Patrol robot based on remote cooperation and consultation support

Abstract

The invention relates to the technical field of remote medical treatment, in particular to a patrol robot based on remote cooperation and consultation support, and the robot comprises an instruction analysis module, an authority scheduling module, a path exchange module, a task execution module and an instruction linkage module. According to the method, the action direction, the trigger time period, the target coordinate and the structure of the equipment part in the patrol instruction are disassembled, the doctor patrol behavior execution record is subjected to combined classification of continuous task frequency and target range coverage, and after the doctor behaviors with high control tendency are clearly sorted, the gravity center is controlled to be gradually focused on the actual master control doctor end; in a scene of simultaneous intervention of multiple doctors, through a synchronous pairing and priority ordering mechanism of voice and focus instruction fields, a master control and cooperative control instruction relation is determined, a device response link under multi-instruction input is kept clear and independent, and interaction stability of cooperation of the multiple doctors and structural smoothness of instruction execution in remote patrol diagnosis are comprehensively enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of telemedicine, in particular to a patrol robot based on remote collaboration and consultation support. BACKGROUND

[0002] The technical field of telemedicine includes a system for realizing the transmission and sharing of medical resources at a distance by using information technology and communication networks. The core content of this technical field includes remote diagnosis, remote consultation, remote monitoring, remote surgical guidance, and chronic disease management. Through the construction of a service platform based on the Internet, the Internet of Things, and mobile communication, patients in non-medical institutions or primary medical institutions can obtain high-quality medical resources. Telemedicine relies on audio and video transmission technology, clinical data acquisition and transmission equipment, and intelligent scheduling and management methods to cover multiple links such as patient health data acquisition, transmission, storage, analysis, and interaction. It is widely used in cross-regional collaborative medical services between urban and remote areas, and between primary and superior medical institutions, promoting the balanced allocation of medical resources and the construction of a hierarchical diagnosis and treatment system.

[0003] Among them, the patrol robot based on remote collaboration and consultation support refers to a mobile medical auxiliary device that realizes real-time interaction between doctors and patients at different locations through wireless communication and remote control technology. It has the ability to conduct examinations and assist in preliminary diagnosis on site. The technical matters covered by this patent topic include multi-source audio and video information acquisition, mobile terminal remote navigation, doctor-end control instruction analysis, real-time consultation data transmission, etc. By integrating variable focal length high-definition camera components, high-sensitivity sound pickup devices, image compression and encoding modules, and remote transmission interfaces in the robot platform, doctors can control the camera angle and microphone sound pickup direction in real time through the remote interface, obtain on-site image and voice information, and realize information access of remote consultation experts and multi-party voice and video interaction.

[0004] Although the basic operation of the remote medical device is remotely implemented in the prior art, in the multi-expert synchronous instruction input scene, there is no differential reception of instruction processing, and there is no pre-judgment mechanism for the overlap of physical paths between instruction targets and device focus, resulting in frequent interruption or error response of device action. In the consultation site, when one doctor operates the robot to fix the view, another doctor instructs it to navigate and shift, the two action instruction targets overlap but the directions are opposite, which can easily cause task execution conflict. The current system does not establish a unified behavior pattern recognition path after the doctor's instruction input, ignores the reflection of instruction frequency and coverage area on actual control intention, often processes the instructions of multiple doctors equally, causes the lack of primary and secondary stratification of resource scheduling, and reduces the consultation instruction processing efficiency. In terms of path regulation, the existing system mainly uses fixed path scheduling, does not consider multi-dimensional service factors such as patient condition and waiting time in the node optimization link, resulting in the lack of real-time and medical priority response ability in the selection of patrol diagnosis nodes. In the remote collaboration scene, multiple doctors' voices and control focus are prone to contention, and it is difficult to establish a clear interaction entrance according to the control priority relationship, which affects the overall collaborative control effect. SUMMARY

[0005] In order to solve the technical problems of the prior art that instruction processing is received without discrimination, and there is no pre-judgment mechanism for the overlap of physical paths between instruction targets and device focus, resulting in frequent interruption or error response of device action. In the consultation site, when one doctor operates the robot to fix the view, another doctor instructs it to navigate and shift, the two action instruction targets overlap but the directions are opposite, which can easily cause task execution conflict. The current system does not establish a unified behavior pattern recognition path after the doctor's instruction input, ignores the reflection of instruction frequency and coverage area on actual control intention, often processes the instructions of multiple doctors equally, causes the lack of primary and secondary stratification of resource scheduling, and reduces the consultation instruction processing efficiency. In terms of path regulation, the existing system mainly uses fixed path scheduling, does not consider multi-dimensional service factors such as patient condition and waiting time in the node optimization link, resulting in the lack of real-time and medical priority response ability in the selection of patrol diagnosis nodes. In the remote collaboration scene, multiple doctors' voices and control focus are prone to contention, and it is difficult to establish a clear interaction entrance according to the control priority relationship, which affects the overall collaborative control effect, the present application provides a patrol diagnosis robot based on remote collaboration and consultation support. The technical solution is as follows: On the one hand, a patrol diagnosis robot based on remote collaboration and consultation support is provided, which comprises: An instruction analysis module is used to extract the action direction field, trigger time period, target position coordinate field and target device part field based on the patrol diagnosis instruction submitted by the remote consultation end, perform direction vector comparison on the action direction field and the navigation direction value in the robot state, and generate a parallel control instruction list; The authority scheduling module is configured to extract the doctor number and the consultation stage task execution record through the parallel control instruction list, mark the doctor whose continuous task number exceeds two as an active execution state, mark the doctor whose patient number range coverage area is lower than two wards as a concentrated task source doctor, perform task control behavior classification, and generate a doctor patrol task control distribution table; The path replacement module is configured to call the doctor patrol task control distribution table, perform priority comparison between the service state field of the navigation node and the target node state field in the set, and generate patrol node change identification content. The task execution module is configured to extract node path segment instructions using the patrol node change identification content, perform available state checking on the camera component, the sound pickup component, and the signal sending component, perform device activation operation on the callable components, and generate a patrol task start state registration item.

