Method and system for agent group scheduling, processor and storage medium
By determining the anti-collision path between candidate agents and shared space agents in the intellectual group, the problem of insufficient flexibility in scheduling of intelligent group in the prior art is solved, and the flexibility and security of dynamic scheduling are achieved.
Patent Information
- Application Number
- CN202411402250.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-09
- Publication Date
- 2025-02-14
AI Technical Summary
In the prior art, the flexibility of intelligent group scheduling is low and cannot flexibly deal with dynamic time during the tower crane process, which is suitable for insufficient dynamic scheduling scenarios.
By obtaining the task parameters of the task to be allocated, determine the candidate intellectual group and the shared space intellectual group that can execute the task to be allocated, and determine the anti-collision between the candidate intellectual and the shared space intellectual group based on the task parameters and the current task path. The path is ultimately determined based on the anti-collision path.
It realizes the flexibility of dynamic scheduling of intelligent groups, can be applicable to dynamic scheduling scenarios, improves the real-time and adaptability of scheduling, and avoids collisions between agents.
Smart Images
Figure CN119439900A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of operating machinery, and in particular to a method, system, processor and storage medium for intelligent group scheduling. Background Art
[0002] An intelligent group refers to a motion system composed of multiple intelligent agents. For example, in an engineering operation scenario, an intelligent group can include truck cranes, crawler cranes, robots, tower cranes, and so on. Taking a complex tower group as an example, a complex tower group is a motion system composed of multiple or even hundreds of tower cranes. When dealing with large projects and super-high building projects, due to limited time and space, it is necessary to deploy multiple or even more tower cranes with overlapping operating areas, thus forming a complex tower group system. At present, the optimization of scheduling for complex tower groups mainly focuses on the optimization of tower crane service sequence. Mixed integer programming, heuristic optimization, and machine learning methods are used to optimize the service sequence of multiple tower cranes with overlapping areas, specifically optimizing the task allocation, loading point selection, and tower crane operation sequence in the overlapping area. However, the real-time performance of service scheduling optimization is poor, and it cannot flexibly respond to the dynamic time in the tower crane process, and is not suitable for dynamic scheduling scenarios. Therefore, the existing technology has the problem of low flexibility in intelligent group scheduling. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide a method, system, processor and machine-readable storage medium for intelligent group scheduling, so as to solve the problem of low flexibility in intelligent group scheduling existing in the prior art.
[0004] In order to achieve the above-mentioned object, a first aspect of an embodiment of the present application provides a method for scheduling an agent group, the method comprising:
[0005] Obtaining task parameters of the task to be assigned, including the task start point and task end point;
[0006] Determine a group of candidate agents that can perform the task to be assigned based on the task parameters;
[0007] Determine a shared space agent group corresponding to each candidate agent in the candidate agent group, wherein the shared space agents in the shared space agent group are in operation and the operation areas overlap with the operation areas of the candidate agents;
[0008] Determine, according to the task parameters, an anti-collision path between each candidate agent and each shared space agent in the corresponding shared space agent group;
[0009] A target path is determined based on the collision-avoiding path and a target agent executes the target path.
[0010] In an embodiment of the present application, a target path and a target intelligent agent that executes the target path are determined based on an anti-collision path, including: for any candidate intelligent agent, determining whether there is a common anti-collision path between all shared space intelligent agents of the candidate intelligent agent according to the anti-collision path; determining the candidate intelligent agent with a common anti-collision path as the target intelligent agent, and determining the common anti-collision path as the target path executed by the target intelligent agent.
[0011] In an embodiment of the present application, an anti-collision path between each candidate intelligent agent and each shared space intelligent agent in the corresponding shared space intelligent agent group is determined according to the task parameters, including: for any candidate intelligent agent, determining the shared operation area of the candidate intelligent agent and the corresponding shared space intelligent agents; determining an initial path set of the candidate intelligent agent according to the task parameters; obtaining the current task path of each shared space intelligent agent; and determining an anti-collision path between the candidate intelligent agent and each corresponding shared space intelligent agent in the initial path set based on the task parameters, the current task path and the shared operation area.
[0012] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, an anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, including: when the current task path passes through the shared work area and the task starting point and / or the task end point are located in the shared work area, determining that there is no anti-collision path in the initial path set.
[0013] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, an anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, including: when the current task path passes through the shared work area, and both the task starting point and the task end point are not located in the shared work area, all paths in the initial path set that do not pass through the shared work area are determined as anti-collision paths.
[0014] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, the anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, including: when the current task path does not pass through the shared work area, all paths in the initial path set are determined as anti-collision paths.
