, each AGV has no stop action before reaching the target coordinate point. In a single scheduling task, when the AGV reaches its target coordinate point, it stops at this point until all AGVs reach their respective target coordinate points.

[0072] Step S4: Calculate the common activity area of any two AGVs during this scheduling process.

[0073] For the convenience of using data formulas to explain the problem in detail, the two AGVs here are respectively denoted as and . The starting coordinate point of is ; The starting coordinate point of is . It can be known from the above conditions that and .

[0074] From the structure of the two-dimensional grid map and the Manhattan distance walking rule, it can be known that The activity area of is the rectangular area formed from its starting coordinate point to its target coordinate point The activity area of is the rectangular area formed from its starting coordinate point to its target coordinate point and In this embodiment, and The common activity area of and is the intersection of on the two-dimensional grid map and is denoted as

[0075] The mathematical expressions are respectively:

[0076] ;

[0077] ;

[0078] .

[0079] Step S5: Classify the types of path conflict determination according to the number of traveling directions of these two AGVs.

[0080] Specifically, The traveling direction of along the x-axis is denoted as ; the traveling direction along the y-axis is denoted as The traveling direction of along the x-axis is denoted as .

[0081] Among them, the sign function is used to return the sign of a numerical value. It returns 1 for a positive number, 0 for zero, and -1 for a negative number. The rule is as follows: when the input numerical value is positive, it returns 1; when the input numerical value is zero, it returns 0; when the input numerical value is negative, it returns -1. Thus, it can be seen that taking as an example, the values 1, -1, and 0 represent moving along the axis in the positive direction, axis in the negative direction, and no axis direction movement.

[0082] For the convenience of discussion, the number of moving directions of and is represented by the parameter B. The mathematical expression is . Because for , and take values of 0 or 1, that is, , so . According to the structure discussed above, it can be known that and The possible values of the number of moving directions B of

[0083] on the two-dimensional grid graph are 2, 3, or 4, that is, B ∈ {2, 3, 4}.

[0084] Step S6: Based on the classification result and the common activity area of the two AGVs, judge the conflict relationship between the two AGVs. and Since the Manhattan distances of and on the two-dimensional grid graph may be different, starting from the same moment, The time from its starting coordinate point to its target coordinate point is denoted as , and The time from its starting coordinate point to its target coordinate point is denoted as . Because and can only move one grid on the two-dimensional grid graph per unit time, so and can be respectively expressed as and The Manhattan distances from their respective starting coordinate points to their respective target coordinate points, that is, The movement duration of is The movement duration of , where represents the Manhattan distance between two nodes on the two-dimensional grid graph.

[0085] Because the movement durations of the two AGVs may be different, within the entire cycle, and the path conflict determination for and can be divided into two stages, namely the first stage and the second stage. In the first stage, and are both in the moving state. In the second stage, i, T j} and i, T j}, one of them has reached the target coordinate point and is in the stop state after reaching the target coordinate point. That is, the first stage is from time 0 to min{T i, T j}, and the second stage is from the moment of min{T

[0086] According to the Manhattan distance rule, on a two-dimensional grid map, there may be multiple path choices for an AGV to move from its starting coordinate point to its target coordinate point. For the convenience of discussion, the set of the shortest paths (i.e., Manhattan distance paths) from its starting coordinate point to its target coordinate point is denoted as and the set of Manhattan distance paths from its starting coordinate point to its target coordinate point is denoted as .

[0087] Meanwhile, for the convenience of analysis, on the two-dimensional grid map and the path conflict types can be divided into point conflict and edge conflict. Point conflict means that there exists a certain moment such that when moves t steps along , moves t steps along and they reach the same position in and simultaneously, that is, i, T j} and two adjacent positions (x,y) ϵ D ij and (x’,y’) ϵ D ij such that at time t is at (x,y) and is at (x’,y’), while at time t+1 is at (x’,y’) and is at (x,y), that is, the positions of and are swapped at adjacent times.

