A task scheduling method, device, computer device, and medium

Abstract

The present invention discloses a task scheduling method, apparatus, computer device, and medium. The task scheduling method includes: in response to receiving a scheduling task sent by a device, obtaining available idle robot information; generating corresponding demand identification information according to the scheduling task, where the demand identification information includes a task type, a destination location, and material information; determining a matching target robot from the obtained idle robots according to the task type and the material information; and controlling the target robot to move to the destination location according to the task type and the destination location.

Description

Technical Field

[0001] The present invention relates to the field of control technologies, and in particular, to a task scheduling method, apparatus, computer device, and medium. Background Art

[0002] An AGV (Automated Guided Vehicle), that is, an automated guided vehicle, is an intelligent logistics device that realizes autonomous movement through navigation technology. An AGV is an intelligent transport vehicle and an important part of intelligent equipment, and can well complete specified tasks in environments such as unmanned factories and unmanned warehouses.

[0003] In the prior art, the scheduling efficiency of AGVs is low during material transportation, and the working efficiency is low.

[0004] The above content is only used to assist in understanding the technical solution of the present invention, and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of the present invention is to provide a task scheduling method, apparatus, computer device, and medium, aiming to solve the problem that the scheduling efficiency of AGVs is low and the working efficiency is low during material transportation in the prior art.

[0006] To achieve the above objective, the present invention provides a task scheduling method, and the task scheduling method includes:

[0007] In response to receiving a scheduling task sent by a device, obtain available idle robot information;

[0008] According to the scheduling task, generate corresponding demand identification information, where the demand identification information includes a task type, a destination location, and material information;

[0009] According to the task type and the material information, determine a matching target robot from the obtained idle robots;

[0010] According to the task type and the destination location, control the target robot to move to the destination location.

[0011] Preferably, in the task scheduling method, the target robot and the destination location are on the same floor;

[0012] Correspondingly, it further includes:

[0013] In response to receiving a path request service sent by the target robot, obtain the current position of the target robot, status data near the current position, and historical path data;

[0014] Construct a network structure based on the status data and historical path data, where the key positions in the status data serve as grid points of the network structure, and the data information between two adjacent grid points includes the length between the two adjacent grid points and the probability of passage;

[0015] Take the network structure and the current position as inputs and input them into a pre-trained neural network structure to output the next position;

[0016] Control the target robot to move to the next position.

[0017] Preferably, in the task scheduling method, the step of taking the network structure and the current position as inputs and inputting them into a pre-trained neural network structure to output the next position includes:

[0018] Generate an attribute representation of network points according to the network structure, and the generation formula is:

[0019] ;

[0020] Take the attribute representation of the historical path data, the attribute representation of the status data, and the features of the edges connected to the grid point where the current position is located as inputs and input them into a pre-trained neural network structure to output the next position;

[0021] Where,

[0022] P i k+1 is the attribute representation of grid point i at layer k + 1;

[0023] P t k is the attribute representation of grid point t at layer k;

[0024] r i is the number of edges connected to grid point i;

[0025] r t is the number of edges connected to grid point t;

[0026] The adjacent node of grid point i is grid point t;

[0027] Q k is the matrix of connection weights between grid points at layer k;

[0028] λ is the activation function.

[0029] Preferably, in the task scheduling method, the target robot and the destination position are not on the same floor;

[0030] Accordingly, after the step of controlling the target robot to move to the destination position according to the task type and the destination position, the following steps are further included:

[0031] In response to receiving the path request service sent by the target robot, obtain the current position of the target robot;

[0032] Determine whether the current position is a preset path point;

[0033] When the judgment result is yes, obtain the communication information sent by the lifting device;

[0034] When the judgment result is no, control the target robot to continue moving forward according to the path planning between the target robot and the destination position.

[0035] Preferably, in the task scheduling method, the path point is located outside the lifting device;

[0036] Accordingly, the step of when the judgment result is yes and obtaining the communication information sent by the lifting device includes:

[0037] When the current position is a preset path point, determine whether the target robot needs to enter the lifting device according to the planned path between the target robot and the destination position;

[0038] When it is determined that the target robot needs to enter the lifting device, obtain the communication information sent by the lifting device, and control the target robot to enter the lifting device according to the communication information;

[0039] When it is determined that the target robot does not need to enter the lifting device, control the target robot to continue moving forward according to the path planning between the target robot and the destination position.

[0040] Preferably, in the task scheduling method, the path point is located inside the lifting device;

[0041] Accordingly, the step of when the judgment result is yes and obtaining the communication information sent by the lifting device includes:

[0042] When the current position is a preset path point, obtain the communication information sent by the lifting device;

[0043] According to the communication information, judge whether the lifting device has reached the preset destination floor;

[0044] When the lifting device reaches the preset destination floor, control the target robot to leave the lifting device.

[0045] Preferably, in the task scheduling method, the demand identification information further includes the previous process section, and the task type is a feeding task;

[0046] Correspondingly, determining a matching target robot from the obtained idle robots according to the task type and the material information includes:

[0047] Determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information.

