Underwater platform navigation driver attention assessment method and system

Abstract

The invention discloses an attention assessment method and system for an underwater platform navigation driver, and relates to the field of driver attention assessment, and the method comprises the steps: carrying out the statistics of the operation frequency of the driver for the navigation of an underwater platform within a period of time, and automatically selecting a corresponding attention assessment means and alarm according to the level of the operation frequency; wherein the attention assessment mode integrates face fatigue monitoring and voice response monitoring. According to the method, the attention evaluation method is carried out in a mode of combining face fatigue monitoring and voice response monitoring, so that the attention of a navigation driver can be evaluated more reasonably, and the voice response type frequency is adaptively selected according to the operation frequency; the situation that the total task operation load of operators is too high due to attention evaluation in a voice response mode is avoided, and sailing driving safety can be improved.

Description

Technical Field

[0001] The present invention relates to the field of driver attention assessment, and in particular to a method and system for assessing the attention of underwater platform navigation drivers. Background Art

[0002] Underwater navigation platforms are one of the important tools used in current ocean development and are widely applied in military and civilian fields, such as underwater target exploration, mapping and positioning, marine scientific surveys, undersea oil development, special disposal operations, etc. With the progress of technology and the increasing complexity of underwater operation tasks, it is an inevitable trend to develop large-scale mobile deep-sea platforms to achieve long-term, large-scale multi-functional exploration and fine engineering operations in the deep sea. To support the smooth completion of tasks by mobile deep-sea platforms, drivers are required to conduct long-term driving control of underwater platforms and all-weather status safety monitoring. Currently, there are many mature achievements in the state monitoring of automobile drivers, mainly through means such as facial vision detection, physiological feature detection, and analysis of personnel operation characteristics for fatigue monitoring. However, the driving control of underwater platforms has certain particularities. Its driving control frequency is closely related to the navigation conditions. The driving control cycle is very long but the average load is not large. In some conditions, there is basically no operation, but the driver must be required to concentrate on observing important navigation information throughout the process and detect early features of abnormal situations as early as possible. The irregular driving control characteristics of underwater navigation platforms cannot meet the requirements for judging fatigue through operation characteristics. Only through means such as visual monitoring and physiological feature monitoring, it is impossible to judge whether the driver's attention is focused on effective important information. Summary of the Invention

[0003] In view of the above problems and technical requirements, the inventor of the present invention has proposed a method and system for assessing the attention of underwater platform navigation drivers. The technical solution of the present invention is as follows:

[0004] In a first aspect, the present application provides a method for assessing the attention of underwater platform navigation drivers, including the following steps:

[0005] Statistically analyze the operation frequency of the driver for underwater platform navigation within a period of time, and select corresponding attention assessment means and alarms according to the operation frequency;

[0006] Among them, the attention assessment method adopts face fatigue monitoring and voice response monitoring.

[0007] A further technical solution thereof is that selecting corresponding attention assessment means and alarms according to the operation frequency includes:

[0008] If the operation frequency is high, only enable face fatigue monitoring for attention assessment and output an alarm according to the assessment result;

[0009] If the operating frequency is medium frequency and low frequency, then both face fatigue monitoring and voice response monitoring are enabled for attention assessment. Then, the two assessment results are combined for comprehensive determination, and an alarm is output.

[0010] Among them, for the comprehensive determination corresponding to medium frequency, the evaluation result of face fatigue monitoring has a greater weight, and for the comprehensive determination corresponding to low frequency, the evaluation result of voice response monitoring has a greater weight.

[0011] Whether the operating frequency is high frequency, medium frequency or low frequency is determined by comparing with a set threshold.

[0012] A further technical solution thereof is to enable voice response monitoring for attention assessment, including:

[0013] Randomly announce the name corresponding to a certain navigation status data concerned by the driver according to the voice interaction frequency, and obtain the response duration of the driver's feedback on the specific value of the navigation status data, the average feedback time for each digit, and the correctness of the feedback.

[0014] Statistically calculate the average response duration, the average value of the average feedback time, and the feedback correct rate of the driver's feedback on the specific values of all navigation status data within a period of time, conduct a comprehensive determination, and give an evaluation result.