[0015] In an embodiment of the present application, when the intelligent agent is a tower crane, an anti-collision path between each candidate intelligent agent and each shared space intelligent agent in the corresponding shared space intelligent agent group is determined according to the task parameters, including: for any candidate tower crane and any corresponding shared space tower crane, determining the type of twin towers composed of the candidate tower crane and each shared space tower crane, the twin tower type includes an equal-height twin tower type and a high-low twin tower type; obtaining the current operating parameters of the shared space tower crane, the operating parameters include a task starting point, a task end point and a current task path; when the twin tower type is a high-low twin tower type, determining a first shared operating area, the first shared operating area being the overlapping area of the outer circle of the higher tower in the candidate tower crane and the shared space tower crane and the boom operating area of the lower tower, the outer circle being the center of the higher tower and the task The line connecting the starting point and the task end point is used as the larger circle in the circle with the radius; the initial anti-collision path of the candidate tower crane relative to the shared space tower crane is determined according to the operation parameters, task parameters and the first shared operation area; the second shared operation area is determined, and the second shared operation area is the overlapping area of the outer circle of the higher tower and the balance arm operation area of the lower tower; the first distance is compared with the second distance, the first distance is the sum of the radius of the outer circle of the higher tower and the length of the balance arm of the lower tower, and the second distance is the distance between the tower centers of the candidate tower crane and the shared space tower crane; when the first distance is less than the second distance, the path passing through the second shared operation area in the initial anti-collision path is eliminated to obtain the anti-collision path; when the first distance is greater than or equal to the second distance, the initial anti-collision path is determined as the anti-collision path.
[0016] A second aspect of an embodiment of the present application provides a processor configured to execute the above-mentioned method for scheduling a group of intelligent agents.
[0017] A third aspect of an embodiment of the present application provides a system for scheduling an agent group, comprising: an agent group; and the above-mentioned processor.
[0018] A fourth aspect of an embodiment of the present application provides a machine-readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the above-mentioned method for scheduling a group of intelligent agents is implemented.
[0019] The above technical scheme obtains the task parameters of the task to be assigned, including the task starting point and the task end point, and then determines the candidate agent group that can execute the task to be assigned according to the task parameters, and determines the shared space agent group corresponding to each candidate agent in the candidate agent group, the shared space agent in the shared space agent group is in operation and the operation area overlaps with the operation area of the candidate agent, and then determines the anti-collision path between each candidate agent and each shared space agent in the corresponding shared space agent group according to the task parameters, and finally determines the target path and the target agent that executes the target path based on the anti-collision path. The present application can determine the agent and path that can execute the task to be assigned based on the task parameters of the task to be assigned, combined with the relative relationship and actual operation status of the operation areas between each agent in the current agent group, and can be applied to the scenario of dynamic scheduling of agent groups, with high flexibility.
[0020] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the accompanying drawings:
[0022] Figure 1 A flowchart of a method for scheduling a group of intelligent agents provided in an embodiment of the present application;
[0023] Figure 2 A schematic diagram of a bird's-eye view of the tower crane movement provided in a specific embodiment of the present application;
[0024] Figure 3 A schematic diagram of the omnidirectional rotation scheduling of a tower crane provided in an embodiment of the present application;
[0025] Figure 4 A schematic diagram of a kinematic geometry model of a climbing "double tower" system provided in a specific embodiment of the present application;
[0026] Figure 5 A schematic diagram of a geometric model of a high-low "double-tower" system provided in an embodiment of the present application;
[0027] Figure 6 A schematic diagram of an algorithm flow for dynamic scheduling of complex tower groups provided in an embodiment of the present application. DETAILED DESCRIPTION
[0028] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0029] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0030] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0031] Figure 1 A flow chart of a method for scheduling a group of intelligent agents provided in an embodiment of the present application. Figure 1 As shown, the embodiment of the present application provides a method for agent group scheduling, which is described by taking the method applied to a processor as an example. The method may include the following steps:
[0032] Step S101, obtaining task parameters of the task to be assigned, the task parameters including the task start point and the task end point.
[0033] Step S102, determining a group of candidate agents that can execute the task to be assigned according to the task parameters.
[0034] Step S103, determining a shared space agent group corresponding to each candidate agent in the candidate agent group, wherein the shared space agents in the shared space agent group are in operation and their operation areas overlap with the operation areas of the candidate agents.
[0035] Step S104, determining the anti-collision path between each candidate agent and each shared space agent in the corresponding shared space agent group according to the task parameters.
[0036] Step S105 , determining a target path and a target agent that executes the target path based on the anti-collision path.
[0037] Specifically, the agent can be an operating equipment such as a truck crane, a crawler crane, a robot and a tower crane, and the agent group can refer to a motion system composed of multiple agents. It can be understood that each agent in the embodiment of the present application corresponds to a fixed operating area, each agent operates in the corresponding operating area, and the operating areas of each agent can overlap partially. During the operation of the agent group, after the processor receives the task parameters of the task to be assigned input by the user, it can preliminarily determine the candidate agent group that can execute the task to be assigned according to the task parameters of the task to be assigned. Among them, the task parameters include the task starting point and the task end point, that is, the starting position and the end point of the task to be assigned. In an example, the processor can determine the candidate agent group that can execute the task to be assigned according to the task parameters of the task to be assigned, as well as the operating area and operating state of each agent. The operating state of the agent includes an idle state and a running state, that is, the candidate agent group consists of multiple agents including the task starting point and the task end point in the corresponding operating area and the operating state is an idle state.