[0048] Preferably, in the task scheduling method, determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information includes:

[0049] Determining a plurality of matching robots from the obtained idle robots according to the task type, the previous process section, and the material information;

[0050] Screening out the target robot according to the material feeding times of the determined plurality of robots.

[0051] Preferably, in the task scheduling method, the task type is a feeding task;

[0052] Correspondingly, determining a matching target robot from the obtained idle robots according to the task type and the material information includes:

[0053] Determining the previous process section of the required material according to the task type and the device information for sending the scheduling task;

[0054] Determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information.

[0055] Preferably, in the task scheduling method, the task type is a material taking task;

[0056] Correspondingly, determining a matching target robot from the obtained idle robots according to the task type and the material information includes:

[0057] Determining a plurality of matching robots from the obtained idle robots according to the task type and the material information;

[0058] Screening out the target robot according to the states of the determined plurality of robots, where the states of the robot include the charging state and the power state.

[0059] To achieve the above object, the present invention further provides a task scheduling device, and the task scheduling device includes:

[0060] A response unit, configured to obtain available idle robot information in response to receiving a scheduling task sent by a device;

[0061] A generating unit, configured to generate corresponding requirement identification information according to the scheduling task, where the requirement identification information includes a task type, a destination location, and material information;

[0062] A screening unit, configured to determine a matching target robot from the obtained idle robots according to the task type and the material information;

[0063] A control unit, configured to control the target robot to move to the destination location according to the task type and the destination location.

[0064] To achieve the above object, the present invention further provides a computer device, where the computer device includes:

[0065] At least one processor; and,

[0066] A memory communicatively connected to the at least one processor; wherein,

[0067] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above task scheduling method.

[0068] To achieve the above object, the present invention further provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the above task scheduling method.

[0069] The present invention has at least the following beneficial effects:

[0070] The task scheduling method provided by the present invention obtains available idle robot information in response to receiving a scheduling task sent by a device; generates corresponding requirement identification information according to the scheduling task, where the requirement identification information includes a task type, a destination location, and material information; determines a matching target robot from the obtained idle robots according to the task type and the material information; and controls the target robot to move to the destination location according to the task type and the destination location, so that resources can be accurately and efficiently allocated, and the efficiency of material scheduling can be improved.

[0071] Furthermore, since in a factory environment, condition data (such as obstacles in the environment) may change at any time, the present invention can optimize the current path in real time according to the actual environmental conditions by combining a neural network structure, thereby improving work efficiency. Description of the Drawings

[0072] Figure 1 Schematic diagram of the task scheduling method provided by the present invention in the first embodiment;

[0073] Figure 2 Schematic diagram of the task scheduling method provided by the present invention in the second embodiment;

[0074] Figure 3 Schematic diagram of the path points at the lifting equipment provided by the present invention;

[0075] Figure 4 Schematic diagram of the task scheduling device provided by the present invention;

[0076] Figure 5 Schematic diagram of the computer device provided by the present invention.

[0077] The realization of the object, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0078] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0079] In the embodiments of the present invention, the term "and / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0080] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence.

[0081] In the embodiments of the present invention, the term "multiple" refers to two or more, and other quantifiers are similar thereto.

[0082] In the present invention, unless otherwise stated, the orientation terms such as "upper", "lower", "top", "bottom" are usually in the direction shown in the drawings, or in the vertical, perpendicular or gravitational direction of the component itself; similarly, for the sake of easy understanding and description, "inner" and "outer" refer to the inner and outer of the contour of each component itself, but the above orientation terms do not limit the present invention.

[0083] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will elaborate on each embodiment of the present invention in conjunction with the accompanying drawings. However, those of ordinary skill in the art can understand that in each embodiment of the present invention, many technical details are provided to help readers better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed by the present invention can still be achieved. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and cross-referenced with each other on the premise of not being contradictory.

[0084] To solve the above problems, this embodiment relates to a task scheduling method that can be applied to a computer device. The computer device can be a desktop computer, a tablet computer, a notebook, a mobile terminal, or other electronic devices with data processing capabilities. In other embodiments, it can also be other electronic devices with data processing capabilities, which are not specifically limited herein.

[0085] The following describes the implementation details of the task scheduling method according to the first embodiment of the present invention. The following content is only provided for easy understanding of the implementation details and is not necessary for implementing this solution. Figure 1 The schematic diagram of the task scheduling method of the present invention is shown.

[0086] The specific process of this embodiment is as Figure 1 shown and specifically includes:

[0087] Step S100, in response to receiving a scheduling task sent by a device, obtain available idle robot information;

[0088] It should be understood that the robots in the present invention can be, but are not limited to, AGVs. The devices in the present invention can be, but are not limited to, production devices, such as texturing devices, diffusion devices, and etching devices. Taking the process sequence of texturing, diffusion, and etching in turn as an example, the normal process flow is usually that the materials produced by the texturing device are sent to the entrance of the diffusion device. After the diffusion device finishes production, the materials produced by the diffusion device are sent to the entrance of the etching device.

