Intelligent management method for fitness shelter based on multi-source data fusion

Abstract

The application discloses a fitness shelter intelligent management method based on multi-source data fusion and relates to the technical field of fitness shelter data management.The application comprises the following steps: acquiring a target user training plan and current task end group load, displacement, return speed and contact data, constructing an unloading track and determining a residual control state;when the target equipment of the next task is occupied, determining a predicted release interval according to the current user's group end unloading track and preloading track, updating the residual control state in combination with a candidate release time, identifying a receiving gap, and forming a waiting path, a bridging insertion path or a replacement equipment path; further checking an equipment configuration reset state and a configuration adjustment section, and transmitting and truncating the occupation disturbance caused by the replacement occupation along the equipment occupation sequence. The application can improve the judgment accuracy of the equipment release time and the training receiving state, reduce equipment configuration misjudgment, and improve the rationality of the equipment scheduling and training connection in the fitness shelter.

Description

Technical Field

[0001] This invention belongs to the field of data management technology for fitness cabins, and in particular relates to an intelligent management method for fitness cabins based on multi-source data fusion. Background Technology

[0002] A fitness cabin is a facility that integrates aerobic training, strength training, and auxiliary training functions into a limited cabin space. It manages users' fitness activities through user training plans, fitness equipment status, and training process data. Because the types and number of equipment that can be placed in a fitness cabin are limited, multiple users may need to use the same equipment when training at the same time.

[0003] Currently, fitness cabins typically guide users to use different fitness equipment in a corresponding order based on pre-generated training plans. When a user completes the current training task, and the equipment for the next training task is occupied by another user, existing systems generally ask the user to wait, directly adjust the subsequent training order, or recommend other equipment according to the preset equipment correspondence. However, the remaining occupancy time of the target equipment will change with the current user's training progress, the number of remaining sets, and the interval between sets. The appropriate time for the user to start the next training task is also related to the completed task, recovery status, and the connection between the remaining training tasks. If the user waits directly, the waiting time may exceed the reasonable training interval; if other training tasks are arbitrarily advanced, it may affect the connection between subsequent tasks; if other equipment is used directly as a substitute, the original training task may not be effectively completed due to differences in the movement trajectory, load mode, and training effect of the equipment. Therefore, a fixed processing method is difficult to adapt to the dynamic changes in the actual use of equipment and the user's current training status.

[0004] Therefore, existing intelligent management methods for fitness cabins cannot comprehensively consider the dynamic relationship between the real-time usage status of equipment and the user's training progress. It is difficult to reasonably coordinate the user's training schedule when the equipment is occupied, which can easily lead to excessively long waiting times, unreasonable training connections, and low efficiency in the utilization of equipment resources. To address this, this invention proposes an intelligent management method for fitness cabins based on multi-source data fusion. Summary of the Invention

[0005] The purpose of this invention is to provide a smart management method for fitness cabins based on multi-source data fusion, which solves the problems of existing smart management methods for fitness cabins failing to comprehensively consider the dynamic relationship between the real-time usage status of equipment and the user's training progress, making it difficult to reasonably coordinate user training arrangements when equipment is occupied, and easily leading to excessively long waiting times, unreasonable training connections, and low efficiency in the utilization of equipment resources.

[0006] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: This invention relates to a smart management method for fitness cabins based on multi-source data fusion, comprising the following steps: S1. Obtain the target user's training plan and the load, displacement, return speed and contact data of the last group of the current task, construct the unloading trajectory, divide the active and passive unloading segments according to the synchronization relationship between load decrease and displacement return, contact duration and speed stability, and determine the residual control state. S2. When the target device of the next task is occupied, collect the load, displacement and contact data at the end of the current user group, generate the unloading trajectory at the end of the group, and pair it with the preload trajectory of the adjacent action group to form a succession template, determine the reloadable termination domain and the handover termination domain; determine the expected release interval based on the approach order, dwell time and stability of the current trajectory to the two termination domains. S3. Within the expected release range, update the residual control state according to the waiting time corresponding to the candidate release time, and sequentially match it with the contact establishment, transmission preload and resistance activation states of the next task. The first state that does not meet the continuity condition is identified as the receiving gap; the candidate release time that does not form a receiving gap forms a waiting path. S4. When there is a gap, select bridging occupancy units that can be executed by the idle equipment in the mobile cabin from the remaining tasks of the target user's training plan. The loading entry of the bridging occupancy unit accepts the residual control state after the update. Its unloading termination state continues the first continuous state after the gap and is reset before the target equipment is released, forming a bridging insertion path. S5. When there is no bridging occupancy unit, select candidate devices that accept the updated residual control state at the preload entry and whose load process continuously covers the critical load segment of the next task, embed their replacement occupancy period into the device occupancy sequence, exclude candidate devices that prevent the updated residual control state of the affected user from continuing its preload state, determine the replacement device path, and update the device occupancy sequence and user travel guidance.

[0007] Furthermore, the unloading trajectory consists of trajectory points arranged according to the sampling time. Each trajectory point includes the load drop slope, displacement return ratio, return speed, contact state, and load fluctuation amplitude. The first trajectory point where the contact signal changes from continuous to interrupted and the timing deviation between load drop and displacement return begins to increase is determined as the boundary point between the active unloading section and the passive unloading section. The residual control state includes unloading baseline parameters and candidate moment acceptance parameters. The unloading baseline parameters include the duration of the active unloading section, the duration of the passive unloading section, the unloading endpoint load, the remaining return displacement, the unloading endpoint speed, and the contact state. The candidate moment acceptance parameters include the expected contact establishment interval, the expected pre-tightening starting load, the expected pre-tightening starting speed, the expected initial loading displacement, and the expected pre-tightening load holding amplitude.

[0008] Furthermore, when forming the succession template, the reset state of the target device is used as the alignment point to pair the terminal trajectory segment of the unloading trajectory of the previous action group with the starting trajectory segment of the preloading trajectory of the next action group; the set of unloading terminal states that enter the preloading trajectory within the preset inter-group time after pairing is determined as the reloading termination domain, and the set of unloading terminal states that are interrupted by the contact signal and do not enter the preloading trajectory within the preset handover time is determined as the handover termination domain. When determining the expected release interval, the current user's end-of-group unloading trajectory is matched with the trajectory segment in the continuation template to determine the remaining time for the current trajectory to reach the reloadable termination domain and the handover termination domain respectively; the predicted time when the current trajectory reaches the handover termination domain by the handover branch is taken as the start time of the expected release interval; the predicted time when the current trajectory enters the reloadable termination domain by the reloadable branch, continues to complete the remaining action group, and finally reaches the handover termination domain is taken as the end time of the expected release interval.