[0006] As a further scheme of the present application, the parallel control instruction list includes a control target number sequence, a direction consistency label, a coordinate region pairing result, and a conflict exclusion identification. The doctor patrol task control distribution table includes an execution frequency classification label, a target region distribution category, a doctor identity number mapping sequence, and a behavior tendency ordering structure. The patrol node change identification content includes a node update number, a service state label grouping, a navigation redirection identification, and a path priority execution mark. The patrol task start state registration item includes a device activation number list, a component state mapping value, a path segment execution identification, and a task queue update result.

[0007] As a further scheme of the present application, the instruction analysis module includes: The field extraction submodule is configured to extract an action direction field, a trigger time period, a target position coordinate field, and a target device part field based on the patrol instruction submitted by the remote consultation end, divide the data types respectively and record as action vector items and positioning coordinate items, and generate an instruction data structure set. The direction comparison submodule is configured to perform vector splitting processing on the action vector items in the instruction data structure set and the navigation direction data, perform difference judgment according to the included angle between the split direction items and the set direction reference angle, mark the direction items falling into the deviation allowable interval as navigation input, and generate a direction matching label set. The coordinate matching submodule is configured to perform spatial distance measurement according to the positioning coordinate items in the instruction data structure set and the real-time path node coordinate set, calculate the Euclidean distance between the coordinate point pairs and perform node attribution determination combining the trajectory continuous feature, mark the attribution node as a controlled target region, bind the label matching item and the anchor point set, and generate a parallel path control list.

[0008] As a further scheme of the present application, the authority scheduling module includes: The task extraction submodule is used to extract the corresponding doctor number and consultation stage task execution record based on the parallel control instruction list, number and merge the task records of different time periods under the same doctor number according to the time order, and uniformly mark and filter out interrupted tasks to generate a doctor task list. The execution status identification submodule is used to call the length information of each task sequence in the doctor task list, calculate the frequency of consecutive tasks, determine whether the number of consecutive tasks exceeds two, and mark it as an active execution status if the condition is met. It also counts the number of ward numbers covered by the associated patient number in each doctor task, marks doctors with fewer than two ward numbers as centralized task source doctors, and generates a doctor status identification sequence. The behavior classification submodule is used to classify doctors in active execution status and doctors from centralized task sources into differentiated task control behavior classification groups according to the doctor status identifier sequence, and to call the task participation time period, task coverage ward number and task type to calculate the task execution time period deviation value, and to compare the horizontal task distribution ratio based on the behavior classification to generate a doctor rounds task control distribution table.

[0009] As a further aspect of the present invention, the deviation value during the task execution period is calculated using the following formula: ; in, This represents the deviation value during the task execution period. This represents the number of doctors in an active execution state. Representing the The number of real-time task days for each doctor within a real-time task cycle. Representing the The earliest execution time number of the same type of task for each doctor. Representing the The latest execution time number of the same type of task for each doctor. Representing the The doctor's target task time center point.

[0010] As a further aspect of the present invention, the path switching module includes: The data processing submodule is used to call the doctor's rounds task control distribution table, extract the patient waiting time record, symptom level code and dense visit record fields corresponding to the navigation node, and merge the three data according to the navigation node number, arrange them into a continuous arrangement structure in chronological order, and generate a node task load set. The state comparison submodule is configured to compare the service state field of the navigation node with the real-time state field of the target node in the node task load set according to the service occupancy rate, the active consultation frequency, and the queue length in sequence, and generate a node priority switching pair sequence by comparing the corresponding bit positions of the state sequences of the navigation node and the target node. The node adjustment submodule is configured to extract the position information, the task arrangement period, and the consultation intensive frequency of the target node to be replaced in the node priority switching pair sequence, update the execution sequence of the replaceable node according to the priority switching bit position, and generate a patrol node change identification content according to the real-time navigation node path structure and the path segment replacement identification content.

[0011] As a further scheme of the present application, the task execution module comprises: The path number arrangement submodule is configured to extract the marked path segment number and the corresponding node operation instruction by using the patrol node change identification content, arrange the continuous path segments in sequence according to the node index to form a path segment execution sequence, evaluate the trigger index association relationship between the node and the path segment, and generate a path segment instruction sequence set. The device detection submodule is configured to read the state field of the camera component, the sound pickup component, and the signal sending component arranged in the path segment by using the path segment instruction sequence set, determine whether the component state is marked as a callable state, eliminate the component entries with the unusable state field and record the node index, and generate a patrol component callable index group. The task activation submodule is configured to perform device initialization, function interface opening, and signal channel connection operations on the camera, sound pickup, and signal sending components according to the available component number under the node in the patrol component callable index group, bind the device general activation process in the path segment instruction under the node, and generate a patrol task start state registration item.

[0012] As a further scheme of the present application, the robot further comprises an instruction linkage module. The instruction linkage module is configured to extract the voice instruction field and the operation focus field in the active state in the doctor end interactive interface by using the patrol task start state registration item, perform control target pairing on the focus field and the target part field, arrange the voice instruction field, mark the first field in the sequence as the master control, and mark the remaining fields as the collaboration, and generate a remote collaborative control instruction set. The remote collaborative control instruction set comprises a master doctor number, a voice control link, a synchronous operation mapping structure, and an interactive focus association item.

