Method and system for dynamic adjustment of collaborative control authority based on operator state assessment

Abstract

This invention belongs to the field of human-machine collaborative control technology and discloses a method and system for dynamic adjustment of collaborative control rights based on operator state assessment. The method includes: collecting multi-source state data of the operator to generate a continuous state feature vector; performing weighted fusion calculation on the continuous state feature vector to generate a comprehensive state index and trend of the operator; presetting control law parameters and generating a manual control ratio based on the comprehensive state index and trend; calculating the rate of change of the manual control ratio and obtaining a smoothed manual control ratio through smoothing constraint processing; acquiring manual control signals and automatic control signals, performing weighted fusion based on the smoothed manual control ratio to generate a fused control signal for collaborative control; evaluating the collaborative control error index during the collaborative control process and optimizing the control law parameters through feedback. This invention enables stable, reliable, and adaptive dynamic adjustment of manned-unmanned collaborative control rights based on operator state assessment.

Description

Technical Field

[0001] This invention relates to the field of human-machine collaborative control technology, and more specifically, to a method and system for dynamic adjustment of collaborative control rights based on operator state assessment. Background Technology

[0002] With the rapid development of intelligent equipment and autonomous system technologies, the collaborative execution of tasks by manned equipment and unmanned systems has gradually become an important application mode in complex task scenarios. In such collaborative control systems, human operators and automatic control systems need to participate in task execution together. By rationally allocating control rights, human-machine collaborative control can be achieved to improve the overall operating efficiency and safety of the system. Existing manned-unmanned collaborative control systems usually adopt a fixed control rights allocation method or a discrete-level control rights switching method. For example, in some systems, control rights are dominated by manual control for a long time, and automatic control is only triggered to take over under specific conditions. In other systems, several control levels are set to switch between manual and automatic control in stages. However, such control rights allocation methods are usually difficult to dynamically adjust according to the operator's real-time status. When the operator becomes fatigued, loses attention, or the task load changes, the system cannot adjust the control rights ratio in time, which may affect the collaborative control effect.

[0003] Furthermore, in existing technologies, there is usually a lack of effective continuous fusion mechanisms between manual control signals and automatic control signals. When control rights are switched, sudden changes in control rights or control conflicts can easily occur, leading to decreased system stability or even operational risks. At the same time, most existing systems lack the ability to optimize the control rights allocation strategy based on the actual control effect during the collaborative control process, and cannot continuously improve the collaborative control performance during long-term operation. Therefore, how to continuously and smoothly adjust the control rights dynamically according to the operator's real-time state changes, and achieve stable and reliable collaborative control between manual and automatic control, has become a technical problem that urgently needs to be solved in the field of manned-unmanned collaborative control.

[0004] In view of this, the present invention proposes a method and system for dynamic adjustment of collaborative control rights based on operator status assessment to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a method for dynamic adjustment of collaborative control rights based on operator state assessment, comprising: Collect multi-source state data from operators, normalize and construct features from the multi-source state data, and generate continuous state feature vectors; The continuous state feature vectors are weighted and fused to generate the operator's comprehensive state index, and the changing trend of the comprehensive state index is calculated. Preset control law parameters, and combine the comprehensive state index with the corresponding trend of change to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio. Based on the continuously changing artificial control ratio over time, the rate of change of the artificial control ratio is calculated, and a smoothed artificial control ratio is obtained through smoothing constraint processing. The system acquires manual and automatic control signals, performs weighted fusion based on the smoothed manual control ratio, generates a fused control signal for collaborative control, and determines whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection. The comprehensive status index is compared with the preset safety threshold, and the automatic control dominant mode or manual control recovery mode is triggered based on the comparison result. Evaluate the collaborative control error index during the collaborative control process, and optimize and adjust the control law parameters based on the collaborative control error index.

[0006] Furthermore, methods for generating continuous state feature vectors include: Acquire operator physiological state data, operational behavior data, and task response data; integrate the operator's physiological state data, operational behavior data, and task response data to form multi-source state data of the operator; wherein, the multi-source state data contains multiple state data items; perform normalization processing on each state data item in the multi-source state data to obtain the normalized feature value of each state data item; arrange the normalized feature values ​​corresponding to all state data items in sequence to form a continuous state feature vector. The expression for normalization is as follows: ; In the formula, For the first At the sampling time, the first The normalized eigenvalues ​​of each state data item. For the first The first sampling time before normalization Each status data item For the first The minimum value of each state data item. For the first The maximum value of each state data item.

[0007] Furthermore, methods for generating the operator's overall state index include: A preset feature weight set is provided, which contains the feature weight corresponding to each normalized feature value in the continuous state feature vector. Based on each feature weight, the normalized feature values ​​in the continuous state feature vector are weighted and summed to obtain the original comprehensive state index. The expression for the original comprehensive state index is: ; In the formula, For the first The original comprehensive state index at each sampling time point For the first Feature weights of each normalized eigenvalue. This represents the total number of normalized eigenvalues ​​in the continuous state eigenvector. The original composite state index is subjected to interval mapping to obtain the composite state index, and then the composite state index is subjected to moving average smoothing.

[0008] Furthermore, methods for generating artificially controlled proportions that change continuously over time include: The control law parameters are preset, including state index adjustment parameters and trend adjustment parameters. Based on the comprehensive state index, the trend of the comprehensive state index, and the control law parameters, the manual control ratio is obtained by mapping calculation through a preset continuous monotonic function. The expression for a continuous monotonic function is: ; In the formula, For the first The proportion of manual control at each sampling time point For smoothing functions, This is the state index adjustment parameter. For trend adjustment parameters, For the first The comprehensive state index at each sampling time point For the first The trend of the comprehensive state index at each sampling time.

[0009] Furthermore, methods for obtaining smoothed artificial control ratios through smoothing constraint processing include: The rate of change of the manually controlled ratio is marked as the current rate of change; a preset upper limit for the rate of change is set, and the absolute value of the current rate of change is compared with the upper limit for the rate of change. If the absolute value of the current rate of change is less than or equal to the upper limit of the rate of change, then the smoothing constraint processing is not performed, and the manual control rate is directly used as the smoothed manual control rate; if the absolute value of the current rate of change is greater than the upper limit of the rate of change, then the smoothing constraint processing is required. The method for handling smoothing constraints is as follows: If the current rate of change of the scale is positive, then the expression for smoothing constraints is: In the formula, For the first The smoothing manual control ratio at each sampling time point This represents the upper limit of the proportional change rate; if the current proportional change rate is negative, the expression for smoothing constraint processing is: .