[0009] Furthermore, the continuity conditions in step S3 include: the time interval between the termination time of the contact signal in the previous state and the establishment time of the contact in the next state does not exceed the contact interval threshold; the load change direction in the previous state is consistent with the loading direction in the next state; the difference between the unloading endpoint displacement and the starting displacement in the next state does not exceed the displacement threshold; and the load drop amplitude before entering the resistance effective state from the transmission preload state does not exceed the drop threshold; the first state that does not meet at least one continuity condition is determined as the bearing gap.

[0010] Furthermore, the bridging occupancy unit includes the user's travel period to the idle device, the loading period, the training period, the unloading period, and the device reset period; the loading entry state of the bridging occupancy unit and the updated residual control state satisfy the continuity condition. After the bridging occupancy unit is completed, its unloading termination state is used as the new state reference. The residual control state is updated according to the remaining waiting time from the end time of the instrument reset period to the corresponding candidate release time. The updated residual control state is sequentially matched with the contact establishment state, transmission preload state and resistance activation state of the next task. When the receiving gap disappears and the end time of the instrument reset period is not later than the corresponding candidate release time, the bridging occupancy unit is determined to form a bridging insertion path.

[0011] Furthermore, in step S5, the critical load segment is the trajectory segment from the moment the resistance component takes effect until the load drops to a preset proportion of the peak value after reaching its peak value in the next task. The critical load segment is characterized by the load direction, displacement range, and normalized load change curve. When the load direction of the candidate device is consistent with the load direction of the critical load segment, the displacement range coverage ratio is not lower than the coverage threshold, and the deviation of the normalized load change curve does not exceed the curve deviation threshold, the load process of the candidate device is determined to continuously cover the critical load segment.

[0012] Furthermore, other users whose replacement occupation time overlaps with the allocated device occupation time are identified as affected users. The candidate usage time of the affected users is updated according to the overlap duration, and the residual control status of the affected users is updated using the waiting status update template. When the updated residual control status does not meet the continuity condition with the preload status of its next task, the corresponding candidate device is excluded.

[0013] Furthermore, when the current user continues to execute the next action group, the actual unloading trajectory at the end of the group is paired with the actual preload trajectory of the next action group to form a reloadable sample; when the current user ends the use of the target device, the actual final unloading trajectory at the end of the group, the contact interruption status, and the status of not forming a preload trajectory again within the preset handover time are used to form a handover sample; the continuation template, reloadable termination field, and handover termination field are updated based on the reloadable sample and the handover sample.

[0014] Furthermore, the bridging occupancy unit and the alternative occupancy unit corresponding to the alternative device path are respectively identified as occupancy units to be verified. For each occupancy unit to be verified, the first user in the device occupancy sequence following the occupancy unit to be verified is identified as the next affected user. Based on the device configuration state after the occupancy unit to be verified completes the target user's task, and the position of the human support component, the position of the transmission connection component, and the resistance setting state corresponding to the preload trajectory of the next affected user, a device configuration reset trajectory is generated. The state in which the human support component and the transmission connection component complete locking, the resistance setting reaches the target state corresponding to the next affected user, and the load fluctuation and displacement fluctuation return to stability is identified as the configuration handover termination domain. The load, displacement, and contact changes collected before entering the configuration handover termination domain are identified as configuration adjustment sections, and the configuration adjustment sections are prohibited from being added to the continuation template or used for the calculation of the expected release interval as the preload trajectory of the next affected user. The reset time period required for the occupancy unit to be verified to reach the configuration handover termination domain is inserted into the device occupancy sequence, and occupancy units to be verified that cannot enter the configuration handover termination domain before the candidate usage time of the next affected user are excluded.

[0015] Furthermore, after inserting the alternative occupancy period corresponding to the alternative device path into the device occupancy sequence, the device corresponding to the alternative occupancy period is used as the starting point of the occupancy disturbance, and the affected users are determined sequentially according to the subsequent occupancy order of the device. The residual control state of each affected user is updated based on the delayed candidate usage time and corresponding waiting time, and the updated residual control state is continuously matched with its original preload state. When the affected user meets the continuity condition, the updated device reset completion time is determined based on its updated candidate usage time, training duration, unloading duration, and device reset duration, ensuring that the device reset completion time is not later than the original candidate usage time of the next user. When the device reset completion time is later than the next user's original candidate usage time, the time difference between the two is used as the remaining occupancy disturbance and propagated to the next user. When the affected user does not meet the continuity condition, the first state that does not meet the continuity condition is identified as a new acceptance gap. A reconstruction path is determined for the affected user according to the determination order of the bridging insertion path and the alternative device path. The occupancy period corresponding to the reconstruction path is used as a new occupancy disturbance. The residual control state update and continuity matching continue along the subsequent occupancy order of the corresponding device. The propagation is terminated when the remaining occupancy disturbance is absorbed by the device's idle period. The processing is terminated when there are no subsequent affected users for the corresponding device.

[0016] The present invention has the following beneficial effects: 1. This invention constructs the final unloading trajectory by collecting load, displacement, return speed and contact data during user training, and divides it into active unloading segment and passive unloading segment, which can more accurately characterize the residual control state of the user after completing the current task; and combines the final unloading trajectory and preload trajectory of the current user of the target device to determine the expected release interval. Compared with judging the idle state of the device based solely on the preset training time, this invention ultimately improves the accuracy of judging the release time of the device.

[0017] 2. This invention updates the residual control state based on the waiting time corresponding to the candidate release time, and continuously matches it with the contact establishment, transmission preload and resistance activation state of the next task. When there is a gap in the connection, a bridging insertion path is formed, and when there is no bridging occupied unit, an alternative device path is determined. This makes the training task adjustment correspond to the user's actual training state and the device load process, reducing the impact of long waiting time and inappropriate substitution on the continuity of training.

[0018] 3. By establishing a configuration handover termination domain, this invention distinguishes the adjustment process of the instrument seat, support components, transmission connection components, and resistance settings from the actual preloading process of the next user, and incorporates the configuration reset period into the instrument occupancy sequence. This can avoid misjudging the load, displacement, and contact changes caused by component adjustments as preloading data, and reduce the premature handover time of the instrument and subsequent occupancy conflicts.

