Multi-machine coordination conflict resolution method and system based on medical logistics robot scheduling

Abstract

The application provides a multi-machine coordination conflict resolution method and system based on medical logistics robot scheduling, which comprises the following steps: based on the working environment map, road network information and received instructions of multiple medical logistics robots, a coordination scheduling trajectory planning result is generated; wherein the coordination scheduling trajectory planning result is the optimal global static path of the multiple medical logistics robots; the coordination scheduling trajectory planning result is configured to the multiple medical logistics robots; the running data of each medical logistics robot is monitored in real time, and it is judged whether a conflict occurs at present. Through the synergistic effect of the prediction model and the resolution rule, the application realizes the whole-cycle conflict prevention from static planning to dynamic adjustment, ensures the transportation efficiency of high-priority materials, avoids the chain conflict caused by temporary obstacles or path deviation, and is suitable for the efficient, safe and intelligent coordination scheduling scene of multiple medical logistics robots in a complex hospital environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of conflict detection and resolution, in particular to a multi-machine cooperative conflict resolution method and system based on medical logistics robot scheduling. BACKGROUND

[0002] With the rapid development of smart hospital construction, medical logistics robots (such as medicine distribution robots, instrument transportation robots, etc.) play an important role in the transportation of materials within the hospital. The hospital environment has the characteristics of strong dynamics, complex space (such as variable department layout, narrow passageway, scattered distribution of elevators and charging stations), and high task timeliness requirements. When multiple medical logistics robots work cooperatively, they need to efficiently complete the transportation of materials under the premise of safety.

[0003] Existing methods mostly rely on mixed integer programming or genetic algorithms to generate global static optimal paths, which can ensure that the initial path has no static conflicts, but do not consider dynamic environmental changes (such as temporary intrusion of pedestrians, sudden failure of robots, etc.). Once the static path is planned, it is difficult to adjust in real time, leading to conflicts (such as path segment overlap, time window conflict) in actual operation due to dynamic obstacles or robot trajectory deviation, which requires frequent manual intervention and reduces scheduling efficiency. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a multi-machine cooperative conflict resolution method and system based on medical logistics robot scheduling to solve the problems raised in the background art.

[0005] To achieve the above purpose, the present application realizes the following technical scheme: a multi-machine cooperative conflict resolution method based on medical logistics robot scheduling, comprising the following steps:

[0006] S1: Based on the working environment map of multiple medical logistics robots, road network information and received instructions, generate a cooperative scheduling trajectory planning result;

[0007] Wherein, the cooperative scheduling trajectory planning result is the optimal global static path of multiple medical logistics robots;

[0008] S2: Configure the cooperative scheduling trajectory planning result to the multiple medical logistics robots;

[0009] S3: Real-time monitoring of the operation data of each medical logistics robot, and determining whether a conflict has occurred, if so, determining the conflict type of the current medical logistics robot and other medical logistics robots;

[0010] S4: According to the conflict type, resolve the conflict related to the relevant medical robots;

[0011] S5: Based on the digestion result, the current position information of the multiple medical logistics robots, and the cooperative scheduling trajectory planning result, a conflict prediction is performed through a prediction model, and the cooperative scheduling trajectory is updated based on the prediction result and configured to the multiple medical logistics robots.

[0012] As a further optimization, the specific steps of S1 include:

[0013] Based on the medical environment map, a topological map or a grid map is constructed, and key nodes and feasible path edges are defined.

[0014] Based on the road network information, independent path planning is performed for each medical logistics robot, and a speed adjustment strategy is used to coordinate path conflicts for each medical logistics robot.

[0015] A space-time path is generated for each medical logistics robot to ensure no static conflict, and a cooperative scheduling trajectory planning result is obtained.

[0016] As a further optimization, the process of real-time monitoring of the operation data of each medical logistics robot and determining whether a conflict occurs currently includes:

[0017] Through the positioning sensors and communication modules deployed on the medical logistics robots, the operation data of each medical logistics robot is collected in real time.

[0018] Based on the operation data, a real-time state matrix of the multiple medical logistics robots is constructed, and it is dynamically matched with the preset path in the cooperative scheduling trajectory planning result.

[0019] Based on the matching result, a conflict detection is performed, and if a conflict trigger condition is detected, a conflict type determination is performed.