[0013] As a further scheme of the present application, the instruction linkage module comprises: The instruction extraction submodule is configured to extract the voice instruction field and the operation focus field in the active state in the doctor terminal interactive interface by using the patrol task start state registration item, perform field screening based on the active state identifier, calculate a field interaction difference characteristic value, and generate an interface interaction control sequence. The target pairing submodule is configured to call the target part field in the operation focus field and the patrol node change identifier content in the interface interaction control sequence, obtain an overlapping area of the focus and the part according to field position coding comparison, and generate a control target mapping structure. The instruction control submodule is configured to sequentially rearrange the instruction fields according to the control target mapping structure according to the time stamp, mark the first field in the sequence as a master control field, mark the remaining fields as collaborative fields, bind the master control field and the corresponding target part in the control target mapping structure group and link the collaborative fields, and generate a remote collaborative control instruction set.

[0014] As a further scheme of the present application, the formula for calculating the field interaction difference characteristic value is: ; Among them, represents the field interaction difference characteristic value, represents the normalized weight of the th active state identifier value, represents the number of data items with a positive trigger state value in the th voice instruction field, represents the real-time value of the corresponding data item in the th operation focus field, represents the number of times of non-activation state accumulation of the th field, represents the total number of extracted fields.

[0015] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: The parallelism recognition mechanism of the patrol task is established by matching the navigation direction value and the real-time node coordinates through the structure disassembly of the action direction, the trigger time period, the target coordinates and the equipment part in the patrol instruction, introducing the double comparison of the direction vector and the path coordinates in the initial stage of the instruction generation, and making the patrol instructions of multiple doctors in the remote collaboration based on the target conflict avoidance principle to be screened in real time. The combination of the continuous task frequency and the target range coverage is carried out on the execution record of the doctor's patrol behavior, the doctor's behavior with strong control tendency is sorted clearly, and then the control focus is gradually focused on the actual master doctor end. The path target is reversely driven by the behavior characteristics of the task performer, three state data of patient waiting, disease level and patient admission density are extracted to realize the dynamic optimization of the path node, and then the targeted activation is performed combined with the callable state of the equipment to ensure that the equipment involved in the task path instruction is in a state of responding at any time. In the scene of multiple doctors' intervention at the same time, the relationship between the master and the collaborative control instructions is established through the synchronous pairing and the priority sorting mechanism of the voice and the focus instruction field, and the clarity and independence of the equipment response link under multiple instruction inputs are maintained. The processing logic fully strengthens the interactive stability of the multi-doctor collaboration and the structural smoothness of the instruction execution in the remote patrol through the path dynamic reconstruction driven by the behavior characteristics, the conflict elimination in advance of the instruction field, the classification focusing of the control intention and the hierarchical pairing of the response. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort based on these drawings.

[0017] Figure 1 is a schematic diagram of a patrol robot based on remote collaboration and consultation support provided by the embodiment of the present application; Figure 2 is a schematic diagram of a robot frame of the present application; DETAILED DESCRIPTION

[0018] The technical solutions in the present application will be described below with reference to the drawings.

[0019] In the embodiments of the present application, the words such as "example", "for example" are used to represent as an example, illustration or explanation. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0020] In the embodiments of the present application, "image" and "picture" can be used interchangeably, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0021] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the distinction is not emphasized.

[0022] In order to make the technical problems, technical solutions and advantages to be solved by the present application more clear, the following will be described in detail in combination with the drawings and specific embodiments.

[0023] The embodiments of the present application provide a patrol robot based on remote cooperation and consultation support, such as Figures 1-2 The schematic diagram of the patrol robot based on remote cooperation and consultation support is shown in the figure, and the robot comprises: The instruction analysis module is used for extracting the action direction field, the trigger time period, the target position coordinate field and the target device part field based on the patrol instruction submitted by the remote consultation end, performing direction vector comparison on the action direction field and the navigation direction value in the robot state, performing coordinate matching on the target position coordinate field and the real-time path node position field, and generating a parallel control instruction list; The permission scheduling module is used for extracting the doctor number and the consultation stage task execution record through the parallel control instruction list, marking the doctors whose continuous task number exceeds two as active execution state, marking the doctors whose patient number range coverage area is lower than two wards as concentrated task source doctors, classifying the task control behaviors, comparing and sorting the execution tendencies after classification, and generating a doctor patrol task control distribution table; The path exchange module is used for calling the doctor patrol task control distribution table, obtaining the patient waiting time record, the symptom level code and the reception intensive record field corresponding to the node, performing priority comparison on the service state field of the navigation node and the target node state field in the set, and generating patrol node change identification content; The task execution module is used for extracting the node path segment instruction by using the patrol node change identification content, performing available state checking on the camera component, the sound pickup component and the signal sending component, performing device activation operation on the components marked as callable in the state, and generating a patrol task start state registration item; The instruction linkage module is used for registering the doctor's end interactive interface in the active state by using the patrol task start state registration item, extracting the voice instruction field and the operation focus field, pairing the focus field and the target site field, arranging the voice instruction field, marking the first field as the main control, marking the remaining fields as the cooperation, and generating the remote cooperative control instruction set. The parallel control instruction list includes a control target number sequence, a direction consistency label, a coordinate area pairing result, and a conflict exclusion label. The doctor patrol task control distribution table includes an execution frequency classification label, a target area distribution category, a doctor identity number mapping sequence, and a behavior tendency ordering structure. The patrol node change label content includes a node update number, a service state label group, a navigation redirection label, and a path priority execution label. The patrol task start state registration item includes a device activation number list, a component state mapping value, a path segment execution label, and a task queue update result. The remote cooperative control instruction set includes a main control doctor number, a voice control link, a synchronous operation mapping structure, and an interactive focus association item.