[0010] Furthermore, the method for generating the fusion control signal includes: Acquire manual control signals and automatic control signals, and perform weighted fusion of the manual control signals and automatic control signals according to the smoothing manual control ratio to generate a fused control signal; The expression for the fusion control signal is as follows: ; In the formula, For the first The fusion control signal at each sampling time For the first Manual control signal at each sampling time For the first Automatic control signals at each sampling time.

[0011] Furthermore, methods for determining whether a priority arbitration mechanism or security enforcement control logic has been triggered include: The absolute value of the difference between the manual control signal and the automatic control signal is calculated to obtain the signal deviation, which is then compared with a preset signal conflict threshold. If the signal deviation is less than or equal to the signal conflict threshold, the priority arbitration mechanism is not triggered; if the signal deviation is greater than the signal conflict threshold, the priority arbitration mechanism is triggered. A set of predefined risk state triggering conditions is set up, which contains multiple predefined safety risk judgment conditions. The current system state of the collaborative control system is obtained, and it is determined whether the current system state meets any safety risk judgment condition in the risk state triggering condition set. If it does, the safety mandatory control logic is triggered; if it does not, the safety mandatory control logic is not triggered.

[0012] Furthermore, methods for determining whether the automatic control-dominated mode or the manual control recovery mode has been triggered include: Preset safety thresholds, including safety takeover threshold, takeover duration threshold, safety recovery threshold, and stability recovery duration threshold; The overall status index is compared with the safe takeover threshold; if the overall status index is less than the safe takeover threshold, the abnormal duration timer is started; if the overall status index is greater than or equal to the safe takeover threshold, the abnormal duration timer is reset to zero; when the cumulative duration recorded by the abnormal duration timer reaches or exceeds the takeover duration threshold, the automatic control master mode is triggered. In automatic control-dominated mode, the comprehensive status index is continuously monitored; when the cumulative duration of the comprehensive status index being higher than the recovery safety threshold reaches or exceeds the recovery stabilization duration threshold, the manual control recovery mode is triggered.

[0013] Furthermore, methods for feedback optimization and adjustment of control law parameters include: The collaborative control error index is obtained, which specifically includes trajectory tracking error and task completion deviation. The trajectory tracking error and task completion deviation are comprehensively evaluated to obtain the comprehensive control error. An error optimization trigger threshold is preset, and the comprehensive control error is compared with the error optimization trigger threshold. If the comprehensive control error is less than or equal to the error optimization trigger threshold, the parameter optimization process is not triggered. If the comprehensive control error is greater than the error optimization trigger threshold, the parameter optimization process is triggered. The parameter optimization process is as follows: obtain all comprehensive state indices and corresponding comprehensive control errors within a preset evaluation period to form a historical collaborative control dataset; calculate the corresponding low-state interval average error and high-state interval average error based on the historical collaborative control dataset, and compare them with preset interval error large judgment thresholds; determine whether to adjust the state index adjustment parameters and trend adjustment parameters according to the preset parameter adjustment step size based on the comparison results; update the control law parameters based on the adjusted state index adjustment parameters and trend adjustment parameters.

[0014] The collaborative control dynamic adjustment system based on operator state assessment, implementing the aforementioned collaborative control dynamic adjustment method based on operator state assessment, includes: The status acquisition module is used to collect multi-source status data of the operator, normalize the multi-source status data and construct features to generate continuous status feature vectors. The index generation module is used to perform weighted fusion calculation on continuous state feature vectors to generate the operator's comprehensive state index and calculate the trend of the comprehensive state index. The control law calculation module is used to preset control law parameters, and combine the comprehensive state index with the corresponding trend of change to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio. The proportional smoothing module is used to calculate the rate of change of the manual control ratio based on the continuously changing manual control ratio over time, and to obtain the smoothed manual control ratio through smoothing constraint processing. The fusion arbitration module is used to acquire manual control signals and automatic control signals, perform weighted fusion according to the smoothed manual control ratio, generate fused control signals for collaborative control, and determine whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection. The takeover management module is used to compare the comprehensive status index with the preset safety threshold and determine whether to trigger the automatic control master mode or the manual control recovery mode based on the comparison result. The feedback optimization module is used to evaluate the collaborative control error index during the collaborative control process and to optimize and adjust the control law parameters based on the collaborative control error index.

[0015] The technical effects and advantages of the collaborative control dynamic adjustment method and system based on operator status assessment of this invention are as follows: By collecting operator physiological state data, operational behavior data, and task response data, a continuous state feature vector containing multiple state data items is constructed. A comprehensive operator state index is then generated through weighted fusion calculation. This enables the system to accurately assess the operator's overall control capability level in real time, providing a reliable quantitative basis for dynamic adjustment of control authority. This effectively overcomes the deficiency of failing to adjust control allocation promptly based on changes in operator state due to a lack of comprehensive assessment of real-time operator state. Furthermore, by performing a moving average smoothing process on the comprehensive state index and calculating its changing trend, the system can filter out instantaneous fluctuations when assessing operator state. It can capture the direction and rate of state changes; by constructing a continuous monotonic function based on the comprehensive state index and its changing trend, it generates a continuously changing manual control ratio over time, realizing continuous dynamic adjustment of the manual control ratio and the automatic control ratio. This effectively overcomes the shortcomings of inflexible control adjustment and inability to make fine-grained continuous adjustments based on operator state changes caused by fixed allocation of control rights or discrete level switching; by setting an upper limit on the rate of change of the ratio to constrain the rate of change of the manual control ratio, the transition process between manual control and automatic control remains smooth and continuous, effectively avoiding system control jitter and operational discomfort caused by sudden changes in control rights. By weighting and fusing manual and automatic control signals according to a smoothed manual control ratio to generate a fused control signal, and combining this with a signal conflict detection and priority arbitration mechanism, the system can automatically select the priority signal based on the operator's current state credibility when conflicts arise between manual and automatic control signals. Simultaneously, a safety-enforced control logic forcibly switches to automatic control when a risky state is detected to ensure system safety. This effectively overcomes the shortcomings of the lack of an effective continuous fusion mechanism between manual and automatic control signals and the ease with which control conflicts can occur during control switching. Furthermore, a takeover management mechanism is set up to automatically trigger automatic takeover when the operator's overall state index remains below the safety takeover threshold. The system adopts a dynamic control-dominated mode and gradually increases the automatic control ratio according to the rate constraint. After the operator's state recovers to a safe level, the manual control ratio is gradually restored, realizing dynamic takeover and smooth recovery of control. Through a feedback optimization mechanism, the trajectory tracking error and task completion deviation are integrated into a comprehensive control error. Based on the average error level within different comprehensive state index intervals, the state index adjustment parameter and trend adjustment parameter in the control law parameters are specifically adjusted to form a closed-loop collaborative control structure. This enables the system to continuously improve the adaptability of the control law parameters during long-term operation, ultimately achieving stable, reliable, and adaptive dynamic adjustment of manned-unmanned collaborative control based on operator state assessment. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the collaborative control dynamic adjustment system based on operator status assessment according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of the collaborative control dynamic adjustment method based on operator status assessment in Embodiment 2 of the present invention. Detailed Implementation

