Method for determining man-machine interaction cognitive performance of sailors

Abstract

The invention discloses a crew man-machine interaction cognitive performance determination method, and relates to the field of crew man-machine interaction cognitive behavior modeling, and the method comprises the steps: a judgment and decision model loads a target task, forms a generation set, searches a generation corresponding to a maximum utility value for execution through the judgment and decision model, and updates a work memory model. The visual attention model is driven to operate, the eye movement preparation time and the eye movement glancing speed are obtained, and the visual attention performance is determined; the operation control model is driven to operate, the hardware environment of the display control equipment and the Fresnel's law parameters of the sailors are obtained, and the operation control performance is determined; updating the working memory model according to the execution result; updating the generative memory model according to the updated working memory model, and determining judgment and decision performance; summing the visual attention performance, the manipulation control performance and the judgment and decision performance, and determining the total cognitive performance of man-machine interaction of the crew. According to the method, the man-machine interaction cognitive performance of the crew can be accurately calculated in a conventional or long-voyage state.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of crew human-computer interaction cognitive behavior modeling, and particularly relates to a crew human-computer interaction cognitive performance determination method. BACKGROUND

[0002] The human-computer interaction process of a command system is a result of the joint action of the system, the crew, and the working environment. The cognitive performance (referring to the total completion time of the human-computer interaction task) and the cognitive load (referring to the total mental load in the cognitive operation process) of the crew in the human-computer interaction process of the command system directly affect the human-computer interaction efficiency, and further affect the generation of human-computer comprehensive combat effectiveness and the exertion of combat effectiveness. Researching and measuring the cognitive performance and the cognitive load in the human-computer interaction process of the crew is helpful to comprehensively understand the level of the human-computer interaction design of the system, find the design defects (such as some design schemes showing poor cognitive performance and high cognitive load) existing therein, and further improve the interaction design of the system and enhance the combat effectiveness.

[0003] The working state of the crew is divided into two types of “regular state” and “long voyage state”. The working task contents of the two types are consistent, but there are differences in the on-duty form, working location, and duration. The “regular state” refers to a human-computer interaction working state in which the crew is not continuously on duty for 24 hours, can take a long rest in the middle, can communicate with the outside, and can carry out daily training under the conditions of land and port. The “long voyage state” refers to another human-computer interaction working state in which the crew is continuously on duty for 24 hours, takes a rest in the middle according to the shift plan, works in a sealed cabin of the ship, cannot communicate with the outside, and carries out training and combat under the conditions of ship voyage. The approximate duration of the long voyage at sea is 7 to 90 days, for example, the longest voyage time of a Chinese nuclear submarine is 90 days, and there are occasionally cases less than 7 days. Unlike the regular state, in the long voyage state, the visual attention, judgment and decision-making, and control ability of the crew will change due to the influence of the duration of the voyage, the sealed cabin, and other factors, thereby causing changes in the cognitive performance of human-computer interaction. Therefore, in order to more accurately and comprehensively study the cognitive performance of crew human-computer interaction, the two working states of regular and long voyage should be fully considered.

[0004] Currently, the cognitive performance of crew's human-machine interaction is usually checked in the form of crew's human-in-the-loop experiment based on the ship command system task. However, there are four difficulties and problems in the human-in-the-loop experiment: first, the experimental test efficiency is low. It is difficult to carry out human-in-the-loop experiments for multiple crews at the same time due to the heavy training tasks of crews, the complex deployment process and other factors. Multiple experiments need to be carried out to ensure the test quality, resulting in long test time and low efficiency. Second, it is difficult to implement experiments under extreme or extreme task conditions. It is difficult to meet the experimental conditions and may cause damage to the health of the crew in the human-in-the-loop experiment under these conditions (such as 90-day long voyage). Third, there is a lack of efficient and accurate methods and means to replace the human-in-the-loop experiment of crew's human-machine interaction. Due to the special nature of crew's ability characteristics, human-machine interaction tasks and environment, and the field of routine and long voyage working state, the existing research foundation cannot fully and accurately replace the human-in-the-loop experiment, resulting in the problems of the first two points. Fourth, it is difficult to modify the problems in the experimental test. The human-in-the-loop experiment can only be carried out in the late stage of the development of the command system. Once the cognitive performance design problem is found in the test, the design and test need to be re-performed, resulting in high time and cost consumption.

[0005] Human-machine interaction cognitive behavior modeling and simulation (hereinafter referred to as "cognitive modeling") is an effective method to solve the problem of human-in-the-loop experiment. By modeling the behavior of crew's visual attention, judgment and decision-making, and control, a quantitative model and simulation method of crew's cognitive behavior is formed. The human-machine interaction process of the crew is constructed and simulated in the computer by combining the command system task, and the cognitive performance of the crew's human-machine interaction under different simulation task conditions is efficiently and accurately predicted. The human-in-the-loop experiment is replaced and the existing problems are solved. The design defects are found as much as possible in the system design stage and improved, and the cost-effectiveness of the development is reduced, which has important research significance.

[0006] At present, the research foundation of crew's cognitive modeling is relatively weak, and due to the special nature of the background of the ship field, the ability characteristics of the crew, the type of task, and the routine and long voyage working state, the existing cognitive model is used to model the cognitive behavior of the crew. There is a lack of quantitative modeling and simulation method of crew's human-machine interaction cognitive performance for routine and long voyage states.

[0007] To realize the quantitative calculation and simulation prediction of the cognitive performance of the crew, a crew human-machine interaction cognitive performance quantitative modeling and simulation method for the two states of the routine and the long voyage needs to be built. However, the existing research has the following three problems: first, the existing HIP (Human Information Processing), ACT-R (Adaptive Control of Thought Rational), MIDAS (Mental Imagery and Decision Analysis System) and other models lack quantitative modeling description of cognitive performance, which makes the existing ACT-R, MIDAS and other models lack of interpretability of the calculation results, and leads to great difficulty in quantitative modeling of the cognitive performance of the crew; second, due to the influence of the ship display and control device conditions, the isolation and airtightness and the long voyage working conditions, the parameters of the visual attention model and the control model of the crew are inaccurate in the routine state, which leads to inaccurate calculation of the cognitive performance in the routine state; the change rule model of the parameters of the visual attention model, the control model and the judgment and decision model in the long voyage state is also lacking, and due to unclear technical path, lack of experimental hardware conditions and scarcity of crew subjects, the research foundation of the quantitative modeling of the cognitive performance in the long voyage state is almost blank, which makes the cognitive performance in the long voyage state unable to be calculated, which is also the core challenge of the long voyage cognitive performance modeling; third, the existing research is insufficient in the task model, simulation architecture and implementation method for the simulation calculation of the cognitive performance, which makes it very difficult to realize the efficient and visual simulation calculation of the cognitive performance. SUMMARY

[0008] The purpose of the present application is to provide a crew human-machine interaction cognitive performance determination method, which can accurately calculate the crew human-machine interaction cognitive performance in the routine or long voyage state.

[0009] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0010] In a first aspect, the present application provides a crew human-machine interaction cognitive performance determination method, comprising:

[0011] The judgment and decision model loads the target task, sets the scheduling period, and subdivides the target task into a plurality of sub-targets; the judgment and decision model comprises a production container, a production rule interpreter and a judgment and decision model parameter; the production container comprises a plurality of productions; the judgment and decision model parameter comprises a scheduling period and a key cognitive ability index set affecting the scheduling period; and the target task is a human-machine interaction task in the routine state or the long voyage state.