[0024] Specifically, as shown in Figure 2 The instruction analysis module includes: The field extraction submodule extracts the action direction field, the trigger time period, the target position coordinate field, and the target device site field based on the patrol instruction submitted by the remote consultation end, respectively divides the data types and records them as action vector items and positioning coordinate items, and generates an instruction data structure set. The text needs to extract multiple key elements including action direction, trigger time period, target position coordinates, and target device site. In the actual execution process, the text is preprocessed, including symbol cleaning, syntax segmentation, and language standardization. The semantic units are identified by a trained named entity recognition model. The verb phrases such as "close", "move to", and "adjust direction" are identified as action direction related expressions by a word vector model. The direction vector is determined based on the context spatial reference. For the description "move to the right side of the operating table", "right side" is identified and combined with a predefined direction dictionary to correspond to a standard direction code. The extraction of the trigger time period relies on the time phrase recognition module. For expressions such as "every Wednesday morning", it can be converted into a specific time period identifier for unified processing of path planning tasks. The target position coordinates need to be based on the indoor layout map and the calibration of the reference position. Setting "30 cm to the left of the operating table" can be converted into relative coordinates with the operating table as the origin. The device site recognition calls the device structure word table, such as "monitor probe" and "syringe pump button". The belonging device unit is determined by joint recognition of keywords and context relationships. The extraction result is packaged into a standardized field and then summarized into a structured data entry to generate the instruction data structure set.

[0025] The direction comparison sub-module calls the action vector item in the instruction data structure set and the navigation direction data for vector splitting processing, judges the difference between the included angle of the split direction items and the set direction reference angle, labels the direction items whose included angle falls within the deviation tolerance interval as navigation input, and generates a direction matching label set; The direction action field in the split tour instruction needs to be split into direction vector components and compared with the current navigation path direction. In specific operations, the extracted direction information is expressed in the form of two-dimensional space, and the direction of each node is extracted according to the direction change between the nodes in the navigation record to establish a basis for direction comparison. The angle between the tour direction and the navigation direction needs to be evaluated. If the angle is within the set tolerance interval, which is set to thirty degrees, it is considered that the direction is consistent with the navigation direction. In application, "move left" is extracted, and there is a northwest movement trajectory in the navigation record. If the direction difference is small, the trajectory point is marked as a direction matching point. The direction matching label is then added to the matching result set to further participate in the path optimization process. In this judgment process, the direction accuracy and path stability need to be considered. To avoid interference items, a minimum action amplitude threshold is set to filter small deviation actions. A sliding window mechanism is introduced to jointly judge a plurality of direction items to improve the accuracy of direction matching, identify the path paragraph with consistent direction, provide a basis for navigation path screening, and generate a direction matching label set.

[0026] The coordinate matching sub-module performs spatial distance calculation according to the positioning coordinate item in the instruction data structure set and the real-time path node coordinate set, calculates the Euclidean distance between the coordinate point pairs, and performs node attribution determination combined with the trajectory continuity feature. The attribution node is labeled as the target area to be controlled. The label matching item is paired and bound with the anchor point set to generate a parallel path control list. The target coordinates extracted in the patrol instruction are compared with the real-time node set in the navigation path in terms of distance to determine whether the target position belongs to the current control range. In the execution process, the target coordinates are taken as input, the path node set is traversed to measure the spatial interval between the target point and the recorded points, and then it is determined whether they belong to the same region according to the set distance tolerance threshold. To further improve the accuracy, the continuity of the trajectory nodes is analyzed to identify the motion change mode between the coordinate points. The position difference change amplitude between the continuous nodes is calculated to determine whether the target point appears on the continuous path segment. For the nodes belonging to the same region but at the path inflection point, it is also determined whether the direction switching is smooth to avoid misidentification. In the example, if the target device part is a “monitor probe” and the target coordinates are close to the anchor point position of the monitor, when there are multiple candidate points in the path, the path point closest to the target anchor point will be preferentially selected for binding to form an effective path control item. The matching relationship also needs to be recorded. In the multi-target device instruction, it is necessary to ensure that there is no intersection between the device positioning points. The path nodes that match successfully will be added to the parallel path control list.

[0027] Specifically, as shown in Figure 2 The permission scheduling module includes: The task extraction submodule extracts the corresponding doctor number and consultation stage task execution record based on the parallel control instruction list, merges the task records of different periods for the same doctor number according to time sequence, and uniformly marks and excludes interrupted tasks to generate a doctor task list. Extract the corresponding doctor number and its associated consultation stage task execution record in each task. In the execution process, the inspection robot parses the task items in the list, identifies the unique doctor number, and extracts the doctor number as the primary key to build a doctor task group. The inspection robot needs to identify the timestamp field in the task record to determine the specific time of each task execution. Under the same doctor number, the inspection robot sorts the task records in different time periods according to time sequence, and then assigns each task a sequential number to form a complete task sequence. For the items in the task record that have an execution interruption identifier, the inspection robot will exclude them from the sequence by identifying the marker field, such as the keywords “has been suspended” or “non-planned termination” in the task status. The completed doctor task list is composed of the doctor number, task sequence, and corresponding timestamp field. In actual application, if doctor A has executed 5 tasks on different days, the inspection robot will arrange them in order according to the record time and remove the 3rd task with an interrupted status to obtain the doctor task list.

[0028] The execution state recognition submodule calls the task sequence length information in the doctor task list, calculates the continuous task frequency, judges whether the number of continuous tasks exceeds two, and if the condition is met, marks it as an active execution state, counts the number of ward numbers covered by the patient number associated with each doctor's task, and marks the doctor with less than two wards as a concentrated task source doctor to generate a doctor state identification sequence; The inspection robot traverses the task sequence of each doctor in the list, extracts the sequence length information and records the number of continuous task execution segments of each doctor, then calculates the continuous task frequency, sets the task record of doctor number D001 to appear three times in a row, and the continuous task execution interval is less than the set time threshold, sets the interval less than one hour as continuous task, then the inspection robot determines that it is in an active execution state, and labels the doctor number as "active", and the inspection robot reads the patient number involved in the task associated with the doctor, then maps the patient number to the ward number where the patient is located, counts the number of wards, and if the result shows that the total number of wards is less than two, the inspection robot marks the doctor as a "concentrated task source" doctor, and sets five tasks of doctor B to the same ward patients, then the doctor is identified as a concentrated task source type, and the whole process forms a doctor state identification sequence after marking.

