Multi-robot operating method, multi-robot operating system, and control apparatus

Abstract

The present invention relates to a multi-robot operating method comprising: a movement path calculation step of calculating a movement path set including respective paths along which a plurality of robots move from starting points to destination points; a collision detection step of determining whether cycle conflicts are included in the movement path set; a path modification step of, if it is determined in the collision detection step that cycle conflicts are included in the movement path set, modifying the movement path set to a corrected path set; and a corrected path execution step of moving the plurality of robots according to the corrected path set.

Description

Multi-robot operation method, multi-robot operation system, and control device

[0001] The present invention relates to a technology for operating multiple robots, and more particularly, to a method for operating multiple robots, a multi-robot operating system including multiple robots and a control device for controlling multiple robots, and a multi-robot control device for operating multiple robots.

[0002] Advances in information processing technology have led to ongoing research on logistics automation. Recently, with increased information processing speeds and advancements in artificial intelligence technology, research is underway to apply these technologies to logistics processes. For example, active research is underway on logistics automation, where unmanned robots autonomously move within a logistics warehouse and move items to planned locations.

[0003] Furthermore, with the increasing use of robots in industrial automation, logistics, and manufacturing, designing collaborative robots is becoming increasingly important. When multiple robots work simultaneously in the same environment, preventing collisions and finding optimal paths becomes essential.

[0004] And as intelligent transportation systems such as autonomous driving advance, the importance of sophisticated route planning to ensure safe and efficient movement of multiple vehicles on the road is also increasing.

[0005] In addition, as the use of multi-agent systems increases in fields such as smart cities, smart logistics, and public safety, efficient collaboration and movement coordination between them have become necessary.

[0006] Multi-agent path finding (MFP) is a concept that emerged against this backdrop. Simply put, MFP is an algorithmic approach to solving complex path planning problems. In other words, MFP is a technique for planning paths for multiple agents to efficiently reach their respective destinations along optimal or near-optimal paths without conflict.

[0007] Multi-agent pathfinding technology can be applied to a wide range of fields, including robotics, automotive operation, airport logistics, video games, smart factories, and unmanned warehouses. Furthermore, multi-agent pathfinding technology can plan paths by considering not only each agent's characteristics, such as location, destination, movement speed, and size, but also the interactions between agents and dynamic changes in the environment.

[0008] In particular, because multiple agents plan their routes in advance, robots can be controlled along optimal routes, minimizing resource waste. In practice, applying this technology can effectively reduce logistics costs through optimal route search.

[0009] However, conventional multi-agent pathfinding technologies were primarily developed for static environments. This limits their ability to respond to unpredictable obstacles or variables that can arise in dynamic environments.

[0010] For example, consider a multi-agent path search for a logistics robot. When performing path search, the physical volume of the logistics robot itself may not be considered. Furthermore, in real-world situations, delays can occur depending on the physical distance along the path the logistics robot travels. However, because multi-agent path search assumes ideal physical conditions and searches under static conditions, there is a high possibility of discrepancies between the path searched and the actual operation of the logistics robot.

[0011] Furthermore, as the number of agents increases, computational complexity increases, making real-time path planning difficult in large-scale systems. Furthermore, as complexity increases, the likelihood of conflicts occurring between agents increases.

[0012] In multi-agent pathfinding, the following types of collisions can occur:

[0013] A vertex conflict occurs when two or more agents move to the same location at the same time. In other words, a vertex conflict can occur when two or more agents are positioned at the same point at the same time in a path derived by multi-agent pathfinding.

[0014] The path function representing the location of a specific agent k at a specific time t can be expressed by the following formula.

[0015]

[0016] The vertex conflict situation can be expressed in a formula as follows.

[0017]

[0018] That is, among the paths derived by multi-agent path search, a conflict may occur in which the positions of agent i and agent j become the same at a specific time x.

[0019] An edge conflict occurs when two or more agents are traveling the same path at a given time. This means that two different agents may be in the same location at one time, but then move together to different locations at a later time.

[0020] The edge conflict can be expressed in a formula as follows:

[0021]

[0022] That is, among the paths derived by multi-agent path search, if the positions of agent i and agent j are the same at a specific time x, and the positions of agent i and agent j are also the same at the next time x+1, it can be seen that an edge collision has occurred.

[0023] A cycle conflict occurs when a group of agents in a group moves in a circular fashion, biting each other's tails. This means that the agents in a group move in a circular fashion, biting each other's tails, from their respective positions.

[0024] The cycle conflict can be expressed in a formula as follows:

[0025] ...

[0026]

[0027] That is, among the paths derived by multi-agent path search, if agents i, i+1, ..., j-1, j move to the location of the next agent, it can be seen that a cycle collision has occurred.

[0028] In addition to the examples described above, various types of conflicts can occur. When a conflict occurs in multi-agent pathfinding, agents must replan their paths to the destination, which consumes additional time and resources. Consequently, the overall efficiency of the system can be reduced.

[0029] Additionally, agents must stop or wait to avoid collisions, which can lead to movement delays. Consequently, there is a problem of wasted time.

[0030] Additionally, frequent collisions can degrade the overall stability and reliability of the system, which can be a significant problem in safety-critical multi-agent systems.

[0031] The examples above are only a few, and the problems that arise when conflicts occur can be quite diverse. Therefore, performing multi-agent pathfinding that takes such conflicts into account is becoming increasingly important.

[0032] Based on these needs, recent research has focused on multi-agent pathfinding methods that consider vertex and edge collisions, among the aforementioned conflict situations. These studies have presented algorithms for optimizing pathfinding by proactively detecting vertex and edge collisions.

[0033] However, cycle conflicts are a characteristic observed when conducting multi-agent path exploration with more than 100 agents. In other words, cycle conflicts occur when managing a very large number of agents, and thus, not much research has been conducted on multi-agent path exploration that takes cycle conflicts into account.

