Method, system, and computer program product for bias correction command simulation including human bias perturbations

Abstract

The application discloses a deviation correction command simulation method, system and computer program product containing human bias disturbance, belongs to the technical field of trajectory deviation correction, and comprises the following steps: constructing bias quantity, determining the center position and width coefficient of the bias grade membership function corresponding to each bias quantity; initializing a judgment disturbance signal generator; constructing a deviation correction instruction priority rule; determining a random distribution function of the deviation correction instruction release delay time; obtaining the cognitive bias quantity grade of the commander by using a river selection method; obtaining the current optimal deviation correction instruction; judging whether the current optimal deviation correction instruction changes, and if yes, a new deviation correction instruction is released; judging whether the simulation ends, and if not, the next simulation step is waited for, the simulation process data is re-accessed, otherwise the simulation ends, and the deviation correction command containing human bias disturbance is realized. The method can comprehensively simulate the deviation correction command process under human bias disturbance, and effectively reflects the cognitive bias and decision-making process of the commander in a complex environment.

Description

Technical Field

[0001] This invention relates to the field of trajectory correction technology, and in particular to a trajectory correction command simulation method, system, and computer program product that incorporates human-caused deviation disturbances. Background Technology

[0002] Corrective command refers to the use of corrective commands to alert and guide operators during the operation of various equipment, ensuring accuracy and safety. Concise and qualitative corrective commands are used in scenarios such as industrial product assembly, pipeline installation, and crane operation. These commands are typically simple, clear, and primarily qualitative, such as "too high," "adjust to the left," or "slow down." Current corrective command simulations lack effective representation of human error, particularly the possibility of errors in the operator's judgment. They fail to reflect the ambiguity in the operator's assessment of deviations and lack sufficient temporal continuity in the instructions, making it impossible to simulate potential misjudgments and different command styles. Summary of the Invention

[0003] The purpose of this application is to overcome the deficiencies of the prior art and provide a method, system and computer program product for corrective command simulation that includes human factor deviation disturbances.

[0004] In a first aspect, this application provides a corrective command simulation method incorporating human factor deviation disturbances, comprising the following steps:

[0005] Construct deviation quantities, and determine the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity;

[0006] Initialize the disturbance signal generator;

[0007] Construct priority rules for error correction instructions;

[0008] Determine the random distribution function for the delay time of the correction instruction issuance;

[0009] By accessing simulation process data, the level of commander's cognitive bias is obtained using the river selection method;

[0010] Obtain the commander's current optimal corrective instructions;

[0011] Determine whether the current optimal correction instruction has changed. If so, issue a new correction instruction and update the issuance delay time of the next commander's correction instruction based on the random distribution function of the correction instruction issuance delay time.

[0012] Determine if the simulation has ended. If not, wait for the next simulation step and reconnect the simulation process data. Otherwise, the simulation ends, thus realizing corrective command that includes human error disturbances.

[0013] Optionally, a deviation quantity is constructed, and the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity are determined, including:

[0014] Determine the amount of deviation that the commander needs to consider;

[0015] Determine the center position of the membership function of the deviation level corresponding to each deviation;

[0016] Determine the width coefficient of the membership function for the deviation level.

[0017] Optionally, the disturbance signal generator is initialized, including:

[0018] Set the filter convolution kernel width coefficient corresponding to each deviation;

[0019] Set the disturbance signal strength coefficient corresponding to each deviation;

[0020] Initialize the white noise signal sequence to obtain the initialized judgment disturbance signal generator.

[0021] Optionally, the random distribution function expression for the delay time of the correction instruction issuance is:

[0022]

[0023] in, This allows for a delay in issuing corrective instructions to the next commander. To delay the random number for the correction instruction, To minimize response latency, This represents the longest response delay time.

[0024] Optionally, by accessing simulation process data and using the river selection method, the level of commander cognitive bias can be obtained, including:

[0025] Access simulation process data;

[0026] The actual deviation level index of each deviation quantity is determined based on the center position of the deviation level membership function;

[0027] Calculate the deviation level membership degree for each deviation quantity based on the center position of the deviation level membership function and the width coefficient of the deviation level membership function;

[0028] Based on the river selection method, determine the proportion of river distribution for each deviation level;

[0029] The position of the current deviation level cognitive pointer is obtained by passing the white noise signal sequence through a convolution filter;

[0030] Update each white noise signal sequence, and select the deviation level index currently recognized by the commander from the river distribution based on the position of the cognitive pointer;

[0031] The commander's cognitive bias level is obtained based on the aforementioned commander cognitive bias level index.

[0032] Optionally, the actual deviation level index expression for each deviation is:

[0033]

[0034] in, For the first i An index of the actual deviation level for each deviation quantity; For the first i The magnitude of the deviation; Indicates the first i The first deviation n The center position of each deviation level.

[0035] Optionally, the membership expression for the deviation level of each deviation is:

[0036]

[0037] in, Indicates the first i Deviation for the first n Membership degree of each deviation level, For the first i The magnitude of the deviation Indicates the first i The first deviation n The center position of each deviation level For the first i The width coefficient of the membership function of the deviation level of each deviation quantity.