[0029] The behavior classification submodule classifies the active execution state doctors and concentrated task source doctors into different task control behavior classification groups according to the doctor state identification sequence, calls the task participation time period, task covered ward number and task type, calculates the task execution time period deviation value, compares the horizontal task distribution proportion based on the behavior classification, and generates a doctor patrol task control distribution table; The task execution time period deviation value is calculated by the formula: ; Wherein, represents the task execution time period deviation value, represents the number of active execution state doctors, represents the real-time task number of the th doctor in the real-time task period, represents the earliest execution time number of the th doctor of the same type task, represents the latest execution time number of the th doctor of the same type task, represents the target task time center point of the th doctor; Parameter meaning and formula calculation derivation process: , represents the total number of active doctors, and the number of doctors in the "active" state is obtained by extracting the execution state from the hospital dispatching task execution state log; represents the actual task days of the th doctor in the current task cycle, and the data is calculated by the task start and end time field in the task scheduling record; The first doctor: the task start and end time is from May 1st to May 3rd, and the task days are 3; The second doctor: the task start and end time is from May 2nd to May 4th, and the task days are 3; The third doctor: the task start and end time is from May 1st to May 2nd, and the task days are 2; represents the earliest execution time number of the th doctor of the same type of task, and the number is accumulated from 1 per day in the cycle, May 1st is 1, and so on. The data comes from the task database: The first doctor: the task starts on May 1st, and the earliest number is 1; The second doctor: May 2nd, the earliest number is 2; The third doctor: May 1st, the earliest number is 1; represents the latest execution time number of the th doctor of the same type of task, and the data is also extracted from the task database: The first doctor: the task ends on May 5th, and the latest number is 5; The second doctor: May 6th, the number is 6; The third doctor: May 3rd, the number is 3; represents the target task time center point of the th doctor in the current cycle, that is, the preset task standard time point, and the data is set by the hospital patrol strategy, and is uniformly set to the midpoint time (number 3) in the cycle: Substitute the parameters into the formula and calculate as follows: The first doctor: ; The intermediate value is: ; The second doctor: ; The intermediate value is: ; The third doctor: ; The intermediate value is: ; The three intermediate values are substituted into the overall formula: ; The result shows that the task execution period deviation value is 6.27, which represents the root mean square difference degree of the overall deviation of the doctor's task participation time from the target center time point. The larger the value, the more dispersed the task execution time, and the value is used as a weight reference for measuring the execution stability in the subsequent horizontal task distribution proportion comparison.

[0030] Specifically, as shown in Figure 2 The path switching module includes: The data arrangement submodule calls the doctor's tour task control distribution table, extracts the patient waiting time record, symptom level code and intensive reception record field corresponding to the navigation node, and merges the three data according to the navigation node number, arranges them in chronological order, and generates a node task load set; The inspection robot reads the navigation node number associated with each tour task in the distribution table, and calls the related database interface for each navigation node to extract the waiting time record, symptom level code and intensive reception record field of the corresponding patient. The waiting time record is based on the queuing inspection robot log at the time of task triggering, and the total time elapsed from registration to the current time for each patient is obtained and recorded in minutes. The symptom level code is extracted from the automatic classification result of the pre-reception inquiry inspection robot on the reception platform, and is divided into five levels, with level I as critical, level II as emergency, level III as ordinary, etc. The intensive reception record is obtained by counting the number of receptions in the node within the last hour. The three types of data are attached to the data structure of each navigation node in the form of fields. The inspection robot merges these three data according to the navigation node number, and arranges the data in chronological order according to the timestamp as the sorting basis to form a time-continuous structure. Each item represents a navigation node, including its corresponding patient waiting time, current symptom level and reception frequency per unit time. In navigation node A, there are three data of waiting time 35 minutes, symptom level II and reception density 8 times / hour. After sorting, a node task load item is constructed, which is an important basis for subsequent state judgment and node optimization, forming a node task load set.

[0031] The state comparison submodule calls the service state field of the navigation node and the real-time state field of the target node in the node task load set based on the target node number, and respectively maps the service occupancy rate, active reception frequency and queue length order code to a unified sequence. The corresponding bit positions of the state sequences of the navigation node and the target node are compared to generate a node priority switching sequence. The target node number needs to be identified from the node task load set. The inspection robot extracts the service state field and real-time state field of the target node in turn. The service state field includes the current doctor on-duty state, device resource occupancy rate, channel occupancy, etc. The real-time state field is provided by the current navigation task scheduling result, mainly including queuing length, reception frequency, etc. The inspection robot maps the two data sources into a unified state evaluation sequence. The mapping method is to set a fixed index order for coding. The first bit is the service occupancy rate, the second bit is the active reception frequency, and the third bit is the queuing length. Then each item is assigned a sequence coding by using a standardized sorting method. The higher the value, the higher the ranking. After mapping, the inspection robot compares the differences in each index ranking between the target node and the current navigation node. If the difference is large, it means that the target node performs better or worse in some aspects. The index difference comparison result between the navigation node and the target node is marked. If the target node is better than the current node in both queuing length and service occupancy rate and the ranking difference reaches two levels or more, the sequence is marked as "suggested switching", and the node pair is recorded as a candidate replacement object. The node priority switching pair sequence is generated.