[0012] The judgment and decision model analyzes all sub-goals in the task goal set to form a production set, the judgment and decision model performs the production corresponding to the maximum utility value based on the formed production set, and after the execution is completed, the judgment and decision model updates the working memory model; the working memory model includes visual attention working memory and manipulation control working memory; the working memory model is used to save the temporary information corresponding to visual attention and operation control;

[0013] The working memory model drives the visual attention model to operate using the visual attention working memory, and performs eye movement preparation and eye movement saccade action;

[0014] According to the eye movement preparation time and the eye movement saccade speed after the operation of the visual attention model is completed, the visual attention performance is determined;

[0015] The working memory model drives the manipulation control model to operate using the manipulation control working memory, and performs mouse movement, clicking and keystroke action;

[0016] According to the Fitts' law parameters of the display control device hardware environment and the Fitts' law parameters of the crew after the operation of the manipulation control model is completed, the manipulation control performance is determined;

[0017] The visual attention model and the manipulation control model return the visual attention execution result and the manipulation control execution result respectively, update the visual attention working memory and the manipulation control working memory, and obtain the updated working memory model;

[0018] According to the updated working memory model, the production memory model of the judgment and decision model is updated, and the current sub-goal execution is completed; the production memory model includes all sub-goals of the current target task and the current state value of the sub-goal;

[0019] The judgment and decision scheduling period of the current sub-goal is obtained, and the judgment and decision performance is determined;

[0020] Until the target task is executed, the visual attention performance, the manipulation control performance and the judgment and decision performance of all sub-goals are summed up, and the total cognitive performance of the crew human-computer interaction is determined.

[0021] According to the specific embodiments provided in the application, the application has the following technical effects:

[0022] The application provides a crew human-computer interaction cognitive performance determination method, a target task is loaded through a judgment and decision model, a task target set and a scheduling period are set; all sub-targets in the task target set are analyzed by the judgment and decision model to form a production set, the judgment and decision model finds a production corresponding to a maximum utility value for execution, and the judgment and decision model updates a working memory model; the working memory model drives a visual attention model to operate using visual attention working memory, and performs eye movement preparation and eye movement saccade actions; eye movement preparation time and eye movement saccade speed are obtained to determine visual attention performance; the working memory model drives a manipulation control model to operate using manipulation control working memory, and performs mouse movement, clicking and keystroke actions; a human-computer interaction device hardware environment and a crew's Fitts law parameters are obtained to determine manipulation control performance; the visual attention model and the manipulation control model return execution results, and the working memory model is updated according to the execution results; the production memory model is updated according to the updated working memory model, and the current sub-target execution is completed; the judgment and decision scheduling period of the current sub-target is obtained to determine judgment and decision performance; the target task is executed, the visual attention performance, the manipulation control performance and the judgment and decision performance of all sub-targets are summed to determine the total cognitive performance of the crew human-computer interaction. The application can accurately calculate the crew human-computer interaction cognitive performance in a conventional or long voyage state. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0024] Figure 1 A schematic diagram of a crew human-computer interaction cognitive behavior architecture;

[0025] Figure 2 A flowchart of a crew human-computer interaction cognitive performance determination method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0027] The above purposes, features and advantages of the present application can be more obvious and easy to understand. The present application will be further described in detail below with reference to the drawings and specific embodiments.

[0028] Human-computer interaction among crew members is the result of the interaction between humans (crew members) and machines (command and control system tasks) under specific environmental (work environment) conditions. This application proposes an Architecture of Commander Human-Computer Interaction (ACHCI), such as... Figure 1 As shown, Figure 1 The blue dashed box represents the research scope of crew member human-computer interaction cognitive behavior modeling. ACHCI describes the entire process of information flow in crew member human-computer interaction, from visual attention to judgment and decision-making, and finally to manipulation and control, while also supporting the description of the cognitive behavior mechanism of crew member human-computer interaction.

[0029] ACHCI generally comprises four parts, the first being the human-machine interaction task module of the ship's command and control system, see... Figure 1 The left side has a yellow background; the second is the crew module, see... Figure 1 The blue background on the right; the third is the environment module, see... Figure 1 The right side features a green background; fourthly, it represents the effectiveness of the crew's human-computer interaction process (including cognitive performance and cognitive load), see... Figure 1 The bottom section is located in the middle. The task module includes the ship's display and control equipment and corresponding command and control system software. Driven by example tasks, it is responsible for presenting human-computer interaction tasks to the crew module sequentially and responding to clicks, inputs, and other operations from the crew module. The crew module is responsible for processing the tasks presented by the task module. The environment module influences crew cognitive behavior, thereby affecting cognitive performance and cognitive load in the performance representation module.

[0030] The cognitive behavior mechanism of crew members' human-computer interaction can be described from three aspects: "cognitive behavior process mechanism," "cognitive behavior ability change mechanism," and "cognitive behavior efficacy representation mechanism." Specifically, the cognitive behavior process mechanism describes the organizational structure and operational principles of crew members' visual attention, judgment and decision-making, and manipulation and control cognitive behavior processes; the cognitive behavior ability change mechanism describes the impact of environmental modules on crew members' visual attention, judgment and decision-making, and manipulation and control cognitive behavior abilities; and the cognitive behavior efficacy representation mechanism describes the overall performance of crew members' cognitive behavior processes from the perspectives of cognitive performance and cognitive load.

[0031] This application proposes a Commandercognitive behavior model of human-computer interaction (CCBMHCI) based on cognitive behavior mechanisms to describe the cognitive behavior process of crew members in human-computer interaction and output cognitive performance and cognitive load.

[0032] The CCBM HCI is defined as a 3-tuple Ccbpm, Ccbacm, Ccberm. Wherein, Ccbpm (Commander cognitive behavioral process model) is a model for describing the cognitive behavioral process of the crew, Ccbacm (Commander cognitive behavioral ability change model) is a model for describing the change of the cognitive behavioral ability of the crew, Ccberm (Commander cognitive behavioral effectiveness representation model) is a model for representing the effectiveness of the cognitive behavior of the crew.

[0033] Ccbpm = {Vam, Jdm, Pmm, Wmm, Lmm, Mcm} is a model for describing the cognitive behavioral process of the crew's visual attention, judgment and decision, and manipulation control. Wherein, Vam (Visual attention model), Jdm (Judgment and decision model), Pmm (Production memory model), Wmm (Working memory model), Lmm (Long-term memory model), Mcm (Manipulation control model) represent the visual attention model, the judgment and decision model, the production memory model, the working memory model, the long-term memory model, and the manipulation control model, respectively.

[0034] Ccbacm = {Lvwsacm} is a model for describing the change of the cognitive behavioral ability of the crew, and Lvwsacm (Long-term voyage working status ability change model) represents a model for describing the change of the cognitive behavioral ability of the crew in the long-term voyage working status.

[0035] Ccberm = {Ccp, Ccl} is a model for representing the effectiveness of the human-machine interaction of the crew. Wherein, Ccp (Commander cognitive performance) and Ccl (Commander cognitive load) represent the cognitive performance model and the cognitive load model of the human-machine interaction of the crew, respectively.