[0019] 4. This invention transmits the time changes caused by substitution occupancy step by step along the device occupancy sequence, and determines whether the occupancy disturbance continues to propagate based on the residual control status of the affected users and the subsequent idle period; the training path is re-determined only when the training succession relationship is disrupted, and the disturbance is terminated after the time delay is absorbed by the idle period, which can avoid continuous adjustment of the occupancy sequence of multiple users and multiple devices caused by local substitution device allocation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the training path determination process in the fitness cabin of this invention. Figure 2 This is a flowchart of the instrument configuration handover and verification process in this invention; Figure 3 This is a flowchart illustrating the multi-user occupancy disturbance propagation process in this invention. Detailed Implementation

[0022] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0023] Example 1 See Figure 1 This invention provides a smart management method for fitness cabins based on multi-source data fusion, comprising the following steps: In this embodiment, the fitness cabin is equipped with multiple resistance training devices, a user identification terminal, and a management terminal. The resistance training devices are configured with load acquisition units, displacement acquisition units, and contact acquisition units. The load acquisition unit is located in the counterweight transmission part, cable transmission part, device lever arm, or resistance drive part, and is used to collect the load acting on the device transmission mechanism during training. The displacement acquisition unit is located in the counterweight, cable, pulley, or device lever arm, and is used to collect the displacement of the moving parts relative to the reset position. The contact acquisition unit is located in the handle, pedal, support pad, or seat contact part, and is used to determine the contact state between the user and the device.

[0024] Each acquisition unit samples data according to a unified clock. The acquired data carries at least the sampling time, device identifier, user identifier, and training task identifier. The management terminal aligns the load, displacement, and contact data according to the sampling time and determines the return speed based on the displacement changes between adjacent sampling times. For abrupt data generated at a single sampling time where no corresponding changes occur in the preceding and following sampling data, it is identified as sensor instantaneous disturbance data and is discarded. The training plan includes at least the training task sequence, the target device corresponding to each training task, the number of action sets, the load level, and the expected training duration.

[0025] In this embodiment, the reset allowable range, displacement threshold, fallback threshold, coverage threshold, and curve deviation threshold are determined based on the effective motion stroke, idle stroke, sensor resolution, device factory calibration data, and historical successful training records of the corresponding device. Historical successful training records refer to records in which the user was able to sequentially complete contact establishment, transmission preload, and resistance activation, and complete the corresponding training task.

[0026] S1. Obtain the target user's training plan and the load, displacement, return speed and contact data of the last group of the current task, construct the unloading trajectory, divide the active and passive unloading segments according to the synchronization relationship between load decrease and displacement return, contact duration and speed stability, and determine the residual control state. Specifically, the management terminal extracts data from the last valid action of the target user's current task group, from the time the load reaches the peak of this action until the moving parts of the device return to the reset allowable range. The sampling time when the load reaches the peak of this action is taken as the start time of the unloading trajectory, and the sampling time when the moving parts of the device enter the reset allowable range, the return speed decreases, and the load fluctuation returns to stability is taken as the end time of the unloading trajectory.

[0027] The displacement return ratio is determined based on the position of the current displacement relative to the effective movement stroke of the instrument. For instruments with different structures, the reset position of the instrument is taken as the zero point of displacement, and the maximum effective displacement of the instrument in this action is taken as the end point of the stroke, so that the displacement under different action amplitudes can be represented by proportional position. The load fluctuation amplitude is determined based on the difference between the maximum load and the minimum load within a preset number of trajectory points before and after the current trajectory point.

[0028] The unloading trajectory consists of trajectory points arranged according to the sampling time. Each trajectory point includes the load drop slope, displacement return ratio, return speed, contact state, and load fluctuation amplitude. The first trajectory point where the contact signal changes from continuous to interrupted and the timing deviation between load drop and displacement return begins to increase is determined as the boundary point between the active unloading section and the passive unloading section. In the active unloading section, the user remains in contact with the instrument, and the load descent process and the displacement return process of the moving parts are relatively synchronized, with the return speed variation of adjacent trajectory points within the allowable range; in the passive unloading section, the user's control over the moving parts of the instrument is weakened, and the instrument continues to return mainly under the action of counterweights, elastic reset components or drive mechanisms, and the timing deviation between load descent and displacement return increases accordingly.

[0029] When determining the boundary point, the continuous trajectory points after the contact signal is interrupted are verified. When the contact signal is continuously interrupted and the timing deviation between the load drop and the displacement return is continuously increased, the trajectory point where the contact signal is first interrupted is determined as the boundary point. When the contact signal is interrupted and then recovers in a short time, and the load drop and the displacement return are synchronized again, the interruption is considered as the user adjusting the grip position and is not used as the boundary between the active unloading section and the passive unloading section.

[0030] When no trajectory point meeting the boundary condition is detected, the complete unloading trajectory is determined as the active unloading section, and the duration of the passive unloading section is recorded as zero. When the contact signal is continuously interrupted after the load reaches its peak and the subsequent displacement is mainly generated by the instrument reset action, the duration of the active unloading section is recorded as zero, and the trajectory after the peak load is determined as the passive unloading section.

[0031] The residual control state includes unloading baseline parameters and candidate moment acceptance parameters. The unloading baseline parameters include the duration of the active unloading section, the duration of the passive unloading section, the unloading endpoint load, the remaining return displacement, the unloading endpoint speed, and the contact state. The candidate moment acceptance parameters include the expected contact establishment interval, the expected pre-tightening starting load, the expected pre-tightening starting speed, the expected initial loading displacement, and the expected pre-tightening load holding amplitude.

[0032] Among them, the duration of the active unloading section, the duration of the passive unloading section, the unloading endpoint load, the remaining return displacement, the unloading endpoint speed and the contact status are the actual state parameters formed when the current task ends. The residual control status is updated according to the waiting time. It does not change the unloading trajectory data that has been collected, but uses the state parameters as state indexes to determine the corresponding state representation when the user enters the next task after the corresponding waiting time in the waiting state update template.

[0033] When creating a waiting state update template, the target user's historical adjacent training tasks are categorized according to training body part, equipment resistance type, load level, and movement range. For each historical training under the same category, the residual control state at the end of the previous task, the actual waiting time, the contact establishment time, preload start load, preload start speed, initial load displacement, and load holding status at the first loading of the next task are saved to form a waiting state sample.

[0034] The waiting state samples under the same category are arranged according to the waiting time. For the current residual control state, historical samples with active unloading section duration, passive unloading section duration, unloading end speed, and remaining return displacement that are close to the current state are selected. Based on the waiting time corresponding to the candidate release time, the expected contact establishment interval, expected pre-tightening start load, expected pre-tightening start speed, and expected first loading displacement at the candidate release time are determined from the adjacent waiting state samples. This forms the updated residual control state corresponding to the candidate release time.

[0035] When the historical waiting state samples of the target user are insufficient, a common initial template corresponding to the same instrument resistance form, the same training task, and similar load level is called. After the target user completes the actual training, the residual control state before this waiting, the actual waiting time, and the actual first loading state are added to the user's waiting state sample for subsequent updates.