[0034] However, the trend in multi-agent technologies is shifting toward managing a larger number of agents. Consequently, the importance of methods for optimizing paths found through multi-agent path search, when they contain cycle conflicts, is increasing.

[0035] The present invention aims to provide a multi-robot operation method, a multi-robot operation system, and a control device capable of searching for a path of multiple robots while considering cycle collisions when searching for a path of multiple robots by utilizing multi-agent path search.

[0036] In addition, the present invention aims to solve the problem of providing a multi-robot operation method, a multi-robot operation system, and a control device capable of adding an operation to modify an existing path when a path including a cycle collision is searched in a multi-agent path search.

[0037] In addition, the present invention aims to solve the problem of providing a multi-robot operation method, a multi-robot operation system, and a control device capable of minimizing resource waste by optimally modifying a searched path when a cycle collision is detected in a multi-agent path search.

[0038] In addition, the present invention aims to solve the problem of providing a multi-robot operation method, a multi-robot operation system, and a control device that can utilize an existing searched path when modifying a searched path in a multi-agent path search.

[0039] In addition, the present invention aims to solve the problem of providing a multi-robot operation method, a multi-robot operation system, and a control device that can shorten the overall required time by simultaneously executing path re-planning and movement of multiple robots when path re-planning is required in multi-agent path search.

[0040] In order to solve the above-described problem, a multi-robot operation method according to one embodiment of the present invention includes a movement path calculation step of calculating a movement path set including each path along which a plurality of robots move from a starting point to a destination point, a collision detection step of determining whether a cycle conflict is included in the movement path set, a path modification step of modifying the movement path set to a correction path set if it is determined in the collision detection step that a cycle conflict is included in the movement path set, and a correction path execution step of moving the plurality of robots according to the correction path set.

[0041] In addition, the method may further include a replanning step that is performed together with the correction path execution step and calculates a replanning path set that includes each path along which the plurality of robots move from a new starting point determined according to the correction path set to the destination point.

[0042] Additionally, when the replanning step is completed, the collision detection step can determine whether a cycle collision is included in the replanning path set.

[0043] In addition, if it is determined in the collision detection step that a cycle collision is included in the replanned path set, the replanned path set may be modified into a corrected replanned path set in the path modification step, and the plurality of robots may be moved according to the corrected replanned path set in the corrected path execution step.

[0044] Additionally, the path correction step, the correction path execution step, the replanning step, and the collision detection step may be repeated until it is determined that no cycle collision is included in the replanned path set in the collision detection step.

[0045] Additionally, if it is determined in the collision detection step that no cycle collision is included in the replanned path set, the plurality of robots can be moved according to the replanned path set.

[0046] In addition, the path correction step may include a collision robot designation step of defining one of the collision configuration robots determined to constitute a cycle collision in the collision detection step among the plurality of robots as a collision robot, a collision robot correction step of correcting a path in which the collision robot moves, an influence robot correction step of correcting a path of another robot affected by the collision robot correction step, and a correction path acquisition step of obtaining the set of correction paths derived through the collision robot correction step and the influence robot correction step.

[0047] In addition, the collision robot designation step may include a disruption effect determination step for determining a disruption effect defined as the degree to which the movement of other robots is affected if the collision configuration robots stop at the expected collision time determined to be a cycle collision to occur in the collision detection step, and a disruption effect comparison step for selecting a robot determined to have the lowest disruption effect in the disruption effect determination step as the collision robot.

[0048] In addition, the collision robot correction step may include a stop time determination step for determining a stop time for stopping the collision robot by comparing the expected collision time determined to be a cycle collision to occur in the collision detection step with a preset reference collision time, and a step for obtaining a correction path of the collision robot by correcting the movement path of the collision robot by stopping the collision robot at the stop time.

[0049] In addition, the replanning step may include a starting point resetting step of resetting the stopping points of the plurality of robots determined according to the correction path set to the new starting points, a step of calculating the paths along which the plurality of robots move based on the new starting points, and a step of obtaining the replanned path set through the calculating step.

[0050] Meanwhile, in order to solve the above-described problem, a multi-robot operating system according to one embodiment of the present invention may include a plurality of robots and a control device that controls the plurality of robots.

[0051] In addition, the plurality of robots each include a robot main body, a driving unit provided in the robot main body to move the robot main body, a robot control unit to control the operation of the driving unit, and a robot communication unit provided to communicate with the outside, and the control device includes a communication unit provided to be able to communicate with the robot communication unit, a storage unit provided to store data, and a control unit provided to process communication data acquired through the communication unit or data stored in the storage unit, and the control unit calculates a movement path set including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the movement path set, and if it is determined that a cycle conflict is included in the movement path set, modifies the movement path set to a correction path set, and the communication unit can transmit a signal to the robot communication unit to move the plurality of robots according to the correction path set.

[0052] In addition, the control unit may calculate a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the corrected path set to the destination point after modifying the set of movement paths to the set of correction paths, and the communication unit may transmit a signal to the robot communication unit to move the plurality of robots according to the set of replanned paths.

[0053] Meanwhile, in order to solve the above-described problem, a control device according to an embodiment of the present invention controls a plurality of robots, and includes a communication unit that is provided to be able to communicate with the outside, and a control unit that is provided to process communication data obtained through the communication unit, and the control unit calculates a movement path set including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the movement path set, and if it is determined that a cycle conflict is included in the movement path set, modifies the movement path set into a correction path set, and moves the plurality of robots according to the correction path set.

[0054] Additionally, the control unit may calculate a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the set of corrected paths to the destination point, and may move the plurality of robots according to the replanned path set.