[0036] The crew's human-computer interaction cognitive behavior process is generally divided into three stages: the first is the visual attention stage, which includes a visual processor for identifying specific objects in the human-computer interaction task; the second is the judgment and decision-making stage, which includes a judgment and decision-making processor, working memory, long-term memory, and production memory, etc., for judging and processing various types of information in the cognitive behavior process; and the third is the control and command stage, which includes a manipulation processor for executing the human-computer interaction task. The operation mechanism of the three stages will be described below.

[0037] The core work of the crew in the visual attention stage is to fixate and identify specific objects at specific locations on the screen. Therefore, the visual processor of ACHCI mainly processes the information of visual fixation, encodes the target image observed by the crew into specific content, and provides input to the judgment and decision-making processor. From the nature of the crew's visual fixation, the visual processor is more inclined to simulate the crew's visual attention theory rather than simply the visual perception theory. This is because the judgment and decision-making processor is more concerned about what the specific content of the target image provided by the visual processor is, rather than what the specific image representation theory and representation details are. The crew's visual attention runs through the entire human-computer interaction task execution process.

[0038] A visual attention model Vam = {Vpm, Vem, Vamp} is defined to describe the model of the crew's visual attention process, Vpm (Prepare model) and Vem (Execution model) represent the crew's eye movement preparation model and eye movement saccade execution model respectively, both of which return the execution time. Vamp (Visual attention model parameters) represents the crew's visual attention model parameters. Vamp = {Tprep, v}, Tprep and v represent the eye movement preparation time and eye movement saccade speed in the normal state respectively. The crew's visual model is called in the whole process of human-computer interaction.

[0039] The judgment and decision-making stage includes a judgment and decision-making processor, a production memory, a working memory, and a long-term memory.

[0040] (1) Judgment and decision-making processor.

[0041] The human-machine interaction task of the crewman is executed according to certain order and rules, and the main function of the judgment and decision processor is to schedule the human-machine interaction task. The judgment and decision processor contains a set of production rules, which, according to the order of the human-machine interaction task, specifies what operation is performed under what condition to complete the corresponding task, and the judgment and decision processor also contains a production rule interpreter. The production rule interpreter comprehensively judges the information in the working memory, and then selects the current production rule that meets the conditions for execution, and further issues instructions to the visual processor and the manipulation processor. In the whole process of human-machine interaction, the judgment and decision processor is periodically cycled, at the beginning of each cycle, the judgment and decision processor uses the output of the visual processor, the knowledge of the long-term memory and the working memory of the previous cycle to comprehensively judge and update the content of the working memory, at the end of each cycle, the content of the production memory is updated, and the command is sent to the visual processor and the manipulation processor to execute the corresponding action. The cycle of the judgment and decision processor is different from the external stimulation to the visual processor, and the output of the visual processor must wait for the current cycle of the judgment and decision processor to be completed.

[0042] A judgment and decision model Jdm={Pc, Psi, Jdmp} is defined for describing the model of the judgment and decision process of the crewman, which cyclically schedules the visual attention model information, the manipulation control model information, and the information of the working memory, the long-term memory, and the production memory in a certain time period. Pc (Product in container), Psi (Production system interpreter), and Jdmp (Judgment and decision model parameters) represent the production container, the production rule interpreter, and the judgment and decision model parameters respectively. Jdmp={Dat, Kca}, Dat (Default action time) and Kca (Key cognitive abilities) represent the time of a single judgment and decision in a regular state (hereinafter referred to as "scheduling period") and the set of key cognitive ability indicators that affect the scheduling period in a regular state respectively. The crewman judgment and decision model continuously runs in the whole process of human-machine interaction.

[0043] Pc is composed of a set of productions, that is, Pc={Pc1, Pc2,..., Pc N}. Psi represents the utility value of each production in Pc in the current judgment and decision cycle, that is, Psi={u pv1 , u pv2 ,..., u pvN}, because of the seriality of the production rule execution, only one production can be executed in one decision-making and scheduling cycle, and the production corresponding to max(Psi) is selected in the current cycle.

[0044] (2) Production memory.

[0045] The crew can make different responses to different target items depending on the understanding of the current target and the ability to maintain the target without any changes in the external environment. Suppose that two numbers must be added through a series of steps to obtain the answer, and the partial results, such as the sum of the tens, are tracked at all times while performing these steps.

[0046] Production memory is the "control library" of the decision-making processor, and it contains all the sub-goals for achieving the current goal and the current step in the program for achieving these sub-goals. The production rule interpreter realizes the reading and control of the visual processor, the manipulation processor, and the long-term memory by analyzing the goals and programs in the production memory, and the production memory plays a crucial role in ensuring the correctness of the human-machine interaction goals, sub-goals, and task steps.

[0047] A production memory model Pmm = {Pg, Pgv} is defined to describe the model of the crew's production memory, where Pg (Production goal) and Pgv (Production goal value) represent the sub-goals of the crew's human-machine interaction task and the current state value of the sub-goals, respectively.

[0048] Pg is composed of a set of task sub-goals, i.e., Pg = {g1, g2,..., g N}. Pgv saves the state value of each goal in Pg under the current scheduling cycle, i.e., Pgv = {gv1, gv2,..., gv N}.

[0049] (3) Working memory.

[0050] The temporary information corresponding to the crew's visual attention and manipulation control is saved by the working memory. The working memory includes visual working memory and manipulation working memory. The visual working memory and the manipulation working memory are the bridges for information communication between the visual processor, the manipulation processor, and the decision-making processor. The visual working memory and the manipulation working memory save the current information generated by the visual processor and the manipulation processor, as well as the decision-making and scheduling information of the decision-making processor for the visual processor and the manipulation processor. The information in the visual working memory and the manipulation working memory is updated in each cycle of the decision-making processor to provide decision-making and scheduling for the decision-making processor, ensuring that the human-machine interaction task can be completed correctly.

[0051] A working memory model Wmm = {Vwm, Mwm} is defined to describe the model of the crew's working memory. Vwm (Visual working memory) and Mwm (Manipulation working memory) represent the crew's visual attention working memory and manipulation control working memory respectively, and each of them stores the current information in a plurality of Vam and Mcm. Vwm and Mwm are the bridges for information communication between Jdm and Vam and Mcm, and Vwm and Mwm are updated once in each judgment and decision cycle. Jdm only processes the information related to the current task in Vwm and Mwm, rather than all the information, which is consistent with the cognitive behavior of the crew (for example, the crew can see a lot of information on the screen, but only process the information related to the current task).

[0052] (4) Long-term memory.

[0053] Although the production memory ensures the continuity of the process of human-computer interaction of the crew, the information stored in the production memory is temporary. For example, the crew needs to track the target with the batch number "1206" at present, and may forget after a period of time after the current task is completed. However, some information will become long-term memory information after a long time of learning, frequent use or professional training, such as the maximum speed of the ship is 45 knots per hour or 3+4=7. The long-term memory information remains consistent in the calculation.