[0089] Taking the diffusion device as an example, the scheduling task sent by the diffusion device may be to send the materials produced by the previous process (texturing device) to the entrance of the diffusion device; the scheduling task sent by the diffusion device may also be to send the materials produced by the diffusion device to the entrance of the next process etching device after the diffusion device finishes production; of course, in some embodiments, the scheduling task sent by the diffusion device may also be to transport the materials produced by the diffusion device to the buffer station for caching.

[0090] It should be noted that taking device A as an example, when the scheduling task sent by device A is to send the materials produced in the previous process to device A, the idle robot required is the robot carrying the materials produced in the previous process; of course, in some embodiments, it can also be to match the empty-load robot, and when the empty-load robot executes the task, the empty-load robot first arrives at the preset point to obtain the materials produced in the previous process and then continues to execute the task.

[0091] Step S200, generate corresponding demand identification information according to the scheduling task, where the demand identification information includes task type, destination location, and material information;

[0092] It should be noted that the task type can include, but is not limited to, material taking tasks, material sending tasks, etc. The material taking task is to take out the materials on the device, and the taken-out materials can be sent to the buffer station, or temporarily placed on the robot waiting for the next process section, or sent to other preset points; the material sending task is to send the materials produced in the previous process section to the device. In some embodiments, the material taking task can be further divided into a buffer task of sending to the buffer station and a material taking task of simply taking the materials and storing them on the robot waiting for the next process section.

[0093] The destination location can be the location where the device sending the demand is located. More specifically, each device may include a material sending port and a discharging port, and the destination location may be the material sending port of the device or the discharging port of the device, which can be specifically determined according to the task type.

[0094] During specific operations, the demand identification information can be formed into an array, and the array can include, but is not limited to, the identifier of the previous process section to determine which process section the taken materials can be sent to. The array can include, but is not limited to, the destination device (i.e., the device sending the demand); more specifically, it can also include the specific location of the destination device. For example, the material sending port 01 of the texturing device 101 can be recorded as 10101. The destination location can store the absolute position coordinates or the position number of the specific required materials of the destination device. For example, if the material sending port 01 of the texturing device 101 needs materials, the corresponding destination location is 10101, and according to the destination location number, its absolute coordinates can be queried.

[0095] In addition, it is worth noting that even in the same process section, there may be multiple devices, so the identifier corresponding to each device is uniquely corresponding. For example, there may be four texturing devices, corresponding to the numbers 101, 102, 103, and 104 respectively; there may be three diffusion devices, corresponding to the numbers 201, 202, and 203 respectively; there may be four etching devices, corresponding to the numbers 301, 302, 303, and 304 respectively.

[0096] It should be noted that the demand identification information may or may not include the previous process section. When the previous process section is not included in the demand identification information and the task type is a feeding task, the process section of the current device can be determined based on the device information for sending the scheduling task, so as to determine the corresponding previous process section according to the process section of the current device. Since each device information is unique and the process section of the current device is clear, the previous process section can be determined. Of course, in some embodiments, if the current device is the first process section, there is no previous process section, and in this case, it can also be determined whether there is a previous process section based on the device information.

[0097] Step S300: Determine a matching target robot from the obtained idle robots according to the task type and the material information.

[0098] It should be understood that when the task type is a feeding task, the matching condition for the target robot can be a robot whose material carried on the idle robot matches the material information in the demand identification information. When there are multiple matching robots, it can also be to select the most matching one as the target robot according to other preset rules, such as screening the robots in the order of first in, first out of the materials on the robots.

[0099] Suppose device A needs the material 01 produced by the previous process 1. Then the target robot can be screened according to the feeding order of the material 01 on the robot.

[0100] It should be noted that during the matching, if the device requires the material produced by the previous process, it is necessary to judge whether it is the material produced by the previous process during the screening. In some embodiments, the screening conditions can also be set according to specific requirements.

[0101] Specifically, the demand identification information further includes the previous process section, and the task type is a feeding task.

[0102] Correspondingly, step S300 includes: determining a matching target robot from the obtained idle robots according to the task type, the previous process section and the material information.

[0103] Taking the idle robot R1 as an example, when the material on the idle robot R1 is the material produced by the texturing equipment, and the diffusion equipment A that sends the demand requires exactly the material of its previous process, the texturing equipment, then it can be considered that the material and its corresponding process match the demand identification information, and the idle robot R1 can be used as the target robot.

[0104] More specifically, step S300 includes: determining a plurality of matching robots from the obtained idle robots according to the task type, the previous process section, and the material information; and screening out a target robot according to the material feeding times of the determined plurality of robots.

[0105] Taking equipment A as an example, when equipment A issues a scheduling task, the process corresponding to equipment A is the current process section. Then, the raw materials processed by equipment A are usually the materials produced in the previous process section (of course, except when equipment A is the first process). When the scheduling task issued by equipment A is a feeding task, if the material carried by an idle robot is the material produced in the previous process section, then this idle robot is considered a matching robot; when there are multiple matching robots, the most appropriate one is selected from them as the target robot.