[0038] Optionally, the position expression for the current deviation level cognitive pointer is:

[0039]

[0040] in, For the first i Each deviation in the current simulation step The position of the cognitive indicator of the deviation level. For the first i The deviation corresponds to the filter convolution kernel strength coefficient of the disturbance signal generator. For the first i The first of the white noise signal sequences j One element, For the first iThe deviation amount corresponds to the filter convolution kernel width coefficient of the judgment disturbance signal generator. This is the simulation time step;

[0041] Commander's Cognitive Bias Level The expression is:

[0042]

[0043] in, Index of cognitive bias levels for commanders Indicates the first i The number of deviation levels for each deviation quantity.

[0044] Secondly, this application also provides a corrective command simulation system including human factor deviation disturbances, for executing the corrective command simulation method including human factor deviation disturbances as described in any of the first aspects, comprising: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the corrective command simulation method including human factor deviation disturbances as described in any of the first aspects.

[0045] Thirdly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the corrective command simulation method including human factor deviation disturbances as described in any one of the first aspects.

[0046] This application provides a method, system, and computer program product for simulating command correction under human bias disturbances. By combining the river selection method with a deviation level membership function, and comprehensively considering both the actual deviation level and the disturbance signal, it accurately obtains the level of cognitive bias of the commander. By setting a random distribution function for the delay time of the correction command issuance, it simulates the response delay in actual command, realistically recreating command scenarios with human bias disturbances. This method can comprehensively simulate the command correction process under human bias disturbances, effectively reflecting the commander's cognitive biases and decision-making processes in complex environments. It provides a scientific and accurate simulation tool for studying command decision-making behavior, and is of great significance for improving the reliability and practicality of command decision-making simulation.

[0047] To make the above-mentioned features and advantages of the invention more apparent and understandable, specific embodiments are described below, and detailed descriptions are provided in conjunction with the accompanying drawings. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart of a command simulation method for correcting human bias disturbances provided in one embodiment of this application.

[0050] Figure 2 This is a flowchart of step S1 in a command simulation method for correcting human bias disturbances provided in one embodiment of this application.

[0051] Figure 3 This is a flowchart of step S2 in a command simulation method for correcting human bias disturbances provided in one embodiment of this application.

[0052] Figure 4 This is a flowchart of step S5 in a corrective command simulation method that includes human factor deviation disturbances, provided in one embodiment of this application.

[0053] Figure 5 This is a membership distribution diagram of the seven deviation levels corresponding to the height deviation in the correction command simulation method containing human factor deviation disturbance provided in another embodiment of this application.

[0054] Figure 6 This is a schematic diagram showing the location of the height deviation level cognitive pointer in the river distribution in a correction command simulation method that includes human factor deviation disturbances provided in another embodiment of this application.

[0055] Figure 7 This is a schematic diagram showing the position of the lateral deviation level cognitive pointer in the river distribution in a deviation correction command simulation method that includes human factor deviation disturbances provided in another embodiment of this application.

[0056] Figure 8 This is a schematic diagram showing the river distribution and deviation command issuance status of the deviation level in a deviation correction command simulation method that includes human-caused deviation disturbances, provided in another embodiment of this application.

[0057] Figure 9 This is a schematic diagram of the deviation level, river distribution, and deviation command issuance in a deviation correction command simulation method that includes human-caused deviation disturbances, provided in another embodiment of this application. Detailed Implementation

[0058] To make the objectives and technical solutions of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the described embodiments of this application without creative effort are within the scope of protection of this application.

[0059] In one embodiment, see Figure 1 This application provides a corrective command simulation method that includes human error disturbances, which may include the following steps: steps S1 to S8.

[0060] Step S1: Construct the deviation quantity and determine the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity.

[0061] Step S2: Initialize the disturbance signal generator.

[0062] Step S3: Construct the priority rules for correction instructions.

[0063] Step S4: Determine the random distribution function of the correction instruction issuance delay time.

[0064] Step S5: Input simulation process data and use the river selection method to obtain the level of commander's cognitive bias.

[0065] Step S6: Obtain the current optimal correction instruction as perceived by the commander.

[0066] Step S7: Determine whether the current optimal correction instruction has changed. If so, issue a new correction instruction and update the issuance delay time of the next commander's correction instruction based on the random distribution function of the correction instruction issuance delay time.

[0067] Step S8: Determine whether the simulation has ended. If it has not ended, wait for the next simulation step and reconnect the simulation process data. Otherwise, the simulation ends, realizing the corrective command that includes human error disturbances.

[0068] This application's command simulation method for correcting human bias disturbances scientifically simulates the uncertainty of human cognitive bias by constructing a bias quantity and membership function, combined with an initialized judgment disturbance signal generator. It utilizes the river selection method to process simulation data, effectively integrating actual biases and disturbance factors to accurately obtain the level of commander cognitive bias, enabling random disturbance of discrete signal sequences in the continuous time domain. By constructing a priority rule for corrective instructions and setting a random distribution function for the release delay time, it standardizes the instruction decision-making logic and restores the response delay characteristics in real command scenarios. This method can realistically reproduce the command decision-making process under complex human influences, providing reliable simulation support for analyzing commander behavior patterns and verifying the effectiveness of corrective strategies, significantly improving the realism and practicality of command simulation systems.

[0069] In step S1, please refer to Figure 1 In step S1, construct the deviation quantity and determine the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity.