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

[0018] Example 1

[0019] Please see Figure 1 As shown in this embodiment, the collaborative control dynamic adjustment system based on operator status assessment includes a status acquisition module, an index generation module, a control law calculation module, a proportional smoothing module, a fusion arbitration module, a takeover management module, and a feedback optimization module. The modules are connected via wired and / or wireless means to realize data transmission between the modules.

[0020] The status acquisition module is used to collect multi-source status data from the operator, normalize the multi-source status data and construct features to generate continuous status feature vectors.

[0021] Methods for collecting multi-source status data from operators include: Physiological state data of the operator is acquired from physiological signal acquisition equipment (i.e., sensor equipment used to collect the operator's physiological signs in real time) deployed in the control cabin where the operator is located. The physiological state data is used to record the operator's real-time physiological signs information, specifically including heart rate data, eye movement data, and electroencephalogram (EEG) data. The heart rate data is the operator's real-time heart rate value collected by the heart rate sensor; the eye movement data is the operator's gaze duration and pupil diameter collected by the eye tracker; and the EEG data is the frequency domain characteristic value of the operator's EEG signal collected by the EEG sensor. The operator's operational behavior data is acquired from an operational behavior acquisition device (i.e., a data acquisition device used to record the operator's input information to the control device) deployed on the operator's control device. This operational behavior data records the behavioral characteristics of the operator performing control actions through control devices such as joysticks, steering wheels, or control panels. Specifically, it includes control input frequency, control input amplitude, and control input regularity. Control input frequency is the number of times the operator performs a control input action within a preset sampling period. Control input amplitude is the displacement or force value of a single control input action. Control input regularity is the standard deviation of the time interval between multiple consecutive control input actions, reflecting the stability of the operator's control rhythm; a smaller control input regularity indicates a more stable control rhythm. The sampling period is preset by those skilled in the art based on actual conditions. The task response data of the operator is obtained from the task management platform of the collaborative control system (i.e., the task scheduling management platform used to manage and record the task instruction issuance and operator response information during the collaborative task execution process). The task response data is used to record the operator's response characteristics to the task prompt information issued by the system, specifically including task response time and task execution accuracy. The task response time is the time interval between the system issuing the task prompt information and the operator completing the corresponding control action. The task execution accuracy is the ratio of the number of times the operator correctly completes the task instruction within the preset evaluation period to the total number of task instructions. The operator's physiological state data, operational behavior data, and task response data are integrated to form multi-source state data of the operator. Among them, the multi-source state data contains multiple state data items, each of which corresponds to a specific type of state acquisition data, including real-time heart rate value, fixation duration, pupil diameter, frequency domain feature value, control input frequency, control input amplitude, control input regularity, task response time, and task execution accuracy.

[0022] Methods for generating continuous state feature vectors include: Normalization is performed on each state data item in the multi-source state data to map state data with different dimensions and numerical ranges to a unified numerical interval, resulting in the normalized feature value of each state data item; the specific expression for the normalization process is as follows: ; In the formula, For the first At the sampling time, the first The normalized eigenvalues ​​of each state data item. For the first The first sampling time before normalization Each status data item For the first The minimum value of each state data item. For the first The maximum value of each state data item.

[0023] For positive state data items where larger values ​​indicate stronger control capabilities, the calculation is performed according to the aforementioned minimum-maximum normalization formula. For negative state data items where larger values ​​indicate higher levels of fatigue, load, or instability, reverse normalization is used. This involves dividing the difference between the maximum value and the current sampled value by the difference between the maximum value and the minimum value, ensuring that each normalized feature value satisfies the condition that larger values ​​indicate stronger operator control capabilities. This avoids reverse calculation biases caused by negative physical quantities such as excessively high heart rate, excessively large pupil diameter, and excessively long task response time during weighted fusion.

[0024] The normalized feature values ​​corresponding to all state data items are arranged sequentially to form a continuous state feature vector; the continuous state feature vector is used to comprehensively represent the multi-dimensional state information of the operator at the current moment in a standardized numerical vector form.

[0025] The index generation module is used to perform weighted fusion calculations on continuous state feature vectors to generate the operator's comprehensive state index and calculate the trend of the comprehensive state index.

[0026] Methods for generating the operator's overall state index include: A preset feature weight set is provided, which contains the feature weight corresponding to each normalized feature value in the continuous state feature vector. Each feature weight is preset by a person skilled in the art based on the degree of influence of different state data items on the operator's control ability, and the sum of all feature weights is one. Based on the weights of each feature, the normalized eigenvalues ​​in the continuous state feature vector are weighted and summed to obtain the original comprehensive state index; the specific expression of the original comprehensive state index is as follows: ; In the formula, For the first The original comprehensive state index at each sampling time point For the first Feature weights of each normalized eigenvalue. This represents the total number of normalized eigenvalues ​​in the continuous state eigenvector.

[0027] The original comprehensive state index is subjected to interval mapping processing, which limits the value range of the original comprehensive state index to between zero and one, thus obtaining the comprehensive state index. The comprehensive state index is used to comprehensively reflect the operator's current overall control ability level with a single value. The closer the comprehensive state index is to one, the better the operator's current state and the stronger the control ability. The closer the comprehensive state index is to zero, the worse the operator's current state and the weaker the control ability. To avoid interference from transient fluctuations in the operator's state on the comprehensive state index, a moving average smoothing process is performed on the comprehensive state index; specifically, a preset smoothing window length is used, which is pre-set by those skilled in the art based on the stability requirements of the state assessment; the most recent... The comprehensive state index corresponds to each sampling time point, and the mean is calculated. The comprehensive state index is then updated based on the mean calculation result. To smooth out window length.