[0055] According to another disclosed embodiment, a multi-robot operating system for controlling a plurality of robots comprises: a communication unit for communicating with the robots; a storage unit for storing data necessary for controlling the robots; and a control unit for processing communication data acquired through the communication unit or data stored in the storage unit, wherein the control unit calculates a movement path set including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the movement path set, and, if it is determined that a cycle conflict is included in the movement path set, modifies the movement path set into a correction path set, and the communication unit transmits a signal to the robots for moving the plurality of robots according to the correction path set.

[0056] According to one embodiment of the present invention, the present invention has the effect of providing a multi-robot operation method, a multi-robot operation system, and a control device capable of searching for a path of multiple robots while considering cycle collisions when searching for a path of multiple robots by utilizing multi-agent path search.

[0057] In addition, the present invention has the effect of providing a multi-robot operation method, a multi-robot operation system, and a control device that can add an operation for modifying an existing path when a path including a cycle collision is searched in a multi-agent path search.

[0058] In addition, the present invention has the effect of providing a multi-robot operation method, a multi-robot operation system, and a control device capable of minimizing resource waste by optimally modifying a searched path when a cycle collision is detected in a multi-agent path search.

[0059] In addition, the present invention has the effect of providing a multi-robot operation method, a multi-robot operation system, and a control device that can utilize a previously searched path when modifying a searched path in multi-agent path search.

[0060] In addition, the present invention has the effect of providing a multi-robot operation method, a multi-robot operation system, and a control device that can shorten the overall required time by simultaneously executing path re-planning and movement of multiple robots when path re-planning is required in multi-agent path search.

[0061] FIG. 1 is a diagram schematically illustrating a path search process that can be applied to a multi-robot operation method, a multi-robot operation system, and a control device according to one embodiment of the present invention.

[0062] Figure 2 is a conceptual diagram schematically illustrating components of a control device according to one embodiment of the present invention.

[0063] Figure 3 is a conceptual diagram schematically illustrating components of a robot according to one embodiment of the present invention.

[0064] FIG. 4 illustrates a method for moving a plurality of robots illustrated in FIG. 1 of the present invention.

[0065] Figure 5 is a flowchart illustrating a multi-robot operation method according to one embodiment of the present invention.

[0066] Figure 6 specifically illustrates a movement path calculation step according to one embodiment of the present invention.

[0067] Figure 7 specifically illustrates a path modification step according to one embodiment of the present invention.

[0068] Figure 8 specifically illustrates a collision robot designation step according to one embodiment of the present invention.

[0069] FIG. 9 specifically illustrates a collision robot correction step according to one embodiment of the present invention.

[0070] Figure 10 specifically illustrates a replanning step according to one embodiment of the present invention.

[0071] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement them. However, the present invention can be implemented in various different forms. Therefore, the present invention is not limited to the embodiments described below. Furthermore, in the drawings included in this specification, parts irrelevant to the description are omitted for clarity of explanation.

[0072] In this specification, even in different embodiments, identical or similar configurations are given identical or similar reference numbers, and their descriptions are replaced with the first description.

[0073] As used herein, singular expressions include plural expressions unless the context clearly dictates otherwise. Furthermore, when describing embodiments disclosed herein, detailed descriptions of related known technologies are omitted if they are deemed to obscure the gist of the embodiments disclosed herein.

[0074] In addition, it should be noted that the attached drawings are only intended to facilitate easy understanding of the embodiments disclosed in this specification, and should not be construed as limiting the technical ideas disclosed in this specification by the attached drawings.

[0075] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement the invention. However, the present invention can be implemented in various different forms and is not limited to the embodiments described herein. In addition, in the drawings, parts irrelevant to the description are omitted for clarity of description, and similar parts are designated with similar reference numerals throughout the specification.

[0076] FIG. 1 is a diagram schematically illustrating a path search process that can be applied to a multi-robot operation method, a multi-robot operation system, and a control device according to one embodiment of the present invention.

[0077] Figure 2 is a conceptual diagram schematically illustrating components of a control device according to one embodiment of the present invention.

[0078] Figure 3 is a conceptual diagram schematically illustrating components of a robot according to one embodiment of the present invention.

[0079] Referring to FIGS. 1 to 3, a multi-robot operation method, a multi-robot operation system, and a control device according to one embodiment of the present invention are briefly described.

[0080] A multi-robot operation system (S) may be configured to include multiple robots (2a, 2b) and a control device (1). The drawing illustrates a first robot (2a) and a second robot (2b), but only two robots are depicted for convenience of explanation. In practice, the present invention can be applied to multiple robots of 100 or fewer or more.

[0081] In multi-robot operation, a plurality of robots (2) are provided. The description of the robots (2) can be applied to all of the multiple robots actually applied, including the first robot (2a) and the second robot (2b).

[0082] The above plurality of robots (2) may each include a robot body (24), a driving unit (22) provided in the robot body (24) to move the robot body (24), a robot control unit (20) that controls the operation of the driving unit (22), and a robot communication unit (21) provided to communicate with the outside.

[0083] The robot communication unit (21) can communicate with the outside world in various ways, such as data communication, Bluetooth communication, and network communication. The communication method of the robot communication unit in this specification is not limited, and any method for transmitting or receiving data or signals to or from the outside world can be applied. The robot (2) can receive information transmitted from the outside world or receive information from the outside world through the robot communication unit (21).

[0084] The robot control unit (20) can interact with the robot communication unit (21). In addition, the robot control unit (20) can transmit a command signal to the driving unit (22) to operate the driving unit (22). The robot body (24) can be moved to a predetermined position by the driving unit (22).

[0085] Additionally, the robot (2) may include an operating unit (23). The operating unit (23) may be a part related to the function of the robot (2). For example, when the robot (2) is used for logistics, the operating unit (23) may correspond to a part that loads or unloads objects onto the robot. The robot control unit (20) may control the operating unit (23) to perform the function of the robot (2).