[0106] When there are multiple matching robots, to screen out the most appropriate one, it can be determined by the power status, charging status, etc. of the robots. Usually, robots that are not charging and / or robots with the highest power are preferably selected for allocation. In some embodiments, it can also be determined whether to select a robot that is not charging or a robot with the highest power according to the distance between the robot and the destination location. When the distance between the robot and the destination location exceeds a preset distance, the robot with the highest power is selected as the target robot; when the distance between the robot and the destination location does not exceed the preset distance and the power of the robot is not lower than a preset value, a robot that is not charging is selected as the target robot.

[0107] In some embodiments, the task type is a material fetching task; correspondingly, step S300 includes: determining a plurality of matching robots from the obtained idle robots according to the task type and the material information; and screening out a target robot according to the status of the determined plurality of robots, where the status of the robot includes the charging status and the power status.

[0108] In addition, there will also be a situation. If there are no idle robots, but there are robots whose tasks are about to be completed, and the next task type, the next process section, and the material information corresponding to the next process section of this robot match the demand identification information, then this robot can also be considered as the target robot at this time.

[0109] When there is no idle robot, the task currently being executed by robot R is a material picking task and it is in the process of distribution (for example, after picking the material and sending it to a preset point to wait for the next process). At this time, when the next process corresponding to the material on the robot, the material information, the next task type of the robot match the demand identification information, robot R can also be determined as the target robot. At this time, it can be to control robot R to directly start executing a new task, or to wait for robot R to reach the preset point and then control it to start executing a new task.

[0110] In addition, in some embodiments, after determining the target robot from the screened idle robots, it can be to control the remaining idle robots to perform charging, or it can be to do nothing.

[0111] When controlling the remaining idle robots to charge, the method further includes obtaining available charging piles; and allocating the available charging piles to the remaining idle robots. Specifically, how to allocate can be based on the principle of proximity. In some embodiments, it can also be preferentially allocated according to the battery levels of the remaining idle robots. The lower the battery level of an idle robot, the higher its charging priority.

[0112] Step S400, according to the task type and the destination location, control the target robot to move to the destination location.

[0113] It should be noted that different devices may be located on different floors, or they may be on the same floor; the feeding port and the discharging port of the same device may also be on the same floor or on different floors, depending on the actual situation.

[0114] For example, the machine platform of device A can be double-layered. It can be that the upper layer is the feeding port and the lower layer is the discharging port; or it can be that the lower layer is the feeding port and the upper layer is the discharging port. Of course, it is also possible that the machine platform of device A is single-layered, that is, the feeding port and the discharging port are on the same layer.

[0115] In addition, the target robot and the destination location may be on the same floor or on different floors. When the target robot and the destination location are on different floors, then a lifting device (such as an elevator) is needed to carry the target robot for floor switching.

[0116] In some embodiments, the target robot and the destination location are on the same floor; correspondingly, the method further includes steps S410 to S410.

[0117] Step S410, in response to receiving a path request service sent by the target robot, obtain the current position of the target robot, the status data near the current position, and the historical path data;

[0118] It should be noted that the status data near the current location is the data of the environment near the current location, such as road conditions, obstacle conditions, etc. The historical path data may include, but is not limited to, the path of the target robot moving from the initial position to the current position. In some embodiments, the historical path data may also include the historical path from the current position to the destination position.

[0119] Step S420: Construct a network structure according to the status data and the historical path data, where the key positions in the status data are used as the grid points of the network structure, and the data information between two adjacent grid points includes the length between the two adjacent grid points and the magnitude of the passing possibility.

[0120] It should be understood that the key positions can be determined according to the required accuracy, which can be the grid points obtained by dividing the smallest unit size in the map, or the grid points formed by dividing the preset unit size in the map. In some embodiments, the key positions can also be the key points marked by the user on the map.

[0121] Step S430: Take the network structure and the current position as inputs and input them into a pre-trained neural network structure to output the next position.

[0122] In some embodiments, step S430 includes step S431 and step S432.

[0123] Step S431: Generate the attribute representation of the network points according to the network structure, and the generation formula is:

[0124] ;

[0125] Step S432: Take the attribute representation of the historical path data, the attribute representation of the status data, and the features of the edges connected to the grid point where the current position is located as inputs and input them into a pre-trained neural network structure to output the next position.

[0126] In some embodiments, the neural network structure is a graph neural network structure. In some other embodiments, any neural network structure that can achieve the above functions can also be used.

[0127] It should be noted that the attribute representations of the historical path data and the status data can both be represented by the attribute features of the grid points. The specific representation method can refer to the above formula.

[0128] It should be noted that when the current position and the destination position are on the same floor, the above method can be used to optimize the path; in some embodiments, although the current position and the destination position are not on the same floor, the path from the current position to the destination position can be decomposed into a path between the current position and the elevator point on the same floor, and the above method can also be used to optimize this path.

