Process scheduling system suitable for different service robots to take elevator

Abstract

The invention discloses a process scheduling system suitable for elevator taking of different service robots, and relates to the technical field of industrial control. According to the invention, through a protocol adaptation sub-module, an industrial control field bus adaptation logic is associated, a communication protocol of a multi-brand service robot and an elevator is unified, a standardized data system is constructed, efficient interaction of elevator taking task data of different types of robots is guaranteed, and the system compatibility is improved; based on an intelligent matching algorithm of a task allocation decision module and historical data fusion modeling, robot and elevator combinations are screened, task conflicts are dynamically adjusted, elevator running prediction is simulated through a dynamic model, an elevator taking path and a scheduling strategy are optimized, elevator congestion idling is reduced, the elevator resource utilization rate is increased, emergency tasks are guaranteed to be responded preferentially, and the service life of the elevator is prolonged. The complex logistics requirements of hospitals are met; the robot path planning gives consideration to obstacle avoidance and motion characteristics, the whole process of robot elevator taking is efficient and safe, a hospital is assisted to construct an intelligent logistics collaborative system, and the operation cost and instability are reduced.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, and in particular to a process scheduling system applicable to elevator use by different service robots. Background Technology

[0002] The demand for logistics robots within hospitals is growing rapidly, covering services such as delivery, cleaning, inspection, patient guidance, and triage. Taking delivery as an example, there are various types, including integrated cabinet robots, split-type pile-mounting robots, clamping integrated robots, and rolling integrated robots. However, robot manufacturers, limited by resources, typically focus on developing only one or two scenarios. Currently, there is no unified standard protocol for elevator control of hospital logistics robots, requiring each robot manufacturer to adapt its own elevator control protocol. Hospital elevator resources are limited, and due to the lack of a unified protocol, each manufacturer applies for elevators separately for its own robots. This not only wastes costs but also increases the instability of robot operation due to multiple elevator control systems running in parallel, making it difficult to efficiently coordinate and serve the hospital's logistics needs. There is an urgent need for a scheduling solution that can accommodate multiple types of robots and coordinate elevator resources. Summary of the Invention

[0003] The purpose of this invention is to provide a process scheduling system suitable for elevator use by different service robots, to build a collaborative foundation through unified protocol adaptation, to break down communication barriers between multi-brand robots and elevators, to achieve resource sharing, to enhance the collaborative scheduling capabilities of different service robots, to ensure the efficient flow of elevator use tasks, to optimize resource allocation, to avoid congestion and idle time, to prioritize the response to emergency tasks, and to collaboratively ensure the safety and efficiency of the entire elevator use process for robots, thereby reducing costs and improving stability, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] This system is applicable to elevator scheduling for different service robots, comprising an elevator task acquisition module, a task allocation decision module, an elevator scheduling module, and a robot path planning module. The elevator task acquisition module receives elevator task requests from different service robots, parses the task content, extracts key target information, and marks task priorities. The task allocation decision module, based on key target information, combined with current elevator and robot status data, allocates elevator tasks to the corresponding service robot and elevator combinations and sends task allocation instructions to the corresponding service robots and elevators. The elevator scheduling module schedules the corresponding elevators according to the task allocation instructions, controls the elevator's operating status, and provides data feedback based on the elevator's real-time operating status, updating the status data of each elevator. The robot path planning module plans the optimal path for the service robot from its current location to the elevator lobby, and the path from the elevator to the target floor to its final destination, based on the task allocation instructions, and provides data feedback based on the service robot's status information.

[0006] Furthermore, the elevator task acquisition module specifically includes: a protocol adaptation submodule, which collects the communication protocol specifications of each service robot and the elevator control system, establishes a protocol adaptation library, performs protocol conversion on the elevator task request data sent by the service robot and the status data fed back by the elevator, and obtains the converted standardized data; a real-time data acquisition submodule, which extracts the unique identifier of the service robot, the current floor location information, and the target floor information from the converted standardized data, obtains the task type based on the elevator task request data, and determines the corresponding task priority identifier according to the task type; and a historical data collection submodule, which collects historical robot data sets and historical elevator data sets within a preset time period based on the standardized data converted by the protocol adaptation module, and sends the collected historical robot data sets and historical elevator data sets to the task allocation decision module.

[0007] Furthermore, the task allocation decision module sends task allocation instructions as follows: It acquires real-time operating status data of each elevator and status information of each service robot, and prioritizes each elevator-riding task based on the acquired task priority identifier; it constructs a matching degree calculation model to calculate the matching degree between each elevator-riding task and each service robot-elevator combination, and mediates allocation conflicts caused by multiple elevator-riding tasks simultaneously matching the same elevator; based on the matching degree and mediation results, it allocates the elevator-riding task to the service robot-elevator combination with the highest matching degree, records the allocation result in the task allocation table, updates the allocation status of the corresponding elevator-riding task, and sends task allocation instructions to the corresponding service robot and elevator.

[0008] Furthermore, the historical robot dataset includes robot position time curves, task priority time curves, and elevator request time curves within each time unit; the historical elevator dataset includes elevator operation status time curves and elevator response time curves within each time unit.

[0009] Furthermore, the service robot and elevator combination is based on preset constraints, combined with the status information of each service robot and the real-time operating status data of the elevator to filter out service robots and elevators that meet the conditions, and then the service robots and elevators that meet the conditions are associated and matched to generate a service robot and elevator combination.

[0010] Furthermore, the task allocation decision module also includes: splitting the historical elevator data set by time unit to obtain a time series dataset, and fusing it with real-time operating status data to generate an elevator model dataset, where the timestamp of the real-time data is the current moment, and the timestamp of the historical data is the start moment of the corresponding time unit; preprocessing the elevator model dataset, using the preprocessed elevator model input as input data to construct an elevator model, performing elevator operation simulation prediction based on the elevator model, and obtaining the elevator model prediction results.