[0070] For example, please refer to Figure 2 Step S1 may include the following steps: Step S11 to Step S13.

[0071] Step S11: Determine the amount of deviation that the commander needs to consider.

[0072] Step S12: Determine the center position of the membership function of the deviation level corresponding to each deviation.

[0073] Step S13: Determine the width coefficient of the deviation level membership function.

[0074] As an example, in step S11, the number of deviations that the commander needs to consider is determined, and the number of deviations is denoted as K. Each deviation is defined with multiple deviation levels. The name of the deviation level is the corrective instruction that the commander can issue during the command process. The commander needs to issue the corresponding corrective instruction to the operator or driver based on the currently observed degree of deviation to remind them of the current deviation.

[0075] As an example, deviations may include: height deviation, lateral deviation, and diagonal deviation.

[0076] As an example, in step S12, set K Each deviation corresponds to K The center position of the group deviation level membership function is A clear standard deviation is established for each deviation level as a benchmark for judging the degree of membership of the actual deviation. This allows the actual deviation to be quantified by its distance from the center position to determine the degree of membership of each level, thereby enabling the modeling of the commander's fuzzy judgment logic.

[0077] As an example, if the first i Each deviation has N The deviation level is then the first i The center position of the membership function of each deviation level ,in, Indicates the first i The first deviation n The center position of each deviation level.

[0078] As an example, the first The deviation level is set to 0, and the... The deviation level is set to 1, and the... The deviation level is set to 2, and the first deviation level is... The deviation level is set to -1, and the... Each deviation level is set to -2, and so on, resulting in... K All deviation levels corresponding to each deviation quantity provide a computable mathematical benchmark for the abstract fuzzy level.

[0079] As an example, the number of deviation levels N It must be an odd number.

[0080] As an example, in step S13, set K Each deviation corresponds to K The width coefficient of the membership function of the group deviation level is .

[0081] As an example, the probability of a commander making a mistake in judging the deviation level can be controlled by adjusting the membership function width coefficient. i Membership function width coefficient for each deviation level The larger, the more i The wider the membership function curve for each deviation level, the closer the calculated membership degrees for each deviation level, and the greater the likelihood that the commander will misjudge the deviation level; conversely, the wider the membership function curve for the first deviation level, the more likely the commander is to misjudge the deviation level. i Membership function width coefficient for each deviation level The smaller, the first i The narrower the membership function curve for each deviation level, the farther apart the calculated membership degree for each deviation level, and the less likely the commander is to misjudge the deviation level, meaning the commander's judgment of the deviation level is more accurate.

[0082] As an example, the membership function can be a Gaussian membership function.

[0083] In step S2, please refer to Figure 1 In step S2, the disturbance signal generator is initialized.

[0084] For example, please refer to Figure 3 Step S2 may include the following steps: Step S21 to Step S23.

[0085] Step S21: Set the filter convolution kernel width coefficient corresponding to each deviation.

[0086] Step S22: Set the disturbance signal strength coefficient corresponding to each deviation.

[0087] Step S23: Initialize the white noise signal sequence to obtain the initialized judgment disturbance signal generator.

[0088] As an example, in step S21, set K Each deviation corresponds to K The width coefficient of each filter convolution kernel is To control the rate of change of the disturbance, wherein, To indicate the first i The deviation amount corresponds to the filter convolution kernel width coefficient of the judgment disturbance signal generator.

[0089] As an example, the larger the filter convolution kernel width coefficient, the wider the convolution kernel, and the smaller the rate at which the commander model changes its judgment of the deviation amount perturbation; conversely, the smaller the filter convolution kernel width coefficient, the more noise samples the convolution kernel contains, the lower the signal smoothness, the more high-frequency components are retained, and the faster the commander model changes its judgment of the deviation amount perturbation.

[0090] As an example, in step S22, set K Each deviation corresponds to K The intensity coefficient of each disturbance signal is To control the amplitude of the disturbance, among which, Indicates the first i The deviation corresponds to the filter convolution kernel strength coefficient of the judgment disturbance signal generator.

[0091] As an example, the larger the filter convolution kernel strength coefficient, the greater the degree of perturbation of the commander model's judgment deviation of the deviation amount; conversely, the smaller the filter convolution kernel strength coefficient, the smaller the degree of perturbation of the commander model's judgment deviation of the deviation amount.

[0092] As an example, in step S23, the white noise signal sequence is initialized as follows: ,in, Indicates the first i The white noise signal sequence corresponding to each deviation value for the judgment disturbance signal generator is expressed as follows: , h The length of the white noise sequence. For the first iThe white noise signal sequence corresponding to each deviation value is the judgment disturbance signal generator. The j Each element is used. Based on the filter convolution kernel width coefficient, the disturbance signal strength coefficient, and the initialized white noise signal sequence, an initialized judgment disturbance signal generator is obtained to generate a stationary random signal that conforms to the characteristics of human bias, which is used to simulate the commander's judgment disturbance of the deviation level.

[0093] As an example, a Gaussian random number generator with a mean of 0 and a variance of 1 can be used to sample and obtain a white noise signal sequence.

[0094] As an example, by adjusting the center position of the membership function, the width coefficient, and the spectral characteristics of the stationary random signal, the probability of the commander issuing error correction instructions and the frequency of changing correction instructions can be adjusted to reflect the command process of commanders with different levels and styles.