[0129] Among them,

[0130] P i k+1 is the attribute representation of grid point i on the (k + 1)-th floor;

[0131] P t k is the attribute representation of grid point t on the k-th floor;

[0132] r i is the number of edges connected to grid point i;

[0133] r t is the number of edges connected to grid point t;

[0134] The adjacent node of grid point i is grid point t;

[0135] Q k is the matrix of connection weights between grid points on the k-th floor;

[0136] λ is the activation function.

[0137] It should be noted that the attribute representation of the historical path data is the attribute representation of the grid point where the current position is located on the last floor. The attribute representation of the status data is the attribute representation of the status data corresponding to the historical path data, that is, the attribute representation of the neighbor grid points of grid point i. The feature of the edge connected to the grid point where the current position is located is the feature of the edge connected to grid point i.

[0138] Step S440, control the target robot to move to the next position.

[0139] The task scheduling method provided by the present invention, in response to receiving a scheduling task sent by a device, obtains available idle robot information; according to the scheduling task, generates corresponding demand identification information, the demand identification information includes a task type, a destination position, and material information; according to the task type and the material information, determines a matching target robot from the obtained idle robots; according to the task type and the destination position, controls the target robot to move to the destination position, so that resources can be accurately and efficiently allocated, and the efficiency of material scheduling can be improved.

[0140] As Figure 2 shown, Figure 2Illustrated is a second embodiment provided by the present invention based on the first embodiment, where the target robot and the destination location are not on the same floor; correspondingly, after the step S400, the following steps are further included:

[0141] Step S510, in response to receiving a path request service sent by the target robot, obtain the current position of the target robot;

[0142] It should be understood that during the traveling process, the target robot will continuously request path services to determine the specific walking direction.

[0143] Step S520, determine whether the current position is a preset path point;

[0144] It should be understood that when the target robot is located at the lifting device, it is necessary to consider whether to continue sending the next path. Therefore, path points are usually set outside and inside the lifting device.

[0145] For example, if the target robot is on the first floor, it is necessary to judge whether to enter the lifting device or when to enter the lifting device based on the path point and the communication situation with the lifting device; after the target robot enters the lifting device, the service of continuing to send the path needs to be cut off at this time. After the target robot reaches the specified floor, control the target robot to drive out of the lifting device.

[0146] As Figure 3 shown, taking the example that the target robot enters the lifting device from floor L1 and drives out of the lifting device when reaching L2, a path point P1 needs to be set outside the door of the lifting device at floor L1, a path point P2 needs to be set when entering the lifting device, a path point P3 needs to be set inside the lifting device when reaching L2, and a path point P4 needs to be set outside the lifting device at L2.

[0147] When it is detected that the target robot reaches the path point P1, it can be judged whether the target robot needs to enter the lifting device according to the pre-established path planning between the target robot and the destination location. When the judgment result is yes, communication with the lifting device needs to be established. When it is received that the lifting device reaches L1 and the lifting device is in the open door state, control the target robot to enter the lifting device. After the target robot enters the lifting device and reaches the path point P2, the service of continuing to send the path needs to be cut off. When it is received that the target robot reaches the path point P3 and the lifting device reaches L2, when the lifting device is in the open door state, control the target robot to drive out of the lifting device. When it is received that the target robot reaches the path point P4, control the target robot to continue moving forward according to the pre-established path planning between the target robot and the destination location.

[0148] It should be noted that

[0149] Step S530, when the judgment result is yes, obtain the communication information sent by the lifting device;

[0150] It should be understood that in this embodiment, the execution subject is a computer device. At this time, it is necessary for the computer device to establish communication with the lifting device, and further accurately control the target robot based on the arrival situation and door opening situation of the lifting device.

[0151] Specifically, the path point is located outside the lifting device; Step S530 includes: when the current position is a preset path point, determine whether the target robot needs to enter the lifting device according to the planned path between the target robot and the destination position; when it is determined that the target robot needs to enter the lifting device, obtain the communication information sent by the lifting device, and control the target robot to enter the lifting device according to the communication information; when it is determined that the target robot does not need to enter the lifting device, control the target robot to continue moving forward according to the path planning established in advance between the target robot and the destination position.

[0152] In some embodiments, the path point is located inside the lifting device; Step S530 includes: when the current position is a preset path point, obtain the communication information sent by the lifting device; judge whether the lifting device has reached the preset destination floor according to the communication information; when the lifting device reaches the preset destination floor, control the target robot to leave the lifting device.

[0153] Since when the target robot enters the lifting device, if the target robot is continuously controlled according to the preset path planning, it may easily cause failures. At this time, it is necessary to cut off the control service. After the lifting device reaches the destination floor and is in the door-open state, then continue to control the target robot to improve the control accuracy.

[0154] Step S540, when the judgment result is no, control the target robot to continue moving forward according to the path planning established in advance between the target robot and the destination position.

[0155] To achieve the above object, the present invention also provides a task scheduling device, as Figure 4 shown. The task scheduling device includes a response unit 610, a generation unit 620, a screening unit 630, and a control unit 640.