[0011] Furthermore, the elevator scheduling module controls the elevator's operating status as follows: It parses the service robot identifier, starting floor, target floor, and task priority information from the task allocation instructions; based on the elevator model prediction results, combined with the starting and target floors, it plans the optimal elevator path and generates elevator operation control instructions, including the order of stopping floors, speed adjustment parameters, and door opening / closing time settings; it monitors the actual elevator operating status in real time and compares it with the elevator model prediction results, generating correction instructions based on the comparison results to dynamically adjust the elevator's operating status; when the elevator reaches the starting floor, it sends a positioning signal to the corresponding service robot, controls the elevator door to remain open for a preset time, and after receiving feedback that the robot has entered the car, it closes the elevator door and runs towards the target floor according to the path planned by the robot path planning module; when the elevator reaches the target floor, it controls the elevator door to open, and after receiving feedback that the service robot has left the car, it closes the elevator door and updates the elevator status data, including the current floor, load rate, and operating status identifier, and feeds it back to the task allocation decision module.

[0012] Furthermore, the robot path planning module plans the optimal path as follows: When the service robot receives the task instruction to go to the elevator lobby, the path planning module uses the robot's current position as the starting point and the elevator lobby position as the ending point to calculate the path using a path planning algorithm; when the service robot reaches the target floor and exits the elevator, it uses the elevator lobby exit position as the starting point and the task target position as the ending point to calculate the path again, generating the optimal path for the service robot; during the calculation process, obstacle information and conflict areas are extracted, and the optimal path is smoothed according to the kinematic characteristics of the service robot.

[0013] Furthermore, the matching degree between each elevator riding task and each service robot and elevator combination is calculated, including: obtaining the service robot and elevator combinations to obtain several combinations; calculating the reliability between each elevator riding task and each combination to obtain the first matching degree; .in, This represents the first matching degree between the current elevator riding task and the i-th combination; This represents the historical success rate of the i-th combination executing the current elevator task type; This represents the real-time fault risk value for the i-th combination; This represents the reliability requirement coefficient for the current elevator riding task; This represents the idle rate of the i-th combination; This represents the fault repair time for the i-th combination; calculate the functional compatibility between each elevator ride task and each combination to obtain the second matching degree; .in, This represents the second matching degree between the current elevator riding task and the i-th combination; , , Indicates the weighting coefficient; This represents the maximum safe load-bearing capacity in the i-th combination; This indicates the weight of goods to be transported in the current elevator ride. This represents the estimated time for the entire process of the current elevator ride task to be executed by the i-th combination; Indicates the end time of the entire process for the current elevator ride task; Let represent the floor permission adaptation coefficient of the i-th combination; calculate the resource adaptation degree between each elevator ride task and each combination to obtain the third matching degree; .in, This represents the third degree of matching between the current elevator riding task and the i-th combination; , , Indicates the weighting coefficient for resource adaptation; This indicates the current remaining load of the elevator in the i-th combination; This represents the remaining number of stops for the elevator in the current operating cycle of the i-th combination; The standard stop redundancy threshold for the i-th combination is represented; the target matching degree between the current elevator ride task and each combination is determined based on the first matching degree, the second matching degree, and the third matching degree. in, This indicates the degree of target matching between the current elevator riding task and the i-th combination; , , This represents the weighting coefficient.

[0014] Furthermore, the mediation of allocation conflicts caused by multiple elevator rides simultaneously matching the same elevator includes: obtaining the information set required for conflict mediation; preprocessing the information set to obtain a standardized information set; using the task deadline in the standardized information set as the target dimension, setting a dynamic time window based on the task density of the time period, and clustering the service robot tasks within the dynamic time window to obtain task groups; for each task group, obtaining the candidate task set for the elevator corresponding to each task group; performing conflict prediction on each task group based on the candidate task sets of each elevator, and determining whether each task group has a potential conflict identifier; if the potential conflict identifier exists, then the task group is taken as the target task group; constructing and solving a multi-dimensional adaptation optimization model based on the target matching degree to obtain the initial mediation result corresponding to the target task group; dynamically adjusting the initial mediation result based on real-time status data to obtain the adjusted mediation result corresponding to the target task group; if the adjusted mediation result still has conflicts, using a relaxation constraint algorithm to resolve the conflicts, and obtaining the final mediation result of the target task group.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By using a protocol adaptation submodule to associate with bus adaptation logic in the industrial control field, the communication protocols between multi-brand service robots and elevators are unified, enabling the interoperability of position data, task requirements, and status information of different types of robots. This constructs a standardized data system, solves the problem of inconsistent robot elevator control protocols in hospital scenarios, ensures efficient interaction of elevator task data for different types of robots, and improves system compatibility. Furthermore, relying on the intelligent matching algorithm of the task allocation decision module and historical data fusion modeling, robot-elevator combinations are selected, task conflicts are dynamically adjusted, and collaborative scheduling of different types of robots is achieved. The matching dimension is dynamically adjusted according to the characteristics of different types of robots. By employing weighted algorithms to achieve differentiated collaboration, and through dynamic model simulation of elevator operation prediction, the system optimizes elevator routes and scheduling strategies, reducing elevator congestion and idle runs, improving elevator resource utilization, reducing elevator stops, increasing collaborative efficiency, ensuring priority response to emergency tasks, and adapting to the complex logistics needs of hospitals. Robot path planning considers both obstacle avoidance and motion characteristics, with differentiated matching and collaborative avoidance mechanisms, significantly reducing the task conflict rate between different types of robots, effectively improving task completion rate and safety, optimizing collaborative efficiency, and achieving high efficiency and safety throughout the entire robot elevator process. This helps hospitals build an intelligent logistics collaboration system, flexibly adapting to the multi-robot collaboration needs of different scenarios, reducing operating costs and instability. Attached Figure Description