[0020] The specific process of conflict detection includes:

[0021] According to the real-time state matrix, a deviation value between the predicted trajectory and the current trajectory of each medical logistics robot is calculated.

[0022] Based on a preset conflict threshold, it is determined whether a substantial conflict is formed.

[0023] If a conflict is formed, the conflict occurrence time, position, and involved robot number are marked, and a conflict type identification process is triggered.

[0024] As a further optimization, the process of conflict type determination includes conflict type classification and type matching.

[0025] The conflict type classification includes node conflict, path segment conflict, direction conflict, and dynamic obstacle conflict.

[0026] The type matching specifically matches the conflict type according to the conflict detection result, and records the conflict log.

[0027] As a further preferred, the specific flow of S5 includes:

[0028] S51: Obtain real-time running data of each medical logistics robot, and filter and denoise the real-time running data;

[0029] Wherein, the real-time running data includes current position coordinates, current speed vector, task target position and kinematics constraint of medical logistics robot;

[0030] S52: Input the processed real-time running data into a spatio-temporal graph convolution network prediction model to predict the spatio-temporal trajectory distribution of each medical logistics robot in the future time window;

[0031] S53: Calculate the conflict probability of any two medical logistics robots at any time in the future, and determine whether the two robots collide;

[0032] ;

[0033] Wherein, P ij represents the conflict probability of medical logistics robots i and j, and Ω represents the safe distance area of two robots, 、 respectively represent the two-dimensional Gaussian distribution of medical logistics robots i and j whose probability distribution is unknown at future time, 、 respectively represent the expected position of medical logistics robots i and j;

[0034] S54: If it is determined that there is a conflict, the trajectory of the relevant robot is re-planned, and a new trajectory is generated by using a spatio-temporal algorithm, and the objective function of the new trajectory is:

[0035] ;

[0036] Wherein, α and β are weight coefficients, represents the total time of trajectory, represents the conflict probability of medical logistics robots i and j, and

[0037] The constraint condition of the objective function is: kinematics constraint of robot and path smoothness;

[0038] The path smoothness is specifically: .

[0039] As a further preferred, a multi-machine cooperative conflict resolution system based on medical logistics robot scheduling is used to realize the multi-machine cooperative conflict resolution method based on medical logistics robot scheduling, and the system is characterized by comprising:

[0040] The trajectory generation module is configured to obtain a working environment map of the multiple medical logistics robots, road network information, and received instructions, and generate a cooperative scheduling trajectory planning result.

[0041] The cooperative scheduling trajectory planning result is an optimal global static path for the multiple medical logistics robots.

[0042] The trajectory configuration module is configured to configure the cooperative scheduling trajectory planning result to the multiple medical logistics robots.

[0043] The conflict determination module is configured to monitor running data of each medical logistics robot, and determine whether a conflict occurs at present. If yes, the conflict type of the current medical logistics robot and other medical logistics robots is determined.

[0044] The conflict resolution module is configured to resolve the conflict of the relevant medical robot related to the conflict according to the conflict type.

[0045] The configuration update module is configured to perform conflict prediction through a prediction model based on the resolution result, position information of the multiple medical logistics robots at present, and the cooperative scheduling trajectory planning result, update the cooperative scheduling trajectory based on the prediction result, and configure the multiple medical logistics robots.

[0046] As a further optimization, the specific process of the trajectory generation module to generate the cooperative scheduling trajectory planning result includes:

[0047] Based on the medical environment map, a topological map or a grid map is constructed, and key nodes and feasible path edges are defined.

[0048] Based on the road network information, independent path planning is performed for each medical logistics robot, and path conflict is coordinated for each medical logistics robot through a speed adjustment strategy.

[0049] A space-time path is generated for each medical logistics robot to ensure no static conflict, and the cooperative scheduling trajectory planning result is obtained.

[0050] As a further optimization, the process of real-time monitoring of the running data of each medical logistics robot and determination of whether a conflict occurs at present includes:

[0051] The running data of each medical logistics robot is collected in real time through a positioning sensor and a communication module deployed on the medical logistics robot.

[0052] Based on the running data, a real-time state matrix of the multiple medical logistics robots is constructed, and is dynamically matched with a preset path in the cooperative scheduling trajectory planning result.

[0053] Based on the matching result, conflict detection is performed. If a conflict trigger condition is detected, conflict type determination is performed.