[0086] The robot control unit (20) can control the driving unit (22) or the operating unit (23) based on a signal received from the robot communication unit (21) or a result calculated by the robot control unit (20) itself.

[0087] The control device (1) may include a communication unit (11) equipped to enable communication with the outside, a storage unit (12) equipped to store data, and a control unit (10) equipped to process communication data acquired through the communication unit or data stored in the storage unit. The communication unit (11) may exchange signals with the robot communication unit (21).

[0088] Additionally, the storage unit (12) may be equipped as a data storage device. The storage unit (12) may store data required for control, data received through the communication unit (11), data derived through computational operations in the control unit (10), etc. The storage unit may interact with the control unit (10) to receive or provide data.

[0089] The above control unit (10) may include a calculation unit (13) that performs a calculation process based on data. In addition, the control unit (10) may include a command unit (14) that generates a control command that controls the configuration of another control device (1). The calculation unit (13) and the command unit (14) constitute the control unit and may perform the function of a processing unit that calculates and generates control commands necessary for controlling the robot (2).

[0090] When performing path search for multi-robot operation, the calculation unit (13) can calculate the path along which the robot moves based on data stored in the storage unit (12) or data received from the communication unit (11). When the calculation unit (13) completes the path calculation, a movement command signal can be generated through the command unit (14). The communication unit (11) can transmit the movement command signal generated by the command unit (14) to the robot communication unit (21).

[0091] The robot communication unit (21) can receive a movement command signal and transmit it to the robot control unit (20). Thereafter, the robot control unit (20) can control the driving unit (22) or the operating unit (23) according to the movement command signal. Accordingly, the robot main body (24) can move and perform operations according to the command of the control device (1).

[0092] The above description only briefly describes the interaction between the control unit and the robot. The robot and the control unit interact in real time, modifying and supplementing command signals to operate.

[0093] Referring again to Fig. 1, a first robot (2a) and a second robot (2b) and a control device (1) are illustrated. The points where each robot can move and be positioned can be referred to as nodes (3a to 3i). Nine nodes are illustrated in Fig. 1 for convenience of explanation. They can be defined in alphabetical order as the first node (3a), the second node (3b), the eighth node (3h), and the ninth node (3i).

[0094] When determining the path along which multiple robots move in a multi-robot operation method and system, multi-agent path finding (MAPF) technology can be utilized.

[0095] Based on Figure 1, the MAPF technology will be briefly described. The lines (A, B) indicated by broken arrows represent the paths of each robot derived by MAPF. The first robot path (A) may represent the path along which the first robot (2a) moves, and the second robot path (B) may represent the path along which the second robot (2b) moves.

[0096] That is, the first robot path (A) has the first node (3a) as the starting point and the eighth node (3h) as the destination point. To be more specific, the first robot path (A) may mean a path along which the first robot (2a) sequentially moves through the first node (3a), the fourth node (3d), the fifth node (3e), and the eighth node (3h).

[0097] The second robot path (B) starts from the third node (3c) and ends at the sixth node (3f). To be more specific, the second robot path (B) may refer to a path in which the second robot (2b) sequentially moves through the third node (3c), the second node (3b), the fifth node (3e), and the sixth node (3f). In practice, MAPF can search for paths in which more than 100 robots move simultaneously.

[0098] A control device (1) can calculate a set of movement paths (A, B) including each path taken by a plurality of robots from a starting point to a destination point, and then transmit a command signal to move each robot. A specific method for moving the robots according to the set of movement paths (A, B) can be applied using an Agent Priority Queue (hereinafter, APQ).

[0099] FIG. 4 illustrates a method for moving a plurality of robots illustrated in FIG. 1 of the present invention.

[0100] Table 1(T1) tabulates the APQ for moving the robot in Fig. 1. The first column of Table 1(T1) lists the nodes to which the robot can move. The second column of Table 1(T1) lists the robots located at each node in order of priority.

[0101] As described above, the first robot path (A) in FIG. 1 is expressed as a path along which the first robot (2a) sequentially moves through the first node (3a), the fourth node (3d), the fifth node (3e), and the eighth node (3h). Accordingly, in Table 1 (T1) of FIG. 4, the first robot (2a) is indicated in the second column corresponding to the first node (3a), the fourth node (3d), the fifth node (3e), and the eighth node (3h).

[0102] Additionally, according to the second robot path (B), the second robot (2b) is displayed in the second column corresponding to the third node (3c), the second node (3b), the fifth node (3e), and the sixth node (3f).

[0103] Referring to Table 1 (T1), both the first robot (2a) and the second robot (2b) are indicated at node 5 (3e). In fact, in the first robot path (A) and the second robot path (B), when each robot moves twice from its starting point, the first robot (2a) and the second robot (2b) arrive at node 5 (3e) simultaneously. In such cases, APQ can designate priorities for each robot.

[0104] In the second column corresponding to node 5 (3e) of Table 1 (T1), the order is expressed as robot 1 (2a), robot 2 (2b). That is, in node 5 (3e), robot 1 (2a) can be seen as having a higher priority than robot 2 (2b).

[0105] Based on APQ as shown in Table 1 (T1), since the first robot (2a) has priority at node 5 (3e), the robots can be controlled so that the first robot (2a) passes through node 5 (3e) first, and then the second robot (2b) passes through node 5 (3e) in that order.

[0106] As above, by specifying the priorities of multiple robots in each node and controlling the movement of the robots accordingly, it is possible to prevent the robots from colliding with each other.

[0107] If the first robot path (A) and the second robot path (B) described above are expressed as a path set, they can be expressed as in Table 2 (T2). Table 2 (T2) can mean a path set including each path along which a plurality of robots move.

[0108] The first column of Table 2 (T2) represents time steps. At each step, the robot's position can be moved. The second column of Table 2 (T2) indicates the node where the first robot (2a) is located for each time step. The third column of Table 2 (T2) indicates the node where the second robot (2b) is located for each time step.