[0016] Figure 1 This is a block diagram of the process scheduling system for elevator use by different service robots according to the present invention; Figure 2 This is a flowchart of the present invention for using different service robots to ride elevators. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] To address the technical issues of the lack of a unified standard protocol for robotic elevator control, requiring manufacturers to adapt their own protocols, and the limited elevator resources in hospitals leading to cost waste and operational instability, please refer to [the relevant documentation]. Figure 1This system is applicable to the process scheduling of elevator use by different service robots. It includes an elevator task acquisition module, a task allocation decision module, an elevator scheduling module, and a robot path planning module. These modules interact in real-time via a fieldbus control system, ensuring compatibility with the existing fieldbus control system communication environment in hospitals. Data interaction is achieved through a wireless network, ensuring coordinated operation of all system components and meeting the efficient elevator use needs of multiple service robots in a hospital environment. The elevator task acquisition module receives elevator task requests from different service robots, parses the task content, extracts key target information such as the robot's current position, target floor, and task priority, and marks the task priority. In a preferred embodiment, the elevator task acquisition module specifically includes a protocol adaptation submodule, which collects the communication protocol specifications of each service robot and the elevator control system, associates them with industrial bus adaptation logic in the industrial control field, establishes a protocol adaptation library, and performs protocol conversion on the elevator task request data sent by the service robots and the status data fed back by the elevator to obtain the converted standardized data.

[0019] The communication protocol specifications include the service robot's location data transmission protocol, task priority encoding protocol, elevator request instruction format, and elevator operating status (floor, direction of travel, load rate) transmission protocol and response instruction format.

[0020] The protocol adaptation library includes a protocol identifier: representing the service robot brand and model or elevator control model; and a conversion rule: mapping non-standard data fields to system-unified fields. The present invention provides the following example:

[0021] Map "Service Robot X's Task Level A" to "System Priority 3" and "Elevator Y's Load Percentage" to "System Load Rate Value";

[0022] The beneficial effects of the above technical solution are as follows: by collecting service robot location data transmission protocols, task priority encoding protocols, elevator request instruction formats, elevator operation status transmission protocols, and response instruction formats, and associating them with mature industrial bus adaptation logic in the field of industrial control, a protocol adaptation library covering multiple brands and types of equipment is established. Differentiated data is uniformly converted into standardized data that the system can recognize, making the system compatible with multiple mainstream brands, service robot types, and various elevator control systems. There is no need to deploy separate elevator control systems for different devices, reducing hardware costs per scenario, breaking down communication barriers between multiple brands and types of robots, and significantly improving system compatibility.

[0023] The real-time data acquisition submodule extracts the service robot's unique identifier, current floor location information, and target floor information from the converted standardized data. It also obtains the task type based on the elevator ride request data and determines the corresponding task priority identifier according to the task type. The historical data collection submodule collects historical robot and elevator data sets within a preset time period based on the standardized data converted by the protocol adaptation module, and sends the collected historical robot and elevator data sets to the task allocation decision module. In a preferred embodiment, the historical robot data set includes the robot position-time curve and task priority for each time unit. The system includes time curves and elevator request time curves. The robot position time curve shows the robot's real-time position changing over time within a time unit; the position data is obtained from robot positioning data after protocol adaptation. The task priority time curve shows the priority of the robot's currently executing task changing over time within a time unit; the priority data is obtained through task encoding conversion after protocol adaptation. The elevator request time curve shows the time and target floor of the robot initiating an elevator request changing over time within a time unit; the request data is obtained through elevator command parsing after protocol adaptation. In a preferred embodiment, the historical elevator data set includes each group... The time unit includes elevator operation status time curves and elevator response time curves. The elevator operation status time curve shows the elevator's current floor, direction of travel, and load rate changing over time within the time unit; this status data is obtained from elevator feedback data after protocol adaptation. The elevator response time curve shows the time interval between receiving a passenger request and arriving at the requested floor changing over time within the time unit; the response time is obtained by calculating the difference between the "request reception time" and the "elevator arrival time at the requested floor." The task allocation decision module is used to allocate passenger tasks to corresponding elevators based on key target information, combined with current elevator status data and robot status information. The system combines service robots with elevators and sends task allocation instructions to the corresponding service robots and elevators. The elevator scheduling module schedules the elevators according to the task allocation instructions, controls their operation, ensures the corresponding robot reaches its target location via the elevator, and provides real-time data feedback based on the elevator's operating status to update the status data of each elevator and prevent congestion or idle running. The robot path planning module plans the optimal path for the service robot from its current location to the elevator lobby, and the path from the elevator to the target floor to its final destination, based on the task allocation instructions. It also provides data feedback based on the service robot's status information.

[0024] The beneficial effects of the above technical solution are as follows: by combining the robot position time curve with the elevator request time curve, the peak demand for elevators of different types of robots in each time period can be predicted, and the supply and demand of elevator resources can be predicted. Based on this pattern, the task allocation decision module can plan the elevator window for different types of robots in advance, and predict that a certain elevator is more suitable for a certain type of robot in a certain time period, thereby reducing the conflict of multiple robots vying for elevators, avoiding the mismatch between elevator resources and robot demand, and improving the efficiency of collaboration.

[0025] The task allocation decision module sends task allocation instructions by: acquiring real-time operating status data of each elevator, including its floor, direction of travel, availability, and load status; and the current location, busy status, and battery level of each service robot. Based on the acquired task priority identifiers, each elevator-riding task is prioritized, with higher-priority tasks processed first to ensure timely execution of emergency tasks. An intelligent matching algorithm is used to construct a matching degree calculation model, calculating the matching degree between each elevator-riding task and each service robot and elevator combination, and mediating allocation conflicts caused by multiple elevator-riding tasks simultaneously matching the same elevator.