[0054] The specific process of conflict detection includes:

[0055] According to the real-time state matrix, the deviation value of the predicted trajectory and the current trajectory of each medical logistics robot is calculated;

[0056] Based on the preset conflict threshold, it is judged whether a substantial conflict is constituted;

[0057] If a conflict is constituted, the conflict occurrence time, position and involved robot number are marked, and a conflict type identification process is triggered.

[0058] As a further optimization, the conflict type determination process includes conflict type classification and type matching;

[0059] The conflict type classification includes node conflict, path segment conflict, direction conflict and dynamic obstacle conflict.

[0060] The type matching is specifically matching the conflict type according to the conflict detection result, and recording the conflict log.

[0061] As a further optimization, the specific process of updating the collaborative scheduling trajectory based on the resolution result, the current position information of the multiple medical logistics robots and the collaborative scheduling trajectory planning result, and configuring the multiple medical logistics robots includes:

[0062] Obtain the real-time running data of each medical logistics robot, and perform filtering and noise reduction processing on the real-time running data;

[0063] The real-time running data includes the current position coordinates, the current speed vector, the task target position and the kinematic constraints of the medical logistics robot.

[0064] The processed real-time running data is input into a spatio-temporal graph convolution network prediction model to predict the spatio-temporal trajectory distribution of each medical logistics robot in a future time window.

[0065] The conflict probability of any two medical logistics robots at any time in the future is calculated, and it is determined whether the two robots have a conflict.

[0066] ;

[0067] Wherein, P ij represents the conflict probability of medical logistics robots i and j, and Ω represents the safe distance region of the two robots, 、 respectively represent the two-dimensional Gaussian distribution of the probability distribution of medical logistics robots i and j at a future time, 、 respectively represent the expected positions of medical logistics robots i and j.

[0068] If it is determined that there is a conflict, the trajectory of the relevant robot is re-planned, a space-time algorithm is used to generate a new trajectory, and the objective function of the new trajectory is:

[0069] ;

[0070] wherein, and are weight coefficients, denotes the total time of the trajectory, denotes the conflict probability of the medical logistics robots i and j, and

[0071] The constraint condition of the objective function is: the kinematics constraint of the robot and the path smoothness;

[0072] The path smoothness is specifically: .

[0073] The application provides a multi-machine cooperative conflict resolution method and system based on medical logistics robot scheduling, which has the following beneficial effects: through the fusion optimization of multi-machine cooperative static path planning and real-time dynamic conflict prediction, the global optimal initial trajectory is generated based on the medical environment map and road network information, and the mixed integer programming or genetic algorithm is used to ensure that the path is free of static conflicts, thereby realizing the cooperative guarantee of initial scheduling efficiency and safety, effectively avoiding resource waste and collision risk caused by path overlap; and through the deployment of positioning sensors and communication modules to collect robot operation data in real time, a real-time state matrix is constructed and the preset path is dynamically matched, the types of node conflict, path segment conflict and dynamic obstacle conflict are accurately identified, and conflict rapid response is realized according to the material priority, task remaining time and other resolution rules; at the same time, the future trajectory distribution of the operation data filtered and denoised in real time is predicted by means of a space-time graph convolution network prediction model, the conflict probability is quantified and the trajectory is dynamically adjusted, the kinematics of the robot and the path smoothness are constrained by the objective function, and iterative optimization is performed until the conflict probability is lower than a threshold, thereby significantly improving the robustness and dynamic adaptability of the scheduling system; in addition, the cooperative action of the prediction model and the resolution rule realizes the whole-cycle conflict prevention from static planning to dynamic adjustment, thereby guaranteeing the transportation efficiency of high-priority materials while avoiding the chain reaction of conflicts caused by temporary obstacles or path deviation, and being suitable for efficient, safe and intelligent cooperative scheduling of multiple medical logistics robots in a complex hospital environment. BRIEF DESCRIPTION OF DRAWINGS

[0074] Figure 1 The flowchart of the multi-machine cooperative conflict resolution method based on medical logistics robot scheduling of the application is shown in the figure;

[0075] Figure 2 The block diagram of the multi-machine cooperative conflict resolution system based on medical logistics robot scheduling of the application is shown in the figure. DETAILED DESCRIPTION

[0076] Embodiments of the present application are described below in the context of fully functional methods and devices, systems, and applications. Those of ordinary skill in the art will realize that the embodiments of the present application are capable of other different embodiments, and their equivalents, and that the embodiments of the present application are not limited to the particular embodiments described herein. In the following description, for the purposes of explanation and not limitation, specific details are set forth such as particular embodiments. However, it will be apparent to those skilled in the art that the embodiments of the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods and devices are omitted so as not to obscure the description of the embodiments of the present application with details that are well known to those skilled in the art.