[0028] Methods for calculating the changing trend of the comprehensive state index include: A preset trend calculation time step is established, which is pre-set by those skilled in the art based on the detection sensitivity requirements of the state change rate. Based on the trend calculation time step, the changing trend of the comprehensive state index is calculated, specifically expressed as follows: ; In the formula, For the first The trend of the comprehensive state index at each sampling time point For the first The comprehensive state index at each sampling time point For the first The comprehensive state index corresponding to the trend calculation time step before each sampling time. Calculate the time step for the trend.

[0029] The trend of the comprehensive status index reflects the rate and direction of change of the operator's status over time. A positive trend of the comprehensive status index indicates that the operator's status is improving, while a negative trend indicates that the operator's status is deteriorating.

[0030] The control law calculation module is used to preset control law parameters, and combine the comprehensive state index with the corresponding change trend to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio.

[0031] Methods for generating artificially controlled proportions that change continuously over time include: The control law parameters are preset, including state index adjustment parameters and trend adjustment parameters. The state index adjustment parameters are used to control the influence of the comprehensive state index on the manual control ratio. The trend adjustment parameters are used to control the influence of the change trend of the comprehensive state index on the manual control ratio. Each control law parameter is preset by a person skilled in the art based on the response characteristics of the collaborative control system. Based on the comprehensive state index, its changing trend, and the control law parameters, the manual control ratio is obtained through mapping calculation using a preset continuous monotonic function; the specific expression of the continuous monotonic function is as follows: In the formula, For the first The proportion of manual control at each sampling time point For smoothing functions, This is the state index adjustment parameter. Here, the trend adjustment parameter is used; the manual control ratio is a value between zero and one, which represents the proportion of manual control in the collaborative control. The closer the manual control ratio is to one, the more dominant the manual control is, and the closer the manual control ratio is to zero, the more dominant the automatic control is. The smoothing function is a conventional mathematical function in this field, such as the Sigmoid function, etc. The specific function form will not be elaborated here. Methods for determining the automatic control ratio include: Calculate the automatic control ratio based on the manual control ratio; the specific expression for the automatic control ratio is: In the formula, For the first Automatic control ratio at each sampling time.

[0032] The proportional smoothing module is used to calculate the rate of change of the manual control ratio based on the continuously changing manual control ratio over time, and to obtain the smoothed manual control ratio through smoothing constraint processing.

[0033] Methods for obtaining smoothed artificial control ratios through smoothing constraint processing include: A preset upper limit for the proportional change rate is established, which is pre-set by those skilled in the art based on the requirements of the collaborative control system for the smoothness of control transition. The rate of change of the manual control ratio is marked as the current proportional change rate, which reflects the magnitude of change of the manual control ratio per unit time. The specific expression for the current proportional change rate is as follows: ; In the formula, For the first The current rate of change of the scale at each sampling time. For the first The proportion of manual control at each sampling time point The time interval between two sampling times.

[0034] The absolute value of the current proportional change rate is compared with the upper limit of the proportional change rate. If the absolute value of the current proportional change rate is greater than the upper limit, it is determined that the manually controlled proportional change is too fast, and smoothing constraint processing needs to be performed. The method of smoothing constraint processing is as follows: if the current proportional change rate is positive, the expression for smoothing constraint processing is: In the formula, For the first The smoothing manual control ratio at each sampling time point This represents the upper limit of the proportional change rate; if the current proportional change rate is negative, the expression for smoothing constraint processing is: If the absolute value of the current proportional change rate is less than or equal to the upper limit of the proportional change rate, then the smoothing constraint processing is not performed, and the manual control ratio is directly used as the smoothed manual control ratio; calculate the difference between the first and the smoothed manual control ratio to obtain the smoothed automatic control ratio. It should be noted that by applying a smooth constraint to the manual control ratio, the transition between manual and automatic control is kept smooth and continuous, avoiding control jitter or operational discomfort caused by sudden changes in the control ratio.

[0035] The fusion arbitration module is used to acquire manual control signals and automatic control signals, perform weighted fusion based on the smoothing ratio of manual control, generate fused control signals for collaborative control, and determine whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection.

[0036] Methods for generating fusion control signals include: The system acquires the current manual control signal from the operator's control device; wherein the manual control signal is the control command signal output by the operator through control devices such as joysticks, steering wheels, or control panels; and acquires the current automatic control signal from the automatic control system; wherein the automatic control signal is the control command signal autonomously generated by the automatic control system based on task planning and environmental perception. Based on the smoothing ratio of manual control, the manual control signal and the automatic control signal are weighted and fused to generate a fused control signal; the specific expression of the fused control signal is as follows: ; In the formula, For the first The fused control signal at each sampling time is used as the final output collaborative control command to the actuator for collaborative control. For the first Manual control signal at each sampling time For the first Automatic control signals at each sampling time.

[0037] Methods for determining whether a priority arbitration mechanism or security enforcement control logic has been triggered include: A preset signal conflict threshold is established, which is pre-set by those skilled in the art based on the safety margin requirements of the collaborative control system. The absolute value of the difference between the manual control signal and the automatic control signal is calculated to obtain the signal deviation. The signal deviation is compared with the signal conflict threshold. If the signal deviation is less than or equal to the signal conflict threshold, it is determined that there is no conflict between the manual control signal and the automatic control signal, and the priority arbitration mechanism is not triggered. If the signal deviation is greater than the signal conflict threshold, it is determined that there is a conflict between the manual control signal and the automatic control signal, and the priority arbitration mechanism is triggered. The priority arbitration mechanism is as follows: the current comprehensive status index is compared with a preset arbitration status threshold, which is preset by a person skilled in the art based on the minimum requirements for the operator's credible control capability; if the comprehensive status index is greater than or equal to the arbitration status threshold, the operator's current status is determined to be within the credible range, and the manual control signal is used as the priority signal, replacing the fusion control signal in the output to the execution mechanism; if the comprehensive status index is less than the arbitration status threshold, the operator's current status is determined to be untrustworthy, and the automatic control signal is used as the priority signal, replacing the fusion control signal in the output to the execution mechanism.

[0038] A set of predefined risk state triggering conditions is established, which includes multiple predefined safety risk judgment conditions. Each safety risk judgment condition is pre-set by a person skilled in the art according to the safety operation specifications of the collaborative control system. The current system state of the collaborative control system is obtained, and it is determined whether the current system state meets any safety risk judgment condition in the risk state triggering condition set. If it does, the collaborative control system is determined to be in a risk state, and the safety mandatory control logic is triggered. If it does not meet, the collaborative control system is determined not to be in a risk state, and the safety mandatory control logic is not triggered. The safety mandatory control logic specifically involves: forcibly replacing the fusion control signal with the automatic control signal output to the actuator, and temporarily setting the manual control ratio to zero.