[0015] A further technical solution thereof is to use voice response monitoring for attention assessment, and further include:

[0016] Adaptive adjustment of the voice interaction frequency f according to the operating frequency is as follows:

[0017]

[0018] Among them, H max is the set maximum operating frequency threshold, H min is the set minimum operating frequency threshold, H is the statistically calculated operating frequency, f max is the set upper limit of the voice interaction frequency, f min is the set lower limit of the voice interaction frequency.

[0019] A further technical solution thereof is to obtain the response duration of the driver's feedback on the specific value of the navigation status data, the average feedback time for each digit, and the correctness of the feedback, including:

[0020] The time interval from the end moment of the voice announcement of the name corresponding to a certain navigation status data to the start moment of the driver's voice feedback is used as the driver's response duration.

[0021] Record the time t2 used from the start moment to the end moment of the driver's voice feedback, and identify the voice information of the feedback, and extract the specific value of the navigation status data recognized.

[0022] Compare the specific values of the navigation status data with the relevant real navigation status data recorded during the underwater platform's navigation to determine whether the feedback is correct;

[0023] If the feedback is correct, divide t2 by the length of the specific values of the navigation status data to obtain the average feedback time for each digit in the specific values.

[0024] Its further technical solution is to comprehensively determine the average reaction time ta1, the average value of the average feedback time ta3, and the feedback correct rate P, and give an evaluation result, including:

[0025] If P > Pv, and ta1 < tav1 & ta3 < tav3, then the driver's attention is concentrated;

[0026] If P < Pv, then the driver's attention is not concentrated;

[0027] If P > Pv, and tav1 < ta1 || tav3 < ta3, then the driver's attention is not concentrated;

[0028] Among them, Pv is the set feedback correct rate threshold, tav1 is the set average reaction time threshold, and tav3 is the threshold of the average value of the average feedback time.

[0029] Its further technical solution is that if the operation frequency is medium frequency, combine the two evaluation results for comprehensive determination and output an alarm, including:

[0030] If the evaluation result of face fatigue monitoring is that the driver is fatigued, and the evaluation result of voice response monitoring is that the driver's attention is not concentrated, output a first-level alarm;

[0031] If the evaluation result of face fatigue monitoring is that the driver is fatigued, and the evaluation result of voice response monitoring is that the driver's attention is concentrated, output a second-level alarm;

[0032] If the evaluation result of face fatigue monitoring is that the driver is not fatigued, and the evaluation result of voice response monitoring is that the driver's attention is not concentrated, output a third-level alarm;

[0033] Otherwise, no alarm is given.

[0034] Its further technical solution is that if the operation frequency is low frequency, combine the two evaluation results for comprehensive determination and output an alarm, including:

[0035] If the evaluation result of face fatigue monitoring is that the driver is fatigued, and the evaluation result of voice response monitoring is that the driver's attention is not concentrated, output a first-level alarm;

[0036] If the evaluation result of face fatigue monitoring indicates that the driver is not fatigued, and the evaluation result of voice response monitoring indicates that the driver's attention is not concentrated, then a secondary alarm is output;

[0037] If the evaluation result of face fatigue monitoring indicates that the driver is fatigued, and the evaluation result of voice response monitoring indicates that the driver's attention is concentrated, then a tertiary alarm is output;

[0038] Otherwise, no alarm is given.

[0039] In a second aspect, the present application also provides an attention evaluation system for underwater platform navigation drivers, including a comprehensive evaluation computer, a camera, a voice output module, and a voice input module;

[0040] The comprehensive evaluation computer is communicatively connected to the underwater platform navigation console for obtaining the navigation status data of the underwater platform and the driving operation instructions input by the driver;

[0041] The comprehensive evaluation computer is communicatively connected to the camera for obtaining the facial video information of the driver and performing face fatigue monitoring;

[0042] The comprehensive evaluation computer is communicatively connected to the voice output module for outputting voice broadcast instructions for the driver to execute;

[0043] The comprehensive evaluation computer is communicatively connected to the voice input module for obtaining the voice information feedback by the driver;

[0044] The comprehensive evaluation computer internally executes the attention evaluation method for underwater platform navigation drivers as described in the first aspect, for evaluating the attention of the navigation driver and giving appropriate alarms.