[0032] The node adjustment submodule extracts the position information, task arrangement period, and reception intensive frequency of the target node in each suggested replacement node pair. According to the real-time navigation node path structure, the replaceable node execution order is updated and marked with path segment replacement identification content, and the inspection robot generates the inspection node change identification content. The inspection robot extracts the spatial position information, task arrangement period, and reception intensive frequency of the target node in each suggested replacement node pair. The spatial position information is used to determine whether the target node is within the current navigation path range. The task arrangement period needs to determine whether there is a task conflict or time overlap. The reception intensive frequency is used to determine whether the workload of the node in the corresponding time period has been saturated. After integrating these three types of information, the inspection robot matches and analyzes the real-time navigation path structure to determine the replaceable node list. The replaceable node inspection robot updates its original execution order and adjusts the path structure according to the ranking order recommended in the priority switching pair sequence. The original node execution order in the updated navigation path is replaced by the new target node. At the same time, a marked field is inserted in the path segment structure to mark that this path segment has been replaced and adjusted. This facilitates the synchronous update of subsequent task scheduling records. The original order of a certain inspection path is set as node A-B-C. After comparison, the inspection robot determines that the execution efficiency of node B is lower than that of node D, and node D is located in the adjacent ward without task overlap. The path is updated to A-D-C, and the "B→D" identification content is inserted at the position of path segment B to generate the inspection node change identification content.

[0033] Specifically, as Figure 2As shown, the task execution module includes: The path number arrangement submodule extracts the marked path segment number and corresponding node operation instruction according to the patrol node change identification content, and arranges the continuous path segments in the node index order to form a path segment execution sequence, evaluates the trigger index association relationship between the nodes and the path segments, and generates a path segment instruction sequence set; According to the path data carding of the patrol node change identification content, the inspection robot extracts the marked path segment number from the change identification content, and obtains the corresponding node operation instruction in each path segment. The node exchange information is set to mark that "path segment 3 is replaced by node B with node D". The inspection robot updates the operation instruction of path segment 3 to the action and task content corresponding to node D. The inspection robot sorts the path segment execution information according to the node index order to ensure the sequential execution coherence of the patrol task on the physical path. In the sorting process, it is necessary to identify whether there is a segment skipping phenomenon between path segments. If there is a breakpoint in the index of a node operation instruction, the corresponding path segment needs to be traced back and an placeholder is inserted to maintain the sequence integrity. The inspection robot evaluates the node trigger index association relationship of the arranged path segment sequence. This evaluation is based on the trigger condition of the operation instruction in the path segment and the time record of the previous node task completion signal. If the trigger time difference between path segments exceeds the set delay threshold, a delay risk field needs to be marked in the path sequence, and a path segment instruction sequence set is generated.

[0034] The device detection submodule calls the path segment instruction sequence set, reads the state field of the camera component, the sound pickup component and the signal sending component configured in the path segment, judges whether the component state is marked as a callable state, eliminates the component entries with the state field marked as an unusable state and records the node index, and generates a patrol component callable index group; The inspection robot traverses each path segment in the sequence set, reads the state field information of the camera component, the sound pickup component and the signal sending component configured in the path segment, and the state information is derived from the front-end device control interface or the device management platform API. The state field includes state identifiers such as "available", "initialization failure", "communication interruption" and "limited authority". The inspection robot marks the state field as callable or not callable through matching identification rules. For the component entries marked as "not callable", the inspection robot excludes them from the available list and records the corresponding node index in the path segment for subsequent processing process to perform a skip operation. The components with the state field of "available" are included in the patrol component callable index group. In the index group, each item contains a node number and a list of callable component numbers corresponding thereto. In actual application, if the state of the camera configured in a path segment node is "communication interruption", the sound pickup module is "available", and the signal sending module is "available", the inspection robot will exclude the camera device under the node to generate the patrol component callable index group.

[0035] The task activation submodule binds the device general activation process under the node in the path segment instruction according to the available component number under the node in the callable index group of the patrol component, and generates a patrol task start state registration item. According to the callable component number of each node, the video camera, audio pickup and signal sending devices are processed item by item. The video camera device initialization process includes device number confirmation, lens calibration and image stream transmission channel establishment. The audio pickup device needs to perform audio channel initialization, input gain standard setting and echo suppression parameter adjustment. The signal sending component needs to complete channel binding, sending format setting and communication channel connection process. Each type of device is opened by the patrol robot calling the hardware driver module to open the functional interface, and is marked as activation completed after the interface response returns to the normal state. The patrol robot binds the activation identification of the activated devices under each node in the path segment instruction as a pre-starting condition for the task execution stage and writes it into the task control table. The activation timestamp and response code are recorded for each device initialization process. Each record contains node number, device type, initialization result and whether the connection is successful or not, which is used to track the task preparation state. If the camera and audio module are activated successfully but the signal sending module fails to initialize at a certain path segment node E, the node will be marked as partially successful in the start state registration item, providing a recovery mechanism interface for the subsequent module, and obtaining the patrol task start state registration item.

[0036] Specifically, as shown in Figure 2 the instruction linkage module includes: The instruction extraction submodule extracts the voice instruction field and operation focus field in the activation state in the doctor end interactive interface using the patrol task start state registration item, filters the fields based on the activation state identification, calculates the field interaction difference characteristic value, and generates an interface interaction control sequence. The formula for calculating the field interaction difference characteristic value is: ; Wherein, represents the field interaction difference characteristic value, represents the normalized weight of the th activation state identification value, represents the number of data items with positive trigger state value in the th voice instruction field, represents the real-time value of the corresponding data item in the th operation focus field, represents the number of non-activation state accumulations of the th field, represents the total number of extracted fields; Parameter meaning and formula calculation derivation process: Voice instruction field monitors 5 groups of interaction data, and the total number of fields is Each group of data is based on the voice recognition results and focus operation records automatically collected in the doctor's end interaction monitoring period. The monitoring period is collected every 10 seconds, continuously sampled for 10 minutes, and a total of 60 field data is collected. After averaging through the data window, the following sampling values are obtained; Group 1 data: Activation state identification normalized weight The normalization basis is the average trigger frequency of each field in the same task, and the normalized value is between 0 and 1. The higher the frequency, the greater the weight; Positive trigger state value in the voice instruction field It is obtained by counting the number of instructions that are successfully recognized and match the positive action; The current value corresponding to the operation focus field It is obtained by accumulating the count of the identification field of the current focus operation area in the interface; Field non-activated state cumulative number It is obtained by accumulating the number of times the same field appears in the non-activated state in the past tasks; Intermediate calculation: ; ; ; ; ; Group 2 data: , , , ; ; ; ; ; ; Group 3 data: , , , ; ; ; ; ; ; Group 4 data: , , , ; ; ; ; ; ; Group 5 data: , , , ; ; ; ; ; ; Substitute the above calculated values into the main formula: ; The result shows that the field interaction difference eigenvalue is 0.889, which represents the difference between the voice instruction field and the operation focus field in the doctor's interaction interface based on the current tour diagnosis task. The larger the value, the weaker the interaction relevance between the fields in the task activation process, the lower the proportion of the reserved field into the control sequence, and the higher the field screening strictness. The data used to generate the interface interaction control sequence will be more concentrated on the instruction-focus highly matched content.