[0039] The takeover management module is used to compare the comprehensive status index with the preset safety threshold and determine whether to trigger the automatic control master mode or the manual control recovery mode based on the comparison result.

[0040] Methods for determining whether automatic control master mode or manual control recovery mode has been triggered include: Preset safety thresholds, including safety takeover threshold, takeover duration threshold, safety recovery threshold, and stability recovery duration threshold; Both the safety takeover threshold and the takeover duration threshold are preset by those skilled in the art according to the safety operation standards of the collaborative control system; wherein, the safety takeover threshold is the lower limit of the comprehensive state index required to trigger automatic takeover; the takeover duration threshold is the shortest duration required to trigger automatic takeover when the comprehensive state index is continuously lower than the safety takeover threshold; Both the safety recovery threshold and the stabilization recovery time threshold are preset by those skilled in the art based on the operator's status recovery judgment criteria. The safety recovery threshold is the lower limit of the comprehensive status index required to determine the operator's status recovery, and the safety recovery threshold is greater than the safe takeover threshold. The stabilization recovery time threshold is the shortest stabilization time required for the comprehensive status index to remain higher than the safety recovery threshold.

[0041] The current comprehensive status index is compared with the safety takeover threshold. If the comprehensive status index is less than the safety takeover threshold, an abnormality duration timer is started to record the cumulative duration for which the comprehensive status index is continuously lower than the safety takeover threshold. If the comprehensive status index is greater than or equal to the safety takeover threshold, the abnormality duration timer is reset to zero. When the cumulative duration recorded by the abnormality duration timer reaches or exceeds the takeover duration threshold, the automatic control master mode is triggered. The automatic control dominant mode is as follows: the smooth manual control ratio is gradually reduced to the preset minimum manual control ratio according to the upper limit of the ratio change rate, and the smooth automatic control ratio is increased accordingly to ensure that the sum of the smooth manual control ratio and the smooth automatic control ratio is always one; wherein, the minimum manual control ratio is preset by those skilled in the art based on the minimum level of control participation retained by the operator in the takeover state.

[0042] In automatic control-dominated mode, the comprehensive status index is continuously monitored; when the cumulative duration of the comprehensive status index being higher than the recovery safety threshold reaches or exceeds the recovery stabilization duration threshold, it is determined that the operator's status has been restored to a safe level, triggering the manual control recovery mode. The manual control recovery mode is as follows: exit the automatic control dominant mode, recalculate the smoothed manual control ratio and mark it as the recovery manual control ratio; gradually increase the smoothed manual control ratio, which has been reduced to the minimum manual control ratio, to the recovery manual control ratio according to the upper limit of the ratio change rate, and reduce the smoothed automatic control ratio accordingly.

[0043] The feedback optimization module is used to evaluate the collaborative control error index during the collaborative control process and to optimize and adjust the control law parameters based on the collaborative control error index.

[0044] Methods for feedback optimization and adjustment of control law parameters include: The collaborative control error index is obtained from the task execution monitoring platform of the collaborative control system (i.e., the system operation monitoring platform used to monitor the actual motion state of the actuators of the collaborative control system in real time and record the task execution results). The collaborative control error index reflects the degree of deviation between the actual control effect of the collaborative control system under the current control law parameters and the expected control target. The collaborative control error index specifically includes trajectory tracking error and task completion deviation. The trajectory tracking error is the deviation between the actual motion trajectory and the expected motion trajectory after the actuator is driven by the fused control signal. The task completion deviation is the deviation between the actual task execution result and the task planning target. A comprehensive control error is obtained by comprehensively evaluating the trajectory tracking error and the task completion deviation. Specifically, corresponding error weights are set for the trajectory tracking error and the task completion deviation, and each error weight is preset by those skilled in the art based on the relative importance that the collaborative control system attaches to trajectory accuracy and task completion. Based on each error weight, the trajectory tracking error and the task completion deviation are weighted and summed to obtain the comprehensive control error. The comprehensive control error is used to comprehensively reflect the overall control effect of the current collaborative control with a single value. A preset error optimization trigger threshold is established, which is pre-set by those skilled in the art based on the control accuracy requirements of the collaborative control system. The comprehensive control error is compared with the error optimization trigger threshold. If the comprehensive control error is less than or equal to the error optimization trigger threshold, the control law parameters are determined to meet the control accuracy requirements, and the parameter optimization process is not triggered. If the comprehensive control error is greater than the error optimization trigger threshold, the control law parameters are determined to need to be optimized and adjusted, and the parameter optimization process is triggered. The parameter optimization process specifically includes: All comprehensive state indices and corresponding comprehensive control errors within a preset evaluation period are obtained to form a historical collaborative control dataset; wherein, the evaluation period is preset by those skilled in the art based on the amount of data accumulation required for optimizing control law parameters; A preset state index segmentation threshold is set, which is pre-set by those skilled in the art based on the classification standard of the operator's control ability level. The value range of the comprehensive state index is divided into a low state range and a high state range according to the state index segmentation threshold. The low state range is the range where the comprehensive state index is less than the state index segmentation threshold, and the high state range is the range where the comprehensive state index is greater than or equal to the state index segmentation threshold. From the historical collaborative control dataset, all comprehensive state indices in the low-state range are selected, and the average of the corresponding comprehensive control errors is calculated to obtain the average error of the low-state range; from the historical collaborative control dataset, all comprehensive state indices in the high-state range are selected, and the average of the corresponding comprehensive control errors is calculated to obtain the average error of the high-state range. The threshold for judging excessive interval error and the parameter adjustment step size are preset. The threshold for judging excessive interval error is preset by those skilled in the art based on the acceptable control error level in each state interval. The parameter adjustment step size is preset by those skilled in the art based on the convergence stability requirements of the control law parameter optimization. The average error of the low-state interval is compared with the threshold for judging excessive interval error. If the average error of the low-state interval is greater than the threshold for judging excessive interval error, it is determined that the control law parameters are not sensitive enough to the reduction of the manual control ratio when the operator's condition is poor. In this case, the state index adjustment parameter in the control law parameters needs to be adjusted by one step size in the direction of increasing its absolute value so that the manual control ratio can be reduced more quickly when the comprehensive state index decreases. If the average error of the low-state interval is less than or equal to the threshold for judging excessive interval error, the state index adjustment parameter is not adjusted. The average error of the high-state interval is compared with the threshold for determining excessive interval error. If the average error of the high-state interval is greater than the threshold for determining excessive interval error, it is determined that the control law parameters are insufficient in allocating manual control power when the operator is in a good state. In this case, the trend adjustment parameter in the control law parameters needs to be adjusted by one step in the direction of increasing the proportion of manual control response, so that a higher proportion of manual control can be obtained when the operator is in a good state. If the average error of the high-state interval is less than or equal to the threshold for determining excessive interval error, the trend adjustment parameter is not adjusted. Based on the adjusted state index adjustment parameters and trend adjustment parameters, the control law parameters are updated to complete the feedback optimization of the control law parameters.