[0045] A further technical solution thereof is that the voice broadcast instruction is to randomly broadcast the name corresponding to a certain navigation status data concerned by the driver, and the list of names corresponding to all the navigation status data concerned by the driver is stored in the comprehensive evaluation computer in the form of a configuration table;

[0046] Among them, the name list includes depth, depth rate, height, height rate, course, trim, roll, remaining power, and navigation position.

[0047] The beneficial technical effects of the present invention are:

[0048] (1) The method of combining face fatigue monitoring using video and attention evaluation through voice response monitoring can more reasonably evaluate the attention of navigation drivers.

[0049] (2) During the process of attention assessment through voice response monitoring, the operation frequency of the navigation driver is considered, and the frequency of voice response is adaptively selected according to the operation frequency, so as to avoid the total task operation load of the operator being too high due to attention assessment by voice response.

[0050] (3) When the driver is at a medium or low operation frequency, the evaluation system can also improve the safety of navigation driving by actively requiring the operator to pay attention to key navigation safety information. Description of the Drawings

[0051] Figure 1 It is a schematic diagram of an attention evaluation system for underwater platform navigation drivers provided by this application.

[0052] Figure 2 It is a flowchart of a method for evaluating the attention of underwater platform navigation drivers provided by this application.

[0053] Figure 3 It is a flowchart of a method for attention evaluation by voice response monitoring provided by this application. Detailed Embodiments

[0054] The following further describes the detailed embodiments of the present invention with reference to the drawings.

[0055] An embodiment of this application provides an attention evaluation system for underwater platform navigation drivers, as Figure 1 shown. This system is deployed on the underwater platform 301 and specifically includes a comprehensive evaluation computer 101, a camera 102, a voice output module 103, and a voice input module 104. Among them:

[0056] The comprehensive evaluation computer 101 is communicatively connected to the underwater platform navigation console 201 for obtaining the navigation status data of the underwater platform and the operation instructions input by the driver. The comprehensive evaluation computer 101 is communicatively connected to the camera 102 for obtaining the facial video information of the driver captured by the camera and performing face fatigue monitoring. In this embodiment, the camera 102 is deployed in front of the driver and is aimed at the driver's face. The voice output module 103 and the voice input module 104 cooperate to achieve voice response monitoring. Specifically, the comprehensive evaluation computer 101 is communicatively connected to the voice output module 103 for outputting voice broadcast instructions to prompt the driver to execute. The comprehensive evaluation computer 101 is communicatively connected to the voice input module 104 for obtaining the voice information feedback by the driver. Finally, the comprehensive evaluation computer 101 summarizes the above information and executes an attention evaluation method for underwater platform navigation drivers internally to evaluate the attention of the navigation driver and give appropriate warnings.

[0057] Another embodiment of the present application provides a method for evaluating the attention of underwater platform navigators, which is implemented based on the above attention evaluation system, as Figure 2 shown, and specifically includes the following steps:

[0058] Step 1: First, obtain the operation instructions input by the underwater navigator, continuously count the operation frequency H of the navigator for the underwater platform navigation in the recent period of time, and divide the operation frequency into three levels: high, medium, and low. If H is greater than the set maximum operation frequency threshold H max (such as 10 times / minute), the operation frequency is high; if H is less than or equal to the set minimum operation frequency threshold H min (such as 3 times / minute), the operation frequency is low; otherwise, the operation frequency is medium, that is, H min <H≤H max .

[0059] Step 2: Select the corresponding attention evaluation means and alarms from Steps 2.1 to 2.3 according to the operation frequency. The attention evaluation methods include face fatigue monitoring and voice response monitoring.

[0060] Step 2.1: If the operation frequency is high, only enable face fatigue monitoring for attention evaluation and output an alarm according to the evaluation result. This can avoid the additional operation burden on the navigator during the attention evaluation process of voice response monitoring. If the visual monitoring detects that the navigator's face is fatigued at this time, a first-level alarm is output through the voice output module; otherwise, no alarm is given.