[0038] Furthermore, considering that the operating areas of the agents partially overlap, in order to prevent collisions from occurring during the execution of the task, the processor may further determine a shared space agent group of each candidate agent in the candidate agent group, where the shared space agents in the shared space agent group are in operation and their operating areas overlap with the operating areas of the corresponding candidate agents. It is understood that the candidate agent group may have no corresponding shared space agent, or may have one or more shared space agents.
[0039] Next, based on the task parameters, the anti-collision paths between each candidate agent and each shared space agent in the corresponding shared space agent group are determined. Among them, the anti-collision path between the candidate agent and a shared space agent refers to a path that can perform the task to be assigned and will not collide with the shared space agent. In one example, the processor can determine the shared working area of the candidate agent's working area with the shared space agent based on the working areas of the candidate agent and each shared space agent in the corresponding shared space agent group, and then determine all the paths that the candidate agent can perform the task to be assigned and do not pass through the shared working area based on the task parameters, that is, obtain the anti-collision path of the candidate agent. It can be understood that the anti-collision path of the candidate agent relative to any shared space agent can be zero, one, or more.
[0040] Finally, the target path and the target agent that executes the target path are determined based on the anti-collision path. Among them, the target path refers to an optional path that can execute the task to be assigned and will not collide with any agent, and the target agent refers to a candidate agent that can execute the task to be assigned according to the target path. In one example, the processor can determine the anti-collision path of all shared working areas that do not pass through the working area of the corresponding candidate agent as the target path, and determine the corresponding candidate agent as the target agent that executes the target path. It can be understood that the processor can output each target agent and the target path corresponding to each target agent for user selection, and then execute the task to be assigned in response to the user's selection.
[0041] Traditional research on scheduling optimization of intelligent agent groups mainly focuses on the optimization of the service order of intelligent agents. Usually, methods such as mixed integer programming, heuristic optimization and machine learning are used to optimize the task allocation and service order in the overlapping areas of intelligent agent groups with overlapping operation areas. However, it mainly focuses on the optimization of the overall task, with poor real-time performance, and cannot flexibly respond to the dynamic scheduling needs during the operation of the intelligent agent, that is, any dynamic event will affect the scheduling plan generated in advance. However, during the operation of the intelligent agent group, the embodiment of the present application responds to the task requirements in a timely manner according to the task parameters of the task to be assigned and the operation parameters of each intelligent agent, determines the target intelligent agent and target path that can execute the task to be assigned, and outputs or further provides it to the user for selection. Compared with traditional methods, the present application can meet the dynamic scheduling needs during the operation of the intelligent agent group and has higher flexibility.
[0042] The above technical scheme obtains the task parameters of the task to be assigned, including the task starting point and the task end point, and then determines the candidate agent group that can execute the task to be assigned according to the task parameters, and determines the shared space agent group corresponding to each candidate agent in the candidate agent group, the shared space agent in the shared space agent group is in operation and the operation area overlaps with the operation area of the candidate agent, and then determines the anti-collision path between each candidate agent and each shared space agent in the corresponding shared space agent group according to the task parameters, and finally determines the target path and the target agent that executes the target path based on the anti-collision path. The embodiment of the present application can determine the agent and path that can execute the task to be assigned based on the task parameters of the task to be assigned, combined with the relative relationship and actual operation status of the operation areas between each agent in the current agent group, and can be applied to the scenario of dynamic scheduling of agent groups, with high flexibility.
[0043] In an embodiment of the present application, determining a target path based on an anti-collision path and a target agent that executes the target path may include: for any candidate agent, determining whether there is a common anti-collision path between all shared space agents of the candidate agent based on the anti-collision path; determining the candidate agent with a common anti-collision path as the target agent, and determining the common anti-collision path as the target path for the target agent to execute.
[0044] It can be understood that the anti-collision path between a candidate agent and a shared space agent refers to a path that can perform the task to be assigned and will not collide with the shared space agent. The target path refers to an optional path that can perform the task to be assigned and will not collide with any agent, and the target agent refers to a candidate agent that can perform the task to be assigned according to the target path.
[0045] Specifically, in order to determine the target path and target agent of the task to be assigned, after determining the anti-collision path of each candidate agent relative to the corresponding shared space agent, considering that the target path must meet the requirement of not colliding with any agent, for each candidate agent, the processor can use the intersection method to determine whether there is an intersection between the anti-collision paths of the candidate agent relative to each shared space agent, that is, whether there is a common anti-collision path. It can be understood that the common anti-collision path will not collide with any shared space agent corresponding to the candidate agent. Therefore, in the case where the candidate agent has the common anti-collision path, the common anti-collision path can be determined as the target path, and the candidate agent can be determined as the target agent. It can be understood that for a candidate agent that does not have a common anti-collision path, it is directly determined that the candidate agent cannot be assigned to perform the task to be assigned. In this way, it is possible to prevent collision incidents from occurring during the operation of the agent, thereby improving the safety of the operation.