[0095] In step S3, please refer to Figure 1 In step S3, construct the priority rules for correction instructions.

[0096] As an example, we establish the correspondence between deviation levels and corrective commands for each deviation quantity. At any given time, a certain deviation quantity will have one and only one corresponding deviation level, and each deviation level will correspond to one corrective command.

[0097] As an example, the correspondence between deviation levels and corrective instructions can be represented in the form of Table 1, which quantifies abstract deviations into specific command instructions and establishes a one-to-one correspondence between the deviation level classification and the commander's command.

[0098] Table 1. Correspondence between Deviation Levels and Corrective Action Instructions

[0099]

[0100] Furthermore, the first can be set i The priority of the correction command corresponding to each deviation. The expression is:

[0101]

[0102] in, This refers to the current operation status information; Indicates the first i The priority calculation rules for each deviation include Each judgment condition and its corresponding numerical formula; For the first i The first deviation m Numerical formulas; For the first i The deviation level of each deviation quantity; Indicates the first i The first deviation m Each judgment condition has an output value of 0 or 1. For all judgment conditions of a deviation, there is exactly one judgment condition with a value of 1. t For time.

[0103] As an example, in the priority calculation rules, each deviation must have a default judgment condition. When other conditions are not met, the default judgment condition is true.

[0104] As an example, the priority calculation rules can be shown in Table 2.

[0105] Table 2 Priority Calculation Rules

[0106]

[0107] In step S4, please refer to Figure 1 In step S4, the random distribution function of the correction instruction issuance delay time is determined.

[0108] As an example, after recognizing a deviation, the commander will delay issuing a corrective order for a period of time. The length of the delay is related to the commander's skill level, and the delay time is not a fixed value, but a random distribution that follows a certain pattern.

[0109] As an example, a Beta random distribution can be used to simulate the delay in issuing corrective instructions by the commander.

[0110] Specifically, the two parameters of the Beta random distribution can be set as follows: α and β Based on the two parameters set α and β Initialize the Beta distribution random number generator to generate random numbers that conform to the probability distribution of human factors bias; set the minimum response delay time based on human factors experimental data, engineering constraints, and bias level characteristics. Longest response latency Sample a correction instruction delayed random number using an initialized Beta-distributed random number generator. random numbers The shape of the distribution is determined by two parameters of the Beta random distribution. α and β Control; delay random numbers according to correction instructions Shortest response delay time Longest response latency Determine the delay time for the next commander's corrective order. The expression for the random distribution function of the correction instruction issuance delay time is: .

[0111] In step S5, please refer to Figure 1 In step S5, the simulation process data is accessed, and the commander's cognitive bias level is obtained using the river selection method.

[0112] For example, please refer to Figure 4 Step S5 may include the following steps: Step S51 to Step S57.

[0113] Step S51: Input simulation process data.

[0114] Step S52: Determine the actual deviation level index for each deviation quantity based on the center position of the deviation level membership function.

[0115] Step S53: Calculate the deviation level membership degree for each deviation quantity based on the center position of the deviation level membership function and the width coefficient of the deviation level membership function.

[0116] Step S54: Determine the proportion of river distribution for each deviation level according to the river selection method.

[0117] Step S55: Pass the white noise signal sequence through a convolution filter to obtain the position of the current deviation level cognitive pointer.

[0118] Step S56: Update each white noise signal sequence and select the deviation level index currently recognized by the commander from the river distribution based on the position of the cognitive pointer.

[0119] Step S57: Obtain the commander's cognitive bias level based on the commander's cognitive bias level index.

[0120] As an example, in step S51, simulation process data is accessed, for each simulation step... Read and update once for each, ensuring that each Get the latest deviation data internally to avoid the correction command from failing due to data lag.

[0121] As an example, the simulation process data may include: the magnitude of the deviation, the rate of change of the deviation, and the control status information.

[0122] As an example, in step S52, the actual deviation level index of the deviation amount is calculated based on the center position of the deviation level membership function, and the expression is:

[0123]

[0124] in, For the first i The index of the actual deviation level of the deviation quantity, that is, the actual deviation level is the first deviation of the corresponding deviation quantity. The deviation level is an index value, not the level itself; For the first i The magnitude of the deviation; Indicates the first i The first deviation n The center position of each deviation level.

[0125] As an example, in step S53, the deviation level membership degree of each deviation quantity is calculated based on the center position of the deviation level membership function and the width coefficient of the deviation level membership function. i Deviation for the first n Membership of each deviation level The expression is:

[0126]

[0127] in, For the first i The magnitude of the deviation Indicates the first i The first deviation n The center position of each deviation level For the first i The width coefficient of the membership function of the deviation level of each deviation quantity.

[0128] As an example, in step S54, the proportion of river distribution for each deviation level is determined according to the river selection method.

[0129] As an example, the proportion of the river distribution may include: the proportion in the upper half of the river distribution and the proportion in the lower half of the river distribution.

[0130] As an example, for the first i There are several deviations. There are deviation grades that are greater than or equal to the actual deviation grade (including the actual deviation grade itself). If the deviation level is less than or equal to the actual deviation level (including the actual deviation level itself), then the... i The first deviation u The proportion of deviation levels greater than or equal to the actual deviation level in the upper half of the river distribution is: The expression is:

[0131]

[0132] in, j Indicates the first j A deviation level that is greater than or equal to the actual deviation level. For the first i Deviation for the firstn Membership degree of each deviation level, For the first i An index of the actual deviation level for each deviation quantity.