[0088] According to the degree of path conflict, the relationship between and can be defined as hostile, free, and compatible. The hostile relationship means that the path conflict between and is inevitable, that is, if and only if , , then conflicts with . The free relationship means that there is no path conflict between and , that is, if and only if , , then does not conflict with . The compatible relationship means that it is neither a hostile relationship nor a free relationship between and . Under the compatible relationship, actual path conflicts between and can be avoided through reasonable path planning.

[0089] The conflict relationship between two AGVs during the entire cycle is jointly determined by the conflict relationships in the first stage and the second stage. Specifically: if the first stage or the second stage is a hostile relationship, then the overall relationship is hostile; if the first stage and the second stage are not hostile relationships and there is a compatible relationship, the overall relationship is compatible; if the first stage and the second stage are both free relationships, the overall relationship is free. The conflict relationship during the entire cycle is shown in the following table:

[0090]

[0091] It can be understood that if the common activity area between and is empty, that is, , then there must be no path conflict between and , so it is free.

[0092] When and have a non-empty common activity area, that is, , according to the value of the number B of the traveling directions of and in the two-dimensional grid graph, it is necessary to specifically analyze their path conflicts. The analysis process is as follows:

[0093] When B = 2, since , there must be . Let's assume , , that is It only moves in the positive X-axis direction. Below, the discussion will be based on the traveling direction.

[0094] Scenario 1.1. When , :

[0095] At this time, and both move in the positive X-axis direction. Therefore, on the two-dimensional grid map, and 's active areas and are a straight line, and their common active area is also a straight line.

[0096] In the first stage, and 's traveling directions are both in the positive X-axis direction. They both move one step in the positive X-axis direction per unit time, and their relative positions remain unchanged. Therefore, and are free.

[0097] In the second stage:

[0098] When (that is, reaches the target coordinate point before ), if is located on the 's necessary path, then blocks , that is, if and then blocks . At this time, and are hostile, otherwise and are free.

[0099] When (that is, reaches the target coordinate point before ), if is located on the 's necessary path, then blocks , that is, if and then blocks . At this time, and are hostile, otherwise and are free.

[0100] As described above, in the second stage, if an AGV reaches its target coordinate point first and this target coordinate point blocks the only path of another AGV, then they are in a hostile relationship; otherwise, they are in a free relationship.

[0101] Scenario 1.2: When , :

[0102] At this time, moves in the positive X-axis direction while both move in the negative X-axis direction, that is, and are moving towards each other, so a conflict will definitely occur within . Therefore, and are hostile.

[0103] Scenario 1.3: When , :

[0104] At this time, moves in the positive X-axis direction, moves in the positive or negative Y-axis direction, and intersect in a "cross" shape, and there is only 1 intersection node within . Since only contains a single position, there will be no edge conflict, and only the case of point conflict needs to be considered.

[0105] The analysis process in the first stage is as follows:

[0106] When , that is, and enter the node simultaneously, so a point conflict occurs. Therefore, and are in a hostile relationship;

[0107] When , at this time and cannot enter the intersection node simultaneously. Therefore, and are in a free relationship.

[0108] The analysis process in the second stage is as follows:

[0109] When , they were hostile in the first stage, so and are in a hostile relationship, and there is no need to analyze here.

[0110] When and if and , then the only path of is blocked. Therefore, and are hostile, otherwise free; if and , then the only path of is blocked. Therefore, and are hostile, otherwise free.

[0111] When B = 2, in the case of moving in other directions, such as the traveling direction of is along the negative X-axis ( ), along the positive Y-axis ( ), and along the negative Y-axis ( when moving along the positive X-axis ( ), it is just a simple coordinate transformation of the above . Specifically, if , the coordinate system is flipped along the axis; if , the coordinate system is rotated 90 degrees clockwise; if and , the coordinate system is rotated 90 degrees counterclockwise. After the change, the method for determining the conflict type of

[0112] When B = 3, without loss of generality, , otherwise swap the subscripts . Further, assume (i.e., moving along the positive X-axis and Y-axis), otherwise it can be transformed into this case through coordinate system transformation (rotation or mirroring). The following will discuss based on the traveling direction.