[0045] To illustrate that the above formula is not merely a functional description, the following is an example of a manned-unmanned collaborative control task with a sampling period of 1 second. Nine types of state data items are collected, including real-time heart rate, gaze duration, pupil diameter, EEG theta / β ratio, control input frequency, control input amplitude, control input regularity, task response time, and task execution accuracy. Continuous state feature vectors are obtained according to forward or reverse normalization rules, as shown in Table 1.

[0046] Table 1. Substitution of physical quantities and normalization calculations x1 Real-time heart rate / bpm 55-125 Reverse 92 0.471 x2 gaze duration / s 0.20-1.40 positive 0.95 0.625 x3 Pupil diameter / mm 2.5-6.0 Reverse 4.1 0.543 x4 EEG theta / β ratio 1.0-4.0 Reverse 2.3 0.567 x5 Control input frequency / times per minute 6-24 positive 18 0.667 x6 Control input amplitude / normalized stroke 0.20-1.00 positive 0.72 0.650 x7 Control input regularity / s 0.05-0.80 Reverse 0.41 0.520 x8 Task response time / s 0.40-3.00 Reverse 1.80 0.462 x9 Task execution accuracy 0.50-1.00 positive 0.82 0.640 Taking the feature weight set W=(0.12,0.10,0.10,0.10,0.12,0.10,0.10,0.14,0.12) as an example, the sum of the weights of each feature is 1. Substituting the normalized eigenvalues ​​in Table 1 into the weighted summation formula, the original comprehensive state index can be obtained. If the composite state index of the previous sampling period is 0.640, and the trend calculation time step is 1 second, then the trend of the composite state index is as follows: This indicates that the operator's condition is declining.

[0047] The control law parameters can be taken as state exponential adjustment parameters. Trend adjustment parameters And use the Sigmoid smoothing function. Substitute the composite state index and its changing trend into... ,get Manual control ratio Automatic control ratio The calculation results indicate that the operator can still participate in control, but due to the decline in the state, the proportion of automatic control is increased accordingly.

[0048] If the manual control ratio in the previous sampling period was 0.720, and the currently calculated manual control ratio is 0.536, with a sampling interval of 1 second, then the current ratio change rate is -0.184. If the upper limit of the ratio change rate is set to 0.080, then a smoothing constraint is applied, resulting in a smoothed manual control ratio of 0.720 - 0.080 × 1 = 0.640 and a smoothed automatic control ratio of 0.360. If the current manual control signal is a joystick angular velocity command of 0.35 rad / s and the automatic control signal is a path tracking angular velocity command of 0.12 rad / s, then the fused control signal is 0.640 × 0.35 + 0.360 × 0.12 = 0.267 rad / s. This output retains the operator's input while avoiding a sudden drop in control from 0.720 to 0.536, which could cause abrupt changes in the actuator's commands.

[0049] When the signal deviation between manual control signals and automatic control signals exceeds the signal conflict threshold, the system further reads the comprehensive status index for arbitration. For example, if the signal conflict threshold is 0.20 rad / s and the arbitration status threshold is 0.55, and the signal deviation is |0.35-0.12|=0.23 rad / s and the comprehensive status index is 0.569, then the operator's status is determined to be within the reliable range, and manual control signals can be used preferentially or the weight of manual control signals in the fusion can be increased. If the comprehensive status index is lower than the safety takeover threshold of 0.50 for 3 consecutive seconds, the automatic control dominant mode is triggered, and the proportion of manual control is gradually reduced at the upper limit of the change rate of 0.080 / s until the minimum manual control proportion is reached.

[0050] Table 2. Examples of Comparison of Experimental Results of Cooperative Control Fixed human dominance 0.84 0.00 18 86.7% Discrete three-level switching 0.62 0.35 11 90.0% This application is dynamically adjusted. 0.39 0.08 5 96.7% Table 2 shows examples of the results of a 30-minute comparative experiment conducted under the same task path, the same unmanned platform dynamics model, and the same operator state disturbance conditions. Compared with the fixed human-dominated approach, this application reduces the weight of untrusted human input in real time through the state index, thereby reducing the root mean square error of trajectory tracking from 0.84 m to 0.39 m. Compared with the discrete three-level switching approach, this application limits the maximum control proportional jump variable to 0.08 by increasing the upper limit of the proportional change rate, avoiding the 0.35 level jump in discrete switching, thus reducing the number of sudden changes in actuator commands and conflict triggers. The above experimental data demonstrate that the technical solution of this application can improve the stability, safety, and task completion effect of collaborative control when the operator's state fluctuates.

[0051] During the feedback optimization process, the preset evaluation period can be set to 10 minutes, the state index segmentation threshold can be set to 0.60, and the threshold for judging excessive interval error can be set to 0.50 m. If the average error of the low state interval is 0.58 m, the absolute value of the state index adjustment parameter is increased so that the manual control ratio decreases faster when the comprehensive state index decreases. If the average error of the high state interval is 0.31 m, the trend adjustment parameter is not adjusted. Through the above parameter update based on interval error feedback, the control law parameters can be optimized in a closed loop according to the actual operating effect, rather than relying solely on preset experience values.