[0026] Based on the matching degree and mediation results, the elevator riding task is assigned to the service robot and elevator combination with the highest matching degree. The assignment result is recorded in the task assignment table. At the same time, the assignment status of the corresponding elevator riding task is updated, and the task assignment instruction is sent to the corresponding service robot and elevator.

[0027] The service robot and elevator combination is based on preset constraints, combined with the status information of each service robot and the real-time operation status data of the elevator to filter out qualified service robots and elevators. These qualified service robots and elevators are then matched to generate service robot and elevator combinations. In a preferred embodiment, service robots that are currently idle / low-priority tasks that can be interrupted, have sufficient power to travel to and from the target floor, and whose current location matches the starting location of the elevator task are selected. Elevators whose running direction matches the task requirements, whose current load rate does not exceed a threshold, and which are idle or can respond to the starting floor within a short time are selected based on the starting floor, target floor, and task priority of the elevator task. The matching process involves spatially matching the current location of candidate robots with the current location of candidate elevators. If the time difference between the robot's arrival time in the elevator lobby and the elevator's arrival time at the robot's floor is less than a preset threshold, a combination is generated. Further filtering is performed to check if the robot's size / weight exceeds the elevator car limit and the elevator load rate. If the elevator task is high-priority, combinations where the elevator currently has no low-priority tasks and can respond first are prioritized to ensure adaptability for emergency tasks.

[0028] In a preferred embodiment, the task allocation decision module further includes: splitting the historical elevator data set by time unit to obtain a time series dataset, and fusing it with real-time operating status data to generate an elevator model dataset, wherein the timestamp of the real-time data is the current moment, and the timestamp of the historical data is the start moment of the corresponding time unit; preprocessing the elevator model dataset, using the preprocessed elevator model input as input data to construct an elevator model, performing elevator operation simulation prediction based on the elevator model, and obtaining elevator model prediction results; wherein the elevator model prediction results include the elevator's current floor, real-time running direction, instantaneous load rate, door opening / closing status, and estimated arrival time at each floor.

[0029] Please see Figure 2 The elevator scheduling module controls the elevator's operation status by: parsing the service robot identifier, starting floor, target floor, and task priority information from the task allocation instruction; based on the elevator model prediction results, combined with the starting floor and target floor, planning the optimal elevator running path, and generating elevator operation control instructions, including the order of stopping floors, running speed adjustment parameters, and door opening / closing time settings; in a preferred embodiment, for high-priority tasks, without violating elevator safety operation regulations, the current elevator operation plan can be adjusted to prioritize responding to the starting floor corresponding to the high-priority task, and if necessary, issuing a floor skipping instruction to skip some low-priority task stopping requests; real-time monitoring of the elevator's actual operation... The system monitors the elevator's operating status and compares it with the elevator model's prediction results. Based on the comparison results, it generates correction instructions to dynamically adjust the elevator's operating status, ensuring that the elevator runs along the planned path. When the elevator reaches the starting floor, it sends a positioning signal to the corresponding service robot, controls the elevator door to remain open for a preset time, and closes the elevator door after receiving feedback that the robot has entered the car. The elevator then runs towards the target floor according to the path planned by the robot path planning module. When the elevator reaches the target floor, it controls the elevator door to open. After receiving feedback that the service robot has left the car, it closes the elevator door and updates the elevator status data, including the current floor, load rate, and operating status indicator, and feeds it back to the task allocation decision module.

[0030] The robot path planning module plans the optimal path as follows: When the service robot receives the task instruction to go to the elevator lobby, the path planning module uses the robot's current position as the starting point and the elevator lobby position as the ending point to calculate the path using a path planning algorithm. After the service robot reaches the target floor and exits the elevator, it uses the elevator lobby exit position as the starting point and the task target position as the ending point to calculate the path again, generating the optimal path for the service robot. During the calculation process, obstacle information and conflict areas are extracted to avoid planning paths that collide with obstacles. At the same time, the optimal path is smoothed according to the service robot's kinematic characteristics (such as turning radius, maximum speed, etc.) to ensure that the robot can move safely and smoothly.

[0031] Calculate the matching degree between each elevator riding task and each service robot and elevator combination, including: obtaining service robot and elevator combinations to obtain several combinations; calculating the reliability between each elevator riding task and each combination to obtain the first matching degree; .in, This represents the first matching degree between the current elevator riding task and the i-th combination; This represents the historical success rate of the i-th combination executing the current elevator task type; This represents the real-time fault risk value for the i-th combination; This represents the reliability requirement coefficient for the current elevator riding task; This represents the idle rate of the i-th combination; This represents the fault repair time for the i-th combination; calculate the functional compatibility between each elevator ride task and each combination to obtain the second matching degree; .in, This represents the second matching degree between the current elevator riding task and the i-th combination; , , Indicates the weighting coefficient; This represents the maximum safe load-bearing capacity in the i-th combination; This indicates the weight of goods to be transported in the current elevator ride. This represents the estimated time for the entire process of the current elevator ride task to be executed by the i-th combination; Indicates the end time of the entire process for the current elevator ride task; Let represent the floor permission adaptation coefficient of the i-th combination; calculate the resource adaptation degree between each elevator ride task and each combination to obtain the third matching degree; .

[0032] in, This represents the third degree of matching between the current elevator riding task and the i-th combination; , , Indicates the weighting coefficient for resource adaptation; This indicates the current remaining load of the elevator in the i-th combination; This represents the remaining number of stops for the elevator in the current operating cycle of the i-th combination; The standard stop redundancy threshold for the i-th combination is represented; the target matching degree between the current elevator ride task and each combination is determined based on the first matching degree, the second matching degree, and the third matching degree. in, This indicates the degree of target matching between the current elevator riding task and the i-th combination; , , This represents the weighting coefficient.