[0113] Scenario 2.1: When , :

[0114] At this time, moves along the positive X-axis and Y-axis, moves along the positive X-axis.

[0115] In the first stage:

[0116] First, consider the case of edge conflict. Since when an edge conflict occurs, and There must be a set of opposite traveling directions, and at this time and have the same traveling direction along the X-axis, so there is no edge conflict.

[0117] Then consider the case of point conflict. Since at this time , let the lower left corner point of the rectangular area be . Here, according to the plane rectangular coordinate system, the positive direction of the X-axis is defined as "right" and the positive direction of the Y-axis is defined as "up". At this time, T is the "lower left corner" point. For ,

[0118]

[0119] When and conflict at the point , there is , and it can be obtained that and . At this time, for , , so are all conflict points.

[0120] Based on the above analysis, it can be known that in the first stage:

[0121] When , and do not conflict within , so they are free.

[0122] When , at this time each point in is a conflict point. When and , "penetrates" , then must pass through one of the points in , and reach this point at the same time, so and are hostile; otherwise "does not penetrate" , then can choose the shortest path that does not pass through , so and are compatible.

[0123] The above analysis results based on mathematical formulas can be summarized as follows: In the first stage, when the sum of the coordinate values of the starting coordinate points of two AGVs is not equal, it is a free relationship; when the sum of the coordinate values of the starting coordinate points of two AGVs is equal and the activity area of one AGV penetrates the activity area of the other AGV, it is a hostile relationship, otherwise it is a compatible relationship.

[0124] In the second stage:

[0125] When (that is, arrives at the target coordinate point before ), because can move forward along both the X-axis and the Y-axis. If and , then falls within the area where arrives at . Because , therefore is at least the size of . In \[ \], there must be a shortest path. Therefore, and are not hostile, but compatible; otherwise, will not interfere with , so the two are free.

[0126] When (that is, arrives at the target coordinate point before ), because can only move forward along the positive Y-axis. When and , blocks , so the two are hostile; otherwise and are free.

[0127] The above analysis results based on mathematical formulas can be summarized as follows: In the second stage, when the second AGV reaches the target coordinate point first, if the target coordinate point of the second AGV is within the common activity area of the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the first AGV and that of the second AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are compatible; otherwise, they are free. When the first AGV reaches the target coordinate point first, if the target coordinate point of the first AGV is within the common activity area of the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the second AGV and that of the first AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are in a hostile relationship; otherwise, the two AGVs are in a free relationship.

[0128] Scenario 2.2. When , :

[0129] At this time, Move along the positive directions of the X-axis and Y-axis, Move along the negative X-axis.

[0130] In the first stage:

[0131] First, consider the case of point conflict. When and have a point conflict at a point , . Because , on the X-axis, as x u increases, increases, while decreases. So u is the only point in such that holds. In short, in \u, and will not have a point conflict. Therefore, in the region \u, there must be the shortest path, resulting in and not being hostile.

[0132] Then, consider the case of edge conflict. When and have an edge conflict at two adjacent points , and , that is, . However, reaches u at , Arrival at time +1 For this path, there must be one of the following two ways to resolve this edge conflict. Because if the following way a) is not feasible, then , if the following way b) is not feasible, then . Therefore, when both ways are not feasible, then , but from it can be known that , which leads to a contradiction. Therefore and are not hostile.

[0133] a) Arrival at time , Arrival at time +1 .

[0134] b) Arrival at u at time Arrival at time +1 .

[0135] Based on the above analysis, it can be known that in the first stage:

[0136] When there exists u ϵ such that = , and are compatible; when there are two adjacent points and in , such that and are compatible; in other cases and are both free.