[0156] The response unit 610 is used to obtain available idle robot information in response to receiving a scheduling task sent by a device. It should be understood that the robots in the present invention can be, but are not limited to, AGVs. The devices in the present invention can be, but are not limited to, production devices, such as texturing devices, diffusion devices, and etching devices. Taking the process sequence of texturing, diffusion, and etching in turn as an example, the normal process flow is usually that the materials produced by the texturing device are sent to the inlet of the diffusion device. After the diffusion device finishes production, the materials produced by the diffusion device are sent to the inlet of the etching device.

[0157] Taking the diffusion device as an example, the scheduling task sent by the diffusion device may be to send the materials produced by the previous process (texturing device) to the inlet of the diffusion device; the scheduling task sent by the diffusion device may also be to send the materials produced by the diffusion device to the inlet of the next process etching device after the diffusion device finishes production; of course, in some embodiments, the scheduling task sent by the diffusion device may also be to transport the materials produced by the diffusion device to the buffer station for buffering and placement.

[0158] It should be noted that taking device A as an example, when the scheduling task sent by device A is to send the materials produced by the previous process to device A, the idle robot required is the robot carrying the materials produced by the previous process; of course, in some embodiments, it can also be to match an empty-load robot. When the empty-load robot executes the task, the empty-load robot first arrives at a preset point to obtain the materials produced by the previous process and then continues to execute the task.

[0159] The generation unit 620 is used to generate corresponding demand identification information according to the scheduling task. The demand identification information includes task type, destination location, and material information. It should be noted that the task type can be, but is not limited to, including material-taking tasks, material-sending tasks, etc. The material-taking task is to take out the materials on the device. The taken-out materials can be sent to the buffer station, temporarily placed on the robot waiting for the next process section, or sent to other preset points; the material-sending task is to send the materials produced by the previous process section to the device. In some embodiments, the material-taking task can be further divided into a buffering task of sending to the buffer station and a material-taking task of simply taking the materials and storing them on the robot waiting for the next process section.

[0160] The destination location can be the location where the device sending the demand is located. More specifically, each device may include a material-sending port and a material-discharging port. The destination location may be the material-sending port of the device or the material-discharging port of the device, which can be specifically determined according to the task type.

[0161] During specific operations, the requirement identification information can be formed into an array. The array can but is not limited to containing the identification of the previous process section to determine which process section the obtained materials can be sent to. The array can but is not limited to containing the destination device (i.e., the device sending the requirement); more specifically, it can also be the specific location of the destination device. For example, the feeding port 01 of the texturing device 101 can be recorded as 10101. The destination location can store the absolute position coordinates or the position number of the specific required materials of the destination device. For example, if the feeding port 01 of the texturing device 101 requires materials, the corresponding destination location is 10101. Based on the destination location number, its absolute coordinates can be queried.

[0162] In addition, it is worth noting that even in the same process section, there may be multiple devices. Therefore, the identification corresponding to each device is uniquely corresponding. For example, there may be four texturing devices, corresponding to the numbers 101, 102, 103, and 104 respectively; there may be three diffusion devices, corresponding to the numbers 201, 202, and 203 respectively; there may be four etching devices, corresponding to the numbers 301, 302, 303, and 304 respectively.

[0163] It is worth noting that the requirement identification information can include the previous process section or not. When the requirement identification information does not include the previous process section and the task type is a feeding task, the process section of the current device can be determined according to the device information of the sending scheduling task, so as to determine the corresponding previous process section according to the process section of the current device. Since each device information is unique and the process section of the current device is clear, the previous process section can be determined. Of course, in some embodiments, if the current device is the first process section, there is no previous process section. At this time, it can also be determined whether there is a previous process section according to the device information.

[0164] The screening unit 630 is used to determine a matching target robot from the obtained idle robots according to the task type and the material information. It should be understood that when the task type is a material taking task, the matching condition of the target robot can be a robot whose materials carried on the idle robot match the material information in the requirement identification information. When there are multiple matching robots, it can also be to select the most matching one as the target robot according to other preset rules, such as screening the robots in the order of first in first out of the materials on the robot.

[0165] Suppose device A requires the material 01 produced in the previous process 1. Then the target robot can be screened according to the feeding order of the material 01 on the robot.

[0166] It should be noted that during matching, if the device requires the materials produced in the previous process, then it is necessary to determine whether the materials are those produced in the previous process during screening. In some embodiments, the screening conditions can also be set according to specific requirements.

[0167] Specifically, the demand identification information further includes the previous process section, and the task type is a feeding task;

[0168] Correspondingly, the screening unit 630 is further configured to: determine a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information.

[0169] Taking the idle robot R1 as an example, when the material on the idle robot R1 is the material produced by the texturing equipment, and the diffusion equipment A that sends the demand exactly requires the material of its previous process, the texturing equipment, then it can be considered that the material and its corresponding process match the demand identification information, and the idle robot R1 can be used as the target robot.

[0170] More specifically, the screening unit 630 is further configured to: determine multiple matching robots from the obtained idle robots according to the task type, the previous process section, and the material information; and screen out the target robot according to the material feeding time of the determined multiple robots.