[0037] The target pairing sub-module calls the operation focus field in the interface interaction control sequence and the target part field in the tour node change identifier content, obtains the overlapping area of the focus and the part according to the field position coding comparison, and generates a control target mapping structure; The operation focus field is compared with the target site field in the patrol node change identification content. Each operation focus field corresponds to an interactive element actually focused on by the doctor in the interface. The field position code is identified by the two-dimensional or three-dimensional coordinate block number recorded by the graphical interface component. The patrol robot compares the position code with the site position code contained in the target site field recorded in the patrol node change identification content at the field level to identify whether there is an overlapping area or covering relationship between the two. For items with an overlapping or overlapping relationship, the patrol robot marks them as valid pairing items, records the mapping relationship between the interface position pointed by the operation focus and the actual device site corresponding to it in the patrol task, and sets the operation focus of the doctor in the interface to be located in the "upper abdominal region". The patrol robot identifies that the region corresponds to the "liver examination region" specified in the patrol task change node through comparison, considers that the two are matched successfully, and registers the pairing item in the structure group. The pairing successful item will become the object of subsequent instruction rearrangement and control linkage to ensure that the interactive behavior is consistent with the actual device task and build the control target mapping structure.

[0038] The instruction control sub-module rearranges the instruction fields in sequence according to the control target mapping structure according to the time stamp. The first field in the sequence is marked as the main control field, and the remaining fields are marked as collaborative fields. The main control field is bound to the corresponding target site in the control target mapping structure group and linked with the collaborative fields to generate a set of remote collaborative control instructions. The patrol robot rearranges the voice instruction fields corresponding to each control target group in sequence according to the task record timestamp. The earliest triggered instruction is extracted as the current main control field, and the remaining fields are arranged in sequence according to the trigger time. The rearrangement mechanism ensures that the core intention of the doctor in the task execution is responded to by the patrol robot first. The main control field is the main execution content in the instruction sequence, which will be bound to the paired target site in the control target mapping structure group to form a direct control path. The collaborative field is an auxiliary instruction that needs to be executed after the main control field responds and is linked with it to jointly affect the execution mode of the target node device. If the first voice instruction of the doctor is "start image acquisition" and the second is "synchronize sound recording", the patrol robot will set "start image acquisition" as the main control field and bind it to the "camera device" target site of the node. The subsequent "synchronize sound recording" is set as the collaborative field to link and execute the "sound pickup device". The binding relationship between the main control field and its supporting collaborative field and the corresponding target site is analyzed to drive multiple hardware modules to respond to the operation instruction of the doctor in the patrol task and ensure that the task is executed in the set process to generate a set of remote collaborative control instructions.

[0039] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A mobile clinic robot based on remote collaboration and consultation support, characterized in that: The robot includes: The instruction parsing module is used to extract the action direction field, trigger time period, target position coordinate field and target device part field based on the rounds instructions submitted by the remote consultation terminal, perform direction vector comparison between the action direction field and the navigation direction value in the robot state, and generate a list of parallel control instructions. The permission scheduling module is used to extract doctor numbers and consultation stage task execution records through the parallel control instruction list, mark doctors with more than two consecutive tasks as active execution status, mark doctors whose patient number range covers less than two wards as centralized task source doctors, classify task control behaviors, and generate a doctor rounds task control distribution table. The path switching module is used to call the doctor's rounds task control distribution table, compare the service status field of the navigation node with the status field of the target node in the set, and generate the change identifier of the rounds node. The task execution module is used to use the change identifier content of the patrol node, extract the node path segment instructions, check the availability status of the camera component, the audio pickup component and the signal transmission component, perform device activation operation on the callable components, and generate a patrol task start status registration item.

2. The mobile clinic robot based on remote collaboration and consultation support according to claim 1, characterized in that, The parallel control instruction list includes a control target number sequence, direction consistency label, coordinate region pairing result, and conflict resolution identifier. The doctor's rounds task control distribution table includes execution frequency classification label, target region distribution category, doctor identity number mapping sequence, and behavior tendency sorting structure. The rounds node change identifier includes node update number, service status label grouping, navigation redirection identifier, and path priority execution mark. The rounds task start status registration item includes a device activation number list, component status mapping value, path segment execution identifier, and task queue update result.

3. The mobile clinic robot based on remote collaboration and consultation support according to claim 1, characterized in that, The instruction parsing module includes: The field extraction submodule is used to extract the action direction field, trigger time period, target location coordinate field and target device part field based on the rounds instructions submitted by the remote consultation terminal. The data types are divided and recorded as action vector items and positioning coordinate items respectively, and an instruction data structure set is generated. The direction comparison submodule is used to call the action vector items in the instruction data structure set and the navigation direction data to perform vector splitting processing. Based on the difference between the angle between the split direction items and the set direction reference angle, the direction items whose angle falls into the allowable deviation range are marked as navigation inputs, and a direction matching tag set is generated. The coordinate matching submodule is used to perform spatial distance calculation based on the positioning coordinate items in the instruction data structure set and the real-time path node coordinate set, calculate the Euclidean distance between coordinate point pairs and perform node attribution determination in combination with trajectory continuity features, mark the attribution node as the target area to be controlled, pair and bind the label matching items with the anchor point set, and generate a parallel path control list.