[0137] The above analysis results based on mathematical formulas can be summarized as follows: In the first stage, when there is a coordinate point in the common activity area of two AGVs and the Manhattan distances from the coordinate point to the two AGVs are equal, the two AGVs are in a compatible relationship; when there are two adjacent coordinate points on the X-axis in the common activity area of two AGVs, and the sum of the X coordinate value and the Y coordinate value of the starting coordinate point of the first AGV plus the difference between the X coordinate value and the Y coordinate value of the starting coordinate point of the second AGV is equal to the sum of the X coordinate values of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0138] In the second stage:

[0139] In this case, in the second stage and The conflict situation is exactly the same as the analysis process in the second stage of Scenario 2.1, which will not be elaborated here.

[0140] Scenario 2.3. When , :

[0141] At this time, moves along the positive directions of the X-axis and Y-axis, moves along the positive direction of the Y-axis.

[0142] After exchanging the x and y coordinates, it is transformed into the situation in Section 1.1 of Scenario 1. The analysis process is the same and will not be elaborated here.

[0143] Scenario 2.4. When , :

[0144] At this time, moves along the positive directions of the X-axis and Y-axis, moves along the negative direction of the Y-axis.

[0145] Exchange coordinates and it is transformed into the situation in Section 2.2 of Scenario 2. The analysis process is the same and will not be elaborated here.

[0146] In other cases, such as when , analyze , or in the case of and The path conflict situations are all the same as the analysis process in the above moving along the X-axis and the positive Y-axis ( ), and only simple coordinate transformations are needed. Similarly, analyzing in the case of and The path conflict analysis process is also the same as the analysis process in the case, and will not be elaborated here.

[0147] When B = 4, at this time . Without loss of generality, assume (that is, moves along the positive directions of the X-axis and Y-axis), otherwise it can be transformed into this situation through coordinate transformations (rotation or mirroring). The following will discuss based on the travel direction.

[0148] Scenario 3.1. When , :

[0149] At this time, and Both move in the positive directions of the X-axis and Y-axis.

[0150] In the first stage:

[0151] First, consider the case of edge conflict. Since and have exactly the same traveling direction, there will be no edge conflict.

[0152] Then, consider the case of point conflict. If and have a point conflict, that is, there exists a point such that = , then there must be and for , = , so all the points in

[0153] are conflict points.

[0154] In summary, in the first stage: When and are not in conflict within , they are free.

[0155] When , at this time, each point in is a conflict point. At this time, when or when , and "penetrate" each other, so and must conflict at a point in , so and are hostile, otherwise there exists a shortest path in \ or there exists a shortest path in \ , so and are compatible.

[0156] The above analysis results based on mathematical formulas can be summarized as: In the first stage, when the sum of the coordinate values of the starting coordinate points of two AGVs is not equal, the two AGVs are in a free relationship When the sum of the coordinate values of the starting coordinate points of two AGVs is equal and the active areas of the AGVs penetrate each other, the two AGVs are in a hostile relationship, otherwise they are in a compatible relationship.

[0157] In the second stage:

[0158] When (i.e., preceding to reach the target coordinate point), if and , then falls within reaches the area of, because , thus the size of is at least . In \[ , there must be the shortest path, so and are not adversarial, but compatible; otherwise, will not cause any interference to , so the two are free.

[0159] When (i.e., preceding to reach the target coordinate point), when and , blocks , but at this time has two moving directions. In \[ , there is the shortest path, so the two are compatible; otherwise will not cause any interference to , so the two are free.

[0160] The above analysis results based on mathematical formulas can be summarized as: In the second stage, the target coordinate point of the AGV that reaches first is located in the common activity area of the two AGVs, and when the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the later AGV and the Manhattan distance from the starting coordinate point to the target coordinate point of the earlier AGV is not less than the Manhattan distance between the target coordinate points of the two AGVs, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0161] Scenario 3.2. When , :

[0162] At this time, moves along the positive X-axis and Y-axis, moves along the positive X-axis and negative Y-axis.

[0163] In the first stage:

[0164] First, consider the case of point conflict. When there exists such that at this time, , that is , then all the nodes in .