[0052] This embodiment collects operator physiological state data, operational behavior data, and task response data to construct a continuous state feature vector containing multiple state data items. A weighted fusion calculation is then used to generate a comprehensive operator state index. This enables the system to accurately assess the operator's overall control capability level in real time, providing a reliable quantitative basis for dynamic adjustment of control authority. This effectively overcomes the deficiency of failing to adjust control allocation promptly based on operator state changes due to a lack of comprehensive assessment of the operator's real-time state. By performing a moving average smoothing process on the comprehensive state index and calculating its changing trend, the system can filter out transient fluctuations when assessing the operator's state. The system can capture the direction and rate of state changes during dynamic disturbances. By constructing a continuous monotonic function based on the comprehensive state index and its changing trend, it generates a continuously changing manual control ratio over time, thereby achieving continuous dynamic adjustment of the manual and automatic control ratios. This effectively overcomes the shortcomings of inflexible control adjustment and inability to make fine-grained continuous adjustments based on operator state changes caused by fixed allocation of control rights or discrete level switching. By setting an upper limit on the rate of change of the ratio, the rate of change of the manual control ratio is constrained, ensuring a smooth and continuous transition between manual and automatic control, effectively avoiding system control jitter and operational discomfort caused by sudden changes in control rights. This embodiment generates a fused control signal by weighting and fusing manual and automatic control signals according to a smoothed manual control ratio. Combined with signal conflict detection and priority arbitration mechanisms, the system can automatically select the priority signal based on the operator's current credibility when conflicts arise between manual and automatic control signals. Simultaneously, a safety-enforced control logic forcibly switches to automatic control when a risky state is detected to ensure system safety. This effectively overcomes the shortcomings of lacking an effective continuous fusion mechanism between manual and automatic control signals and the ease with which control conflicts can occur during control switching. Furthermore, a takeover management mechanism is established, automatically triggering a takeover when the operator's overall status index remains below a safety takeover threshold. The system adopts an automatic control-dominant mode and gradually increases the automatic control ratio according to the rate constraint. After the operator's condition returns to a safe level, the manual control ratio is gradually restored, realizing dynamic takeover and smooth recovery of control. Through a feedback optimization mechanism, the trajectory tracking error and task completion deviation are integrated into a comprehensive control error. Based on the average error level within different comprehensive state index intervals, the state index adjustment parameters and trend adjustment parameters in the control law parameters are specifically adjusted to form a closed-loop collaborative control structure. This enables the system to continuously improve the adaptability of the control law parameters during long-term operation, ultimately achieving stable, reliable, and adaptive dynamic adjustment of manned-unmanned collaborative control based on operator condition assessment.

[0053] Example 2

[0054] Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A method for dynamically adjusting collaborative control rights based on operator status assessment is provided. The method includes: Collect multi-source state data from operators, normalize and construct features from the multi-source state data, and generate continuous state feature vectors; The continuous state feature vectors are weighted and fused to generate the operator's comprehensive state index, and the changing trend of the comprehensive state index is calculated. Preset control law parameters, and combine the comprehensive state index with the corresponding trend of change to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio. Based on the continuously changing artificial control ratio over time, the rate of change of the artificial control ratio is calculated, and a smoothed artificial control ratio is obtained through smoothing constraint processing. The system acquires manual and automatic control signals, performs weighted fusion based on the smoothed manual control ratio, generates a fused control signal for collaborative control, and determines whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection. The comprehensive status index is compared with the preset safety threshold, and the automatic control dominant mode or manual control recovery mode is triggered based on the comparison result. Evaluate the collaborative control error index during the collaborative control process, and optimize and adjust the control law parameters based on the collaborative control error index.

[0055] Example 3

[0056] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code that, when executed by the one or more processors, can perform the collaborative control dynamic adjustment method based on operator state assessment as described above.

[0057] The method or system according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as ROM or hard disk, may store the collaborative control dynamic adjustment method based on operator state assessment provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components in the electronic device shown in this application may be omitted according to actual needs.

[0058] Example 4

[0059] One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the cooperative control dynamic adjustment method based on operator state assessment according to an embodiment of this application, as described with reference to the above figures, can be performed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0060] Furthermore, according to embodiments of this application, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a method for dynamic adjustment of cooperative control rights based on operator state assessment. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0061] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0062] The normalization, weighted summation, state feedback mapping, rate limiting, and weighted fusion formulas in this specification are derived from mature mathematical models in min-max normalization, multi-attribute comprehensive evaluation, state feedback / adaptive control, rate limiting, and shared control, respectively. This application does not merely abstractly limit functionality using formulas, but rather applies the aforementioned models to a manned-unmanned collaborative control scenario, using collectable physical quantities such as heart rate, eye movement, EEG, control input, task response time, task accuracy, control signal amplitude, trajectory error, and task completion deviation as inputs or verification quantities. Each preset threshold and parameter can be determined through offline calibration, initial values ​​based on expert experience, and online feedback optimization. In specific implementations, the comprehensive state index, manual control ratio, smoothing control ratio, and fused control signal can be obtained through the substitution calculation process according to the embodiments.

[0063] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for dynamic adjustment of collaborative control rights based on operator status assessment, characterized in that, include: Collect multi-source state data from operators, normalize and construct features from the multi-source state data, and generate continuous state feature vectors; The continuous state feature vectors are weighted and fused to generate the operator's comprehensive state index, and the changing trend of the comprehensive state index is calculated. Preset control law parameters, and combine the comprehensive state index with the corresponding trend of change to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio. Based on the continuously changing artificial control ratio over time, the rate of change of the artificial control ratio is calculated, and a smoothed artificial control ratio is obtained through smoothing constraint processing. The system acquires manual and automatic control signals, performs weighted fusion based on the smoothed manual control ratio, generates a fused control signal for collaborative control, and determines whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection. The comprehensive status index is compared with the preset safety threshold, and the automatic control dominant mode or manual control recovery mode is triggered based on the comparison result. Evaluate the collaborative control error index during the collaborative control process, and optimize and adjust the control law parameters based on the collaborative control error index.

2. The method for dynamic adjustment of collaborative control rights based on operator state assessment according to claim 1, characterized in that, Methods for generating continuous state feature vectors include: Acquire operator physiological state data, operational behavior data, and task response data; integrate the operator's physiological state data, operational behavior data, and task response data to form multi-source state data of the operator; wherein, the multi-source state data contains multiple state data items; perform normalization processing on each state data item in the multi-source state data to obtain the normalized feature value of each state data item; arrange the normalized feature values ​​corresponding to all state data items in sequence to form a continuous state feature vector. The expression for normalization is: ； In the formula, For the first At the sampling time, the first The normalized eigenvalues ​​of each state data item. For the first The first sampling time before normalization Each status data item For the first The minimum value of each state data item. For the first The maximum value of each state data item.

3. The method for dynamic adjustment of collaborative control rights based on operator state assessment according to claim 2, characterized in that, Methods for generating operator comprehensive state index include: A preset feature weight set is provided, which contains the feature weights corresponding to each normalized feature value in the continuous state feature vector. Based on the weights of each feature, the normalized eigenvalues ​​in the continuous state feature vector are weighted and summed to obtain the original comprehensive state index. The expression for the original comprehensive state index is: ； In the formula, For the first The original comprehensive state index at each sampling time point For the first Feature weights of normalized eigenvalues, This represents the total number of normalized eigenvalues ​​in the continuous state eigenvector. The original composite state index is subjected to interval mapping to obtain the composite state index, and then the composite state index is subjected to moving average smoothing.