[0036] S2. When the target device of the next task is occupied, collect the load, displacement and contact data at the end of the current user group, generate the unloading trajectory at the end of the group, and pair it with the preload trajectory of the adjacent action group to form a succession template, determine the reloadable termination domain and the handover termination domain; determine the expected release interval based on the approach order, dwell time and stability of the current trajectory to the two termination domains. When the contact acquisition unit of the target device remains triggered, the load data continues to change, or the control terminal of the target device records an incomplete action group, it is determined that the target device is in an occupied state. The management terminal reads the number of remaining action groups of the current user on the target device, the historical single-group training time and the historical inter-group rest time, and continuously collects the load, displacement and contact data of the end of the current action group.

[0037] The preload trajectory is the trajectory formed by the current user re-contacting the target device, driving the transmission mechanism to eliminate the empty stroke and gradually build up the load before the subsequent action set officially generates effective training resistance. The preload trajectory includes at least the contact start time, preload displacement, preload speed, load building direction and resistance take effect time.

[0038] When forming a succession template, the reset state of the target device is used as the alignment point. The terminal trajectory segment of the unloading trajectory of the previous action group is paired with the starting trajectory segment of the preloading trajectory of the next action group. The set of unloading terminal states that enter the preloading trajectory within the preset inter-group time after pairing is determined as the reloading termination domain. The set of unloading terminal states that are interrupted by the contact signal and do not enter the preloading trajectory within the preset handover time is determined as the handover termination domain. During pairing, the displacements in the unloading trajectory and the preloading trajectory are uniformly converted into proportional positions relative to the effective movement stroke of the target device, and the alignment time is taken as the moment when the moving part of the device enters the reset allowable range. When the unloading trajectory enters the reset state, the current user regenerates a continuous contact signal within the normal rest period between groups, and when the displacement and load change according to the historical preloading direction, the corresponding unloading terminal state is assigned to the reloadable termination domain.

[0039] When the unloading trajectory enters the reset state, the contact signal is continuously interrupted. If no load and displacement changes corresponding to the historical preload trajectory are formed within the preset handover time, the corresponding unloading terminal status will be classified into the handover termination domain. When the user identification terminal detects that the current user has left the area corresponding to the target device, it can be used as an auxiliary judgment basis for entering the handover termination domain.

[0040] The same unloading terminal state may enter the reloadable termination domain and the handover termination domain in different historical training processes. The succession template records the number of times it subsequently enters the two termination domains, the dwell time, and the stability of the trajectory. The current user's device release result is not directly determined by a single historical training record.

[0041] When determining the expected release interval, the current user's end-of-group unloading trajectory is matched with the trajectory segment in the continuation template to determine the remaining time for the current trajectory to reach the reloadable termination domain and the handover termination domain respectively; the predicted time when the current trajectory reaches the handover termination domain by the handover branch is taken as the start time of the expected release interval; the predicted time when the current trajectory enters the reloadable termination domain by the reloadable branch, continues to complete the remaining action group, and finally reaches the handover termination domain is taken as the end time of the expected release interval.

[0042] Specifically, the terminal trajectory segment of the current unloading trajectory is matched with the trajectory segments in the continuation template that lead to the reloadable termination domain and the handover termination domain. Based on the similarity between the current trajectory and the corresponding trajectory segment in terms of load reduction trend, displacement return ratio, contact state and dwell time, the trajectory branch that is closer to the current trajectory is determined.

[0043] Along the trajectory branch leading to the handover termination domain, determine the first remaining time based on the current trajectory change speed and the remaining trajectory length to reach the handover termination domain. Use the sum of the current time and the first remaining time as the start time of the expected release interval. Along the trajectory branch that first enters the reloadable termination domain, determine the latest release time of the target device based on the current user's remaining number of action sets, the historical single-set training time, the rest time between sets, and the time taken to enter the handover termination domain after the last set, and use this as the end time of the expected release interval.

[0044] When the current trajectory changes from approaching the handover termination domain to approaching the reloadable termination domain, or from approaching the reloadable termination domain to approaching the handover termination domain, the continuation template is rematched and the expected release interval is updated.

[0045] S3. Within the expected release range, update the residual control state according to the waiting time corresponding to the candidate release time, and sequentially match it with the contact establishment, transmission preload and resistance activation states of the next task. The first state that does not meet the continuity condition is identified as the receiving gap; the candidate release time that does not form a receiving gap forms a waiting path. Within the expected release interval, multiple candidate release times are set according to a preset time interval, and the time length from the current time to each candidate release time is used as the corresponding waiting time. For each candidate release time, the updated residual control state of the target user is determined using the waiting state update template.

[0046] The contact establishment state for the next task is determined based on the position of the target user in contact with the target device, the duration of contact, and the initial direction of action; the transmission preload state is determined based on the changes in load, displacement, and velocity during the period from contact establishment to the elimination of the empty stroke of the target device; and the resistance activation state is determined based on the load, displacement, and contact state when the load begins to rise continuously and enters the effective training load range.

[0047] When the current task and the next task are performed by different instruments, the reset position of the corresponding instruments is used as the zero point of displacement. The unloading endpoint displacement of the current task and the preloading start displacement of the next task are converted into proportional positions in the effective motion stroke of the corresponding instruments. Continuity matching is performed using proportional positions, and the absolute displacement values ​​of different instruments are not directly compared.

[0048] The continuity conditions include: the time interval between the termination time of the contact signal in the previous state and the establishment time of the contact in the next state does not exceed the contact interval threshold; the load change direction in the previous state is consistent with the loading direction in the next state; the difference between the unloading endpoint displacement and the starting displacement in the next state does not exceed the displacement threshold; and the load drop amplitude before entering the resistance effective state from the transmission preload state does not exceed the drop threshold. The first state that does not meet at least one continuity condition is identified as a connection gap.

[0049] The contact interval threshold is determined based on the contact interval when the target user successfully completed adjacent training tasks in the past; the displacement threshold is determined based on the idle stroke and preload stroke of the target device in the next task; and the fallback threshold is determined based on the allowable load fluctuation range for the target device's transmission mechanism to maintain effective connection after preload is completed.

[0050] Continuous matching is performed according to the actual occurrence sequence of the contact establishment state, transmission preload state, and resistance activation state. When the residual control state after the update cannot enter the contact establishment state, the contact establishment state is identified as the acceptance gap; when it can enter the contact establishment state but cannot stably enter the transmission preload state, the transmission preload state is identified as the acceptance gap; when it can complete the transmission preload but a load drop exceeding the fallback threshold occurs before the resistance activates, the resistance activation state is identified as the acceptance gap.

[0051] When the updated residual control state corresponding to a candidate release time meets the continuity condition with all three states, the candidate release time, the target user's waiting position, and the travel time to the target device are jointly determined as the waiting path. When multiple candidate release times do not form a gap, the candidate release time with a shorter waiting time and higher target device release stability is selected first.