[0165] Then, consider the case of edge conflict. When there is an edge conflict in , without loss of generality, let reach at this time, reach , then . Denote

[0166] In summary, it can be seen from the above analysis that in the first stage:

[0167] When there exists at this time, and are compatible. That is, non-conflicting paths can be found for and . When , control to reach the end point with the action , and find a shortest path for \ in , so at this time and have non-conflicting paths; when , control to reach the end point with the action , and find a shortest path for \ in , so at this time and have non-conflicting paths; when , control to only pass through a point p in , that is, enter p with the action ↑ and pass through from p with the action ↑, and find a shortest path for in \ p, so at this time and have non-conflicting paths.

[0168] When there exists at this time, and are compatible. Similar to the above the situation at this time, when it is always possible to and find non - conflicting paths.

[0169] Otherwise, when neither of the above two situations is satisfied, and are free.

[0170] The above - mentioned analysis results based on mathematical formulas can be summarized as follows: In the first stage, when there is a coordinate point in the common activity area of two AGVs such that the Manhattan distances from the starting coordinate points of the first AGV and the second AGV to the said coordinate point are equal, the two AGVs are compatible; when there are two adjacent coordinate points on the Y - axis in the common activity area of two AGVs and the sum of the X - coordinate value and the Y - coordinate value of the starting coordinate point of the first AGV plus the difference between the Y - coordinate value and the X - coordinate value of the starting coordinate point of the second AGV is equal to the sum of the Y - coordinate values of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0171] In the second stage:

[0172] In this case, in the second stage and the conflict situation is exactly the same as the analysis process in the second stage of Scenario 3.1, which will not be elaborated here.

[0173] Scenario 3.3. When , at this time:

[0174] At this time, moves along the positive X - axis and Y - axis, moves along the negative X - axis and positive Y - axis.

[0175] After exchanging coordinates, it is transformed into the situation in Section 3.2 of Scenario 3, and the analysis process is the same, which will not be elaborated here.

[0176] Scenario 4.4. When , at this time:

[0177] At this time, moves along the positive X - axis and Y - axis, both move along the negative X - axis and Y - axis.

[0178] In the first stage:

[0179] First, consider the case of point conflict. If there exists a point such that , then , that is, , at this time, the conflict point is on a 45-degree straight line, let .

[0180] Then consider the case of edge conflict. If there exist three points such that or , then , where and are 's neighbors. Denote .

[0181] In summary, from the above analysis, in the first stage:

[0182] When there exists , since , therefore can only pass through one of the points (assumed to be p), and plan the shortest path in \ p. Therefore and and there exists a non-conflicting path, so at this time and are compatible

[0183] When there exists , and are compatible. Similar to the above case, at this time, a non-conflicting path can always be found for and .

[0184] Otherwise, when neither of the above two cases is satisfied, and are free.

[0185] The above analysis results based on mathematical formulas can be summarized as follows: In the first stage, when there is a coordinate point in the common activity area of two AGVs and the Manhattan distances from this coordinate point to the starting coordinate points of the two AGVs are equal, the two AGVs are in a compatible relationship; when there are two adjacent coordinate points in the common activity area of two AGVs and the sum of the coordinate values of the starting coordinate point of the first AGV and the starting coordinate point of the second AGV is equal to the sum of the coordinate values of these two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship.

[0186] In the second stage:

[0187] In this case, in the second stage and The conflict situation is exactly the same as the analysis process in the second stage of Scenario 3.1, and will not be elaborated here.

[0188] In other cases, such as When analyzing , or In the case of and The path conflict situations are all the same as Moving along the positive directions of the X-axis and Y-axis ( ), and the analysis process is the same, only a simple coordinate transformation is required.

[0189] According to the method provided by the embodiments of the present application, the present invention also provides an AGV conflict determination system, including a network construction module, a coordinate point acquisition template, a task scheduling module, a conflict type classification module, and a conflict determination module.

[0190] Among them, the network construction module is used to establish a two-dimensional grid map and define that each node in the two-dimensional grid map corresponds to a unique coordinate.