[0054] A long-term memory model Lmm = {Lmi, Lmu, Lmcv} is defined to describe the model of the crew's long-term memory. Lmi (Long-term memory information), Lmu (Long-term memory utility) and Lmcv (Long-term memory current value) represent the set of long-term memory information, the set of long-term memory information utility and the query result returned to Jdm in the current judgment and decision cycle respectively.

[0055] Lmi is represented by a set of long-term memory information, that is: Lmi = {Lmi1, Lmi2,..., Lmi N}. Lmu stores the utility value corresponding to each information in Lmi, that is: Lmu = {Lmu1, Lmu2,..., Lmu N}.

[0056] The human-machine interface for crew members is controlled by a manipulation processor. The manipulation processor controls the hands based on commands from the judgment and decision processor, simulating mouse clicks or keystrokes. The manipulation processor is only invoked when a manipulation task is involved. The simulation of manipulation by the manipulation processor is based on motion characteristics, and the timing of motion generation depends on its characteristic structure and mechanical properties. Completing an action consists of a preparation phase and an execution phase. The preparation phase primarily involves the manipulation processor parsing the instructions from the judgment and decision processor to generate specific operational content. The execution phase mainly involves executing the action based on the operational content from the preparation phase. The manipulation processor can execute only one operation at a time, but it can prepare for the next action while executing the previous one, thus ensuring the continuity of operations.

[0057] A manipulation control model Mcm = {T} is defined. mou T key Mcmp} is a model used to describe crew maneuvering control. mou (Mouse time), T key (Keytime) represents the mouse movement / click model and the keyboard keystroke model, respectively, both of which return the execution time. key ={T str T poi}, T str (Keystroke time), T poi (Finger pointing time) represents the time it takes for the finger to type and the time it takes for the finger to point, respectively. str relative to T poi The movements are simple and take little time. poi It takes a long time and is related to T mou All parameters follow Fitts' Law. Mcmp (Manipulation control model parameters) represents the parameters of the crew maneuvering control model. The maneuvering model is only invoked when maneuvering tasks are involved.

[0058] In order to fully characterize the influence of long voyage working state on the cognitive behavior ability of the crew(including visual attention, judgment and decision, and manipulation control ability), a Commander Cognitive Behavior Ability Change Processor(CCBACP) is proposed in the ACHCI. The CCBACP is used to simulate the change law of the cognitive behavior ability of the crew in the long voyage working state, which changes the current situation that the traditional cognitive modeling does not consider the long voyage working state. The CCBACP cooperates with the visual processor, the judgment and decision processor, and the manipulation processor to simulate the change law of the visual attention ability, the judgment and decision ability, and the manipulation control ability of the crew in the long voyage state. The CCBACP also indirectly affects the cognitive performance and cognitive load of the crew, so that they can adapt to the long voyage working state of the ship.

[0059] A long voyage working state commander cognitive behavior ability change model Lvwsacm(Long-term voyage working status ability change model)={Lvaif, Lvapcm} is defined to describe the change model of the cognitive behavior ability of the crew in the long voyage working state. Lvaif(Long-term voyage ability influence factors) and Lvapcm(Long-term voyage ability parameters change model) represent the influence factors of the cognitive behavior ability in the long voyage state and the change model of the cognitive behavior ability parameters, respectively. Lvaif={Time}, Time represents the sailing time. Lvapcm={Lvamp, Ljdmp, Lmcmp}, Lvamp(Long visual attention model parameters), Ljdmp(Long judgment and decision model parameters), and Lmcmp(Long manipulation control model parameters) represent the change model of the visual attention ability, the judgment and decision ability, and the manipulation control ability parameters in the long voyage working state, respectively.

[0060] In terms of the visual attention ability of the crew, the visual attention model parameters Vam.Vamp.T prep and Vam.Vamp.v are combined with the change mechanism of the cognitive behavior ability of the crew and the visual attention model parameters Vam.Vamp.T prep and Vam.Vamp.v to represent the change model of the visual attention ability parameters of the crew in the long voyage state. Vam.Vamp.T prep and Vam.Vamp.v represent the change model of the visual attention ability parameters of the crew in the long voyage state.

[0061] In the aspect of the crew's steering control ability, the steering control model parameters Mcm, Mcmp, T are combined with the crew's cognitive behavior ability change mechanism a and Mcm, Mcmp, T b , the crew's steering control ability change model respectively represent the change model of the parameters Mcm, Mcmp, T a and Mcm, Mcmp, T b under the long voyage state.

[0062] In the aspect of the crew's judgment and decision ability, the judgment and decision model parameters Jdm, Jdmp are combined with the crew's cognitive behavior ability change mechanism, the judgment and decision ability change model f dat (t) represents the change rule model of Jdm, Jdmp, Dat under the long voyage state, represents the change model of the key cognitive ability Jdm, Jdmp, Kca under the long voyage state, respectively represent the change model of Kca1, Kca2,..., Kca N in the key cognitive ability Jdm, Jdmp, Kca under the long voyage state.

[0063] The human-computer interaction efficiency is an important index for evaluating the design level of equipment information system, and the high or low efficiency directly affects the human-computer comprehensive combat effectiveness. The crew's human-computer interaction efficiency refers to the overall efficiency and quality in the whole process of human-computer interaction. The human-computer interaction efficiency is represented by cognitive performance (CP) and cognitive load (CL). In the crew's ACHCI architecture, CP and CL represent the efficiency of human-computer interaction from the time dimension and the crew's load dimension of human-computer interaction respectively.

[0064] The representation of cognitive performance is researched in the application.

[0065] CP represents the total completion time of the whole human-machine interaction cognitive behavior process of the crew in macroscopic view, and represents the visual attention ability, judgment and decision ability, and manipulation control ability characteristics of the crew in microscopic view (the stronger the ability, the shorter the time), and the microscopic ability characteristics drive the macroscopic performance. CP is composed of visual attention performance, judgment and decision performance, and manipulation control performance, which respectively represent the visual attention time, judgment and decision time, and manipulation control time of the crew. Visual attention performance, judgment and decision performance, and manipulation control performance represent the efficiency of human-machine interaction from a local perspective, while CP represents the efficiency of human-machine interaction from a whole perspective. Visual attention performance, judgment and decision performance, and manipulation control performance are determined by visual attention ability, judgment and decision ability, and manipulation control ability, and all of them are affected by the long voyage working state.

[0066] A commander cognitive performance model Ccp (Commander Cognitive Performance) = {Rtcp, Lvtcp(t)} is defined, which is a model for describing the total cognitive performance of the human-machine interaction cognitive behavior of the crew. According to the cognitive behavior efficiency representation mechanism, performance represents the time consumed, that is, the total time for completing the task. Rtcp (Regular total cognitive performance) and Lvtcp(t) (Long-term voyage total cognitive performance) represent the total cognitive performance models in the regular and long voyage states, respectively.

[0067] From the perspective of cognitive performance calculation, combined with the crew visual attention model Vap, judgment and decision model Jdm, and manipulation control model Mcm, Ccp can also be represented as: Ccp = f tcp (Vap, Jdp, Mcp). Wherein, Vap (Visual attention performance), Jdp (Judgment and decision performance), Mcp (Manipulation control performance), f tcp represent visual attention performance (time), judgment and decision performance (time), manipulation control performance (time), and human-machine interaction task cognitive performance calculation model, respectively.