[0052] S4. When there is a gap, select bridging occupancy units that can be executed by the idle equipment in the mobile cabin from the remaining tasks of the target user's training plan. The loading entry of the bridging occupancy unit accepts the residual control state after the update. Its unloading termination state continues the first continuous state after the gap and is reset before the target equipment is released, forming a bridging insertion path. The management terminal reads the training tasks that have not yet been executed in the target user's training plan, excludes tasks that can only be executed after the next task is completed, and excludes tasks whose corresponding devices are occupied during the target waiting period. For the remaining tasks, it obtains the loading entry status, training duration, unloading termination status, and device reset duration of the corresponding devices.

[0053] The bridging occupancy unit includes the user's travel period to the idle device, the loading period, the training period, the unloading period, and the device reset period. The loading entry state of the bridging occupancy unit and the updated residual control state meet the continuity condition, and its unloading termination state and the first continuous state after the bridging gap meet the continuity condition. Moreover, the termination time of the device reset period is not later than the corresponding candidate release time.

[0054] The travel period is determined based on the traversable path from the target user's current location to the location of the idle device; the loading period begins when the target user contacts the idle device and ends when the device's resistance component takes effect; the training period is determined based on the number of planned action sets and the target user's historical completion time; the unloading period begins at the peak load moment of the last action of the bridging task and ends when the device's moving parts enter the reset allowable range; the device reset period includes the time required for the user to release contact and for the device to return to a state usable by subsequent users.

[0055] The unloading termination state of the bridging occupancy unit is the first continuous state after the connection gap. It does not omit the contact establishment, transmission preload or resistance activation process in the next task. After the bridging occupancy unit is completed, the unloading termination state formed by the bridging task is used as the new state reference. Based on the remaining time between the completion time of the bridging occupancy unit reset and the release time of the target device candidate, the waiting state update template is used again to determine the updated residual control state after bridging.

[0056] After bridging, the updated residual control state is re-matched sequentially with the contact establishment state, transmission preload state, and resistance activation state of the next task to achieve complete continuity. When the original bearing gap no longer appears and the first continuous state after the original bearing gap still meets the continuity condition, it is determined that the unloading termination state of the bridging occupied unit can continue the first continuous state after the bearing gap. When the original bearing gap still exists after re-matching, or a new bearing gap is formed at a more forward position, the corresponding bridging occupied unit is excluded.

[0057] When multiple bridging units are occupied, the bridging unit with the device reset completion time closest to and no later than the target device candidate release time is selected first, so as to reduce the target user's waiting time after completing the bridging task.

[0058] In this embodiment, it is assumed that after the target user completes the seated rowing task, he / she waits to perform the lifting task. After the expected waiting time, the target user is able to establish contact with the lifting device, but cannot stably enter the transmission preload state. At this time, the transmission preload state is identified as the receiving gap. The management terminal selects a bridging task that can be performed by an idle device from the target user's remaining tasks. After the target user completes the bridging task, a new unloading termination state is formed. After rematching, the target user can sequentially enter the contact establishment, transmission preload, and resistance activation states of the lifting task. When the bridging device can complete the reset before the lifting device is released, the bridging task is identified as the bridging occupied unit.

[0059] S5. When there is no bridging occupancy unit, select candidate devices that accept the updated residual control state at the preload entry and whose load process continuously covers the critical load segment of the next task, embed their replacement occupancy period into the device occupancy sequence, exclude candidate devices that prevent the updated residual control state of the affected user from continuing its preload state, determine the replacement device path, and update the device occupancy sequence and user travel guidance.

[0060] There are no bridging units that meet the conditions, including situations where there are no remaining tasks in the target user's training plan that can fill the gap, the devices corresponding to the remaining tasks are all occupied during the waiting period, and the bridging task cannot be reset before the target device's candidate release time.

[0061] The management terminal extracts the human body's stress points, human body support posture, equipment contact points, main limb movement direction, resistance direction, effective movement range, and load change trajectory from other equipment in the cabin. It first performs preliminary screening based on the human body's stress points, human body support posture, equipment contact points, and main limb movement direction corresponding to the next task. The human body's stress points and contact points of candidate equipment should correspond to the next task, the human body support posture should belong to the same posture category, and the main limb movement direction should be within the range of movement directions allowed by the next task. Equipment that does not meet the above conditions will not be included in the critical load segment coverage judgment.

[0062] The critical load segment is the trajectory segment in the next task from the activation of the resistance component to the decrease of the load to a preset proportion of the peak value after reaching its peak value. The critical load segment is characterized by the load direction, displacement range, and normalized load change curve. When the load direction of the candidate device is consistent with the load direction of the critical load segment, the displacement range coverage ratio is not lower than the coverage threshold, and the deviation of the normalized load change curve does not exceed the curve deviation threshold, the load process of the candidate device is determined to continuously cover the critical load segment.

[0063] When determining the critical load segment of the next task, extract the trajectory from the load trajectory of the target user who has effectively completed the task in history, from the time the resistance component starts to take effect until the load reaches its peak and then drops to a preset ratio. Convert the displacement in the trajectory into the proportional position in the effective motion stroke, and arrange the corresponding loads according to the proportional position to form a normalized load change curve.

[0064] When judging candidate devices, the effective motion stroke of the corresponding action of the candidate device is normalized in the same way, and the load change trend of the candidate device and the load of the critical load segment are compared at the same proportional position. If the difference between the load change curves is within the curve deviation threshold, and the effective displacement range of the candidate device covers the proportional position required by the critical load segment, it is determined that its load process can continuously cover the critical load segment.

[0065] In addition to meeting the critical load segment coverage requirement, the candidate device's preload entry should also meet the continuity requirement with the target user's updated residual control state, thereby avoiding determining the substitution relationship solely based on the device's training location or device name.

[0066] Other users whose alternative usage time overlaps with the allocated device usage time are identified as affected users. The candidate usage time of the affected users is updated according to the overlap duration, and the residual control status of the affected users is updated using the waiting status update template. When the updated residual control status does not meet the continuity condition with the preload status of its next task, the corresponding candidate device is excluded.

[0067] The alternative occupancy period includes the travel period of the target user to the candidate device, the preloading period, the training period, the unloading period, and the device reset period. The alternative occupancy period is inserted into the original device occupancy sequence of the candidate device according to the start and end times, and the overlap between the alternative occupancy period and the allocated occupancy period is checked.

[0068] When the replacement occupancy period overlaps with the allocated occupancy period of other users, the reset completion time of the previous occupancy period is used as the earliest time when the next occupancy period can be used. The candidate usage time of the affected users is updated in turn. The waiting time of the affected users is re-determined based on the updated candidate usage time, and the waiting status update template of the corresponding user is called to determine its updated residual control status.