[0191] The coordinate point acquisition template is used to acquire the starting coordinate point and the target coordinate point of each AGV.

[0192] The task scheduling module is used to execute and schedule the movement task of each AGV from the starting coordinate point to the target coordinate point.

[0193] The conflict type classification module can classify the conflict types of two AGVs based on the number of traveling directions of the two AGVs. As described above, the number of traveling directions of the two AGVs can be 2, 3, or 4. Classifying the conflict types of the two AGVs based on the number of traveling directions, the specific classification process has been described in detail above.

[0194] The conflict determination module can determine the conflict relationship between two AGVs based on the common activity area and conflict type of the two AGVs, and the specific determination execution process has been described in detail above.

[0195] Advantages of the implementation of the present invention: First, the present invention takes the starting coordinate point and the target coordinate point of the AGV as inputs, and determines the conflict type through elementary mathematical tools, reducing the computational resource requirements and improving the real-time performance of the system. Second, by enhancing the algorithm performance, the priority and direction of conflict resolution are clarified for the CBS solution strategy of the MAPF problem, and a globally better solution can be found more quickly during the high-conflict path planning process. Third, by providing an accurate basis for conflict determination, a reasonable path can be planned in advance for the AGV, reducing the waiting and avoidance time of the AGV, thereby improving the cargo handling efficiency and the utilization rate of the storage space. Finally, the present invention is applicable to multi-AGV systems of different scales and complexities, has strong versatility and can meet the diverse scenario requirements. In summary, the conflict determination method of the AGV of the present invention has high industrial utilization value.

[0196] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An AGV conflict determination method, applied to an AGV scheduling method, characterized in that: include: Establish a two-dimensional grid graph and define unique coordinates corresponding to each node in the two-dimensional grid graph; Get the starting and target coordinates of each AGV; In a single dispatch process, all AGVs only complete the movement task from the starting coordinate point to the target coordinate point once; Calculate the common activity area of ​​any two AGVs during this scheduling process; The types of path conflict judgments of the two AGVs are classified according to the number of their travel directions; The conflict relationship between the two AGVs is determined as hostile, compatible, or free based on the classification results and the common activity area of ​​the two AGVs.

2. The AGV conflict determination method according to claim 1, characterized in that: Each AGV moves from the starting coordinate point to the target coordinate point according to the Manhattan distance and cannot stop before reaching the target coordinate point. When the AGV reaches the target coordinate point, it stops at the point until all AGVs reach their respective target coordinate points.

3. The AGV conflict determination method according to claim 1, characterized in that: The calculation of the common activity area of ​​any two AGVs in this scheduling process includes: determining the area formed by the starting coordinate points and the target coordinate points of the two AGVs on the two-dimensional grid map, and the common activity area of ​​the two AGVs is the intersection of the activity areas of the two AGVs on the two-dimensional grid map.

4. The AGV conflict determination method according to claim 1, characterized in that: The classification of the type of path conflict judgment according to the number of travel directions of the two AGVs includes: calculating the sum of the absolute values ​​of the movement direction values ​​of the two AGVs along the X-axis and the Y-axis on the two-dimensional grid diagram, and the movement direction values ​​include 1, -1 and 0 and respectively represent the positive, negative and non-axial movement of the AGV along the axis.

5. The AGV conflict determination method according to claim 1, characterized in that: The method of determining whether the conflict relationship between the two AGVs is hostile, compatible or free based on the classification results and the common activity area of ​​the two AGVs includes: dividing the determination process into a first stage and a second stage according to the states of the two AGVs, the first stage being a time period in which the two AGVs move together, and the second stage being a time period in which one AGV reaches a target coordinate point and the other AGV continues to move until it reaches the target coordinate point.