[0133] As an example, the first i The first deviation v The proportion of deviation levels less than or equal to the actual deviation level in the lower half of the river distribution is: The expression is:

[0134]

[0135] in, j Indicates the first j A deviation level that is greater than or equal to the actual deviation level. For the first i Deviation for the first n Membership degree of each deviation level, For the first i An index of the actual deviation level for each deviation quantity.

[0136] As an example, the river distribution map, which determines the deviation level corresponding to each deviation based on the proportion of river distribution, can represent the decision intervals obtained at each time step according to the membership functions of different deviation levels with different colored intervals. The decision intervals can be obtained by normalizing the membership values ​​of each deviation level using the SoftMax function. The distribution of the decision intervals changes with the actual deviation level; the closest deviation level is in the middle, the deviation level further upwards is located above, and the deviation level further downwards is located below. The width of the decision interval is positively correlated with the membership value. The upper bound of all decision intervals is 1, and the lower bound is -1.

[0137] As an example, in step S55, the white noise signal sequence is passed through a convolution filter to obtain the position of the current deviation level cognitive pointer.

[0138] As an example, the commander on the first i Each deviation in the current simulation step The position of the deviation level cognitive pointer is The expression is:

[0139]

[0140] in, For the first i The deviation corresponds to the filter convolution kernel strength coefficient of the disturbance signal generator. For the first i The first of the white noise signal sequences j One element, For the first i The deviation amount corresponds to the filter convolution kernel width coefficient of the judgment disturbance signal generator. This is the simulation time step.

[0141] As an example, the position of the cognitive pointer is determined by both white noise perturbation and Beta distribution membership, to reflect the randomness of human bias.

[0142] As an example, in step S56, each white noise signal sequence is updated.

[0143] As an example, each white noise signal sequence can be removed. The first element In white noise signal sequence Insert a new random number at the end.

[0144] Furthermore, based on the location of the cognitive pointer, the deviation level index currently recognized by the commander is selected from the river distribution. The expression is:

[0145]

[0146] in, Indicates the first i The first deviation j The proportion of deviation levels less than or equal to the actual deviation level in the lower half of the river distribution. Indicates the first i The first deviation j The percentage of deviation levels less than or equal to the actual deviation level in the lower half of the river distribution. Indicates the first i The number of deviation grades corresponding to each deviation quantity that are greater than or equal to the actual deviation grade is 1. Indicates the first i The number of deviation grades corresponding to each deviation quantity that are less than or equal to the actual deviation grade. Indicates the commander's opinion on the first i Each deviation in the current simulation step The position of the cognitive indicator of the deviation level. v Indicates the first v A deviation level that is less than or equal to the actual deviation level. u Indicates the first u A deviation level that is greater than or equal to the actual deviation level.

[0147] As an example, in step S57, based on the commander's cognitive bias level index... Obtain the commander's cognitive bias level The expression is:

[0148]

[0149] in, Indicates the first i The number of deviation levels for each deviation quantity.

[0150] In step S6, please refer to Figure 1 In step S6, the current optimal correction instruction in the commander's perception is obtained.

[0151] As an example, based on the various deviation levels perceived by the commander, the priority of corrective instructions corresponding to each deviation quantity in the commander's perception is calculated. The expression is:

[0152]

[0153] in, This refers to the current operation status information; Indicates the first i The priority calculation rules for each deviation include Each judgment condition and its corresponding numerical formula; For the first i The first deviation m Numerical formulas; For the first i The commander's cognitive bias level for each deviation quantity; Indicates the first i The first deviation m Each judgment condition has an output value of 0 or 1. For all judgment conditions of a deviation, there is exactly one judgment condition with a value of 1. t For time.

[0154] Furthermore, the deviation with the highest priority of the current correction instruction is selected, and the optimal correction instruction is obtained from the table of correspondence between deviation level and correction instruction (Table 1) based on the deviation level of the deviation.

[0155] In step S7, please refer to Figure 1 In step S7, it is determined whether the current optimal correction instruction has changed. If so, a new correction instruction is issued, and the issuance delay time of the next commander's correction instruction is updated according to the random distribution function of the correction instruction issuance delay time.

[0156] As an example, each corrective instruction issued by the commander is the currently optimal corrective instruction in the commander's perception. When the currently optimal corrective instruction changes, the commander issues a new commander instruction. Specifically, if the optimal corrective instruction for the next simulation step remains unchanged, the time point of the optimal corrective instruction is accumulated according to the simulation step size, ensuring that the correction instruction issuance time is continuously accumulated on the timeline, avoiding jumps in time calculation due to unchanged correction instructions. If the duration of the optimal corrective instruction is greater than or equal to the issuance delay time of the next corrective instruction, the corrective instruction is issued in the corresponding simulation step, and a new issuance delay time for the next corrective instruction is recalculated. That is, when the currently optimal corrective instruction has been maintained for TD seconds and the currently optimal corrective instruction is different from the most recently issued corrective instruction, the commander issues a new corrective instruction, and a new corrective instruction delay random number is sampled again using a Beta distribution random number generator. And calculate the delay time for the issuance of the next corrective command from the commander. The expression is: .