[0046] In an embodiment of the present application, based on the task parameters, determining the anti-collision path between each candidate intelligent agent and each shared space intelligent agent in the corresponding shared space intelligent agent group can include: for any candidate intelligent agent, determining the shared operation area of the candidate intelligent agent and the corresponding shared space intelligent agents; determining the initial path set of the candidate intelligent agent based on the task parameters; obtaining the current task path of each shared space intelligent agent; based on the task parameters, the current task path and the shared operation area, determining the anti-collision path between the candidate intelligent agent and the corresponding shared space intelligent agents in the initial path set.
[0047] It can be understood that the shared operation area refers to the area where the operation area of the candidate agent and the corresponding shared space agent overlap. The task path refers to the operation path for performing a certain task. The initial path set of the candidate agent refers to the set of all possible paths that the candidate agent can perform the task to be assigned without considering collision events. Combined with the actual operation situation of the shared space agent, it is considered that the current task path of the shared space agent may not pass through the shared operation area. Therefore, the processor can screen the initial anti-collision path of the candidate agent in combination with the current task path of the shared space agent to determine the anti-collision path of the candidate agent relative to the shared space agent. In this way, combined with the actual operation situation, the optional path is prevented from being missed, and the accuracy of the output results is further improved.
[0048] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, the anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, which may include: when the current task path passes through the shared work area and the task starting point and / or the task end point are located in the shared work area, determining that there is no anti-collision path in the initial path set.
[0049] It can be understood that for a candidate agent and any corresponding shared space agent, when the current task path of the shared space agent passes through the shared operation area, it means that all paths passing through the shared operation area cannot be considered, otherwise a collision may occur. If the task starting point and / or task end point of the task to be assigned is located in the shared operation area, it means that all paths of the candidate agent to execute the task to be assigned will pass through the shared operation area, and there is a risk of collision. Therefore, there is no anti-collision path in the initial path set of the candidate agent.
[0050] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, the anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, which may include: when the current task path passes through the shared work area, and both the task starting point and the task end point are not located in the shared work area, all paths in the initial path set that do not pass through the shared work area are determined as anti-collision paths.
[0051] It can be understood that for a candidate agent and any corresponding shared space agent, when the current task path of the shared space agent passes through the shared operation area, it means that the paths passing through the shared operation area cannot be considered, otherwise a collision may occur. If the task start point and task end point of the task to be assigned are not located in the shared operation area, all paths passing through the shared operation area in the initial path set are eliminated, and all paths that do not pass through the shared operation area are determined as the anti-collision paths of the candidate agent relative to the shared space agent.
[0052] In an embodiment of the present application, based on task parameters, the current task path and the shared work area, the anti-collision path between the candidate intelligent agents in the initial path set and the corresponding shared space intelligent agents is determined, which may include: when the current task path does not pass through the shared work area, all paths in the initial path set are determined as anti-collision paths.
[0053] It can be understood that, for a candidate agent and any corresponding shared space agent, when the current task path of the shared space agent does not pass through the shared work area, it means that all paths in the initial path set corresponding to the candidate agent will not collide with the shared space agent. At this time, all paths in the initial path set can be determined as the anti-collision paths of the candidate agent relative to the shared space agent.
[0054] In an embodiment of the present application, when the intelligent agent is a tower crane, an anti-collision path between each candidate intelligent agent and each shared space intelligent agent in the corresponding shared space intelligent agent group is determined according to the task parameters, including: for any candidate tower crane and any corresponding shared space tower crane, determining the type of twin towers composed of the candidate tower crane and each shared space tower crane, the twin tower type includes an equal-height twin tower type and a high-low twin tower type; obtaining the current operating parameters of the shared space tower crane, the operating parameters include a task starting point, a task end point and a current task path; when the twin tower type is a high-low twin tower type, determining a first shared operating area, the first shared operating area being the overlapping area of the outer circle of the higher tower in the candidate tower crane and the shared space tower crane and the boom operating area of the lower tower, the outer circle being the center of the higher tower and the task The line connecting the starting point and the task end point is used as the larger circle in the circle with the radius; the initial anti-collision path of the candidate tower crane relative to the shared space tower crane is determined according to the operation parameters, task parameters and the first shared operation area; the second shared operation area is determined, and the second shared operation area is the overlapping area of the outer circle of the higher tower and the balance arm operation area of the lower tower; the first distance is compared with the second distance, the first distance is the sum of the radius of the outer circle of the higher tower and the length of the balance arm of the lower tower, and the second distance is the distance between the tower centers of the candidate tower crane and the shared space tower crane; when the first distance is less than the second distance, the path passing through the second shared operation area in the initial anti-collision path is eliminated to obtain the anti-collision path; when the first distance is greater than or equal to the second distance, the initial anti-collision path is determined as the anti-collision path.
[0055] It can be understood that when the agent is a tower crane, considering that the heights of adjacent tower cranes may be consistent or inconsistent during actual operation, that is, there may be a situation of a climbing double tower and a high-low double tower, therefore, the embodiment of the present application can determine the corresponding anti-collision path according to the double tower type composed of the candidate agent and the corresponding shared space agent. Among them, the double tower type includes a climbing double tower and a high-low double tower. The climbing double tower means that the heights of the two tower cranes are consistent, and the high-low double tower means that there is a height difference between the two tower cranes.