[0109] Referring to Table 2 (T2), when the time step is 0, the first robot (2a) is shown as being located at the first node (3a), and the second robot (2b) is shown as being located at the third node (3c). In other words, the starting point of the first robot (2a) can be seen as being set to the first node (3a), and the starting point of the second robot (2b) can be seen as being set to the third node (3c).

[0110] And each robot moves according to its own movement path (A, B). As seen in Table 1 (T1), the first robot (2a) and the second robot (2b) pass through the fifth node (3e). At this time, since the first robot (2a) has priority, the first robot (2a) passes through the fifth node (3e) first.

[0111] As can be seen in Table 2 (T2), at time step 2, the first robot (2a) passes through the fifth node (3e), and the second robot (2b) does not move at time step 2 and can be positioned at the second node (3b) that was positioned at the previous time step 1. And at time step 3, the second robot (2b) can be controlled to pass through the fifth node (3e).

[0112] A set of paths for multiple robots to move can be established in the above manner. However, for convenience of explanation, the above example assumes two robots and nine nodes. In practice, the present invention can be applied to any number of robots or nodes.

[0113] Figure 5 illustrates a flowchart of a multi-robot operation method according to one embodiment of the present invention. Referring to Figure 5, the multi-robot operation method according to one embodiment of the present invention will be described below.

[0114] A multi-robot operation method (M) according to one embodiment of the present invention may include a movement path calculation step (100) of calculating a movement path set including each path along which a plurality of robots move from a starting point to a destination point.

[0115] In the movement path calculation step (100), the starting point or destination point of each of multiple robots is set, and an optimal path that does not cause collisions between robots can be derived. Through the movement path calculation step (100), a path set such as Table 2 (T2) of FIG. 4 can be derived.

[0116] Additionally, the multi-robot operation method may include a collision detection step (200) for determining whether a cycle conflict is included among a set of movement paths.

[0117] As mentioned above, a cycle conflict can occur when some robots in a system move in a tail-chasing manner. When a cycle conflict occurs, some robots may hinder the movement of others by moving in a tail-chasing manner. This can result in inefficiencies in the movement of all robots or even blockage.

[0118] The cycle conflict can be expressed in a formula as follows:

[0119] ...

[0120]

[0121] That is, among the set of paths derived by MAPF, if robots i, i+1, ..., j-1, j move tail-to-tail to the location of an adjacent robot, it can be seen that a cycle collision has occurred.

[0122] In addition, a multi-robot operation method according to one embodiment of the present invention may include a path modification step (300) of modifying the movement path set to a correction path set if it is determined in the collision detection step (200) that a cycle collision is included in the movement path set.

[0123] In addition, the collision detection step (200) can derive the expected collision point at which a cycle collision is expected to occur. Even before actually moving multiple robots, the collision configuration robots that cause a cycle collision and the expected collision point at which a collision is expected to occur can be calculated through the collision detection step (200).

[0124] The path modification step (300) can modify the paths of robots causing cycle collisions. Furthermore, the path modification step (300) can modify the paths of all robots affected by the cycle collision. Furthermore, the path modification step (300) can modify the paths of all robots.

[0125] In addition, a multi-robot operation method according to one embodiment of the present invention may include a correction path execution step (400) of moving the plurality of robots according to the correction path set.

[0126] When the correction path execution step (400) is performed, multiple robots can begin moving. The multiple robots can move along the correction path. The movement of the multiple robots can be implemented through the aforementioned APQ, etc.

[0127] In addition, a multi-robot operation method according to one embodiment of the present invention is performed together with the correction path execution step (400), and may further include a replanning step (500) of calculating a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the correction path set to the destination point.

[0128] The replanning step (500) may be started simultaneously with the correction path execution step (400). Alternatively, the replanning step (500) may be executed prior to the correction path execution step (400). If the replanning step (500) is performed to overlap with the correction path execution step (400), the calculation for replanning the path is performed while the robot is moving, which has the effect of saving time.

[0129] In other words, since the robot continues to move according to the corrected path set even during the path replanning calculation process, there is no need to stop the robot for replanning calculations. This effectively reduces the overall operating time.

[0130] In addition, the multi-robot operation method according to one embodiment of the present invention can determine whether a cycle collision is included in the replanned path set in the collision detection step (200) when the replanning step (500) is completed.

[0131] In addition, in a multi-robot operation method according to one embodiment of the present invention, if it is determined in the collision detection step (200) that a cycle collision is included in the replanned path set, the replanned path set can be modified into a corrected replanned path set in the path modification step (300). In addition, in the corrected path execution step (400), the plurality of robots can be moved according to the corrected replanned path set.

[0132] Meanwhile, a multi-robot operation method according to one embodiment of the present invention may repeatedly perform the path correction step (300), the correction path execution step (400), the replanning step (500), and the collision detection step (200) until it is determined in the collision detection step (200) that no cycle collision is included in the replanned path set.

[0133] On the other hand, a multi-robot operation method according to one embodiment of the present invention is characterized in that, if it is determined in the collision detection step (200) that no cycle collision is included in the replanned path set, the multiple robots are moved according to the replanned path set.

[0134] That is, if it is determined in the collision detection step (200) that no cycle collision is included in the replanned path set, the plurality of robots can move according to the replanned path set and reach the destination point.

[0135] Figure 6 specifically illustrates a movement path calculation step according to one embodiment of the present invention.

[0136] The movement path calculation step (100) of the multi-robot operation method according to one embodiment of the present invention may include a step (110) in which the starting point and the destination point of each of a plurality of robots are input.

[0137] The above starting point may refer to the current location of each of the plurality of robots. The above destination point may refer to the expected location at the end of movement according to the calculated set of movement paths.