6. The AGV conflict determination method according to claim 5, characterized in that: The method of determining whether the conflict relationship between the two AGVs is hostile, compatible or free based on the classification results and the common activity area of ​​the two AGVs includes: if the first stage or the second stage is a hostile relationship, then determining that the entire relationship is a hostile relationship; if the first stage and the second stage are not a hostile relationship and there is a compatible relationship, then determining that the entire relationship is a compatible relationship; if the first stage and the second stage are both free relationships, then determining that the entire relationship is a free relationship.

7. The AGV conflict determination method according to claim 5, characterized in that: The determining of whether the conflict relationship between the two AGVs is hostile, compatible or free based on the classification results and the common activity area of ​​the two AGVs includes: if the two AGVs do not have a common activity area, determining that the conflict relationship between the two AGVs is free; if the two AGVs have a common activity area, further determining is performed based on the number of travel directions of the two AGVs.

8. The AGV conflict determination method according to claim 7, characterized in that: The further determination according to the number of travel directions of the two AGVs includes: if the number of travel directions of the two AGVs is 2, the determination rule is: When the two AGVs are traveling in the same direction, the determination process includes: In the first stage, the two AGVs are in a free relationship; In the second stage, if one AGV reaches its target coordinate point first and the target coordinate point blocks the path that another AGV must pass through, it is a hostile relationship, otherwise it is a free relationship; When two AGVs are traveling in opposite directions, they are in a hostile relationship; When the travel directions of the two AGVs are perpendicular, the determination process includes: In the first stage, if two AGVs enter the only intersection at the same time, they are in a hostile relationship, otherwise they are in a free relationship; In the second stage, if one AGV reaches its target coordinate point first and the target coordinate point blocks the path that another AGV must pass through, it is a hostile relationship, otherwise it is a free relationship.

9. The AGV conflict determination method according to claim 7, characterized in that: The further determination according to the number of travel directions of the two AGVs includes: if the number of travel directions of the two AGVs is 3, the determination rule is: In the first scenario, when the first AGV moves in the positive direction along the X-axis and the Y-axis and the second AGV moves only in the positive direction along the X-axis, the determination process includes: In the first stage, when the sum of the coordinate values ​​of the starting coordinate points of the two AGVs is not equal, it is a free relationship; when the sum of the coordinate values ​​of the starting coordinate points of the two AGVs is equal and the activity area of ​​one AGV runs through the activity area of ​​the other AGV, it is a hostile relationship, otherwise it is a compatible relationship; In the second stage, when the second AGV reaches the target coordinate point first, if the target coordinate point of the second AGV is located in the common activity area of ​​the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the first AGV and the Manhattan distance from the starting coordinate point to the target coordinate point of the second AGV is not less than the Manhattan distance of the target coordinate points of the two AGVs, the two AGVs are compatible, otherwise they are free; when the first AGV reaches the target coordinate point first, if the target coordinate point of the first AGV is located in the common activity area of ​​the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the second AGV and the Manhattan distance from the starting coordinate point to the target coordinate point of the first AGV is not less than the Manhattan distance of the target coordinate points of the two AGVs, the two AGVs are in a hostile relationship, otherwise they are in a free relationship; In the second scenario, when the first AGV moves in the positive direction along the X-axis and the Y-axis and the second AGV moves only in the negative direction along the X-axis, the determination process includes: In the first stage, when there is a coordinate point in the common activity area of ​​the two AGVs and the Manhattan distances from the coordinate point to the two AGVs are equal, the two AGVs are in a compatible relationship; when there are two adjacent coordinate points on the X axis in the common activity area of ​​the two AGVs, and the sum of the X coordinate value and the Y coordinate value of the starting coordinate point of the first AGV plus the difference between the X coordinate value and the Y coordinate value of the starting coordinate point of the second AGV is equal to the sum of the X coordinate values ​​of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship; In the second stage, the determination method is the same as the determination method in the second stage in the first scenario; In the third scenario, when one AGV moves in the positive direction along the X-axis and the Y-axis and the other AGV moves only in the positive direction along the Y-axis, the determination method is the same as that of the first scenario; In the fourth scenario, when one AGV moves in the positive direction along the X-axis and the Y-axis and the other AGV moves only in the negative direction along the Y-axis, the determination method is the same as that in the second scenario.