[0056] Specifically, for any candidate tower crane and any corresponding shared space tower crane, the processor can determine the type of twin towers formed by the two according to the basic parameters of the tower crane, that is, the height data. In one example, when the twin tower type formed by the candidate tower crane and the shared space tower crane is an equal-height twin tower, the shared operating area of the two is the overlapping area of the original operating areas of the two, and the original operating area of the tower crane is the area of a circle with the boom as the radius. Considering that the technicians will not overlap the boom and the balance arm between the two equal-height tower cranes during the actual operation, the anti-collision path of the candidate tower crane relative to the shared space tower crane can be determined according to the method in the above embodiment.
[0057] In another example, when the type of twin towers formed by the candidate tower crane and the shared space tower crane is a high-low twin tower, in order to avoid the intersection operation of complex curves in three-dimensional space, the embodiment of the present application uses an outer circle to determine the shared working area, and the outer circle is a larger circle in the circle with the radius being the line connecting the tower center of the higher tower of the candidate tower crane and the shared space tower crane and the task starting point and the task end point respectively.
[0058] Specifically, when the twin tower type is a high-low twin tower, the first shared operation area is determined, and the first shared operation area is the overlapping area of the outer circle of the higher tower in the candidate tower crane and the shared space tower crane and the original operation area of the lower tower. Based on the first shared operation area, the task parameters and the current task path of the shared space tower crane, the initial anti-collision path of the candidate tower crane relative to the shared tower crane is determined by the method in the above embodiment. In this way, the intersection operation of complex curves in the three-dimensional space can be avoided, the operation process is simplified, and the operation efficiency is improved.
[0059] Furthermore, taking into account that the lifting arms and balancing arms between high and low tower cranes may overlap during actual operation, the processor can determine a second shared operating area, which is the overlapping area of the outer circle of the higher tower in the candidate tower crane and the shared space tower crane and the balancing arm operating area of the lower tower, and the balancing arm operating area is a circle with the balancing arm as the radius.
[0060] Next, the first distance is compared with the second distance, where the first distance is the sum of the outer circle radius of the higher tower and the length of the balance arm of the lower tower, and the second distance is the distance between the tower centers of the candidate tower crane and the shared space tower crane. In one example, when the first distance is less than the second distance, the path passing through the second shared operating area in the initial anti-collision path is eliminated to obtain the anti-collision path of the candidate tower crane relative to the shared space tower crane. In another example, when the first distance is greater than or equal to the second distance, the initial anti-collision path is determined as the anti-collision path of the candidate tower crane relative to the shared space tower crane.
[0061] In this way, the complex tower group system is decomposed into a finite number of dual-tower subsystems, and different path planning methods are proposed for different situations such as equal-height dual towers and high-low dual towers. Finally, the task scheduling and path planning problems of the entire complex tower group are solved by planning the dual-tower subsystem, which can be applied to the dynamic scheduling of complex tower groups and has high flexibility.
[0062] In a specific embodiment of the present application, an intelligent agent is taken as an example of a tower crane. Based on a method for tower group scheduling provided by the present application, it is possible to avoid the risk of collision in overlapping areas during the task planning stage, while having dynamic real-time performance and the function of task and path optimization.
[0063] The tower crane motion is a composite motion consisting of one or more motion forms. For example, Figure 2 A bird's-eye view schematic diagram of the tower crane motion provided for a specific embodiment of the present application shows a composite motion that is a combination of the tower arm rotation motion and the tower crane trolley's amplitude-changing motion. Among them, the motion trajectory of the tower crane trolley is a complex arc trajectory, and its motion speed is synthesized by the tangential velocity perpendicular to the tower arm and the radial velocity along the tower arm. Taking into account the actual situation, the speed form of the tower arm rotation mechanism and the amplitude-changing mechanism is not a simple uniform speed and uniform acceleration motion, so the motion form of the tower crane trolley will be more complex. The complex tower group is a motion system composed of multiple or even hundreds of tower cranes. It is difficult to establish a kinematic model of a complex tower group with multiple tower couplings. Considering the characteristics of electrical control and hydraulic control, its accuracy is also difficult to guarantee. Based on the geometric model of the tower group motion, the embodiment of the present application proposes an algorithm that can be applied to the task scheduling of a complex tower group.
[0064] Figure 3 Schematic diagram of omnidirectional rotation scheduling of tower crane provided in the embodiment of the present application. In order to reduce the waiting time of the tower crane and improve the lifting efficiency, the anti-collision scheduling algorithm proposed in the embodiment of the present application can support omnidirectional rotation scheduling of the tower crane. Figure 3As shown in the figure, S represents the material supply point, i.e. the lifting point of the tower crane (the starting point of the task), and D represents the material demand point, i.e. the installation point of the tower crane (the end point of the task). When the tower arm is in any position (excluding S and D), it can move to the position of point S or point D by rotating clockwise or counterclockwise. Figure 3 As shown, the tower arm rotates from its current position to point S in two ways, and rotates from point S to point D in two ways. Therefore, in this case, the omnidirectional rotation scheduling method includes a total of four rotation motion schemes.