[0077] The following disclosure provides many different embodiments, or examples, for implementing different structures of the present application. For purposes of explanation and not limitation, specific details of certain embodiments are described in the following description. Of course, they are merely examples and other embodiments can be used without departing from the scope of the application. Also, the present application is described in the context of particular embodiments with reference to the attached drawings, which are not to scale, and in which, for purposes of presentation, some features can have exaggerated or simplified in the description and drawings. In addition, the present application provides examples of various specific processes and materials, but one skilled in the art will recognize that other processes and / or materials can be used.

[0078] As shown in Figure 1 The embodiment of the present application provides a multi-machine coordination conflict resolution method based on medical logistics robot scheduling, comprising the following steps:

[0079] S1: Based on the working environment map of the multi-medical logistics robot, the road network information and the received instruction, the cooperative scheduling trajectory planning result is generated;

[0080] The cooperative scheduling trajectory planning result is the optimal global static path of the multi-medical logistics robot.

[0081] Specifically, based on the medical environment map (such as hospital department layout, passage width, elevator position, etc.), a topological map or a grid map is constructed, and key nodes (such as task points, charging stations, intersections) and feasible path edges are defined.

[0082] Based on the road network information, independent path planning is performed for each medical logistics robot, and the speed adjustment strategy is used to coordinate the path conflict of each medical logistics robot.

[0083] The road network information includes path length, expected travel time, priority (such as emergency passage) and other weights.

[0084] A space-time path is generated for each medical logistics robot to ensure no static conflict, and the cooperative scheduling trajectory planning result is obtained.

[0085] It should be noted that when planning the path of the medical logistics robot, a mixed integer programming or genetic algorithm is used to optimize the path, generate a global static optimal trajectory, and ensure that the initial path has no conflict or the conflict probability is the lowest.

[0086] S2: Configure the collaborative scheduling trajectory planning results to multiple medical logistics robots;

[0087] S3: Monitor the operational data of each medical logistics robot in real time and determine whether a conflict has occurred;

[0088] If so, determine the type of conflict between the current medical logistics robot and other medical logistics robots;

[0089] Specifically, the operation data of each medical logistics robot is collected in real time through positioning sensors and communication modules deployed on the medical logistics robots.

[0090] The operational data includes: current geographical coordinates, movement speed and direction, task execution status, and real-time tracking data of the path trajectory.

[0091] A real-time state matrix of multiple medical logistics robots is constructed based on operational data, and it is dynamically matched with the preset path in the collaborative scheduling trajectory planning results.

[0092] Specifically, by aligning timestamps and comparing path nodes, it is determined whether there are path overlaps, time window conflicts, or spatial area occupancy conflicts.

[0093] Conflict detection is performed based on the matching results. If a conflict trigger condition is detected, a conflict type determination is performed.

[0094] The specific process for collision detection includes:

[0095] The deviation between the predicted trajectory and the current trajectory of each medical logistics robot (hereinafter referred to as the robot) is calculated based on the real-time state matrix.

[0096] Whether a substantive conflict exists is determined based on preset conflict thresholds (such as minimum safe distance or time interval).

[0097] If a conflict occurs, mark the time, location, and robot number involved in the conflict, and trigger the conflict type identification process.

[0098] Conflict detection includes: path segment conflict detection, node conflict detection, and dynamic obstacle conflict detection;

[0099] Path segment conflict detection: Compare all future path segments of robots (which can be set to a certain future period); check whether the path segments intersect and the time windows overlap (e.g., two robots enter the same node at the same time).

[0100] Node conflict detection: Extract the node occupancy plan of each robot within the time window; determine whether multiple robots plan to arrive at the same node at the same time.

[0101] Dynamic obstacle conflict: predict collision risk by combining real-time obstacle locations (e.g., pedestrians, other robots).

[0102] The monitored conflict events and operational data are transmitted to the central dispatch system in real time to update the global status database to support subsequent conflict resolution and trajectory prediction.