10. The AGV conflict determination method according to claim 7, characterized in that: The further determination according to the number of travel directions of the two AGVs includes: if the number of travel directions of the two AGVs is 4, the determination rule is: In the first scenario, when both AGVs move in the positive direction along the X-axis and the Y-axis, the determination process includes: In the first stage, when the sum of the coordinate values ​​of the starting coordinate points of the two AGVs is not equal, the two AGVs are in a free relationship; when the sum of the coordinate values ​​of the starting coordinate points of the two AGVs is equal and the activity areas of the two AGVs intersect each other, the two AGVs are in a hostile relationship, otherwise they are in a compatible relationship; In the second stage, if the target coordinate point of the first AGV is located in the common activity area of ​​the two AGVs and the difference between the Manhattan distance from the starting coordinate point to the target coordinate point of the later AGV and the Manhattan distance from the starting coordinate point to the target coordinate point of the first AGV is not less than the Manhattan distance of the target coordinate points of the two AGVs, the two AGVs are in a compatible relationship, otherwise they are in a free relationship; In the second scenario, when the first AGV moves along the positive direction of the X-axis and the Y-axis and the second AGV moves along the positive direction of the X-axis and the negative direction of the Y-axis, the determination process includes: In the first stage, when there is a coordinate point in the common activity area of ​​the two AGVs such that the Manhattan distances from the starting coordinate points of the first AGV and the second AGV to the coordinate point are equal, the two AGVs are compatible; when there are two adjacent coordinate points on the Y axis in the common activity area of ​​the two AGVs and the sum of the X coordinate value and the Y coordinate value of the starting coordinate point of the first AGV plus the difference between the Y coordinate value and the X coordinate value of the starting coordinate point of the second AGV is equal to the sum of the Y coordinate values ​​of the two adjacent coordinate points, the two AGVs are compatible; otherwise, the two AGVs are free; In the second stage, the determination method is the same as the determination method in the second stage in the first scenario; In the third scenario, when the first AGV moves in the positive direction along the X-axis and the Y-axis and the second AGV moves in the negative direction along the X-axis and in the positive direction along the Y-axis, the determination method is the same as that of the second scenario; In the fourth scenario, when the first AGV moves in the positive direction along the X-axis and the Y-axis and the second AGV moves in the negative direction along the X-axis and the Y-axis, the determination process includes: In the first stage, when there is a coordinate point in the common activity area of ​​the two AGVs and the Manhattan distances from the coordinate point to the starting coordinate points of the two AGVs are equal, the two AGVs are in a compatible relationship; when there are two adjacent coordinate points in the common activity area of ​​the two AGVs and the sum of the coordinate values ​​of the starting coordinate point of the first AGV and the starting coordinate point of the second AGV is equal to the sum of the coordinate values ​​of the two adjacent coordinate points, the two AGVs are in a compatible relationship; otherwise, the two AGVs are in a free relationship; In the second stage, the determination method is the same as the determination method in the second stage in the first scenario.

11. The AGV conflict determination method according to any one of claims 1 to 10, characterized in that: The hostile relationship between the two AGVs includes point conflict and edge conflict. The point conflict is that the two AGVs arrive at the same position at the same time, and the edge conflict is that the two AGVs exchange positions with each other at adjacent times.

12. An AGV conflict determination system, characterized in that: include: A network construction module is used to establish a two-dimensional grid graph and define unique coordinates corresponding to each node in the two-dimensional grid graph; Coordinate point acquisition template, used to obtain the starting coordinate point and target coordinate point of each AGV; Task scheduling module, used to execute the task of scheduling each AGV to move from the starting coordinate point to the target coordinate point; The conflict type classification module classifies the conflict types of two AGVs based on the number of travel directions of the two AGVs; The conflict determination module determines the conflict relationship between two AGVs based on the common activity area and conflict type of the two AGVs.

Images