[0068] Vap = {Rvap, Lvvap(t)}, Rvap (Regular visual attention performance) and Lvvap(t) (Long-term voyage visual attention performance) represent the visual attention performance in the regular and long voyage states, respectively.

[0069] Jdp = {Rjdp, Lvjdp(t)}, Rjdp (Regular judgement and decision performance), Lvjdp(t) (Long-term voyage judgement and decision performance) represent judgement and decision performance in regular and long voyage state respectively.

[0070] Mcp = {Rmcp, Lvmcp(t)}, Rmcp (Regular manipulation control performance), Lvmcp(t) (Long-term voyage manipulation control performance) represent manipulation control performance in regular and long voyage state respectively.

[0071] Rtcp = f tcp (Rvap, Rjdp, Rmcp), since the crew's human-machine interaction cognitive behavior process is visual attention, judgement and decision, manipulation control in parallel, Rtcp and Rvap, Rjdp, Rmcp three are not additive relationship, but by f tcp Overall calculation. Similarly, Lvtcp(t) = f tcp (Lvvap(t), Lvjdp(t), Lvmcp(t) ).

[0072] In an exemplary embodiment, as Figure 2 shown, a crew human-machine interaction cognitive performance determination method is provided, comprising the following steps:

[0073] S1: judgement and decision model loads target task, sets scheduling period, and subdivides target task into multiple sub-targets; the judgement and decision model includes production container, production rule interpreter and judgement and decision model parameter; the production container includes multiple productions; the judgement and decision model parameter includes scheduling period and key cognitive ability index set affecting scheduling period; the target task is human-machine interaction task in regular state or long voyage state.

[0074] In this embodiment, Jdm loads human-machine interaction task, and sets task target set Pmm.Pg, sets judgement and decision scheduling period Jdm.Jdmp.Dat.

[0075] As an optional implementation, the determination process of the key cognitive ability index set specifically includes:

[0076] S11: Obtain cognitive ability dimensions in an occupational information network (ONET); the cognitive ability dimensions include reading and language abilities, creative and reasoning abilities, mathematical abilities, recall abilities, visual perception abilities, spatial reasoning abilities, and attention abilities.

[0077] The cognitive abilities in the user ability dimensions under the ONET include seven types of reading and language abilities (VA), creative and reasoning abilities (IGRA), mathematical abilities (QA), recall abilities (MA), visual perception abilities (VPA), spatial reasoning abilities (SA), and attention abilities (AA).

[0078] The cognitive ability indicators of the reading and language abilities VA include oral comprehension, oral expression, written comprehension, and written expression. The cognitive ability indicators of the creative and reasoning abilities IGRA include idea generation, creativity, problem sensitivity, deductive reasoning, inductive reasoning, information ordering, and classification flexibility. The cognitive ability indicators of the mathematical abilities QA include mathematical reasoning and mathematical flexibility. The cognitive ability indicators of the recall abilities MA include memory. The cognitive ability indicators of the visual perception abilities VPA include summary speed, visual search, and perceptual speed. The cognitive ability indicators of the spatial reasoning abilities SA include spatial orientation and spatial transformation. The cognitive ability indicators of the attention abilities AA include selective attention and attention concentration.

[0079] S12: According to the cognitive ability dimensions, a scale survey method is used to determine a crew human-machine interaction cognitive ability basic index system in combination with target tasks; the crew human-machine interaction cognitive ability basic index system includes language abilities, mathematical abilities, attention and memory abilities, visual perception abilities, reaction abilities, and creative and reasoning abilities.

[0080] The language ability VA includes 5 indexes of oral understanding, oral expression, text understanding, chart understanding and written expression; the creative and reasoning ability IGRA includes 7 indexes of thinking fluency, creativity, problem sensitivity, deductive reasoning, inductive reasoning, information ordering and flexibility of classification; the mathematical ability QA includes 2 indexes of mathematical reasoning and mathematical flexibility; the visual perception ability VPA includes 3 indexes of time estimation, visual search and perceptual speed; the attention and memory ability MAA includes 4 indexes of working memory, spatial transformation, selective attention and attention concentration degree; the reaction ability RA includes 3 indexes of simple reaction, discrimination reaction and selection reaction, and the total number of the 6 categories and 24 indexes is 24.

[0081] S13: According to the crew human-computer interaction cognitive ability basic index system, and combining with the 5-level Likert scale theory, a crew human-computer interaction key cognitive ability importance evaluation scale is determined.

[0082] S14: Based on the crew human-computer interaction key cognitive ability importance evaluation scale, cognitive ability importance evaluation data of a plurality of crews are collected.

[0083] S15: Based on the cognitive ability importance evaluation data and the crew human-computer interaction cognitive ability basic index system, a key cognitive ability index is determined by using a factor analysis method, and a key cognitive ability index set is obtained; the key cognitive ability index set includes spatial transformation, working memory, selective attention, attention concentration degree, selection reaction, simple reaction, discrimination reaction, perceptual speed, visual search and time estimation.

[0084] S2: The judgment and decision model analyzes all sub-targets in the task target set to form a production set, the judgment and decision model performs the production corresponding to the maximum utility value based on the formed production set, after the execution is completed, the judgment and decision model updates the working memory model; the working memory model includes visual attention working memory and manipulation control working memory; the working memory model is used to save the temporary information corresponding to visual attention and operation control.

[0085] In this embodiment, Jdm analyzes the current target in Pmm.Pg, forms a production set Jdm.Pc, Jdm finds the serial number (assuming X) corresponding to Max (Jdm.Psi), selects Jdm.Pc X The production is executed, Jdm queries Lmm, and Jdm updates Wmm.

[0086] S3: the working memory model uses a visual attention working memory driven visual attention model to operate, to perform eye movement preparation and eye movement saccade action; the visual attention working memory includes current information generated by a visual attention model and decision scheduling information of a judgment and decision model to the visual attention model; the visual attention model includes an eye movement preparation model, an eye movement saccade execution model and visual attention model parameters; the visual attention model parameters include eye movement preparation time and eye movement saccade speed.

[0087] S4: obtaining the eye movement preparation time and the eye movement saccade speed after the visual attention model operates.

[0088] As an optional implementation, when the target task is a human-computer interaction task in a normal state, the eye movement preparation time and the eye movement saccade speed after the visual attention model operates are obtained, and specifically comprising:

[0089] The eye movement preparation time is determined by the formula T prep = T v -T gexe ; wherein, T prep is the eye movement preparation time; T v is the fixation time measured by the eye tracker, and in this embodiment, T v = 211 ms is measured by experiment; and T gexe is the fixed time of the eye movement saccade, and in this embodiment, T gexe = 70 ms is measured by experiment.

[0090] The eye movement saccade speed is determined by the formula T exe = T1+T2+v×d; wherein, T exe is the eye movement saccade time; T1 is the eye movement saccade programming time, T1 = 50 ms; T2 is the basic time, T2 = 20 ms; v is the eye movement saccade speed; and d is the saccade angle.

[0091] As an optional implementation, when the target task is a human-computer interaction task in a long-haul state, the eye movement preparation time and the eye movement saccade speed after the visual attention model operates are obtained, and specifically comprising:

[0092] The eye movement preparation time is determined by the formula ; wherein, is the eye movement preparation time under the current long-haul days; and t is the current long-haul days.