[0138] Additionally, the movement path calculation step (100) may include a step (120) of calculating the optimal movement path for each robot. The optimal movement path may refer to a path along which multiple robots can move without causing collisions. Furthermore, the optimal movement path may refer to a path that takes the shortest time among the cases through which multiple robots can move. Furthermore, the optimal movement path may refer to a path that minimizes resource waste among the cases.

[0139] The above operation process can be performed in the operation unit (13) of the control device (1).

[0140] Additionally, the movement path calculation step (100) may include a step (130) of deriving a path set including paths along which multiple robots move, through the above-described calculation process.

[0141] The above path set may include all paths traveled by multiple robots. That is, if a total of 100 robots are operated, the above path set may include 100 paths.

[0142] The derived path set can be stored in the storage unit (12) of the control device (1).

[0143] The above movement path calculation step (100) can derive a path set including each movement path of multiple robots through the above process.

[0144] Figure 7 specifically illustrates a path modification step (300) according to one embodiment of the present invention.

[0145] The above path correction step (300) may include a collision robot designation step (310) that defines one of the collision configuration robots determined to constitute a cycle collision in the collision detection step among the plurality of robots as a collision robot.

[0146] The collision robot designation step (310) may refer to a step of determining a specific robot to be stopped in order to proceed with path correction, which will be described later.

[0147] Specifically, if the robots that constitute a cycle collision are Robot 1, Robot 2, Robot 3, and Robot 4, they can be defined as collision-constituting robots. In other words, this can mean that Robot 1, Robot 2, Robot 3, and Robot 4 move in tandem from one time step to the next.

[0148] Based on the example above, in the collision robot designation step (310), any one of the first robot, second robot, third robot, and fourth robot can be designated as a collision robot. A specific method for designating a collision robot will be described later.

[0149] Meanwhile, the path modification step (300) may include a collision robot modification step (320) that modifies the path along which the collision robot specified in the collision robot designation step (310) moves.

[0150] The path correction step (300) may begin with correcting the path of the collision robot. A specific method for correcting the path of the collision robot will also be described later.

[0151] Additionally, the path correction step (300) may include an impact robot correction step (330) that corrects the path of another robot affected by the collision robot correction step (320).

[0152] That is, if the movement path of the collision robot is changed in the collision robot modification step (320), the movement of other robots may be hindered or blocked by the changed path of the collision robot.

[0153] At this time, in the influence robot modification step (330), the paths of the influence robots affected by the changed movement path of the collision robot can be shortened. More specifically, the movement paths of the influence robots already calculated can be modified to the point affected by the changed path of the collision robot. In other words, the arrival point of the influence robot can be modified to the final point where it can proceed without obstruction.

[0154] This method offers the advantage of maximizing the use of previously calculated movement paths. Therefore, by utilizing previously calculated movement paths, you can save resources without having to start over from scratch to modify the movement path.

[0155] Additionally, the path correction step (300) may include a correction path acquisition step (340) for obtaining the correction path set derived through the collision robot correction step (320) and the impact robot correction step (330).

[0156] The above-described correction path acquisition step (340) can derive a path set that includes both the correction path of the collision robot derived in the collision robot correction step (330) and the correction path of the influence robot derived in the influence robot correction step (330). That is, the correction path set can be understood as including all of the correction path of the collision robot described above, the correction path of the influence robot, and the previously calculated path whose path has not been corrected.

[0157] FIG. 8 specifically illustrates a collision robot designation step (310) according to one embodiment of the present invention.

[0158] The above collision robot designation step (310) may include a stop effect determination step (311) for determining the stop effect defined as the degree to which the movement of other robots is affected if the collision configuration robots stop at the expected collision time determined to be a cycle collision to occur in the collision detection step (200).

[0159] The above-mentioned interruption effect determination step (311) can determine the interruption effect based on APQ. As described above, APQ indicates the priority of robots passing through each node for each node.

[0160] Based on APQ, if a specific robot has a high priority in a specific node, then when the specific robot stops, robots with lower priority than the specific robot in the specific node are affected.

[0161] For example, referring to Table 1 (T1) of Fig. 4, it can be seen that the priority of the first robot (2a) in the fifth node (3e) is higher than that of the second robot (2b). In other words, it can be seen that when the first robot (2a) stops moving, the second robot (2b), which is in the lower priority, is also affected.

[0162] Therefore, based on Table 1 (T1), it can be seen that the stopping effect of the first robot (2a) is higher than that of the second robot (2b).

[0163] In the above-mentioned stopping effect judgment step (311), the stopping effect for multiple robots can be calculated using the above method.

[0164] In other words, in the above-mentioned interruption effect determination step (311), the interruption effect of each of the conflicting robots can be calculated and determined. A robot with a higher priority in APQ can be determined to have a higher interruption effect. In other words, a robot with a higher node priority can be determined to have a higher interruption effect.

[0165] In addition, the collision robot designation step (310) may include a disruption effect comparison step (312) for selecting the robot determined to have the lowest disruption effect in the disruption effect determination step (311) as the collision robot.

[0166] That is, the stop effect judgment step (311) can select the robot with the lowest node priority among the above collision configuration robots and designate it as a collision robot.

[0167] As described above, when node priorities are low or the disruption effect is low, the number of affected robots is minimal when the robot's path is modified. Therefore, when collision robots are designated using the method described above, the path modification process described below can be performed with minimal path modifications. Consequently, this method effectively enables efficient multi-robot operation with minimal resources.

[0168] FIG. 9 specifically illustrates a collision robot correction step (320) according to one embodiment of the present invention.

[0169] The above collision robot correction step (320) may include a stop time determination step (322) for determining a stop time for stopping the collision robot by comparing the expected collision time determined to be a cycle collision to occur in the collision detection step (200) with a preset reference collision time.