[0103] In this embodiment, the conflict type determination process includes conflict type classification and type matching;

[0104] The conflict types are categorized as follows:

[0105] Node conflict: Two robots arrive at the same node at the same time.

[0106] Path segment conflict: Two robot path segments overlap within a time window (e.g., moving towards each other).

[0107] Directional conflict: The robot's path direction is opposite and the path segments overlap.

[0108] Dynamic obstacle conflict: Path obstruction caused by pedestrians or temporary obstacles.

[0109] Type matching includes:

[0110] Match the conflict type based on the conflict detection results and record the conflict log.

[0111] The conflict log includes time, robot ID, conflict type, and location;

[0112] It should be noted that the robot ID is specifically assigned to each medical logistics robot during the system initialization phase using preset coding rules.

[0113] S4: Resolve conflicts related to medical robots based on the type of conflict;

[0114] Specifically, based on the tag-based conflict log, the resolution priority is determined according to preset rules;

[0115] Among the preset rules is: robots with higher priority in transporting medical logistics materials will have priority in obtaining the right of way;

[0116] If the priority of the resources is the same, the priority is determined by the remaining time of the robot's current task, with the robot with the shorter remaining time taking priority.

[0117] If all the above conditions are the same, the robot with the smaller number will be given priority.

[0118] S5: Based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results, conflict prediction is performed through a prediction model. Based on the prediction results, the collaborative scheduling trajectory is updated and configured to multiple medical logistics robots.

[0119] The specific process of S5 includes:

[0120] S51: Acquire real-time operating data of each medical logistics robot and perform filtering and noise reduction processing on the real-time operating data;

[0121] The real-time operational data includes the current position coordinates, current velocity vector, target position, and kinematic constraints of the medical logistics robot.

[0122] S52: Input the processed real-time running data into the spatiotemporal graph convolutional network prediction model to predict the spatiotemporal trajectory distribution of each medical logistics robot within the future time window;

[0123] S53: Calculate the probability of conflict between any two medical logistics robots at any future time, and determine whether the two robots will conflict.

[0124] ;

[0125] Among them, P ij Let Ω represent the collision probability between medical logistics robots i and j, and let Ω represent the safe distance between the two robots. , Let i and j be two-dimensional Gaussian distributions with unknown probability distributions in the future. , Let i and j represent the desired positions of medical logistics robots, respectively.

[0126] S54: If a conflict is identified, the trajectory of the relevant robot is replanned, and a new trajectory is generated using a spatiotemporal algorithm. The objective function of the new trajectory is:

[0127] ;

[0128] Where α and β are both weighting coefficients. This represents the total time for the trajectory. Let represent the sum of the conflict probabilities of medical logistics robots i and j.

[0129] The constraints of the objective function are: robot kinematic constraints and path smoothness;

[0130] Path smoothness specifically refers to: ;

[0131] The updated trajectory is sent to the relevant robots, and the collaborative scheduling trajectory is updated.

[0132] It should be noted that if the new trajectory increases the probability of conflict with other robots, iterative optimization will be triggered, and steps S51-S54 will be repeated until the probability of conflict is lower than the threshold.

[0133] Specifically, a predictive model is used to predict potential conflicts among multiple medical logistics robots after conflict resolution, and the collaborative scheduling trajectory is dynamically adjusted to avoid conflicts.

[0134] The multi-robot collaborative conflict resolution method for medical logistics robot scheduling provided in this embodiment integrates and optimizes multi-robot collaborative static path planning with real-time dynamic conflict prediction. It generates a globally optimal initial trajectory based on a medical environment map and road network information, and ensures no static conflicts on the path through mixed integer programming or genetic algorithms. This achieves coordinated assurance of initial scheduling efficiency and safety, effectively avoiding resource waste and collision risks caused by path overlap. Furthermore, by deploying positioning sensors and communication modules to collect robot operation data in real time, a real-time state matrix is ​​constructed and dynamically matched with preset paths. This accurately identifies node conflicts, path segment conflicts, and dynamic obstacle conflict types, and achieves rapid conflict response based on material priority, remaining task time, and other resolution rules. Simultaneously, a spatiotemporal graph convolutional network prediction model is used to predict the future trajectory distribution of the real-time filtered and denoised operation data, quantifying the conflict probability and dynamically adjusting the trajectory. The objective function constrains the robot's kinematics and path smoothness, iteratively optimizing until the conflict probability is below a threshold, significantly improving the robustness and dynamic adaptability of the scheduling system.