[0061] Step 2.2: If the operation frequency is medium, enable both face fatigue monitoring and voice response monitoring for attention evaluation, then make a comprehensive judgment based on the two evaluation results, with the evaluation result of face fatigue monitoring having a greater weight, and then output an alarm. If the evaluation result of face fatigue monitoring is that the driver is fatigued and the evaluation result of voice response monitoring is that the driver's attention is not concentrated, a first-level alarm is output through the voice output module; if the evaluation result of face fatigue monitoring is that the driver is fatigued and the evaluation result of voice response monitoring is that the driver's attention is concentrated, a second-level alarm is output through the voice output module; if the evaluation result of face fatigue monitoring is that the driver is not fatigued and the evaluation result of voice response monitoring is that the driver's attention is not concentrated, a third-level alarm is output through the voice output module; otherwise, no alarm is given.

[0062] Step 2.3: If the operation frequency is low, both face fatigue monitoring and voice response monitoring are enabled for attention assessment. Then, a comprehensive determination is made by combining the two assessment results, with the assessment result of voice response monitoring having a greater weight, and then an alarm is output. If the assessment result of face fatigue monitoring indicates that the driver is fatigued and the assessment result of voice response monitoring indicates that the driver's attention is not concentrated, a first-level alarm is output through the voice output module; if the assessment result of face fatigue monitoring indicates that the driver is not fatigued and the assessment result of voice response monitoring indicates that the driver's attention is not concentrated, a second-level alarm is output through the voice output module; if the assessment result of face fatigue monitoring indicates that the driver is fatigued and the assessment result of voice response monitoring indicates that the driver's attention is concentrated, a third-level alarm is output through the voice output module; otherwise, no alarm is given.

[0063] In this embodiment, enabling face fatigue monitoring for attention assessment includes: The comprehensive assessment computer uses the camera vision to obtain the eye images of the driver and calculates the change in the eye aspect ratio (EAR) to judge the blinking behavior. When the absolute value of the difference between the average EAR value in the current period of time and the initial average EAR value judged by the comprehensive assessment computer exceeds the set threshold, it is considered that the driver is in a fatigued state. Optionally, the initial average EAR value can be calculated by the driver entering normal blinking behaviors before underwater navigation this time.

[0064] In this embodiment, enabling voice response monitoring for attention assessment includes: The comprehensive assessment computer randomly broadcasts the name corresponding to a certain navigation state data that should be concerned by the driver according to the voice interaction frequency f through the voice output module, and obtains the response duration, the average feedback time for each digit, and the correctness of the feedback of the driver's specific value of the navigation state data through the voice input module. The average response duration, the average value of the average feedback time, and the feedback correct rate of the driver's feedback of all specific values of the navigation state data in the recent period of time are statistically analyzed, and a comprehensive determination is made and an assessment result is given. Among them: The voice interaction frequency f is adaptively adjusted dynamically according to the operation frequency of the navigation driver. The list of names corresponding to a certain navigation state data that should be concerned by the driver is configured into the computer in the form of a configuration table before the comprehensive assessment computer runs. The name list generally includes but is not limited to depth, depth rate, height, height rate, course, trim, roll, remaining power, navigation position, etc.

[0065] As Figure 3 shown, the specific steps of the method for enabling voice response monitoring for attention assessment are as follows:

[0066] (1) The comprehensive assessment computer adaptively adjusts the voice interaction frequency f according to the operation frequency of the driver as:

[0067]

[0068] Among them, H is the statistically-operated frequency, and f max is the upper limit of the set voice interaction frequency (such as 1 time per minute), and f min is the lower limit of the set voice interaction frequency (such as 0.2 times per minute).

[0069] (2) The voice interaction period is 1 / f. When the voice interaction period is reached, the comprehensive evaluation computer randomly announces, through the voice output module, the name corresponding to a certain navigation status data that should be concerned by the driver. Record the time interval between the end moment of the voice announcement of the name of a certain navigation status data and the start moment of the driver's voice feedback as the reaction time t1 of the driver.