[0170] Prior to the above-described stop time determination step (322), a reference collision time setting step (321) for modifying the reference collision point may be performed. The reference collision time setting step (321) may be viewed as a process for determining a reference value required for the calculation of the collision robot modification step (320). The reference collision time setting step (321) may be omitted, and the setting value stored in the existing control device (1) may be utilized as is.

[0171] Specifically describing the stop time determination step (322), the stop time determination step (322) can be understood as a step for deriving a stop time to stop the collision robot in order to correct the path of the collision robot.

[0172] Here, "stopping" can include a physical stop. However, it can also mean that the robot is considered to have stopped moving during the computational process. In other words, the collision robot stopping at the stop time can be understood as being considered to have arrived at the point where the collision robot was at the stop time.

[0173] The above stopping time determination step (322) may include a step (3221) of comparing the expected collision time with the reference collision time.

[0174] Additionally, the stop time determination step (322) may include a step (3222) of setting the difference between the expected collision time and the reference collision time as the stop time if the expected collision time is greater than the reference collision time through the comparison step (3221).

[0175] Additionally, the stop time determination step (322) may include a step (3223) of setting the stop time to 0 if the expected collision time is not greater than the reference collision time through the comparison step (3221).

[0176] Additionally, the collision robot correction step (320) may include a step (323) of correcting the movement path of the collision robot by stopping the collision robot at the stop time determined in the stop time determination step (322) to obtain a correction path of the collision robot.

[0177] That is, if the stopping time is determined as the difference between the expected collision time and the reference collision time, the collision robot's path can be corrected so that it stops at the said stopping time. That is, the path can be corrected so that the movement of the collision robot stops at the position at the stopping time. That is, the stopping position of the collision robot can be set as the stopping point according to the corrected path.

[0178] On the other hand, if the stopping time is determined to be 0, it may mean that the collision robot does not move along the calculated path and remains stopped at the starting point. In other words, the collision robot may set the starting point as the stopping point according to the corrected path.

[0179] The correction path of the collision robot can be derived using the method described above. Once the correction path of the collision robot is derived, the affected robot affected by the correction path of the collision robot can be identified. In other words, the affected robot can be defined as a robot affected by the path progression, assuming that the collision robot stops at the aforementioned stopping time.

[0180] In such a case, in the influence robot modification step (330), the path of the influence robots can be modified so that the influence robots move to the final point where each can proceed.

[0181] Figure 10 specifically illustrates a replanning step according to one embodiment of the present invention.

[0182] The above re-planning step (500) may include a starting point resetting step (510) that resets the stopping points of the plurality of robots determined according to the above correction path set to the new starting points.

[0183] In the above-described correction path acquisition step (323), a correction path can be acquired with a stop point at a location different from the previously set destination point. Accordingly, the replanning step (500) can calculate a replanned path by setting the destination point of the correction path as a new starting point.

[0184] Additionally, the replanning step (500) may include a step (530) of calculating a path along which the plurality of robots move based on the new starting point.

[0185] Before calculating a new path, a destination point resetting step (520) may be performed to reset the destination points of multiple robots. However, when calculating a new path, the path may be calculated based on the initially designated destination point. Therefore, the destination point resetting step (520) may be omitted.

[0186] The replanning step (500) may include a step (540) of obtaining a replanned path set including the paths along which each of the plurality of robots moves through the above-described operation step (530). The replanned path set may include the paths of only some of the plurality of robots. Alternatively, the replanned path set may include all paths of each of the plurality of robots.

[0187] The replanned path set derived through the above replanning step (500) can be consistent with the corrected path set. Therefore, even if a cycle collision occurs, the robot does not actually stop and can continue to move according to the replanned path set. This method has the effect of improving efficiency in multi-robot operation.

[0188] Meanwhile, a multi-robot operation system (S) according to another embodiment of the present invention may include a plurality of robots (2) and a control device (1) that controls the plurality of robots (2).

[0189] The above plurality of robots (2) may each include a robot body (24), a driving unit (22) provided in the robot body (24) to move the robot body (24), a robot control unit (20) that controls the operation of the driving unit (22), and a robot communication unit (21) provided to communicate with the outside.

[0190] In addition, the control device (10) may include a communication unit (11) equipped to communicate with the robot communication unit (21), a storage unit (12) equipped to store data, and a control unit (10) equipped to process communication data acquired through the communication unit (11) or data stored in the storage unit.

[0191] In addition, the control unit (10) calculates a set of movement paths including each path along which the plurality of robots (2) move from a starting point to a destination point, determines whether a cycle conflict is included in the set of movement paths, and if it is determined that a cycle conflict is included in the set of movement paths, the set of movement paths can be modified into a correction path set.

[0192] In addition, the communication unit (11) can transmit a signal to the robot communication unit (21) to move the plurality of robots (2) according to the correction path set.

[0193] Meanwhile, the control unit (10) can modify the movement path set to the correction path set, and then calculate a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the correction path set to the destination point.

[0194] In addition, the communication unit (11) can transmit a signal to the robot communication unit (21) to move the plurality of robots (2) according to the replanned path set.

[0195] Meanwhile, a control device according to one embodiment of the present invention includes a communication unit (11) that is provided to enable communication with the outside, and a control unit (10) that is provided to process communication data obtained through the communication unit, wherein the control unit (10) calculates a movement path set including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the movement path set, and, if it is determined that the movement path set includes a cycle conflict, modifies the movement path set into a correction path set, and moves the plurality of robots according to the correction path set.

[0196] In addition, the control unit (11) calculates a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the correction path set to the destination point, and can move the plurality of robots according to the replanned path set.

[0197] The content described above regarding the multi-robot operation method according to one embodiment of the present invention may be equally applied to the multi-robot operation system and control device.