[0157] As an example, each time a commander issues a correction order, a new correction order delay random number must be sampled again using a Beta-distributed random number generator, and a new delay time for the next commander's correction order must be calculated.

[0158] In step S8, please refer to Figure 1 In step S8, it is determined whether the simulation has ended. If it has not ended, it waits for the next simulation step and reconnects the simulation process data. Otherwise, the simulation ends, realizing the corrective command that includes human error disturbances.

[0159] As an example, determine whether the simulation has ended. If it has not ended, wait for the next simulation step and return to step S5 to access the simulation process data. Otherwise, the simulation ends, realizing the corrective command that includes human error disturbances.

[0160] As an example, to reflect the commander's judgment bias, the commander's corrective instructions can be set to have a certain probability of being at a deviation level other than the currently closest deviation level, i.e., generating a misperception. This misperception must satisfy the following conditions: the probability of issuing a corrective instruction for a misperception is positively correlated with the membership degree corresponding to the corrective instruction; the probability of issuing a corrective instruction for a misperception is positively correlated with the difference between the deviation level corresponding to the corrective instruction and the current actual deviation level. For example, if the currently closest deviation level is "high," then even without considering the influence of membership degree, the probability of a misperception being "very high" should be greater than "somewhat high"; the corrective instructions must be relatively continuous in the time sequence. That is, if the commander's corrective instruction is "somewhat high" in the current time step, then multiple adjacent time steps should maintain this corrective instruction continuously, rather than randomly jumping to a new corrective instruction in each current time step.

[0161] In one example, for a pipeline assembly and docking operation, the commander determines the docking deviation based on the position of the pipeline to be docked during the docking process and issues corresponding correction commands in real time. The simulation time step can be set to 0.02s, and the deviations that the commander needs to consider are height deviation and lateral deviation, each of which includes seven deviation levels.

[0162] As an example, height deviation can include seven deviation levels: very high, a little high, high, OK (no height deviation or very small deviation), low, a little low, and very low; lateral deviation can include seven deviation levels: very right, a little right, right, OK (no lateral deviation or very small lateral deviation), left, a little left, and very left.

[0163] Furthermore, assuming that the pipeline is in a uniform translational state during the pipeline assembly and docking process, the initial height deviation can be set to 1.4m, and after 10s the height deviation becomes -1.4m; the initial lateral deviation can be set to -0.2m, and after 10s the lateral deviation becomes -0.5m. Using the method of this application to simulate the pipeline assembly and docking process, the commander issues correction instructions according to the on-site situation.

[0164] Specifically, the deviations that the commander needs to consider can be set to include: altitude deviation and lateral deviation, i.e., the number of deviations is 2; the center position of the membership function of the two sets of deviation levels corresponding to the two deviations can be set as follows. Each deviation includes seven deviation levels.

[0165] As an example, the center position of the membership function of the deviation level corresponding to the height deviation can be:

[0166]

[0167] The unit is meters.

[0168] As an example, the center position of the membership function of the deviation level corresponding to the height deviation can be:

[0169]

[0170] The unit is meters.

[0171] Furthermore, the width coefficient of the membership function for the two deviation levels corresponding to the height deviation and the lateral deviation can be set as follows: Set the width coefficients of the two filter convolution kernels corresponding to the height deviation and lateral deviation to . Set the strength coefficients of the two disturbance signals corresponding to the height deviation and lateral deviation as follows: The initial white noise signal sequence can be set to... ,in, and These are floating-point number queues of length 256, where each element is sampled from a Gaussian random number generator with a mean of 0 and a variance of 1.

[0172] Furthermore, the correspondence between the deviation levels of height deviation and lateral deviation and the correction commands can be constructed as shown in Table 3. The membership distribution of the seven deviation levels corresponding to height deviation is as follows: Figure 5 As shown, from Figure 5 It can be seen that the farther away from the center of the level, the lower the degree of belonging to the level.

[0173] Table 3. Correspondence between deviation levels and correction instructions for height and lateral deviations.

[0174]

[0175] Furthermore, priority calculation rules for height deviation and lateral deviation can be set as shown in Table 4.

[0176] Table 2 Priority Calculation Rules

[0177]

[0178] Furthermore, the two parameters of the Beta random distribution can be set as follows: and Set the minimum response delay time. Longest response latency A correction instruction delay random number is sampled using an initialized Beta-distributed random number generator. This allows us to determine the delay time for the next corrective order from the commander. .

[0179] Furthermore, the simulation process data accessed in the current frame may include: a height deviation of 1.4m and a height deviation change rate of -0.28m / s, a lateral deviation of -0.2m and a lateral deviation change rate of -0.3m / s.