[0033] In this embodiment, .

[0034] In this embodiment, if .

[0035] In this embodiment, .

[0036] In this embodiment, reliability refers to the likelihood that a service robot-elevator combination can complete the task without failure and as expected after the elevator task is assigned to it. It is a quantitative assessment of the stability of the combination in performing the task. Different robot-elevator combinations have different historical performance and current status, which directly affects whether the task can be completed reliably. For example, some combinations may have successfully completed similar tasks many times in the past with few failures; some combinations may frequently fail (such as unstable robot power or stuck elevator doors), causing task interruption. The stability requirements of the tasks themselves are also different (for example, the task of transporting surgical instruments is more susceptible to failure than the task of transporting ordinary documents). Therefore, it is necessary to measure the stability matching degree between the task and the combination through reliability to avoid assigning high-importance tasks to unreliable combinations. Taking the hospital service robot transportation scenario as an example: the historical success rate of a certain combination (robot R1 + elevator E1) =98%, real-time failure risk =0.02; Idle rate =0.8; Fault repair time The allotted time is 5 minutes; however, the current task is to transport emergency medical supplies (reliability requirements are high). =0.9), demand is extremely high), at this time, the combination's A high value indicates that assigning the task to this combination will likely result in a stable completion, with minimal disruption to emergency response due to malfunctions; the historical success rate of a certain combination (Robot R2 + Elevator E2). =70%, real-time failure risk =0.3, idle rate =0.8; Fault repair time =30 minutes (prolonged inability to recover after a malfunction); also for transporting emergency medical supplies. =0.9; at this time, A very low value indicates that assigning the task to this team could result in emergency medicines not being delivered on time due to malfunctions, making it extremely unreliable.

[0037] In this embodiment, functional compatibility refers to whether the core functions of the service robot-elevator combination (such as load-bearing capacity, time efficiency, and floor accessibility) can meet the basic requirements of the elevator riding task. It is a quantitative assessment of the matching between the combination's functions and task requirements. The completion of the task depends on the hardware functional support of the combination. If the functions do not match, the task cannot be completed at all. For example, if the combination's maximum load capacity is only 50kg, it cannot complete the task of transporting 100kg of goods, no matter how high its reliability is. For example, if the task requires delivery within 10 minutes, and the combination estimates that it will take 20 minutes, it will fail due to timeout even without any malfunctions. Therefore, functional compatibility is a prerequisite for the execution of the task and must be given priority. Taking the intelligent warehouse robot transportation scenario as an example: the current task is to transport 80kg of electronic components from the 3rd floor to the 5th floor, which needs to be completed within 20 minutes; the maximum safe load capacity of a certain combination (robot R3 + elevator E3) is... =100kg, estimated time for the entire process =15 minutes, floor permission adaptation coefficient =1. At this time, The high value indicates that the combination's load-bearing capacity, time, and floor access permissions fully meet the task requirements, demonstrating good functional adaptability. For the same task (80kg, 3rd to 5th floors, 20 minutes), if assigned to the combination (Robot R4 + Elevator E4): Maximum safe load-bearing capacity... =60kg, estimated time for the entire process =25 minutes, floor permission adaptation coefficient =0.3 (Elevator E4 cannot directly reach the 5th floor; a transfer is required on the 4th floor, increasing travel time). At this point, A very low value indicates that the combined functions cannot meet the basic requirements of the task, resulting in poor functional adaptability. Functional adaptability is evaluated based on three main categories: load-bearing capacity, time-related functions, and floor access control. Load-bearing capacity: Can the maximum safe load-bearing capacity of the combination (robot + elevator) support the weight of the task goods (avoiding overload damage to equipment or goods)? Time-related functions: Can the entire process time of the combination's task execution be controlled within the task deadline (avoiding timeout failure)? Floor access control: Can the elevator directly reach the starting and destination floors of the task (avoiding detours or inability to reach due to insufficient permissions)?

[0038] In this embodiment, resource adaptability refers to whether the remaining resources (such as remaining load capacity and remaining number of stops) of the service robot-elevator combination match the resource requirements of the elevator task, and whether efficient resource utilization can be achieved. It is a quantitative assessment of the rationality of resource utilization; it represents whether the remaining resources of the combination can be used appropriately by the task: neither wasting resources (such as using a large elevator for small goods) nor overloading (such as the remaining load capacity being just enough for the task, with no redundancy or waste), while also ensuring equipment safety (such as sufficient remaining number of stops to avoid frequent stops leading to malfunctions). Taking an office building service robot delivery scenario as an example: the current task is to deliver 20kg of documents from the 1st floor to the 10th floor; the current remaining load capacity of a certain combination (robot R5 + elevator E5) is... =25kg, number of remaining stops within the current operating cycle =3, Standard docking redundancy threshold =2, real-time failure risk =0.05; at this time, A high value indicates that the remaining resources in the combination match the task requirements, resulting in efficient resource utilization with no waste or shortage. For the same task (20kg, floors 1 through 10), if assigned to the combination (Robot R6 + Elevator E6): the elevator's current remaining load capacity... =200kg, number of remaining stops Standard docking redundancy threshold =2; At this time, A very low value indicates inefficient resource utilization or insufficient resources, and poor adaptability.

[0039] In this embodiment, the focus of reliability is on whether the task can be completed stably, with the core being no faults and minimal interruptions; the focus of functional adaptability is on whether the task can be executed, with the core being that hard requirements such as load capacity, time, and permissions are met; and the focus of resource adaptability is on whether resources can be used efficiently during task execution, with the core being to avoid waste and overload. The reliability, functional adaptability, and resource adaptability together constitute the task-combination matching degree evaluation system to ensure that the service robot's elevator scheduling can be completed, completed stably, and completed efficiently.