[0093] The eye movement saccade speed is determined by the formula ; wherein, is the eye movement saccade speed under the current long-haul days.

[0094] The eye movement saccade speed of the crew in the long-haul state In combination with the seafarer visual attention model, the seafarer visual saccade time is:

[0095]

[0096] Wherein, T lvexe (t) is the long voyage saccade time.

[0097] S5: According to the eye movement preparation time and the eye movement saccade speed after the operation of the visual attention model, determine the visual attention performance.

[0098] In this embodiment, for the regular state visual attention performance Rvap, in combination with the seafarer visual attention model, T va = {Vam, Vamp, T prep , Vam, Vamp, v}.

[0099] For the long voyage state visual attention performance Lvvap(t), based on the regular state visual attention performance Rvap, in combination with the seafarer capability change model Lvwsacm and the seafarer visual attention model Vam, respectively represent the change model of the eye movement preparation time Vam, Vamp, T prep and the saccade speed Vam, Vamp, v in the seafarer visual attention model in the long voyage state, t represents the long voyage time, the tth day.

[0100] S6: The working memory model uses the manipulation control working memory to drive the manipulation control model to operate, and performs mouse movement and click and keystroke actions; the manipulation control working memory includes current information generated by the manipulation control model and decision scheduling information of the judgment and decision model to the manipulation control model; the manipulation control model includes a mouse movement and click model, a keyboard keystroke model, and a manipulation control model parameter.

[0101] S7: Obtain the Fitts' law parameters of the display control device hardware environment and the seafarer's Fitts' law parameters after the operation of the manipulation control model.

[0102] As an optional implementation, when the target task is a human-computer interaction task in a regular state, according to the Fitts' law parameters of the display control device hardware environment and the seafarer's Fitts' law parameters after the operation of the manipulation control model, determine the manipulation control performance, specifically including:

[0103] Determine the manipulation control performance by the formula Tm m = T a + T b log(D / W+0.5); wherein, Tm m is the manipulation control performance; T aThe Fitts' law parameter of the hardware environment of the display and control device; T b The Fitts' law parameter of the crew; D is the mouse movement distance; W is the target width.

[0104] In this embodiment, in the normal state, T a is equal to 157, T b is equal to 202.

[0105] As an optional implementation, when the target task is the human-computer interaction task in the long voyage state, the Fitts' law parameters of the hardware environment of the display and control device and the crew after the operation of the manipulation control model are obtained, and specifically include:

[0106] The Fitts' law parameter of the hardware environment of the display and control device is determined by the formula ; wherein, is the Fitts' law parameter of the hardware environment of the display and control device in the current long voyage days; t is the current long voyage days;

[0107] The Fitts' law parameter of the crew is determined by the formula ; wherein, is the Fitts' law parameter of the crew in the current long voyage days.

[0108] For the long voyage state manipulation control performance Lvmcp(t), the normal state manipulation control performance Rmcp is used as the basis, combined with the crew ability change model Lvwsacm and the crew manipulation control model Mcm, respectively, the parameters a, b change model of the Fitts' law in the long voyage state, t also represents the long voyage time. Consistent with Rmcp, Lvmcp(t) is also composed of the time period sequence of the crew mouse manipulation and the key manipulation.

[0109] S8: According to the Fitts' law parameters of the hardware environment of the display and control device and the crew after the operation of the manipulation control model, the manipulation control performance is determined.

[0110] In this embodiment, Wmm uses Wmm.Vwm to drive Vam to operate, performs eye movement preparation and eye movement saccade action, and forms visual attention performance Vap; Wmm uses Wmm.Mwm to drive Mcm to operate, performs mouse movement click and key action, and forms manipulation control performance Mcp.

[0111] For the normal state manipulation control performance Rmcp, combined with the crew manipulation control model Mcm, T mc = {Mcm.Mcmp.T a , Mcm.Mcmp.T b}, T a , T brespectively represent the Fitts' law parameters a, b applicable to the hardware environment of the display and control device and the crew in the normal state.

[0112] As an optional implementation, when the target task is a human-machine interaction task in the long voyage state, the Fitts' law parameters of the hardware environment of the display and control device and the Fitts' law parameters of the crew after the operation of the control model are determined to determine the control performance, and specifically comprising:

[0113] The control performance is determined by using the formula

[0114] The control performance is determined by using the formula The control performance is determined by using the formula

[0115] S9: The visual attention model and the control model return the visual attention execution result and the control execution result respectively, and update the visual attention working memory according to the visual attention execution result and update the control working memory according to the control execution result, to obtain an updated working memory model.

[0116] S10: The production memory model of the judgment and decision model is updated according to the updated working memory model, and the current sub-target execution is completed; the production memory model includes all sub-targets of the current target task and the current state values of the sub-targets.

[0117] S11: The judgment and decision scheduling period of the current sub-target is obtained, and the judgment and decision performance is determined.

[0118] In this embodiment, Vam and Mcm return the execution results, and Wmm.Vwm, Wmm.Mwm are updated respectively, and the updated results are fed back to Jdm, Pmm.Pgv is updated by Jdm, and the judgment and decision performance Jdp is formed after one cycle.

[0119] In combination with the long voyage state, Ccbacm.Lvwsacm acts on Vam, Mcm, and Jdm, and has an impact on the output results of S3-S11.

[0120] As an optional implementation, when the target task is a human-machine interaction task in the normal state, the judgment and decision scheduling period of the target task is the scheduling period set when the judgment and decision model loads the target task.

[0121] For the judgment and decision performance Rjdp in the normal state, in combination with the crew judgment and decision model Jdm, T jd = Jdm.Jdmp.Dat.

[0122] As an optional implementation, when the target task is a human-computer interaction task in a long voyage state, the judgment and decision scheduling period of the current sub-target is obtained, and the judgment and decision performance is determined, specifically comprising:

[0123] The key cognitive ability indicators in the set of key cognitive ability indicators are subjected to significance analysis to obtain key cognitive ability significance indicators; the key cognitive ability significance indicators include spatial transformation ability, speech working memory ability, spatial working memory ability, selective attention ability, and simple reaction ability.

[0124] In this embodiment, 10 key cognitive ability indicators including spatial transformation, working memory, selective attention, attention concentration degree, and simple reaction are subjected to significance analysis. The five key cognitive abilities of attention concentration degree, selective reaction, discrimination reaction, perceptual speed, and time estimation do not change significantly (P>0.05) over time in the 30-day pre-experiment, and the four abilities of spatial transformation, speech working memory, spatial working memory, selective attention, and simple reaction change significantly (P>0.05) over time. In order to reduce the influence of long cognitive ability test on the cognitive ability of the sailor and thus cause the test results to be biased, and also to reduce the cost of the experiment, the five key cognitive abilities that do not change significantly in the 30-day long voyage experiment are not tested in the 90-day long voyage experiment. Therefore, the key cognitive abilities tested in the 90-day long voyage experiment include spatial transformation, speech working memory, spatial working memory, selective attention, and simple reaction.

[0125] The formula is used to determine the reaction time corresponding to the spatial transformation ability in the long voyage state. The reaction time corresponding to the spatial transformation ability in the long voyage state is determined; wherein, is the reaction time corresponding to the spatial transformation ability under the current long voyage days; t is the number of days of the current long voyage.