[0180] Furthermore, the actual deviation level index for height deviation is determined based on the center position of the deviation level membership function. The actual deviation level index for lateral deviation is Based on the center position of the membership function of the deviation level and the width coefficient of the membership function of the deviation level, calculate the membership degree of the seven deviation levels (-3, -2, -1, 0, 1, 2, 3) of the height deviation. Membership of the seven deviation levels (-3, -2, -1, 0, 1, 2, 3) of the lateral deviation. It can be seen that there is a deviation level in the height deviation that is greater than or equal to the current actual deviation level. Therefore, according to the river selection method, the proportion of height deviation in the upper half of the river distribution is... If there are seven deviation levels in the height deviation that are less than or equal to the current actual deviation level, then according to the river selection method, the proportion of height deviation in the lower half of the river distribution is: If there are four deviation levels in the lateral deviation that are greater than or equal to the current actual deviation level, then according to the river selection method, the proportion of the height deviation in the upper half of the river distribution is: If there are four deviation levels in the lateral deviation that are less than or equal to the current actual deviation level, then according to the river selection method, the proportion of the height deviation in the lower half of the river distribution is: .

[0181] Furthermore, by passing the white noise signal sequence through a convolution filter, the position of the cognitive pointer for the high deviation level is obtained. The position of the lateral deviation level cognitive indicator is .

[0182] Furthermore, updating the white noise signal sequence can be achieved by deleting the white noise signal sequence. The first element in the sequence is sampled and a new random number -0.373431 is inserted into the white noise signal sequence. End of the sequence; remove white noise signal sequence. The first element in the sequence is sampled and a new random number 0.152265 is inserted into the white noise signal sequence. The end of.

[0183] Furthermore, based on the position of the cognitive pointer, the deviation level index currently recognized by the commander is selected from the river distribution. The commander's recognized altitude deviation level index is 7, and the position of the altitude deviation level cognitive pointer in the river distribution for the current simulation step is as follows: Figure 6As shown; if the commander's perceived lateral deviation level index is 3, then the position of the "perceived pointer" of the lateral deviation level in the river distribution for the current simulation step is as follows. Figure 7 As shown.

[0184] Furthermore, based on the deviation level index of the commander's cognition, the vertical deviation level of the commander's cognition is 3, and the horizontal deviation level of the commander's cognition is -1. Based on the vertical deviation level of the commander's cognition, the priority of the correction instruction corresponding to the vertical deviation in the commander's cognition is 2.7, and based on the horizontal deviation level of the commander's cognition, the priority of the correction instruction corresponding to the horizontal deviation in the commander's cognition is 0.4.

[0185] Furthermore, if we obtain the corrective command corresponding to a height deviation level of 3 from Table 3 as "Very High", then the commander's current optimal corrective command is "Very High".

[0186] Furthermore, the current optimal correction instruction has a duration of 0 seconds, and the delay time for the next correction instruction is 0.38 seconds. The current optimal correction instruction is not issued in the current simulation step, meaning it differs from the most recently issued correction instruction. If the optimal correction instruction for the next simulation step remains unchanged (e.g., both are "very high"), the remaining delay time for issuing the optimal correction instruction is accumulated progressively by the simulation step size. If the duration of the optimal correction instruction is greater than or equal to the delay time for issuing the next correction instruction, the correction instruction is issued in the corresponding simulation step, and a new delay time for issuing the next correction instruction is recalculated.

[0187] Furthermore, it determines whether the simulation has ended. If the simulation has not ended, it waits for the next simulation step and returns to step S5 to access the simulation process data; otherwise, the simulation ends, thus realizing corrective command that includes human error disturbances.

[0188] As an example, Figure 8 This diagram illustrates the river distribution and deviation command issuance status of the height deviation levels throughout the simulation process. Figure 9This diagram illustrates the river distribution and deviation command issuance status for lateral deviation levels. The black dashed line represents the no-deviation judgment line, which is always within the current closest membership deviation level judgment interval. The black solid curve represents the deviation judgment curve, which is a stationary random signal curve generated by the disturbance signal generator. This black solid curve fluctuates randomly over time but remains continuous without abrupt changes, and its mean remains 0 while its variance remains constant. The deviation level judgment interval where the deviation judgment curve is located represents the deviation level currently perceived by the commander. This deviation level has a certain probability of not conforming to the closest actual deviation level, constituting a misperception. A misperception must meet the following conditions: the probability of issuing a corrective order is positively correlated with the membership degree corresponding to the corrective order; the probability of issuing a corrective order is positively correlated with the difference between the deviation level corresponding to the corrective order and the current actual deviation level; and the corrective orders must be relatively continuous in time sequence. That is, if the commander's corrective order is "very high" in the current time step, then several adjacent time steps should maintain this corrective order, rather than randomly jumping to a new corrective order in each current time step. Figure 8 It can be seen that for altitude deviation, the actual deviation level changes should be: very high, somewhat high, high, OK, low, somewhat low, very low. However, some commanders, due to deviation disturbances, judged the deviation level changes as: very high, somewhat high, high, OK, OK, low, somewhat low. Among these, OK, low, and somewhat low are incorrect perceptions. From Figure 8 It can be seen that for lateral deviation, the actual deviation level change should be: OK, Left. However, the deviation level change judged by some commanders due to deviation disturbance is: Left, OK, Left, OK, Left. Among them, the first Left and the second OK are incorrect perceptions.