[0040] The working principle and beneficial effects of the above technical solution are as follows: By comprehensively considering reliability, functional adaptability, and resource adaptability, and dynamically adjusting the weights of each dimension according to the core needs of different types of service robots, a differentiated matching logic is constructed. This allows for a more comprehensive assessment of the matching degree between elevator riding tasks and the combination of service robots and elevators, accurately meeting the collaborative needs of different types of robots, thereby improving the accuracy of matching. Through dynamic task priority ranking and conflict resolution mechanisms, global collaborative scheduling of different types of service robots is achieved. The most suitable combination can be selected for each elevator riding task based on the target matching degree, achieving optimized resource allocation and improving the utilization efficiency of elevators and service robots. The matching process considers factors such as the failure risk and historical success rate of the combination, which helps to select a more reliable combination and enhance the reliability of the entire elevator riding system.

[0041] Mediation of allocation conflicts caused by multiple elevator rides simultaneously matching the same elevator includes: obtaining an information set required for conflict mediation; preprocessing the information set to obtain a standardized information set; using the task deadline in the standardized information set as the target dimension, setting a dynamic time window based on the task density of the time period, and clustering the service robot tasks within the dynamic time window to obtain task groups; for each task group, obtaining a candidate task set for the elevator corresponding to each task group; performing conflict prediction on each task group based on the candidate task sets of each elevator, and determining whether each task group has a potential conflict identifier; if the potential conflict identifier exists, then the task group is designated as the target task group; constructing and solving a multi-dimensional adaptation optimization model based on the target matching degree to obtain the initial mediation result corresponding to the target task group; dynamically adjusting the initial mediation result based on real-time status data to obtain the adjusted mediation result corresponding to the target task group; if the adjusted mediation result still has conflicts, using a relaxation constraint algorithm to resolve the conflicts, and obtaining the final mediation result for the target task group.

[0042] In this embodiment, the information set includes: a service robot task information set, an elevator resource information set, and an environmental association information set; the service robot task information set includes: task ID, cargo type, cargo weight, task priority, departure floor, destination floor, task deadline, and remaining robot battery power; the elevator resource information set includes: elevator ID, rated load, real-time load, current floor, single loading / unloading time, remaining maintenance cycle, energy efficiency, and historical task adaptation preferences; the environmental association information set includes: time period task density and floor loading / unloading congestion index.

[0043] In this embodiment, for each task group, a candidate task set for each elevator is calculated, namely: the first screening condition is that the weight of the goods in the task is less than or equal to the elevator's rated load and the elevator's real-time load; the second screening condition is that the robot's remaining battery power is greater than or equal to |task departure floor - elevator current floor|*10 + |elevator current floor - task destination floor|*10 (unit: seconds, each floor movement takes 10 seconds); the third screening condition is that the elevator's single loading and unloading time × the number of tasks already included in the candidate task set is less than or equal to the task deadline - the current time; and the tasks that meet the first, second, and third screening conditions are selected as candidate tasks for the elevator.

[0044] In this embodiment, a dynamic time window is set based on the task density of a time period, including: the task density of a time period is the ratio of the number of tasks in the current 5-minute window to the average number of tasks in the same time period over 7 days; the dynamic time window duration is as follows: for example, if the task density of a time period is ≥1.3, the window duration is 120 seconds; if 0.9≤the task density of a time period is <1.3, the window duration is 300 seconds; if the task density of a time period is <0.9, the window duration is 480 seconds.

[0045] In this embodiment, the service robot tasks within the dynamic time window are clustered using the DBSCAN algorithm.

[0046] In this embodiment, conflict prediction is performed on each task group based on the candidate task set of each elevator, and it is determined whether each task group has a potential conflict identifier. This includes: if the total number of tasks in the task group is greater than the sum of the number of candidate task sets of all elevators, or the total weight of goods in all tasks in the task group is greater than the sum of the remaining load of all elevators, then it is marked as a potential conflict, and a potential conflict identifier is obtained.

[0047] In this embodiment, a multi-dimensional adaptation optimization model is constructed and solved based on the target matching degree to obtain the initial mediation result, including: using an improved non-dominated sorting genetic algorithm to solve the problem with the goal of maximizing the total target matching degree and minimizing the conflict cost; the improved non-dominated sorting genetic algorithm solution includes: initializing the population: generating feasible solutions based on the cargo type and elevator adaptation constraints (e.g., refrigerated cargo only matches refrigerated elevators, and refrigerated elevators are judged by querying the refrigeration function attribute in the database through the elevator ID); crossover operation: introducing an elevator-cargo type affinity factor (affinity = the past 30 days). The success rate of the elevator executing this cargo type is high. Priority is given to elevator-cargo type combinations with a historical fitness rate > 85% for gene exchange. Mutation operation: the population convergence is defined as the standard deviation of the current population fitness / the standard deviation of the initial population fitness. When the convergence is < 40%, the mutation probability is 0.2, and when the convergence is > 80%, the mutation probability is 0.05. Optimal solution selection: the entropy weight method is used to assign weights to the non-dominated solution set (information entropy calculation is based on the distribution of each objective function value), and the solutions in the Pareto front that are "the top 15% of the total objective matching degree + the bottom 15% of the conflict cost" are selected as the initial allocation scheme.