[0171] Taking equipment A as an example, when equipment A issues a scheduling task, the process corresponding to equipment A is the current process section. Then, the raw materials processed by equipment A are usually the materials produced in the previous process section (of course, except when equipment A is the first process). When the scheduling task issued by equipment A is a feeding task, if the material carried by the idle robot is the material produced in the previous process section, then the idle robot is considered a matching robot; when there are multiple matching robots, the most appropriate one is selected from them as the target robot.

[0172] When there are multiple matching robots, to screen out the most appropriate one, it can be determined by the power status, charging status, etc. of the robots. Usually, robots that are not charging and / or robots with the highest power are preferably selected for allocation. In some embodiments, it can also be determined whether to select a robot that is not charging or a robot with the highest power according to the distance between the robot and the destination location. When the distance between the robot and the destination location exceeds the preset distance, the robot with the highest power is selected as the target robot; when the distance between the robot and the destination location does not exceed the preset distance and the power of the robot is not lower than the preset value, a robot that is not charging is selected as the target robot.

[0173] In specific implementation, the task type is a material fetching task; correspondingly, the screening unit 630 is further configured to: include: determining a plurality of matching robots from the obtained idle robots according to the task type and the material information; screening out a target robot according to the states of the determined plurality of robots, where the states of the robot include a charging state and a power state.

[0174] In addition, there will be another situation. If there are no idle robots, but there are robots that are about to complete their tasks, and the next task type, the next process section, and the material information corresponding to the next process section of this robot match the demand identification information, in this case, this robot can also be considered as the target robot.

[0175] When there are no idle robots, the task currently executed by robot R is a material fetching task, and it is in the process of distribution (for example, after fetching the material and sending it to a preset point to wait for the next process). At this time, when the next process, the material information, and the next task type of the robot corresponding to the material on the robot match the demand identification information, robot R can also be determined as the target robot. At this time, it can be to control robot R to directly start executing a new task, or to wait for robot R to reach the preset point and then control it to start executing a new task.

[0176] In addition, in some embodiments, after determining the target robot from the screened idle robots, it can be to control the remaining idle robots to perform charging processing, or it can be to do nothing.

[0177] When controlling the remaining idle robots to charge, the method further includes obtaining available charging piles; allocating the available charging piles to the remaining idle robots. Specifically, how to allocate can be based on the principle of proximity. In some embodiments, it can also be preferentially allocated according to the power ranking of the remaining idle robots. The lower the power of the idle robot, the higher the charging priority.

[0178] The control unit 640 is configured to control the target robot to move to the destination location according to the task type and the destination location. It should be noted that different devices may be on different floors, or they may be on the same floor; the feeding port and the discharging port of the same device may also be on the same floor or on different floors, depending on the actual situation.

[0179] For example, the machine platform of device A can be double-layered, it can be that the upper layer is the feeding port and the lower layer is the discharging port; it can also be that the lower layer is the feeding port and the upper layer is the discharging port. Of course, it is also possible that the machine platform of device A is single-layered, that is, the feeding port and the discharging port are on the same layer.

[0180] In addition, the target robot and the destination location may be on the same floor or on different floors. When the target robot and the destination location are on different floors, a lifting device (such as an elevator) is required to transport the target robot for floor switching.

[0181] To achieve the above object, the present invention further provides a computer device, as Figure 5 shown. The computer device includes at least one processor 701; and a memory 702 communicatively connected to the at least one processor 701. Wherein, the memory 702 stores instructions executable by the at least one processor 701, and the instructions are executed by the at least one processor 701 to enable the at least one processor 701 to execute the above task scheduling method.

[0182] Among them, the memory 702 and the processor 701 are connected by a bus. The bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors 701 and the memory 702 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, so they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be an element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor 701 is transmitted on the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor 701.

[0183] The processor 701 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory 702 can be used to store data used by the processor 701 during operation.

[0184] To achieve the above object, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above task scheduling method is implemented.

[0185] That is, those skilled in the art can understand that all or part of the steps in the above-described implementation methods can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor 701 (processor) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0186] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, those of ordinary skill in the art can make other different forms of changes or variations without making creative efforts, and all of them should fall within the scope of protection of the present invention.

Claims

1. A task scheduling method, characterized in that, Including: Upon receiving a scheduling task sent by a device, obtain information on available idle robots; Generate corresponding demand identification information according to the scheduling task, where the demand identification information includes task type, destination location, and material information; Determine a matching target robot from the obtained idle robots according to the task type and material information; Control the target robot to move to the destination location according to the task type and destination location; Among them, the task type includes a material picking task and a material delivering task. When the task type is a material delivering task, deliver the material produced in the previous process section to the device, and the corresponding idle robot is the robot carrying the material produced in the previous process. The material picking task is to temporarily place the material taken from the device on the robot and wait for the next process section; When the target robot and the destination location are on the same floor; correspondingly, it further includes: Upon receiving a path request service sent by the target robot, obtain the current position of the target robot, the condition data near the current position, and historical path data; Construct a network structure according to the condition data and historical path data, where the key positions in the condition data serve as grid points of the network structure, and the data information between two adjacent grid points includes the length between the two adjacent grid points and the probability of passage; Generate an attribute representation of network points according to the network structure; use the attribute representation of historical path data, the attribute representation of condition data, and the characteristics of the edges connected to the grid point where the current position is located as inputs, input them into a pre-trained neural network structure, and output the next position. The generation formula is: ； Wherein, P i k+1 It is the attribute representation of grid point i at the (k + 1)-th layer; P t k It is the attribute representation of grid point t at layer k; r i The number of edges connected to grid point i; r t The number of edges connected to grid point t; The adjacent node of grid point i is grid point t; Q k is the matrix of connection weights between the grid points of the k-th layer; λ is an activation function; Control the target robot to move to the next position.