[0126] The formula is used to determine the reaction time corresponding to the speech working memory ability in the long voyage state. The reaction time corresponding to the speech working memory ability in the long voyage state is determined; wherein, is the reaction time corresponding to the speech working memory ability under the current long voyage days.

[0127] The formula is used to determine the reaction time corresponding to the spatial working memory ability in the long voyage state. The reaction time corresponding to the spatial working memory ability in the long voyage state is determined; wherein, is the reaction time corresponding to the spatial working memory ability under the current long voyage days.

[0128] The formula is used to determine the reaction time corresponding to the selective attention ability in the long voyage state. The reaction time corresponding to the selective attention ability in the long voyage state is determined; wherein, The reaction time corresponding to the selective attention ability under the current long voyage days.

[0129] The reaction time corresponding to the selective attention ability under the current long voyage days. The reaction time corresponding to the simple reaction ability under the long voyage state is determined by the formula The reaction time corresponding to the simple reaction ability under the current long voyage days.

[0130] The long voyage state key cognitive ability change coefficient is determined according to the reaction time corresponding to the spatial transformation ability under the long voyage state, the reaction time corresponding to the speech working memory ability, the reaction time corresponding to the spatial working memory ability, the reaction time corresponding to the selective attention ability, and the reaction time corresponding to the simple reaction ability.

[0131] The judgment and decision performance is determined according to the long voyage state key cognitive ability change coefficient and the scheduling period.

[0132] As an optional implementation, the judgment and decision performance is determined according to the long voyage state key cognitive ability change coefficient and the scheduling period, specifically including:

[0133] The judgment and decision performance is determined by the formula Lvwsacm.lvapcm.Ljdmp.f dat (t) = Jdm.Jdmp.Dat x Kca(t), wherein Lvwsacm.lvapcm.Ljdmp.f dat (t) is the judgment and decision performance under the current long voyage days; Jdm.Jdmp.Dat is the scheduling period; and Kca(t) is the long voyage state key cognitive ability change coefficient.

[0134] Based on the long voyage experiment results, the long voyage state judgment and decision ability change model Lvwsacm.Lvapcm.Ljdmp is constructed, respectively represent the spatial transformation, working memory, selective attention, and simple reaction ability change models.

[0135] For the long voyage state judgment and decision performance Lvjdp(t), based on the conventional state judgment and decision performance Rjdp, combined with the crew judgment and decision model Jdm and the crew ability change model Lvwsacm, T lvjd = Lvwsacm.Lvapcm.Ljdmp.f dat (t), Lvwsacm.Lvapcm.Ljdmp.f dat (t) represents the change model of Jdm.Jdmp.Dat under the long voyage state, and the size of Lvwsacm.Lvapcm.Ljdmp.f dat (t) is determined by the key cognitive ability change model determination.

[0136] S12: Summing up the visual attention performance, the manipulation control performance and the judgment and decision performance of all sub-targets until the target task execution is completed, determining the total cognitive performance of the crew human-machine interaction.

[0137] In this embodiment, if Pmm.Pg does not execute the last sub-target, repeat execution of S2 to S11 until the end. After the task is completed, the cognitive performances Vap, Jdp and Mcp jointly constitute the total cognitive performance.

[0138] In this embodiment, taking the task Task as input, the cognitive performance in the regular state can be expressed in combination with the crew cognitive performance model Ccp:

[0139] Rtcp=f tcp (Rvap, Rjdp, Rmcp).

[0140] Wherein, Rtcp is the total cognitive performance in the regular state; Rvap, Rjdp, Rmcp, f tcp respectively represent the visual attention performance, the judgment and decision performance, the manipulation control performance and the human-machine interaction task cognitive performance calculation model. Rtcp can be expanded as:

[0141] Rtcp=f tcp (f vap (T va , Task), f jdp (T jd , Task), f mcp (T mc , Task).

[0142] Wherein, f vap , T va respectively represent the visual attention performance calculation model and the model parameter; f jdp , T jd respectively represent the judgment and decision performance calculation model and the model parameter; f mcp , T mc respectively represent the manipulation control performance calculation model and the model parameter. f tcp , f vap , f jdp , f mcp are realized by the ACT-R (Adaptive Control of Thought Rational) cognitive framework. Substituting the above calculated T va , T jd , T mc into the ACT-R model, the total cognitive performance in the regular state can be obtained.

[0143] Taking the task Task as input, in combination with the cognitive performance model Ccp of the seafarer, the cognitive performance in the long voyage state can be expressed as:

[0144] Lvtcp(t)=f tcp (Lvvap(t), Lvjdp(t), Lvmcp(t)).

[0145] Wherein, Lvtcp(t) is the total cognitive performance in the long voyage state; Lvvap(t), Lvjdp(t), Lvmcp(t) represent visual attention performance, judgment and decision-making performance, and manipulation control performance respectively. Lvtcp(t) can be expanded as:

[0146] Lvtcp(t)=f tcp (f vap (T lvva , Task), f jdp (T lvjd , Task), f mcp (T lvmc , Task).

[0147] Wherein, T lvva , T lvjd , T lvmc represent the model parameters of the visual attention model, the judgment and decision-making model, and the manipulation control model in the long voyage state. Substituting the above calculated T lvva , T lvjd , T lvmc into the ACT-R model, the total cognitive performance in the long voyage state can be obtained.

[0148] To solve the problem of lacking of seafarer human-computer interaction cognitive performance quantitative modeling oriented to the two states of routine and long voyage, the application first proposes a seafarer cognitive performance quantitative model considering the two states of routine and long voyage. Secondly, a seafarer judgment and decision-making ability change model based on multi-dimensional cognitive ability and multiple long voyage change models is proposed. Through the verification of the seafarer cognitive performance model in the two states, the simulation calculation error of the cognitive performance in the routine state is within 5.17%, and the simulation calculation error of the cognitive performance in the long voyage state is within 13.47%. The application can improve the quantitative accuracy of seafarer human-computer interaction cognitive performance in the two states of routine and long voyage.

[0149] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0150] The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method of crew human-machine interaction cognitive performance determination, characterized in that, The method comprises the following steps: loading a target task into a judgment and decision model, setting a scheduling period, and subdividing the target task into a plurality of sub-targets; the judgment and decision model comprises a production container, a production rule interpreter, and judgment and decision model parameters; the production container comprises a plurality of productions; the judgment and decision model parameters comprise a scheduling period and a set of key cognitive ability indicators affecting the scheduling period; the target task is a human-machine interaction task in a normal state or a long-haul state; the judgment and decision model analyzes all the sub-targets in the target task set to form a production set, and performs a production corresponding to a maximum utility value based on the formed production set, and after the performance is completed, the judgment and decision model updates a working memory model; the working memory model comprises a visual attention working memory and a manipulation control working memory; the working memory model is used to save temporary information corresponding to visual attention and operation control; the working memory model drives a visual attention model to operate using the visual attention working memory, and performs eye movement preparation and eye movement saccade actions; a visual attention performance is determined according to an eye movement preparation time and an eye movement saccade speed after the visual attention model operates; the working memory model drives a manipulation control model to operate using the manipulation control working memory, and performs mouse movement and click actions and keystroke actions; a manipulation control performance is determined according to a Fitts' law parameter of a hardware environment of a display control device and a Fitts' law parameter of the crew member after the manipulation control model operates; the visual attention model and the manipulation control model return a visual attention execution result and a manipulation control execution result respectively, and update the visual attention working memory and the manipulation control working memory to obtain an updated working memory model; a production memory model of the judgment and decision model is updated according to the updated working memory model, and a current sub-target execution is completed; the production memory model comprises all the sub-targets of the current target task and current state values of the sub-targets; a judgment and decision scheduling period of the current sub-target is obtained, and a judgment and decision performance is determined; until the target task is executed completely, the visual attention performance, the manipulation control performance, and the judgment and decision performance of all the sub-targets are summed to determine a total cognitive performance of the crew member human-machine interaction.