[0065] The embodiment of the present application is based on the kinematic geometry model of a complex tower group, decomposes the tower group system into a finite number of "double-tower" subsystems, converts the complex multi-tower motion problem into a "double-tower" motion problem and solves it. It can be understood that the method proposed in the embodiment of the present application can be applied to the anti-collision scheduling of a tower group system composed of n (n≥2) T-type tower cranes of the same or different heights, and can cope with the scenario of dynamic changes in the tower group, that is, when the distribution of the tower group changes, there is no need to update or rebuild the model.
[0066] Figure 4 The schematic diagram of the kinematic geometric model of the equal-height "double-tower" system provided in a specific embodiment of the present application. The "double-tower" system is the smallest unit of the anti-collision scheduling algorithm, in which the "shared space" is defined, such as Figure 4 The "shared space" between tower crane A and tower crane B is the gray quadrilateral area "ACBD", which is the area where tower crane A and tower crane B may collide. When there is a "shared space" between tower crane A and tower crane B, the scheduling strategy of the "dual tower" system is: the paths of tower crane A and tower crane B to perform tasks will not pass through the "shared space" together. For example, Figure 4 As shown in the figure, when the tower crane B's task execution path 1'→2' passes through the "shared space", the tower crane A will choose a path that avoids the "shared space" when executing the task. In this way, a collision incident can be avoided when the tower cranes A and B work together.
[0067] For ease of explanation, the positive directions specified in the embodiments of the present application are: The angle is 0, and the positive direction is the clockwise direction. Figure 4 As shown, the crane arm is Direction rotates in the positive direction to The direction is recorded as the rotation angle + [∠NAS, ∠NAC], where ∠NAS represents the vector In positive direction to The + sign indicates the rotation direction is positive.
[0068] Based on the above provisions, Figure 4 For example, in the "shared space" ACBD, the coordinates of points C and D are:
[0069]
[0070]
[0071] The rotation angle of tower crane A corresponding to the "shared space" ACBD is: is called the shared space azimuth of A, where
[0072] The rotation angle of tower crane B corresponding to the "shared space" is: is called the shared space azimuth of B, where
[0073] by Figure 4 For example, tower crane A is a tower crane waiting for dispatch (i.e., a candidate tower crane), and tower crane B is a shared space tower crane in operation. The equal-height "dual-tower" system obstacle avoidance logic, that is, the method of determining the anti-collision path of tower crane A relative to tower crane B, can be performed in the following steps:
[0074] Step 1: Determine the shared space azimuth of tower crane A and the shared space azimuth of tower crane B
[0075] Step 2: Determine the current task path of tower crane B and judge whether the current task path is consistent with Is there an intersection? If so, go to step 3.1; otherwise, go to step 3.2.
[0076] Step 3.1: Determine the supply point S and / or demand point D in the task parameters corresponding to the task to be assigned e Is it in If yes, go to step 3.1.1; if supply point S and / or demand point D e None If the number of cells is within 100, proceed to step 3.1.2.
[0077] Step 3.1.1: Returns an empty path, indicating that crane A cannot perform the task.
[0078] Step 3.1.2: Calculate the motion trajectory of tower crane A.
[0079] Calculate the tower crane A from the supply point S to the demand point D e The path includes: judging the positive coverage area of rotation + [∠NAS, ∠NAD e ]Whether There is an intersection. If not, the motion trajectory is determined as If there is an intersection, the motion trajectory is determined as Similarly, the path of tower crane A from safe point O to supply point S can be calculated And by the demand point D e The path back to safe point O Finally, the return trajectory is:
[0080] Step 3.2: Calculate all possible motion trajectories of tower crane A, including:
[0081] The trajectory from O to S includes: From S to D e The tracks include: and By D e The trajectory to O includes: and Finally, 8 feasible trajectories are returned: as well as
[0082] Thus, the embodiment of the present application proposes the above algorithm applicable to the task scheduling of complex tower groups based on the geometric model of tower group motion, thereby avoiding collisions during the task dispatching stage. Meanwhile, the algorithm also has dynamic scheduling and optimization functions.
[0083] It is understandable that in a twin-tower system, the heights of tower crane A and tower crane B are not always the same, that is, there are more complex spatial situations such as high tower A and low tower B. In this regard, in order to face a more complex high-low "twin-tower" system, the embodiment of the present application also proposes a "peripheral circle" method. Among them, the "peripheral circle" is defined as the larger circle in the circle with the center of the high tower and the line connecting the supply point and the demand point as the radius, and the "peripheral circle" changes with the changes of different scheduling tasks.