[0048] In this embodiment, the initial mediation result is dynamically adjusted based on real-time status data to obtain an adjusted mediation result. The dynamic adjustment includes setting three trigger conditions: elevator anomaly trigger: remaining elevator maintenance cycle < 0.5 hours, or fluctuation in elevator single loading / unloading time > 30%; robot status trigger: robot remaining battery power drops sharply by > 15% within 5 minutes, or cargo type is upgraded from ordinary to fragile / refrigerated; task timeliness trigger: remaining task time < total elevator time × 1.2. If any of the trigger conditions are met, real-time status data is re-collected to update the standardized information set, the first matching degree, second matching degree, third matching degree, and target matching degree are recalculated, and the allocation scheme is updated to obtain the adjusted allocation scheme. The impact of the adjusted allocation scheme on the total elevator running time is evaluated. If the impact exceeds a preset tolerance threshold, the process returns to readjustment; otherwise, the adjusted allocation scheme is confirmed.

[0049] In this embodiment, if conflicts still exist in the adjusted mediation results, a relaxed constraint algorithm is used to resolve the conflicts and obtain the final mediation result. This includes: quantifying conflict parameters and calculating the upper limit of the delay time for each conflicting task. in, Indicates the maximum delay time for conflicting tasks; Indicates the remaining operable time for the task; Indicates the remaining execution time of the task; sets the task priority decay factor. Used to quantify the degree to which task latency diminishes its own scheduling priority; sets the system coordination complexity index. Used to quantify the degree of resource consumption and increased complexity of elevator operation coordination when different types of goods are mixed together; ; Indicates the coefficient of difference in goods type; Represent type weights; construct the conflict resolution objective function: ;in, The actual delay time of the task. Represent the logarithm of mixed cargo types; use the branch and bound algorithm to solve the objective function (constraints are as follows). The optimal time window rearrangement scheme is obtained (assuming ≤2 types of mixed goods each time) and used as the final allocation scheme; the comprehensive efficiency index is calculated to verify the effectiveness of the final allocation scheme. The calculation expression is: ;in, For robot task completion rate, Total elevator running time This represents the average actual load on the elevator. This indicates the elevator's rated load. The number represents the number of conflicts. The subscript "old" indicates data before mediation, and the subscript "new" indicates data after mediation.

[0050] The working principle and beneficial effects of the above technical solution are as follows: taking the task deadline as the target dimension, a dynamic time window is set based on the task density of the time period and the service robot tasks are clustered. This method can flexibly adjust the window size according to the density of tasks in different time periods, and group tasks in similar time periods together. For example, during peak periods, the dynamic time window can be reduced so that tasks with close deadlines are grouped together, which facilitates centralized processing, reduces the complexity of task scheduling, and improves the overall operating efficiency of the system. Before task allocation, conflict prediction is performed on each task group to identify potential conflict indicators in advance. This facilitates early intervention for potential problems, enabling the development of targeted collaborative intervention strategies to avoid wasting time and resources on post-conflict handling. It also ensures that the task requirements of different types of robots are met, allowing the system to operate more smoothly and improving task execution efficiency. A multi-dimensional adaptation optimization model is constructed based on target matching, comprehensively considering various matching factors between elevator riding tasks and the combination of service robots and elevators, such as reliability, functional compatibility, and resource compatibility. By solving this model to obtain initial mediation results, the matching relationship between tasks and resources can be more comprehensively and accurately assessed, thereby allocating the most suitable resources to each task and achieving global optimization of elevator and multi-type service robot resources, improving elevator resource utilization.

[0051] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A process scheduling system applicable to elevator use by different service robots, characterized in that, It includes a task acquisition module, a task allocation decision module, an elevator scheduling module, and a robot path planning module; The elevator task acquisition module is used to receive elevator task requests from different service robots, parse the task content, extract key target information, and mark the task priority. The task allocation decision module is used to allocate elevator riding tasks to the corresponding service robot and elevator combination based on key target information, combined with current elevator status data and robot status information, and send task allocation instructions to the corresponding service robot and elevator. Among them, the task allocation sorts the elevator riding tasks according to the task priority identifier, constructs a matching degree calculation model to calculate the matching degree between the elevator riding tasks and each service robot and elevator combination, and mediates the allocation conflict of multiple tasks matching the same elevator. The elevator scheduling module is used to schedule the corresponding elevators according to the task allocation instructions, control the elevator's operating status, and at the same time, provide data feedback based on the real-time operating status of the elevators and update the status data of each elevator. The robot path planning module is used to plan the optimal path for the service robot from its current location to the elevator lobby, and the path from the elevator to the target floor to the final destination, based on the task assignment instructions. At the same time, it provides data feedback based on the service robot's status information.

2. The process scheduling system for elevator use by different service robots as described in claim 1, characterized in that, The elevator task acquisition module specifically includes: The protocol adaptation submodule collects the communication protocol specifications of each service robot and the elevator control system, establishes a protocol adaptation library, performs protocol conversion on the elevator ride request data sent by the service robot and the status data fed back by the elevator, and obtains the standardized data after conversion. The real-time data acquisition submodule extracts the service robot's unique identifier, current floor location information, and target floor information from the converted standardized data. It also obtains the task type based on the elevator ride request data and determines the corresponding task priority identifier according to the task type. The historical data collection submodule collects historical robot data sets and historical elevator data sets within a preset time period based on the standardized data converted by the protocol adaptation module, and sends the collected historical robot data sets and historical elevator data sets to the task allocation decision module.

3. The process scheduling system for elevator use by different service robots as described in claim 2, characterized in that, The task allocation decision module sends task allocation instructions in the following way: The system acquires real-time operating status data of each elevator and status information of each service robot, and prioritizes each elevator ride task based on the acquired task priority identifier. A matching degree calculation model is constructed to calculate the matching degree between each elevator riding task and each service robot and elevator combination, and to mediate the allocation conflict caused by multiple elevator riding tasks being matched with the same elevator at the same time. Based on the matching degree and mediation results, the elevator riding task is assigned to the service robot and elevator combination with the highest matching degree. The assignment result is recorded in the task assignment table. At the same time, the assignment status of the corresponding elevator riding task is updated, and the task assignment instruction is sent to the corresponding service robot and elevator.