2. The crew human-machine interaction cognitive performance determination method of claim 1, wherein, The determination process of the set of key cognitive ability indicators comprises the following steps: obtaining cognitive ability dimensions in a work information network; the cognitive ability dimensions comprise reading and language ability, creativity and reasoning ability, mathematical ability, recall ability, visual perception ability, spatial reasoning ability, and attention ability; determining a crew member human-machine interaction cognitive ability basic index system according to the cognitive ability dimensions, using a scale survey method, and combining a target task; the crew member human-machine interaction cognitive ability basic index system comprises language ability, mathematical ability, attention and memory ability, visual perception ability, reaction ability, and creativity and reasoning ability; determining a crew member human-machine interaction key cognitive ability importance evaluation scale according to the crew member human-machine interaction cognitive ability basic index system and a 5-level Likert scale theory. Based on the crew human-computer interaction key cognitive ability importance evaluation scale, cognitive ability importance evaluation data of a plurality of crew members are collected; Based on the cognitive ability importance evaluation data and the crew human-computer interaction cognitive ability basic index system, a key cognitive ability index is determined by using a factor analysis method, and a key cognitive ability index set is obtained; the key cognitive ability index set includes spatial transformation, working memory, selective attention, attention concentration degree, selection reaction, simple reaction, discrimination reaction, perceptual speed, visual search, and time estimation.

3. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a regular state, the eye movement preparation time and the eye movement saccade speed after the visual attention model is run are obtained, and specifically, the method comprises the following steps: Using the formula T prep = T v - T gexe determine the eye movement preparation time; wherein T prep is the eye movement preparation time; T v is the fixation time measured by the eye tracker; and T gexe is the fixation time of the eye movement saccade. The eye movement saccade velocity is determined by the formula T exe = T1 + T2 + v x d, wherein T exe is the eye movement saccade time; T1 is the eye movement saccade programming time; T2 is the base time; v is the eye movement saccade velocity; and d is the saccade angle.

4. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a long-haul state, the eye movement preparation time and the eye movement saccade speed after the visual attention model is run are obtained, and specifically, the method comprises the following steps: Using the formula determining the eye movement preparation time; wherein, is the eye movement preparation time under the current long-duration spaceflight days; t is the current long-duration spaceflight days; Using the formula The saccadic velocity is determined, wherein is the saccadic velocity for the current long on-orbit days.

5. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a regular state, the Fitts' law parameters of the hardware environment of the display control device and the Fitts' law parameters of the crew after the operation control model is run are used to determine the operation control performance, and specifically, the method comprises the following steps: Tm m = T a + T b log(D / W+0.5) determines the steering control performance; wherein Tm m is the steering control performance; T a is the Fitts' law parameter of the display control device hardware environment; T b is the Fitts' law parameter of the crew; D is the mouse movement distance; and W is the target width.

6. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a long-haul state, the Fitts' law parameters of the hardware environment of the display control device and the Fitts' law parameters of the crew after the operation control model is run are obtained, and specifically, the method comprises the following steps: Using the formula determining a Fitts' Law parameter for the hardware environment of the display control device; wherein, is a Fitts' Law parameter for the hardware environment of the display control device for a current long-duration spaceflight day; t is the current long-duration spaceflight day; Using the formula determining the Fitts' law parameters of the crew; wherein, is the Fitts' law parameters of the crew for the current number of days at sea.

7. The crew human-machine interaction cognitive performance determination method of claim 6, wherein, When the target task is a human-computer interaction task in a long-haul state, the Fitts' law parameters of the hardware environment of the display control device and the Fitts' law parameters of the crew after the operation control model is run are used to determine the operation control performance, and specifically, the method comprises the following steps: The formula is used. determining a maneuver control performance; wherein, is the maneuver control performance for the current number of days in space.

8. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a regular state, the judgment and decision-making scheduling period of the target task is the scheduling period set when the judgment and decision-making model loads the target task.

9. The crew human interaction cognitive performance determination method of claim 1, wherein, When the target task is a human-computer interaction task in a long-haul state, the judgment and decision-making scheduling period of the current sub-target is obtained, and the judgment and decision-making performance is determined, and specifically, the method comprises the following steps: Significance analysis is performed on the key cognitive ability indexes in the key cognitive ability index set, and a key cognitive ability significance index is obtained; the key cognitive ability significance index includes spatial transformation ability, verbal working memory ability, spatial working memory ability, selective attention ability, and simple reaction ability. Using the formula 12483.210 determining the reaction time corresponding to the spatial transformation ability in the long voyage state; wherein, is the spatial transformation reaction time under the current long voyage days; t is the number of days of the current long voyage; The formula is as follows: The reaction time corresponding to the speech working memory capacity in the long voyage state is determined; wherein, is the reaction time corresponding to the speech working memory capacity in the current long voyage state. Using the formula determining the reaction time corresponding to the spatial working memory capacity under the long voyage state; wherein, is the reaction time corresponding to the spatial working memory capacity under the current long voyage days. Using the formula determining the reaction time corresponding to the selective attention ability in the long voyage state; wherein, is the reaction time corresponding to the selective attention ability in the current long voyage state. The reaction time corresponding to the simple reaction ability under the long voyage state is determined by using the formula The reaction time corresponding to the simple reaction ability under the long voyage state is determined by using the formula The reaction time corresponding to the simple reaction ability under the long voyage state is determined by using the formula The long-haul state key cognitive ability change coefficient is determined according to the reaction time corresponding to the spatial transformation ability in the long-haul state, the reaction time corresponding to the verbal working memory ability, the reaction time corresponding to the spatial working memory ability, the reaction time corresponding to the selective attention ability, and the reaction time corresponding to the simple reaction ability. The judgment and decision-making performance is determined according to the long-haul state key cognitive ability change coefficient and the scheduling period.

10. The crew human-machine interaction cognitive performance determination method of claim 9, wherein, The judgment and decision-making performance is determined according to the long-haul state key cognitive ability change coefficient and the scheduling period, and specifically, the method comprises the following steps: Lvwsacm.lvapcm.Ljdmp.f dat (t) = Jdm.Jdmp.Dat x Kca(t) determines the judgment and decision performance; wherein, Lvwsacm.lvapcm.Ljdmp.f dat (t) is the judgment and decision performance under the current long space travel days; Jdm.Jdmp.Dat is the scheduling period; Kca(t) is the change coefficient of the key cognitive ability in the long space travel state.

Images