[0135] Furthermore, the synergistic effect of the predictive model and the resolution rules enables full-cycle conflict prevention from static planning to dynamic adjustment. While ensuring the efficiency of high-priority material transportation, it avoids chain conflicts caused by temporary obstacles or path deviations. It is suitable for efficient, safe, and intelligent collaborative scheduling scenarios of multiple medical logistics robots in complex hospital environments.

[0136] like Figure 2 As shown, this embodiment also provides a multi-machine collaborative conflict resolution system based on medical logistics robot scheduling, used to implement the above-mentioned multi-machine collaborative conflict resolution method based on medical logistics robot scheduling, characterized in that it includes:

[0137] The trajectory generation module is used to acquire the working environment map, road network information and received instructions of multiple medical logistics robots, and generate collaborative scheduling trajectory planning results;

[0138] Among them, the collaborative scheduling trajectory planning result is the optimal global static path for multiple medical logistics robots;

[0139] The trajectory configuration module is used to configure the collaborative scheduling trajectory planning results to multiple medical logistics robots;

[0140] The conflict determination module is used to monitor the operating data of each medical logistics robot and determine whether a conflict has occurred. If so, it determines the type of conflict between the current medical logistics robot and other medical logistics robots.

[0141] The conflict resolution module is used to resolve conflicts between relevant medical robots based on the type of conflict.

[0142] The configuration update module performs conflict prediction based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results through a prediction model. Based on the prediction results, it updates the collaborative scheduling trajectory and configures it to multiple medical logistics robots.

[0143] Furthermore, the specific process for the trajectory generation module to generate collaborative scheduling trajectory planning results includes:

[0144] Based on the medical environment map, construct a topology map or grid map, and define key nodes and feasible path edges;

[0145] Based on road network information, independent path planning is performed for each medical logistics robot, and path conflicts are coordinated for each medical logistics robot through speed adjustment strategies.

[0146] A spatiotemporal path is generated for each medical logistics robot to ensure no static conflicts, thus obtaining the collaborative scheduling trajectory planning results.

[0147] Furthermore, the process of real-time monitoring of the operational data of each medical logistics robot and determining whether a conflict has occurred includes:

[0148] The operation data of each medical logistics robot is collected in real time through positioning sensors and communication modules deployed on the robots.

[0149] A real-time state matrix of multiple medical logistics robots is constructed based on operational data, and it is dynamically matched with the preset path in the collaborative scheduling trajectory planning results.

[0150] Conflict detection is performed based on the matching results. If a conflict trigger condition is detected, a conflict type determination is performed.

[0151] The specific process for collision detection includes:

[0152] The deviation between the predicted trajectory and the current trajectory of each medical logistics robot is calculated based on the real-time state matrix.

[0153] Whether a substantive conflict exists is determined based on a preset conflict threshold;

[0154] If a conflict occurs, mark the time, location, and robot number involved in the conflict, and trigger the conflict type identification process.

[0155] As a further optimization, the conflict type determination process includes conflict type classification and type matching;

[0156] The conflict types are categorized as follows: node conflict, path segment conflict, direction conflict, and dynamic obstacle conflict.

[0157] Type matching specifically involves matching conflict types based on conflict detection results and recording conflict logs.

[0158] Furthermore, based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results, conflict prediction is performed using a prediction model. The specific process of updating the collaborative scheduling trajectory and configuring it to multiple medical logistics robots based on the prediction results includes:

[0159] Acquire real-time operating data of each medical logistics robot and perform filtering and noise reduction processing on the real-time operating data;

[0160] The real-time operational data includes the current position coordinates, current velocity vector, target position, and kinematic constraints of the medical logistics robot.

[0161] The processed real-time running data is input into the spatiotemporal graph convolutional network prediction model to predict the spatiotemporal trajectory distribution of each medical logistics robot within a future time window.

[0162] Calculate the probability of conflict between any two medical logistics robots at any future time, and determine whether the two robots will conflict.

[0163] ;

[0164] Among them, P ij Let Ω represent the collision probability between medical logistics robots i and j, and let Ω represent the safe distance between the two robots. , Let i and j be two-dimensional Gaussian distributions with unknown probability distributions in the future. , Let i and j represent the desired positions of medical logistics robots, respectively.