4. The mobile clinic robot based on remote collaboration and consultation support according to claim 3, characterized in that, The permission scheduling module includes: The task extraction submodule is used to extract the corresponding doctor number and consultation stage task execution record based on the parallel control instruction list, number and merge the task records of different time periods under the same doctor number according to the time order, and uniformly mark and filter out interrupted tasks to generate a doctor task list. The execution status identification submodule is used to call the length information of each task sequence in the doctor task list, calculate the frequency of consecutive tasks, determine whether the number of consecutive tasks exceeds two, and mark it as an active execution status if the condition is met. It also counts the number of ward numbers covered by the associated patient number in each doctor task, marks doctors with fewer than two ward numbers as centralized task source doctors, and generates a doctor status identification sequence. The behavior classification submodule is used to classify doctors in active execution status and doctors from centralized task sources into differentiated task control behavior classification groups according to the doctor status identifier sequence, and to call the task participation time period, task coverage ward number and task type to calculate the task execution time period deviation value, and to compare the horizontal task distribution ratio based on the behavior classification to generate a doctor rounds task control distribution table.

5. The mobile clinic robot based on remote collaboration and consultation support according to claim 4, characterized in that, The deviation value for the task execution period is calculated using the following formula: ； in, This represents the deviation value during the task execution period. This represents the number of doctors in an active execution state. Representing the The number of real-time task days for each doctor within a real-time task cycle. Representing the The earliest execution time number of the same type of task for each doctor. Representing the The latest execution time number of the same type of task for each doctor. Representing the The doctor's target task time center point.

6. The mobile clinic robot based on remote collaboration and consultation support according to claim 4, characterized in that, The path switching module includes: The data processing submodule is used to call the doctor's rounds task control distribution table, extract the patient waiting time record, symptom level code and dense visit record fields corresponding to the navigation node, and merge the three data according to the navigation node number, arrange them into a continuous arrangement structure in chronological order, and generate a node task load set. The status comparison submodule is used to call the service status field of the navigation node and the real-time status field of the target node in the set based on the target node number in the node task load set. The service status field of the navigation node and the real-time status field of the target node in the set are encoded and mapped into a unified sequence according to the service occupancy rate, active reception frequency and queue length respectively. The corresponding positions of the status sequences of the navigation node and the target node are compared to generate a node priority switching pair sequence. The node adjustment submodule is used to extract the location information, task arrangement time period and patient density frequency of the target node to be replaced in the sequence according to the node priority switching pair sequence. Based on the real-time navigation node path structure, the execution order of replaceable nodes is updated according to the priority switching position and the path segment replacement mark content is marked to generate the patrol node change mark content.

7. The mobile clinic robot based on remote collaboration and consultation support according to claim 6, characterized in that, The task execution module includes: The path numbering and sorting submodule is used to extract the marked path segment numbers and corresponding node operation instructions by using the change identifier content of the patrol node, and sort the continuous path segments into a path segment execution sequence according to the node index order, evaluate the trigger index association relationship between the node and the path segment, and generate a path segment instruction sequence set. The equipment detection submodule is used to call the path segment instruction sequence set, read the status fields of the camera components, audio pickup components and signal transmission components configured in the path segment, determine whether the component status is marked as callable, remove component entries with unavailable status fields and record node indexes, and generate a callable index group for inspection components. The task activation submodule is used to perform device initialization, function interface opening and signal channel connection operations on the camera, audio pickup and signal transmission components respectively, according to the available component number under the node in the callable index group of the roving component, bind the general activation process of the device under the node in the path segment instruction, and generate a roving task start status registration item.

8. The mobile clinic robot based on remote collaboration and consultation support according to claim 1, characterized in that, The robot also includes a command linkage module: The instruction linkage module is used to extract the voice instruction fields and operation focus fields that are active in the doctor's interactive interface by using the registration item of the start status of the round of medical tasks, pair the focus fields with the target site fields to control targets, arrange the voice instruction fields, mark the first field of the sort as the master field, mark the remaining fields as the collaboration field, and generate a remote collaborative control instruction set. The remote collaborative control instruction set includes the master doctor number, voice control link, synchronous operation mapping structure, and interactive focus association items.

9. The mobile clinic robot based on remote collaboration and consultation support according to claim 8, characterized in that, The instruction linkage module includes: The instruction extraction submodule is used to extract the voice instruction fields and operation focus fields that are active in the doctor's interactive interface by using the registration item of the initiation status of the round of visits task, and to filter the fields based on the activation status identifier, calculate the field interaction difference feature value, and generate the interface interaction control sequence. The target pairing submodule is used to call the operation focus field in the interface interaction control sequence and the target part field in the patrol node change identifier content, obtain the overlapping area of ​​the focus and part based on the field position code comparison, and generate a control target mapping structure. The instruction control submodule is used to rearrange the instruction fields according to the timestamp based on the control target mapping structure, mark the first field as the main control field, and mark the remaining fields as collaborative fields. The main control field is bound to the corresponding target part in the control target mapping structure group and the collaborative fields are linked to generate a remote collaborative control instruction set.

10. The mobile clinic robot based on remote collaboration and consultation support according to claim 9, characterized in that, The formula for calculating the interaction difference feature value of the fields is: ； in, Represents the interaction difference feature values ​​of the fields. Representing the Normalized weights of the item activation status identifier values Representing the The data item with a positive trigger status value in the voice command field. Representing the The real-time value of the corresponding data item in the focus field of the item operation. Representing the The cumulative number of times an item field is inactive. This represents the total number of fields extracted.

Images