[0069] The updated residual control state of the affected user is sequentially matched with the contact establishment state, transmission preload state and resistance activation state of its original next task. If any affected user has a gap in the connection, it is determined that the target user's use of the candidate device will disrupt the training connection relationship of the assigned user, and the candidate device is excluded.

[0070] Among the remaining candidate devices, the candidate device that can take over the residual control state after the target user's update, whose load process can continuously cover the key load segment, will not disrupt the training acceptance relationship of the affected users, and whose reset completion time is earlier is selected to form an alternative device path. The management terminal updates the device occupancy sequence according to the alternative device path and sends the candidate device location, travel route, expected arrival time and training start time to the target user terminal.

[0071] When the current user continues to execute the next action group, the actual unloading trajectory at the end of the group is paired with the actual preload trajectory of the next action group to form a reloadable sample; when the current user ends the use of the target device, the actual final unloading trajectory at the end of the group, the contact interruption status, and the status of not forming a preload trajectory again within the preset handover time are used to form a handover sample; the continuation template, reloadable termination field, and handover termination field are updated based on the reloadable sample and the handover sample.

[0072] During path execution, the management terminal continues to collect load, displacement, return speed and contact data of the target user, the current user and related instruments. The actual instrument release time is determined by the moment when the current user's final unloading trajectory enters the handover termination domain and remains stable. The actual receiving gap is determined based on the contact establishment, transmission preload and resistance activation process when the target user actually performs the next task.

[0073] When a user completes a set of actions and continues to execute the next set of actions, the actual unloading trajectory of the current set of actions is paired with the actual preloading trajectory of the next set of actions according to the device reset state, and added as a reloadable sample to the continuation template to correct the reloadable termination domain.

[0074] When the current user completes all action groups on the target device and ends use, the final unloading trajectory, contact signal interruption status, device reset status, and status of not forming a preload trajectory again within the preset handover time are recorded and added to the handover template as a handover sample to correct the handover termination domain. For the final unloading trajectory of the next action group, it is not forcibly paired with the preload trajectory.

[0075] When the actual release time of the device exceeds the expected release range, the dwell time of the corresponding trajectory branch, the completion time of the remaining action group, and the remaining time to reach the termination domain are adjusted according to the result that the actual trajectory finally enters the reloadable termination domain or the handover termination domain.

[0076] When a gap is predicted to exist but the target user is actually able to complete the next task continuously, or when a gap is predicted not to exist but the target user is actually interrupted during contact establishment, transmission preload, or resistance activation, the residual control state before this wait, the actual wait time, and the actual initial loading state are added to the wait state update template to correct the state mapping relationship under the corresponding wait time.

[0077] When updating the succession template and the waiting status update template, the new samples are classified according to the equipment type, training task, load level and user training stage. Training trajectories with different transmission forms, resistance directions or human support postures are not directly merged. The updated template is used to determine the expected release range and the gap for subsequent similar training tasks.

[0078] Through the above steps, this embodiment determines whether the target user can continuously enter the next task at different candidate release times based on the target user's current task unloading status, the target device's current user unloading-preloading succession relationship, and the actual loading status of the next task. Based on this, it determines the waiting path, bridging insertion path, and alternative device path, so that the training path adjustment result corresponds to the user's actual control state and the device transmission state.

[0079] Example 2 See Figure 2In the first embodiment, after the bridging or substitution occupancy unit is completed, although the moving parts of the device have returned to their reset positions, the seat, support, transmission connection, and resistance setting may still retain the previous user's configuration, which cannot directly meet the training requirements of the next user. At the same time, the load, displacement, and contact changes generated during component adjustment are easily misidentified as the preloaded data of the next user, leading to the premature judgment of the device handover time and causing conflicts in the subsequent device occupancy sequence. To solve this problem, in this embodiment, the bridging occupancy unit and the substitution occupancy unit corresponding to the substitution device path are respectively determined as occupancy units to be verified.

[0080] Before the execution of the unit to be verified, the first user in the instrument occupancy sequence following the unit to be verified is identified as the next affected user. Based on the expected instrument configuration state after the unit to be verified completes the target user's task, and the position of the human support component, the position of the transmission connection component, and the resistance setting state corresponding to the preloaded trajectory of the next affected user, a predicted configuration reset trajectory is generated. The state in which the human support component and the transmission connection component complete locking, the resistance setting reaches the target state, and the load fluctuation and displacement fluctuation return to stability is identified as the configuration handover termination domain. The reset time period required to reach the configuration handover termination domain is inserted into the instrument occupancy sequence, and the units to be verified that cannot enter the configuration handover termination domain before the candidate usage time of the next affected user are excluded before the path is issued.

[0081] After the actual execution of the unit to be verified, the position, locking status, resistance setting status, load, displacement and contact data of the instrument components are collected to form the actual configuration reset trajectory; the load, displacement and contact changes collected before actually entering the configuration handover termination domain are determined as the configuration adjustment section, and the configuration adjustment section is excluded when extracting the preload trajectory of the next affected user.

[0082] Specifically, the human body support components include a seat, backrest, chest pad, leg pad, and footrest support; the transmission connection components include pulley adjustment components, cable connection components, handle connection components, lever arm adjustment components, and travel limit components. Position detection components and locking detection components are installed on the human body support components and transmission connection components to collect the current position and locking status of the components, respectively. The resistance setting status is read by the device controller.

[0083] Before the unit to be verified is executed, the expected device configuration state after the task is completed is determined according to the device configuration required by the target user when performing the corresponding task. According to the training task and historical preloading record of the next affected user, the position of the target human body support component, the position of the target transmission connection component, and the target resistance setting state are determined. The expected device configuration state is used as the starting state and the target state is used as the ending state. According to the historical adjustment sequence, adjustment duration and locking duration of each component of the corresponding device, the predicted configuration reset trajectory is formed and the expected reset time period is determined.

[0084] When the predicted end time of the reset period is not later than the candidate usage time of the next affected user, the corresponding occupied unit to be verified is retained; when the predicted end time of the reset period is later than the candidate usage time of the next affected user, the corresponding occupied unit to be verified is excluded before the path is issued.

[0085] After the actual execution of the unit to be verified, the position changes, locking status, resistance setting status, load changes, displacement changes and contact changes of each component are recorded according to the sampling time to form the actual configuration reset trajectory. When each component reaches the target position corresponding to the next affected user and completes locking, the resistance setting reaches the target state, and the load fluctuation and displacement fluctuation are within a stable range within a continuous preset time, it is determined that the device has entered the configuration handover termination domain.