[0165] If a conflict is identified, the trajectory of the relevant robot is replanned, and a new trajectory is generated using a spatiotemporal algorithm. The objective function of the new trajectory is:

[0166] ;

[0167] Where α and β are both weighting coefficients, This represents the total time for the trajectory. Let represent the sum of the conflict probabilities of medical logistics robots i and j;

[0168] The constraints of the objective function are: robot kinematic constraints and path smoothness;

[0169] Path smoothness specifically refers to: .

[0170] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for resolving multi-machine collaborative conflicts based on medical logistics robot scheduling, characterized in that, Includes the following steps: S1: Based on the working environment map, road network information and received instructions of multiple medical logistics robots, generate collaborative scheduling trajectory planning results; Among them, the collaborative scheduling trajectory planning result is the optimal global static path for multiple medical logistics robots; S2: Configure the collaborative scheduling trajectory planning results to multiple medical logistics robots; S3: Monitor the operating data of each medical logistics robot in real time and determine whether a conflict has occurred. If so, determine the type of conflict between the current medical logistics robot and other medical logistics robots. S4: Resolve conflicts related to the medical robots involved in the conflict according to the conflict type; S5: Based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results, conflict prediction is performed through a prediction model. The collaborative scheduling trajectory is updated and configured to multiple medical logistics robots based on the prediction results. The specific process of S5 includes: S51: Acquire real-time operating data of each medical logistics robot and perform filtering and noise reduction processing on the real-time operating data; The real-time operational data includes the current position coordinates, current velocity vector, target position, and kinematic constraints of the medical logistics robot. S52: Input the processed real-time running data into the spatiotemporal graph convolutional network prediction model to predict the spatiotemporal trajectory distribution of each medical logistics robot within the future time window; S53: Calculate the probability of conflict between any two medical logistics robots at any future time, and determine whether the two robots will conflict. ； Among them, P ij Let Ω represent the collision probability between medical logistics robots i and j, and let Ω represent the safe distance between the two robots. , Let i and j be two-dimensional Gaussian distributions with unknown probability distributions in the future. , Let i and j represent the desired positions of medical logistics robots, respectively. S54: If a conflict is identified, the trajectory of the relevant robot is replanned, and a new trajectory is generated using a spatiotemporal algorithm. The objective function of the new trajectory is: ； Where α and β are both weighting coefficients, This represents the total time for the trajectory. Let represent the sum of the conflict probabilities of medical logistics robots i and j; The constraints of the objective function are: robot kinematic constraints and path smoothness; Path smoothness specifically refers to: .

2. The method for resolving multi-machine collaborative conflicts based on medical logistics robot scheduling according to claim 1, characterized in that, The specific steps of S1 include: Based on the medical environment map, construct a topology map or grid map, and define key nodes and feasible path edges; Based on road network information, independent path planning is performed for each medical logistics robot, and path conflicts are coordinated for each medical logistics robot through speed adjustment strategies. A spatiotemporal path is generated for each medical logistics robot to ensure no static conflicts, thus obtaining the collaborative scheduling trajectory planning results.

3. The method for resolving multi-machine collaborative conflicts based on medical logistics robot scheduling according to claim 1, characterized in that, The process of real-time monitoring of the operational data of each medical logistics robot and determining whether a conflict has occurred includes: The operation data of each medical logistics robot is collected in real time through positioning sensors and communication modules deployed on the robots. Based on the operational data, a real-time state matrix of multiple medical logistics robots is constructed, and it is dynamically matched with the preset path in the collaborative scheduling trajectory planning result. Conflict detection is performed based on the matching results. If a conflict trigger condition is detected, a conflict type determination is performed. The specific process for collision detection includes: The deviation between the predicted trajectory and the current trajectory of each medical logistics robot is calculated based on the real-time state matrix. Whether a substantive conflict exists is determined based on a preset conflict threshold; If a conflict occurs, mark the time, location, and robot number involved in the conflict, and trigger the conflict type identification process.

4. The method for resolving multi-machine collaborative conflicts based on medical logistics robot scheduling according to claim 3, characterized in that, The conflict type determination process includes conflict type classification and type matching; The conflict types are categorized as follows: node conflict, path segment conflict, direction conflict, and dynamic obstacle conflict. Type matching specifically involves matching conflict types based on conflict detection results and recording conflict logs.