[0084] Specifically, in order to consider the impact of tower crane layout on the anti-collision scheduling algorithm, in the high-low "dual-tower" system, the embodiment of the present application simultaneously considers the impact of the low-tower balance arm and the high-tower tower body on the movement of the tower crane during the anti-collision scheduling process, so as to enhance the stability and adaptability of the algorithm. Figure 5 This is a schematic diagram of a geometric model of a high-low "double tower" system provided in an embodiment of the present application. Figure 5 As shown, tower crane A is a high tower and tower crane B is a low tower. During the hoisting process of tower crane A from S to De, the "outer circle" is: a solid circle with A as the center and r3 as the radius, and r4 is a circle with the balance arm length of the low tower B as the radius.
[0085] When implementing the obstacle avoidance logic of the high-low twin tower system, the logic of the equal-height twin tower system can be used first, that is, filtering operations are performed based on the equal-height twin tower obstacle avoidance path, which is called a filtering condition in the embodiment of the present application. The following are the specific implementation steps of a filtering condition:
[0086] Step 2.1: Use the “peripheral circle” method to construct a double-tower system of equal heights and calculate and return a set of anti-collision paths:
[0087]
[0088] Step 2.2: Based on r3, r4 and L AB , set different filtering conditions for path filtering to achieve the obstacle avoidance logic of the high and low double tower system: AB ≥r3+r4, go to step 2.2.1, otherwise go to step 2.2.2.
[0089] Step 2.2.1: set_output←set_origin.
[0090] Step 2.2.2: Traverse set_origin, if There is a shared spatial azimuth with tower crane A If there is an intersection, remove the complete path from set_origin and assign the remaining complete path to set_output.
[0091] Step 2.3: Return all collision-avoiding paths to set_output.
[0092] In this way, the above-mentioned "outer circle" method can avoid the intersection operation of complex curves in three-dimensional space, and at the same time greatly simplify the path planning of tower crane rotation, amplitude change and lifting movement.
[0093] Figure 6 The following is a schematic diagram of an algorithm flow for dynamic scheduling of complex tower groups provided in the embodiment of the present application. In summary, the anti-collision scheduling algorithm proposed in the embodiment of the present application can allocate tower crane resources for different scheduling tasks according to different optimal conditions and plan the scheduling path. The overall process is as follows: Figure 6 As shown, the following steps are included:
[0094] Step S501: Start.
[0095] Step S502: Input the task to be assigned.
[0096] Step S503: traverse the crane set AT={AT1,…,ATm} of the executable task Task, initialize the index i=1, and initialize setA={}.
[0097] Among them, the tower crane set AT is the candidate intelligent agent group in the above real-time method.
[0098] Step S504: traverse the tower cranes BT={BT1, ..., BTn} that share space with ATi, initialize index j=1, and initialize setB={}.
[0099] The tower crane BT set is the shared space agent group in the above implementation. It can be understood that each candidate tower crane in the tower crane set AT corresponds to a shared space tower crane group.
[0100] Step S505: Determine the type of the "dual tower" system composed of ATi and BTj. If it is an equal-height "dual tower" type, proceed to step S506; otherwise, proceed to step S507.
[0101] Step S506: Calculate the equal-height "double-tower" anti-collision path Tracej, and proceed to step S508.
[0102] Step S507: Calculate the high and low "double tower" anti-collision path Tracej, and proceed to step S508.
[0103] The anti-collision path Tracej refers to the anti-collision path between the candidate tower crane ATi and its corresponding shared space tower crane BTj.
[0104] Step S508: Store Tracej into setB, j=j+1.
[0105] Step S509: Determine whether j>n. If so, proceed to step S510; otherwise, proceed to step S505.
[0106] It can be understood that after traversing all the shared space tower cranes of the candidate tower crane ATi, the anti-collision paths between the candidate tower crane ATi and its shared space tower cranes BTj are obtained by storing in setB.
[0107] Step S510: Traverse setB and find all the paths shared by all traces And its corresponding tower crane Tk is stored in setA, i=i+1.
[0108] Understandably, That is, the common anti-collision path between all shared space agents of the candidate tower crane ATi, that is, the target path for the candidate tower crane ATi to execute the task to be assigned. The candidate tower crane ATi can avoid all shared space tower cranes by executing this path without collision risk.
[0109] Step S511: Determine whether i>m. If so, proceed to step S512; otherwise, proceed to step S504.
[0110] It can be understood that all candidate cranes of the task to be assigned are traversed, and the target paths corresponding to the candidate cranes are stored in set A. That is, set A stores the target paths and their corresponding target cranes.
[0111] Step S512: traverse setA and filter paths using optimization conditions And its corresponding tower crane T k .
[0112] Step S513: Output Task allocation tower crane T and its corresponding path
[0113] It can be understood that in actual application, setA can be further filtered by setting optimization conditions to obtain the final target tower crane T and the corresponding target path of the task to be assigned Task.
[0114] Step S514: End.
[0115] Among them, the optimization conditions can be set according to actual needs. For example, for the optimization condition of operation time, the path with the shortest operation time can be selected. For another example, for the optimization condition of tower crane rotation distance, the path with the shortest tower crane rotation distance can be selected.
[0116] In summary, through the above technical solutions, the complex tower group system is decomposed into a finite number of dual-tower subsystems, and different path planning methods are proposed for different situations of "equal height" and "high and low" dual towers. Finally, the task scheduling and path planning problems of the entire complex tower group are solved by planning the dual-tower subsystem.