4. The method for dynamic adjustment of collaborative control rights based on operator status assessment according to claim 3, characterized in that, Methods for generating artificially controlled proportions that change continuously over time include: The control law parameters are preset, including state index adjustment parameters and trend adjustment parameters. Based on the comprehensive state index, the trend of the comprehensive state index, and the control law parameters, the manual control ratio is obtained by mapping through a preset continuous monotonic function. The expression for a continuous monotonic function is: ; In the formula, For the first The proportion of manual control at each sampling time point For smoothing functions, This is the state index adjustment parameter. For trend adjustment parameters, For the first The comprehensive state index at each sampling time point For the first The trend of the comprehensive state index at each sampling time.

5. The method for dynamic adjustment of collaborative control rights based on operator status assessment according to claim 4, characterized in that, Methods for obtaining smoothed artificial control ratios through smoothing constraint processing include: The rate of change of the manually controlled ratio is marked as the current rate of change; a preset upper limit for the rate of change is set, and the absolute value of the current rate of change is compared with the upper limit for the rate of change. If the absolute value of the current rate of change is less than or equal to the upper limit of the rate of change, then the smoothing constraint processing is not performed, and the manual control rate is directly used as the smoothed manual control rate; if the absolute value of the current rate of change is greater than the upper limit of the rate of change, then the smoothing constraint processing is required. The method for handling smoothing constraints is as follows: If the current rate of change of the scale is positive, then the expression for smoothing constraints is: In the formula, For the first The smoothing manual control ratio at each sampling time point This represents the upper limit of the proportional change rate; if the current proportional change rate is negative, the expression for smoothing constraint processing is: .

6. The method for dynamic adjustment of collaborative control rights based on operator status assessment according to claim 5, characterized in that, Methods for generating fusion control signals include: Acquire manual control signals and automatic control signals, and perform weighted fusion of the manual control signals and automatic control signals according to the smoothing manual control ratio to generate a fused control signal; The expression for the fusion control signal is as follows: ； In the formula, For the first The fusion control signal at each sampling time For the first The manual control signal at each sampling time. For the first Automatic control signals at each sampling time.

7. The method for dynamic adjustment of collaborative control rights based on operator state assessment according to claim 6, characterized in that, Methods for determining whether a priority arbitration mechanism or security enforcement control logic has been triggered include: The absolute value of the difference between the manual control signal and the automatic control signal is calculated to obtain the signal deviation, which is then compared with a preset signal conflict threshold. If the signal deviation is less than or equal to the signal conflict threshold, the priority arbitration mechanism is not triggered; if the signal deviation is greater than the signal conflict threshold, the priority arbitration mechanism is triggered. A set of predefined risk state triggering conditions is set up, which contains multiple predefined safety risk judgment conditions. The current system state of the collaborative control system is obtained, and it is determined whether the current system state meets any safety risk judgment condition in the risk state triggering condition set. If it does, the safety mandatory control logic is triggered; if it does not, the safety mandatory control logic is not triggered.

8. The method for dynamic adjustment of collaborative control rights based on operator state assessment according to claim 7, characterized in that, Methods for determining whether automatic control master mode or manual control recovery mode has been triggered include: Preset safety thresholds, including safety takeover threshold, takeover duration threshold, safety recovery threshold, and stability recovery duration threshold; The overall status index is compared with the safe takeover threshold; if the overall status index is less than the safe takeover threshold, the abnormal duration timer is started; if the overall status index is greater than or equal to the safe takeover threshold, the abnormal duration timer is reset to zero; when the cumulative duration recorded by the abnormal duration timer reaches or exceeds the takeover duration threshold, the automatic control master mode is triggered. In automatic control-dominated mode, the comprehensive status index is continuously monitored; when the cumulative duration of the comprehensive status index being higher than the recovery safety threshold reaches or exceeds the recovery stabilization duration threshold, the manual control recovery mode is triggered.

9. The method for dynamic adjustment of collaborative control rights based on operator state assessment according to claim 8, characterized in that, Methods for feedback optimization and adjustment of control law parameters include: The collaborative control error index is obtained, which specifically includes trajectory tracking error and task completion deviation. The trajectory tracking error and task completion deviation are comprehensively evaluated to obtain the comprehensive control error. An error optimization trigger threshold is preset, and the comprehensive control error is compared with the error optimization trigger threshold. If the comprehensive control error is less than or equal to the error optimization trigger threshold, the parameter optimization process is not triggered. If the comprehensive control error is greater than the error optimization trigger threshold, the parameter optimization process is triggered. The parameter optimization process is as follows: obtain all comprehensive state indices and corresponding comprehensive control errors within a preset evaluation period to form a historical collaborative control dataset; calculate the corresponding low-state interval average error and high-state interval average error based on the historical collaborative control dataset, and compare them with preset interval error large judgment thresholds; determine whether to adjust the state index adjustment parameters and trend adjustment parameters according to the preset parameter adjustment step size based on the comparison results; update the control law parameters based on the adjusted state index adjustment parameters and trend adjustment parameters.

10. A collaborative control dynamic adjustment system based on operator state assessment, implementing the collaborative control dynamic adjustment method based on operator state assessment as described in any one of claims 1-9, characterized in that, include: The status acquisition module is used to collect multi-source status data of the operator, normalize the multi-source status data and construct features to generate continuous status feature vectors. The index generation module is used to perform weighted fusion calculation on continuous state feature vectors to generate the operator's comprehensive state index and calculate the trend of the comprehensive state index. The control law calculation module is used to preset control law parameters, and combine the comprehensive state index with the corresponding trend of change to generate a manual control ratio that changes continuously over time, and determine the automatic control ratio. The proportional smoothing module is used to calculate the rate of change of the manual control ratio based on the continuously changing manual control ratio over time, and to obtain the smoothed manual control ratio through smoothing constraint processing. The fusion arbitration module is used to acquire manual control signals and automatic control signals, perform weighted fusion according to the smoothed manual control ratio, generate fused control signals for collaborative control, and determine whether to trigger the priority arbitration mechanism or safety mandatory control logic through signal conflict detection and risk status detection. The takeover management module is used to compare the comprehensive status index with the preset safety threshold and determine whether to trigger the automatic control master mode or the manual control recovery mode based on the comparison result. The feedback optimization module is used to evaluate the collaborative control error index during the collaborative control process and to optimize and adjust the control law parameters based on the collaborative control error index.

Images