[0189] This application's command simulation method for correcting human bias disturbances transforms abstract command commands into a calculable mathematical form by constructing a deviation quantity and membership function, combined with the setting of center position and width coefficients. This allows the actual deviation level to be quantified by membership measure to accurately characterize the commander's fuzzy judgment logic. A judgment disturbance signal generator, combined with filter convolution kernel width and intensity coefficients and white noise sequences, simulates the randomness and continuity of human cognitive biases, ensuring that judgment results do not drastically change over time, conforming to the inertial characteristics of human decision-making. The river selection method is used to achieve random disturbance of discrete signal sequences in the continuous time domain. This application can systematically reproduce the command decision-making process under complex human disturbances, providing reliable simulation support for analyzing commander behavior patterns and verifying the effectiveness of correction strategies, significantly improving the realism and practicality of command simulation systems.

[0190] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0191] In another embodiment, this application also provides a command simulation system for correcting human bias disturbances, the command simulation system for correcting human bias disturbances comprising: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement any of the above-described command simulation methods for correcting human bias disturbances.

[0192] In another embodiment, this application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer is able to perform the various steps of the correction command simulation analysis method for human factor deviation disturbances provided in the above embodiments.

[0193] The computer-executable instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer-executable instructions may be provided to the processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the computer-executable instructions cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer-executable instructions may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a standalone software package, or entirely on a remote machine or electronic device.

[0194] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices may also be used to provide interaction with the user; feedback provided to the user can be any form of sensory feedback (e.g., visual feedback or haptic feedback); and input from the user can be received in any form, including: sound input, voice input, or haptic input.

[0195] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data electronic devices), or computing systems that include middleware components (e.g., application electronic devices), or computing systems that include front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0196] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0197] Although this application has been disclosed above with reference to embodiments, it is not intended to limit this application. Anyone skilled in the art may make some modifications and refinements without departing from the spirit and scope of this application. Therefore, the scope of protection of this application shall be determined by the appended claims.

Claims

1. A command simulation method for corrective actions incorporating human error perturbations, characterized in that, Includes the following steps: Construct deviation quantities, and determine the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity; Initialize the disturbance signal generator; Construct priority rules for error correction instructions; Determine the random distribution function for the delay time of the correction instruction issuance; By accessing simulation process data, the level of commander's cognitive bias is obtained using the river selection method; Obtain the commander's current optimal corrective instructions; Determine whether the current optimal correction instruction has changed. If so, issue a new correction instruction and update the issuance delay time of the next commander's correction instruction based on the random distribution function of the correction instruction issuance delay time. Determine if the simulation has ended. If not, wait for the next simulation step and reconnect the simulation process data. Otherwise, the simulation ends, thus realizing corrective command that includes human error disturbances. The random distribution function expression for the correction instruction issuance delay time is: in, This allows for a delay in issuing corrective instructions to the next commander. To delay the random number for the correction instruction, To minimize response latency, This is the longest response delay time; The process of accessing simulation data and obtaining the commander's cognitive deviation level using the river selection method includes: accessing simulation data; determining the actual deviation level index for each deviation based on the center position of the deviation level membership function; calculating the deviation level membership degree for each deviation based on the center position and width coefficient of the deviation level membership function; determining the proportion of river distribution for each deviation level using the river selection method; passing the white noise signal sequence through a convolution filter to obtain the position of the current deviation level cognitive pointer; updating each white noise signal sequence and selecting the current commander's cognitive deviation level index from the river distribution based on the position of the cognitive pointer; and obtaining the commander's cognitive deviation level based on the current commander's cognitive deviation level index. The actual deviation level index expression for each deviation is: ； The membership expression for the deviation level of each deviation is: in, For the first i An index of the actual deviation level for each deviation quantity. Indicates the first i Deviation for the first n Membership degree of each deviation level, For the first i The magnitude of the deviation Indicates the first i The first deviation n The center position of each deviation level For the first i The width coefficient of the membership function of the deviation level of each deviation quantity; The position expression for the current deviation level cognitive pointer is: in, For the first i Each deviation in the current simulation step The position of the cognitive indicator of the deviation level. For the first i The deviation corresponds to the filter convolution kernel strength coefficient of the disturbance signal generator. For the first i The first of the white noise signal sequences j One element, For the first i The deviation amount corresponds to the filter convolution kernel width coefficient of the judgment disturbance signal generator. This is the simulation time step; Commander's Cognitive Bias Level The expression is: in, Index of cognitive bias levels for commanders Indicates the first i The number of deviation levels for each deviation quantity.

2. The command simulation method for corrective actions including human error perturbation according to claim 1, characterized in that, Constructing deviation quantities, determining the center position and width coefficient of the deviation level membership function corresponding to each deviation quantity, including: Determine the amount of deviation that the commander needs to consider; Determine the center position of the membership function of the deviation level corresponding to each deviation; Determine the width coefficient of the membership function for the deviation level.

3. The corrective command simulation method including human factor deviation disturbances according to claim 1, characterized in that, Initialize the disturbance signal generator, including: Set the filter convolution kernel width coefficient corresponding to each deviation; Set the disturbance signal strength coefficient corresponding to each deviation; Initialize the white noise signal sequence to obtain the initialized judgment disturbance signal generator.

4. A command simulation system for correcting human error disturbances, characterized in that, The corrective command simulation system incorporating human bias disturbances includes: one or more processors; a storage device for storing one or more programs; and when the one or more programs are executed by the one or more processors, the one or more processors implement the corrective command simulation method incorporating human bias disturbances as described in any one of claims 1 to 3.

5. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the corrective command simulation method comprising human error perturbation as described in any one of claims 1 to 3.

Images