4. The process scheduling system for elevator use by different service robots as described in claim 3, characterized in that, The historical robot dataset includes robot position time curves, task priority time curves, and elevator request time curves for each time unit; the historical elevator dataset includes elevator operation status time curves and elevator response time curves for each time unit.

5. The process scheduling system for elevator use by different service robots as described in claim 4, characterized in that, The service robot and elevator combination is based on preset constraints, combined with the status information of each service robot and the real-time operation status data of the elevator to filter out service robots and elevators that meet the conditions, and then associate and match the service robots and elevators that meet the conditions to generate service robot and elevator combinations.

6. The process scheduling system for elevator use by different service robots as described in claim 5, characterized in that, The task allocation decision module also includes: The historical elevator data set is split into time series datasets by time units, and then fused with real-time operating status data to generate an elevator model dataset. The timestamp of the real-time data is the current time, and the timestamp of the historical data is the start time of the corresponding time unit. The elevator model dataset is preprocessed, and the preprocessed elevator model input is used as input data to construct an elevator model. Based on the elevator model, elevator operation simulation and prediction are performed, and the elevator model prediction results are obtained.

7. The process scheduling system for elevator use by different service robots as described in claim 6, characterized in that, The elevator dispatching module controls the elevator's operating status in the following way: Parse the service robot identifier, elevator start floor, target floor, and task priority information in the task allocation instruction; Based on the elevator model prediction results, combined with the starting floor and the target floor, the optimal running path of the elevator is planned, and elevator operation control instructions are generated, including the order of stopping floors, running speed adjustment parameters and door opening and closing time settings. The system monitors the actual operating status of the elevator in real time and compares it with the prediction results of the elevator model. Based on the comparison results, it generates correction instructions to dynamically adjust the elevator's operating status. When the elevator arrives at the starting floor, a positioning signal is sent to the corresponding service robot, which controls the elevator door to remain open for a preset time. After receiving feedback that the robot has entered the car, the elevator door is closed and the elevator moves to the target floor according to the path planned by the robot path planning module. When the elevator reaches the target floor, the elevator door is opened. After receiving feedback that the service robot has left the car, the elevator door is closed and the elevator status data, including the current floor, load rate and operating status indicator, is updated and fed back to the task allocation decision module.

8. The process scheduling system for elevator use by different service robots as described in claim 7, characterized in that, The robot path planning module plans the optimal path in the following way: When the service robot receives the task instruction to go to the elevator hall, the path planning module uses the robot's current position as the starting point and the elevator hall position as the ending point to calculate the path using the path planning algorithm. Once the service robot reaches the target floor and exits the elevator, the path is calculated again, starting from the elevator lobby exit and ending at the target location, to generate the optimal path for the service robot. During the calculation process, obstacle information and conflict areas are extracted, and the optimal path is smoothed based on the kinematic characteristics of the service robot.

9. The process scheduling system for elevator use by different service robots as described in claim 3, characterized in that, Calculate the matching degree between each elevator riding task and each service robot and elevator combination, including: Obtain the combinations of service robots and elevators, resulting in several combinations; Calculate the reliability between each elevator-riding task and each combination to obtain the first matching degree; ； in, This represents the first matching degree between the current elevator riding task and the i-th combination; This represents the historical success rate of the i-th combination executing the current elevator task type; This represents the real-time fault risk value for the i-th combination; This represents the reliability requirement coefficient for the current elevator riding task; This represents the idle rate of the i-th combination; This represents the repair time for the i-th combined fault; Calculate the functional compatibility between each elevator-riding task and each combination to obtain the second matching degree; ； in, This represents the second matching degree between the current elevator riding task and the i-th combination; , , Indicates the weighting coefficient; This represents the maximum safe load-bearing capacity in the i-th combination; This indicates the weight of goods to be transported in the current elevator ride. This represents the estimated time for the entire process of the current elevator ride task to be executed by the i-th combination; Indicates the end time of the entire process for the current elevator ride task; This represents the permission adaptation coefficient for the i-th combined floor; Calculate the resource compatibility between each elevator-riding task and each combination to obtain the third matching degree; ； in, This represents the third degree of matching between the current elevator ride task and the i-th combination; , , Indicates the weighting coefficient for resource adaptation; This indicates the current remaining load of the elevator in the i-th combination; This represents the remaining number of stops for the elevator in the current operating cycle of the i-th combination; This represents the standard docking redundancy threshold for the i-th combination; The target matching degree between the current elevator riding task and each combination is determined based on the first matching degree, the second matching degree, and the third matching degree; ； in, This indicates the degree of target matching between the current elevator riding task and the i-th combination; , , This represents the weighting coefficient.

10. The process scheduling system for elevator use by different service robots as described in claim 9, characterized in that, Mediation of allocation conflicts caused by multiple elevator ride requests being matched with the same elevator simultaneously, including: Obtain the set of information needed for conflict resolution; The information set is preprocessed to obtain a standardized information set; Using the task deadline in the standardized information set as the target dimension, a dynamic time window is set based on the task density of the time period, and the service robot tasks within the dynamic time window are clustered to obtain task groups. For each task group, obtain the candidate task set for the elevator corresponding to each task group; Based on the candidate task sets of each elevator, conflict prediction is performed for each task group to determine whether each task group has a potential conflict identifier. If the potential conflict identifier exists, then the task group is designated as the target task group. Based on the target matching degree, a multi-dimensional adaptation optimization model is constructed and solved to obtain the initial mediation result corresponding to the target task group; The initial mediation result is dynamically adjusted based on real-time status data to obtain the adjusted mediation result corresponding to the target task group. If conflicts still exist in the adjusted mediation results, a relaxed constraint algorithm is used to resolve the conflicts and obtain the final mediation results of the target task group.

Images