[0086] The data from the end of the unit to be verified to the end of the configuration handover termination domain is marked as the configuration adjustment segment. When extracting the preload trajectory of the next affected user, only the load, displacement and contact data formed by the next affected user contacting the instrument after the instrument enters the configuration handover termination domain are extracted. The actual reset time period is compared with the predicted reset time period, and the component adjustment time and locking time of the corresponding instrument are updated according to the comparison results for the configuration handover judgment of the subsequent unit to be verified.

[0087] Example 3 See Figure 3 In Example 1, inserting the alternative occupancy period of the target user into the device occupancy sequence will delay the candidate usage time of subsequent users of that device. The extended waiting time may prevent the residual control state of the affected user from continuing its original preload state, thus requiring the use of other devices to rearrange training. This causes the device occupancy change to continue to be transmitted between multiple users and multiple devices. In order to avoid the continuous adjustment of the device occupancy sequence caused by local alternative device paths, this example transmits and cuts off the occupancy disturbance caused by alternative occupancy step by step.

[0088] Specifically, the alternative occupancy period corresponding to the alternative device path is inserted into the occupancy sequence of the corresponding device, and the device is determined as the starting point of the occupancy disturbance. According to the occupancy order of the devices after the alternative occupancy period, the users who plan to use the device are identified as the affected users in turn.

[0089] For each affected user, the device reset completion time updated in the previous occupied period is compared with the original candidate usage time of the affected user. If the device reset completion time in the previous occupied period is later, the device reset completion time is used as the updated candidate usage time for the affected user; if the device reset completion time in the previous occupied period is earlier, the original candidate usage time is retained.

[0090] The waiting time of the affected user is determined based on the updated candidate usage time, and the updated residual control state is determined according to the waiting state update method described in Example 1. Then, it is continuously matched with the preload state of the next task originally scheduled for the affected user.

[0091] When the updated residual control state meets the continuity condition, the original training task and device path of the affected user are retained, and the updated device reset completion time of the affected user is determined based on the updated candidate usage time, training duration, unloading duration and device reset duration.

[0092] When the updated device reset completion time is no later than the next user's original candidate usage time, it indicates that the time delay has been absorbed by the device idle time between the two occupied time periods, and the occupancy disturbance is cut off at the affected user; when the updated device reset completion time is later than the next user's original candidate usage time, the time difference between the two is determined as the remaining occupancy disturbance, and it continues to be passed on to the next user along the subsequent occupancy order of the current device.

[0093] When the updated residual control state does not meet the continuity condition, the first state that does not meet the continuity condition is identified as a new acceptance gap according to the continuity matching order described in Example 1. The reconstruction path is determined for the affected user according to the determination order of the bridging insertion path and the alternative device path in Example 1.

[0094] The occupancy period corresponding to the reconstructed path is inserted into the occupancy sequence of the corresponding device, and the delay caused by the occupancy period to the candidate usage time of subsequent users is determined as a new occupancy disturbance. The device corresponding to the reconstructed path is used as the new disturbance starting point, and the affected users are determined according to the subsequent occupancy order of the device. The candidate usage time update, residual control state update, continuity matching and occupancy time judgment are repeated.

[0095] During the same occupancy disturbance handling process, the user ID, device ID, and corresponding occupancy period are recorded. The same user will not be judged repeatedly under the same device and the same occupancy period; when the reconstruction path involves a device that has already been processed, only users who have not been processed after the newly added occupancy period will be judged.

[0096] When the remaining occupancy disturbance is completely absorbed by the idle time in the device occupancy sequence, or when there are no subsequent affected users for the corresponding device, the disturbance branch is terminated; when all disturbance branches are terminated, the updated device occupancy sequence, the candidate usage time of the affected users, the training path, and the user movement guidance are output.

[0097] In this embodiment, it is assumed that user A uses the second device to perform a substitute task, thus delaying the candidate usage time of user B who originally planned to use the second device. If user B's updated residual control state cannot continuously enter the transmission preload state of the original task, the transmission preload state is determined as a new connection gap, and a bridging occupancy unit on the third device is determined for user B. After user B uses the third device, the candidate usage time of subsequent user C is delayed accordingly. When user C can still continue the original preload state, and its device reset completion time is not later than the original candidate usage time of the next user, it is determined that the time delay has been absorbed by the idle period, and the occupancy disturbance is cut off at user C.

[0098] Through the above processing, the training path is re-determined only for users whose training connection is actually disrupted, and the transmission of occupancy disturbance is terminated after the time delay is absorbed by the idle period of the equipment, thus avoiding continuous adjustment of the occupancy sequence of multiple users and multiple equipment caused by a single replacement equipment allocation.

[0099] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A smart management method for fitness cabins based on multi-source data fusion, characterized in that, Includes the following steps: S1. Obtain the target user's training plan and the load, displacement, return speed and contact data of the last group of the current task, construct the unloading trajectory, divide the active and passive unloading segments according to the synchronization relationship between load decrease and displacement return, contact duration and speed stability, and determine the residual control state. S2. When the target device of the next task is occupied, collect the load, displacement and contact data at the end of the current user group, generate the unloading trajectory at the end of the group, and pair it with the preload trajectory of the adjacent action group to form a succession template, determine the reloadable termination domain and the handover termination domain; determine the expected release interval based on the approach order, dwell time and stability of the current trajectory to the two termination domains. S3. Within the expected release range, update the residual control state according to the waiting time corresponding to the candidate release time, and sequentially match it with the contact establishment, transmission preload and resistance activation states of the next task. The first state that does not meet the continuity condition is identified as the receiving gap; the candidate release time that does not form a receiving gap forms a waiting path. S4. When there is a gap, select bridging occupancy units that can be executed by the idle equipment in the mobile cabin from the remaining tasks of the target user's training plan. The loading entry of the bridging occupancy unit accepts the residual control state after the update. Its unloading termination state continues the first continuous state after the gap and is reset before the target equipment is released, forming a bridging insertion path. S5. When there is no bridging occupancy unit, select candidate devices that accept the updated residual control state at the preload entry and whose load process continuously covers the critical load segment of the next task, embed their replacement occupancy period into the device occupancy sequence, exclude candidate devices that prevent the updated residual control state of the affected user from continuing its preload state, determine the replacement device path, and update the device occupancy sequence and user travel guidance.

2. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 1, characterized in that, The unloading trajectory consists of trajectory points arranged according to the sampling time. Each trajectory point includes the load drop slope, displacement return ratio, return speed, contact state, and load fluctuation amplitude. The first trajectory point where the contact signal changes from continuous to interrupted and the timing deviation between load drop and displacement return begins to increase is determined as the boundary point between the active unloading section and the passive unloading section. The residual control state includes unloading reference parameters and candidate moment acceptance parameters. The unloading reference parameters include the duration of the active unloading section, the duration of the passive unloading section, the unloading endpoint load, the remaining return displacement, the unloading endpoint speed, and the contact state. The candidate moment acceptance parameters include the expected contact establishment interval, the expected pre-tightening starting load, the expected pre-tightening starting speed, the expected initial loading displacement, and the expected pre-tightening load holding amplitude.

3. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 1, characterized in that, When forming the connecting template, the reset state of the target device is used as the alignment point, and the terminal trajectory segment of the unloading trajectory of the previous action group is paired with the starting trajectory segment of the preloading trajectory of the next action group. The set of unloading terminal states that enter the preload trajectory within the preset inter-group time after pairing is determined as the reloadable termination domain, and the set of unloading terminal states that are interrupted by the contact signal and do not enter the preload trajectory within the preset handover time is determined as the handover termination domain. When determining the expected release interval, the current user's end-of-group unloading trajectory is matched with the trajectory segment in the continuation template to determine the remaining time for the current trajectory to reach the reloadable termination domain and the handover termination domain respectively; the predicted time when the current trajectory reaches the handover termination domain by the handover branch is taken as the start time of the expected release interval; the predicted time when the current trajectory enters the reloadable termination domain by the reloadable branch, continues to complete the remaining action group, and finally reaches the handover termination domain is taken as the end time of the expected release interval.

4. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 2, characterized in that, The continuity conditions mentioned in step S3 include: the time interval between the contact signal termination time of the previous state and the contact establishment time of the next state does not exceed the contact interval threshold; the load change direction of the previous state is consistent with the loading direction of the next state; the difference between the unloading endpoint displacement and the starting displacement of the next state does not exceed the displacement threshold; and the load drop amplitude before entering the resistance effective state from the transmission preload state does not exceed the drop threshold. The first state that does not meet at least one continuity condition is determined as a connection gap.

5. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 4, characterized in that, The bridging occupancy unit includes the user's travel period to the idle device, the loading period, the training period, the unloading period, and the device reset period; the loading entry state of the bridging occupancy unit and the updated residual control state satisfy the continuity condition. After the bridging occupancy unit is completed, its unloading termination state is used as the new state reference. The residual control state is updated according to the remaining waiting time from the end time of the instrument reset period to the corresponding candidate release time. The updated residual control state is sequentially matched with the contact establishment state, transmission preload state and resistance activation state of the next task. When the receiving gap disappears and the end time of the instrument reset period is not later than the corresponding candidate release time, the bridging occupancy unit is determined to form a bridging insertion path.

6. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 1, characterized in that, The critical load segment mentioned in step S5 is the trajectory segment in the next task from the activation of the resistance component to the load decreasing to a preset proportion of the peak value after reaching its peak value. The critical load segment is characterized by the load direction, displacement range, and normalized load change curve. When the load direction of the candidate device is consistent with the load direction of the critical load segment, the displacement range coverage ratio is not lower than the coverage threshold, and the deviation of the normalized load change curve does not exceed the curve deviation threshold, it is determined that the load process of the candidate device continuously covers the critical load segment. Before the critical load segment coverage judgment is performed, the human force-bearing part, device contact part, and human support posture corresponding to the candidate device are consistent with the next task, and the main movement direction of the limb is within the range of movement directions allowed by the next task.

7. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 4, characterized in that, Other users whose alternative usage time overlaps with the allocated device usage time are identified as affected users. The candidate usage time of the affected users is updated according to the overlap duration, and the residual control status of the affected users is updated using the waiting status update template. When the updated residual control status does not meet the continuity condition with the preload status of its next task, the corresponding candidate device is excluded.

8. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 3, characterized in that, When the current user continues to execute the next action group, the actual unloading trajectory at the end of the group is paired with the actual preload trajectory of the next action group to form a reloadable sample; when the current user ends the use of the target device, the actual final unloading trajectory at the end of the group, the contact interruption status, and the status of not forming a preload trajectory again within the preset handover time are used to form a handover sample; the continuation template, reloadable termination field, and handover termination field are updated according to the reloadable sample and the handover sample.

9. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 1, characterized in that, The bridging occupancy unit and the alternative occupancy unit corresponding to the alternative device path are respectively identified as occupancy units to be verified. For each occupancy unit to be verified, the first user in the device occupancy sequence after the occupancy unit to be verified is identified as the next affected user. Based on the device configuration state after the occupancy unit to be verified completes the target user's task, and the position of the human support component, the position of the transmission connection component, and the resistance setting state corresponding to the preload trajectory of the next affected user, a device configuration reset trajectory is generated. The state in which the human support component and the transmission connection component complete locking, the resistance setting reaches the target state corresponding to the next affected user, and the load fluctuation and displacement fluctuation return to stability is identified as the configuration handover termination domain. The load, displacement, and contact changes collected before entering the configuration handover termination domain are identified as configuration adjustment sections, and the configuration adjustment sections are prohibited from being added to the continuation template or used for the calculation of the expected release interval as the preload trajectory of the next affected user. The reset time period required for the occupancy unit to be verified to reach the configuration handover termination domain is inserted into the device occupancy sequence, and occupancy units to be verified that cannot enter the configuration handover termination domain before the candidate usage time of the next affected user are excluded.

10. The intelligent management method for fitness cabins based on multi-source data fusion according to claim 1, characterized in that, After inserting the alternative occupation period corresponding to the alternative device path into the device occupation sequence, the device corresponding to the alternative occupation period is taken as the starting point of the occupation disturbance, and the affected users are determined in sequence according to the subsequent occupation order of the device. The residual control state of each affected user is updated based on the delayed candidate usage time and corresponding waiting time, and the updated residual control state is matched with its original preloaded state for continuity. When the affected user meets the continuity condition, the updated device reset completion time is determined based on the updated candidate usage time, training time, unloading time, and device reset time. The occupancy disturbance is terminated when the device reset completion time is not later than the original candidate usage time of the next user. When the device reset completion time is later than the original candidate usage time of the next user, the time difference between the two is used as the remaining occupancy disturbance and continues to be propagated. When the affected user does not meet the continuity condition, the first state that does not meet the continuity condition is determined as a new acceptance gap. A reconstruction path is determined for the affected user according to the determination order of the bridging insertion path and the alternative device path, and the occupancy period corresponding to the reconstruction path is used as a new occupancy disturbance. The residual control state is updated and the continuity matching continues along the subsequent occupancy order of the corresponding device. The propagation is terminated when the remaining occupancy disturbance is absorbed by the device idle period, and the processing is terminated when there are no subsequent affected users for the corresponding device.

Images