[0198] The above-described control device (1) may be configured separately from a plurality of robots (2a, 2b) as illustrated in FIG. 1. However, the present invention is not limited thereto, and the control device (1) may be provided to some of the plurality of robots. In addition, the control device (1) may be provided to each of the plurality of robots. The embodiment illustrated in FIG. 1 is merely one embodiment, and the present invention is not limited thereto.

[0199] The embodiments of the present invention described above are not implemented only through devices and methods, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present invention or a recording medium on which the program is recorded, and such implementation can be easily implemented by an expert in the technical field to which the present invention belongs based on the description of the embodiments described above.

[0200] Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

Claims

1. In the method of operating multiple robots, A movement path calculation step for calculating a set of movement paths including each path that a plurality of robots move from a starting point to a destination point; A collision detection step for determining whether a cycle conflict is included among the above set of movement paths; A path modification step for modifying the movement path set to a correction path set when it is determined in the collision detection step that a cycle collision is included in the movement path set; and A multi-robot operation method comprising a correction path execution step of moving the plurality of robots according to the correction path set.

2. In paragraph 1, A multi-robot operation method, which is performed together with the above correction path execution step, further comprising a replanning step of calculating a replanning path set including each path along which the plurality of robots move from a new starting point determined according to the correction path set to the destination point.

3. In paragraph 2, A multi-robot operation method characterized in that, when the above replanning step is completed, it is determined in the collision detection step whether a cycle collision is included in the set of replanned paths.

4. In paragraph 3, If it is determined that a cycle collision is included in the replanned path set in the collision detection step, the replanned path set is modified into a corrected replanned path set in the path modification step, A multi-robot operation method characterized in that, in the above correction path execution step, the plurality of robots are moved according to the correction replanning path set.

5. In paragraph 3, A multi-robot operation method characterized in that the path correction step, the correction path execution step, the replanning step, and the collision detection step are repeated until it is determined that no cycle collision is included in the replanned path set in the collision detection step.

6. In paragraph 3, A multi-robot operation method characterized in that, if it is determined in the collision detection step that no cycle collision is included in the replanned path set, the plurality of robots are moved according to the replanned path set.

7. In paragraph 1, The above path modification steps are A collision robot designation step of defining one of the collision configuration robots determined to constitute a cycle collision in the collision detection step among the plurality of robots as a collision robot; A collision robot modification step for modifying the path along which the collision robot moves; An impact robot correction step for correcting the path of another robot affected by the above collision robot correction step; and A multi-robot operation method, characterized by comprising a correction path acquisition step for acquiring the correction path set derived through the collision robot correction step and the impact robot correction step.

8. In paragraph 7, The above collision robot designation step is A stop effect determination step for determining the stop effect defined as the extent to which the movement of other robots is affected if the collision configuration robots stop at the expected collision point where it is determined that a cycle collision will occur in the collision detection step; and A multi-robot operation method characterized by including a disruption effect comparison step for selecting a robot determined to have the lowest disruption effect as the collision robot in the disruption effect determination step.

9. In paragraph 7, The above collision robot modification steps are A stop time determination step for determining a stop time to stop the collision robot by comparing the expected collision time determined to be likely to occur in the collision detection step with a preset reference collision time; A multi-robot operation method comprising: a step of obtaining a correction path of the collision robot by correcting the movement path of the collision robot by stopping the collision robot at the stop time; 10. In paragraph 2, The above replanning step is a starting point resetting step of resetting the stopping points of the plurality of robots determined according to the set of correction paths to the new starting points; A step of calculating a path along which the plurality of robots move based on the new starting point; and A multi-robot operation method characterized by comprising a step of obtaining the replanned path set through the above-described calculating step.

11. In a multi-robot operating system including a plurality of robots and a control device for controlling the plurality of robots, The above plurality of robots each include a robot body, a driving unit provided in the robot body to move the robot body, a robot control unit that controls the operation of the driving unit, and a robot communication unit provided to communicate with the outside. The above control device includes a communication unit that is equipped to communicate with the robot communication unit, a storage unit that is equipped to store data, and a control unit that is equipped to process communication data obtained through the communication unit or data stored in the storage unit. The control unit calculates a set of movement paths including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the set of movement paths, and if it is determined that a cycle conflict is included in the set of movement paths, modifies the set of movement paths into a correction path set. A multi-robot operating system characterized in that the communication unit transmits a signal to the robot communication unit to move the plurality of robots according to the set of correction paths.

12. In paragraph 11, The control unit modifies the set of movement paths into the set of correction paths, and then calculates a set of replanned paths including each path along which the plurality of robots move from a new starting point determined according to the set of correction paths to the destination point. A multi-robot operating system characterized in that the communication unit transmits a signal to the robot communication unit to move the plurality of robots according to the replanned path set.

13. In a control device that controls multiple robots, The above control device includes a communication unit that is equipped to communicate with the outside, and a control unit that is equipped to process communication data obtained through the communication unit. A control device characterized in that the control unit calculates a set of movement paths including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the set of movement paths, and, if it is determined that the set of movement paths includes a cycle conflict, modifies the set of movement paths into a set of correction paths and moves the plurality of robots according to the set of correction paths.

14. In paragraph 13, A control device characterized in that the control unit calculates a replanned path set including each path along which the plurality of robots move from a new starting point determined according to the set of correction paths to the destination point, and moves the plurality of robots according to the replanned path set.

15. In a multi-robot operating system that controls multiple robots, A communication unit that communicates with the above robot; A storage unit provided to store data required for controlling the above robot; It includes a control unit that is equipped to process communication data obtained through the communication unit or data stored in the storage unit; The control unit calculates a set of movement paths including each path along which the plurality of robots move from a starting point to a destination point, determines whether a cycle conflict is included in the set of movement paths, and if it is determined that a cycle conflict is included in the set of movement paths, modifies the set of movement paths into a correction path set. A multi-robot operating system characterized in that the communication unit transmits a signal to the robots to move the plurality of robots according to the set of correction paths.

Images