5. A multi-machine collaborative conflict resolution system based on medical logistics robot scheduling, used to implement the multi-machine collaborative conflict resolution method based on medical logistics robot scheduling as described in any one of claims 1-4, characterized in that, include: The trajectory generation module is used to acquire the working environment map, road network information and received instructions of multiple medical logistics robots, and generate collaborative scheduling trajectory planning results; Among them, the collaborative scheduling trajectory planning result is the optimal global static path for multiple medical logistics robots; The trajectory configuration module is used to configure the collaborative scheduling trajectory planning results to multiple medical logistics robots; The conflict determination module is used to monitor the operating data of each medical logistics robot and determine whether a conflict has occurred. If so, it determines the type of conflict between the current medical logistics robot and other medical logistics robots. A conflict resolution module is used to resolve conflicts between relevant medical robots associated with the conflict based on the conflict type. The configuration update module performs conflict prediction based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results through a prediction model. Based on the prediction results, it updates the collaborative scheduling trajectory and configures it to multiple medical logistics robots. The specific process of predicting conflicts based on the resolution results, the current location information of multiple medical logistics robots, and the collaborative scheduling trajectory planning results, and updating the collaborative scheduling trajectory and configuring it to multiple medical logistics robots based on the prediction results includes: Acquire real-time operating data of each medical logistics robot and perform filtering and noise reduction processing on the real-time operating data; The real-time operational data includes the current position coordinates, current velocity vector, target position, and kinematic constraints of the medical logistics robot. The processed real-time running data is input into the spatiotemporal graph convolutional network prediction model to predict the spatiotemporal trajectory distribution of each medical logistics robot within a future time window. Calculate the probability of conflict between any two medical logistics robots at any future time, and determine whether the two robots will conflict. ； Among them, P ij Let Ω represent the collision probability between medical logistics robots i and j, and let Ω represent the safe distance between the two robots. , Let i and j be two-dimensional Gaussian distributions with unknown probability distributions in the future. , Let i and j represent the desired positions of medical logistics robots, respectively. If a conflict is identified, the trajectory of the relevant robot is replanned, and a new trajectory is generated using a spatiotemporal algorithm. The objective function of the new trajectory is: ； Where α and β are both weighting coefficients, This represents the total time for the trajectory. Let represent the sum of the conflict probabilities of medical logistics robots i and j; The constraints of the objective function are: robot kinematic constraints and path smoothness; Path smoothness specifically refers to: .

6. A multi-machine collaborative conflict resolution system based on medical logistics robot scheduling according to claim 5, characterized in that, The specific process by which the trajectory generation module generates collaborative scheduling trajectory planning results includes: Based on the medical environment map, construct a topology map or grid map, and define key nodes and feasible path edges; Based on road network information, independent path planning is performed for each medical logistics robot, and path conflicts are coordinated for each medical logistics robot through speed adjustment strategies. A spatiotemporal path is generated for each medical logistics robot to ensure no static conflicts, thus obtaining the collaborative scheduling trajectory planning results.

7. A multi-machine collaborative conflict resolution system based on medical logistics robot scheduling according to claim 5, characterized in that, The process of real-time monitoring of the operational data of each medical logistics robot and determining whether a conflict has occurred includes: The operation data of each medical logistics robot is collected in real time through positioning sensors and communication modules deployed on the robots. Based on the operational data, a real-time state matrix of multiple medical logistics robots is constructed, and it is dynamically matched with the preset path in the collaborative scheduling trajectory planning result. Conflict detection is performed based on the matching results. If a conflict trigger condition is detected, a conflict type determination is performed. The specific process for collision detection includes: The deviation between the predicted trajectory and the current trajectory of each medical logistics robot is calculated based on the real-time state matrix. Whether a substantive conflict exists is determined based on a preset conflict threshold; If a conflict occurs, mark the time, location, and robot number involved in the conflict, and trigger the conflict type identification process.

8. A multi-machine collaborative conflict resolution system based on medical logistics robot scheduling according to claim 7, characterized in that, The conflict type determination process includes conflict type classification and type matching; The conflict types are categorized as follows: node conflict, path segment conflict, direction conflict, and dynamic obstacle conflict. Type matching specifically involves matching conflict types based on conflict detection results and recording conflict logs.

Images