[0117] An embodiment of the present application also provides a processor configured to execute the method for agent group scheduling in the above-mentioned implementation.
[0118] An embodiment of the present application also provides a system for agent group scheduling, including: an agent group; and the processor in the above implementation.
[0119] An embodiment of the present application also provides a machine-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the method for scheduling a group of intelligent agents in the above-mentioned embodiment is implemented.
[0120] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0121] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0122] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0124] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0125] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0127] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0128] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A method for scheduling a group of intelligent agents, characterized in that: The method comprises: Obtaining task parameters of the task to be assigned, wherein the task parameters include a task start point and a task end point; Determine a candidate group of agents that can perform the task to be assigned according to the task parameters; Determine a shared space agent group corresponding to each candidate agent in the candidate agent group, wherein the shared space agents in the shared space agent group are in operation and their operation areas overlap with the operation areas of the candidate agents; Determining, according to the task parameters, an anti-collision path between each candidate agent and each shared space agent in the corresponding shared space agent group; A target path is determined based on the collision-avoiding path and a target agent executes the target path.
2. The method according to claim 1, characterized in that: The target intelligent agent determining a target path based on the anti-collision path and executing the target path includes: For any candidate agent, determining whether there is a common anti-collision path among all shared space agents of the candidate agent according to the anti-collision path; The candidate agent having the common anti-collision path is determined as the target agent, and the common anti-collision path is determined as the target path executed by the target agent.
3. The method according to claim 1, characterized in that The step of determining, according to the task parameters, an anti-collision path between each candidate agent and each shared space agent in a corresponding shared space agent group comprises: For any candidate agent, determining a shared operation area between the candidate agent and corresponding shared space agents; Determine an initial path set of the candidate agent according to the task parameters; Obtaining the current task path of each of the shared space agents; Based on the task parameters, the current task path and the shared operation area, an anti-collision path between the candidate agents and corresponding shared space agents in the initial path set is determined.
4. The method according to claim 3, characterized in that The step of determining, based on the task parameters, the current task path, and the shared operation area, an anti-collision path between the candidate agent and corresponding shared space agents in the initial path set comprises: When the current task path passes through the shared operation area and the task start point and / or the task end point are located in the shared operation area, it is determined that the anti-collision path does not exist in the initial path set.
5. The method according to claim 3, characterized in that: The step of determining, based on the task parameters, the current task path, and the shared operation area, an anti-collision path between the candidate agent and corresponding shared space agents in the initial path set comprises: When the current task path passes through the shared operation area and both the task start point and the task end point are not located in the shared operation area, all paths in the initial path set that do not pass through the shared operation area are determined as the collision-avoiding paths.
6. The method according to claim 3, characterized in that: The step of determining, based on the task parameters, the current task path, and the shared operation area, an anti-collision path between the candidate agent and corresponding shared space agents in the initial path set comprises: In a case where the current task path does not pass through the shared operation area, all paths in the initial path set are determined as the collision-avoiding paths.
7. The method according to claim 1, characterized in that In the case where the agent is a tower crane, determining, according to the task parameters, an anti-collision path between each candidate agent and each shared space agent in a corresponding shared space agent group includes: For any candidate tower crane and any corresponding shared space tower crane, determine a twin tower type formed by the candidate tower crane and the shared space tower crane, wherein the twin tower type includes an equal-height twin tower type and a high-low twin tower type; Acquire the current operating parameters of the shared space tower crane, wherein the operating parameters include a task starting point, a task end point and a current task path; In the case where the twin tower type is a high-low twin tower type, a first shared operation area is determined, wherein the first shared operation area is an overlapping area of an outer circle of a higher tower of the candidate tower crane and the shared space tower crane and a boom operation area of a lower tower, wherein the outer circle is a larger circle of circles whose radii are lines connecting the center of the higher tower and the task starting point and the task end point respectively; Determining an initial anti-collision path of the candidate tower crane relative to the shared space tower crane according to the operation parameters, the task parameters and the first shared operation area; Determine a second shared operation area, where the second shared operation area is an overlapped area of the outer circle of the higher tower and the balance arm operation area of the lower tower; Comparing a first distance with a second distance, wherein the first distance is the sum of the radius of the outer circle of the higher tower and the length of the balance arm of the lower tower, and the second distance is the distance between the tower centers of the candidate tower crane and the shared space tower crane; When the first distance is less than the second distance, eliminating the path passing through the second shared operation area in the initial anti-collision path to obtain the anti-collision path; When the first distance is greater than or equal to the second distance, the initial anti-collision path is determined as the anti-collision path.
8. A processor, characterized in that: The method is configured to execute the method for agent group scheduling according to any one of claims 1 to 7.
9. A system for scheduling a group of intelligent agents, characterized in that: include: Groups of intelligent agents; as well as A processor according to claim 8.
10. A machine-readable storage medium storing a program or an instruction, characterized in that: When the program or the instruction is executed by a processor, the method for agent group scheduling according to any one of claims 1 to 7 is implemented.