2. The task scheduling method according to claim 1, wherein The target robot and the destination location are not on the same floor; Correspondingly, it further includes: Upon receiving a path request service sent by the target robot, obtain the current position of the target robot; Judge whether the current position is a preset path point; When the judgment result is yes, obtain the communication information sent by the lifting device; When the judgment result is no, control the target robot to continue moving forward according to the path planning between the target robot and the destination location established in advance.

3. The task scheduling method according to claim 2, wherein The path point is located outside the lifting device; Correspondingly, when the judgment result is yes, obtaining the communication information sent by the lifting device includes: When the current position is a preset path point, determine whether the target robot needs to enter the lifting device according to the planned path between the target robot and the destination location; When it is determined that the target robot needs to enter the lifting device, obtain the communication information sent by the lifting device, and control the target robot to enter the lifting device according to the communication information; When it is determined that the target robot does not need to enter the lifting device, control the target robot to continue moving forward according to the path planning between the target robot and the destination location established in advance.

4. The task scheduling method according to claim 2, wherein, The path point is located inside the lifting device; Correspondingly, when the judgment result is yes, obtaining the communication information sent by the lifting device includes: When the current position is a preset path point, obtaining the communication information sent by the lifting device; Judging whether the lifting device reaches a preset destination floor according to the communication information; When the lifting device reaches the preset destination floor, controlling the target robot to leave the lifting device.

5. The task scheduling method according to claim 1, wherein, The requirement identification information further includes the previous process section, and the task type is a feeding task; Correspondingly, the determining a matching target robot from the obtained idle robots according to the task type and the material information includes: Determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information.

6. The task scheduling method according to claim 5, wherein, The determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information includes: Determining a plurality of matching robots from the obtained idle robots according to the task type, the previous process section, and the material information; Screening out the target robot according to the material feeding time of the determined plurality of robots.

7. The task scheduling method according to claim 1, wherein The task type is a feeding task; Correspondingly, the determining a matching target robot from the obtained idle robots according to the task type and the material information includes: Determining the previous process section of the required material according to the task type and the device information for sending the scheduling task; Determining a matching target robot from the obtained idle robots according to the task type, the previous process section, and the material information.

8. The task scheduling method according to claim 1, wherein The task type is a material fetching task; Correspondingly, the determining a matching target robot from the obtained idle robots according to the task type and the material information includes: Determining a plurality of matching robots from the obtained idle robots according to the task type and the material information; Screening out the target robot according to the status of the determined plurality of robots, where the status of the robot includes the charging status and the power status.

9. A task scheduling device, characterized in that Includes: A response unit, configured to obtain available idle robot information in response to receiving a scheduling task sent by a device; A generating unit, configured to generate corresponding requirement identification information according to the scheduling task, where the requirement identification information includes a task type, a destination location, and material information; A screening unit, configured to determine a matching target robot from the obtained idle robots according to the task type and the material information; A control unit, configured to control the target robot to move to the destination location according to the task type and the destination location; Wherein, the task type includes a material fetching task and a feeding task. When the task type is a feeding task, the material produced in the previous process section is sent to the device, and the corresponding idle robot is the robot carrying the material produced in the previous process. The material fetching task is to temporarily place the material taken from the device on the robot and wait for the next process section; The control unit is further configured to: when the target robot and the destination location are on the same floor; In response to receiving the path request service sent by the target robot, obtain the current position of the target robot, the condition data near the current position, and the historical path data; Construct a network structure according to the condition data and the historical path data, where the key positions in the condition data are used as grid points of the network structure, and the data information between two adjacent grid points includes the length between the two adjacent grid points and the probability of passage; Generate an attribute representation of network points according to the network structure; use the attribute representation of the historical path data, the attribute representation of the condition data, and the features of the edges connected to the grid point where the current position is located as inputs, and input them into a pre-trained neural network structure to output the next position; the generation formula is: ； Wherein, P i k+1 It represents the attribute of grid point i at the (k + 1)-th layer; P t k It is the attribute representation of grid point t at layer k; r i The number of edges connected to grid point i; r t The number of edges connected to grid point t; The adjacent node of grid point i is grid point t; Q k is the matrix of connection weights between grid points of the k-th layer; λ is an activation function; Control the target robot to move to the next position.

10. A computer device, characterized in that, Includes: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the task scheduling method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the task scheduling method according to any one of claims 1 to 8.

Images