[0070] (3) Record the time t2 used by the driver from the start moment to the end moment of the feedback through the voice input module, and identify the voice information of the feedback to extract the specific value of the identified navigation status data.

[0071] (4) The comprehensive evaluation computer compares the specific value of the navigation status data with the relevant navigation status real data during the underwater platform navigation transmitted by the underwater platform navigation console to determine whether the feedback is correct.

[0072] (5) If the feedback is correct, calculate the average feedback time t3 for each digit in the specific value:

[0073]

[0074] Among them, Num is the length of the specific value of the navigation status data.

[0075] (6) Statistically calculate the average reaction time ta1 of the driver's feedback on the specific values of all navigation status data, the average value ta3 of the average feedback time for each digit, and the feedback correct rate P within a recent period of time (that is, at least including more than one voice interaction period), and comprehensively determine the attention of the navigation driver and give an evaluation result.

[0076] If P > Pv, and ta1 < tav1 & ta3 < tav3, then the driver's attention is concentrated; if P < Pv, then the driver's attention is not concentrated; if P > Pv, and tav1 < ta1 || tav3 < ta3, then the driver's attention is not concentrated. Among them, Pv is the set feedback correct rate threshold, tav1 is the set average reaction time threshold, and tav3 is the set average value threshold of the average feedback time.

[0077] The above is only a preferred embodiment of the present application, and the present invention is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present invention should be considered to be included in the protection scope of the present invention.

Claims

1. An attention evaluation method for underwater platform navigation drivers, characterized in that, The method includes: Counting the operation frequency of the driver for the underwater platform navigation within a period of time, and selecting corresponding attention evaluation means and alarms according to the operation frequency; Among them, the attention evaluation method adopts face fatigue monitoring and voice response monitoring.

2. The underwater platform navigation driver attention evaluation method according to claim 1, wherein The selecting corresponding attention evaluation means and alarms according to the operation frequency includes: If the operation frequency is high, only enable the face fatigue monitoring for attention evaluation, and output an alarm according to the evaluation result; If the operation frequency is medium and low, simultaneously enable the face fatigue monitoring and the voice response monitoring for attention evaluation, then make a comprehensive determination by combining the two evaluation results, and output an alarm; Among them, for the comprehensive determination corresponding to the medium frequency, the evaluation result of the face fatigue monitoring has a greater weight, and for the comprehensive determination corresponding to the low frequency, the evaluation result of the voice response monitoring has a greater weight; Whether the operation frequency is high, medium or low is determined by comparing with a set threshold.

3. The underwater platform navigation driver attention evaluation method according to claim 1, characterized in that, Enabling the voice response monitoring for attention evaluation includes: Randomly broadcasting the name corresponding to a certain navigation state data concerned by the driver according to the voice interaction frequency, and obtaining the response duration of the driver's feedback on the specific value of the navigation state data, the average feedback time for each number, and the correctness of the feedback; Counting the average response duration of the driver's feedback on the specific values of all navigation state data, the mean value of the average feedback time, and the feedback correct rate within a period of time, making a comprehensive determination and giving an evaluation result.

4. The method for evaluating the attention of underwater platform navigation drivers according to claim 3, characterized in that Adopting the voice response monitoring for attention evaluation also includes: Adapting and adjusting the voice interaction frequency f according to the operation frequency as: Among them, H max is the set maximum operation frequency threshold, H min is the set minimum operation frequency threshold, H is the statistically counted operation frequency, f max is the set upper limit of the voice interaction frequency, f min is the set lower limit of the voice interaction frequency.

5. The method for evaluating the attention of underwater platform navigation drivers according to claim 3, characterized in that, The obtaining the response duration of the driver's feedback on the specific value of the navigation state data, the average feedback time for each number, and the correctness of the feedback includes: Taking the time interval between the end moment of the voice broadcast of the name corresponding to a certain navigation state data and the start moment of the driver's voice feedback as the driver's response duration; Recording the time t2 used from the start moment to the end moment of the driver's voice feedback, and identifying the voice information of the feedback to extract the specific value of the navigation state data recognized; Comparing the specific value of the navigation state data with the relevant real navigation state data during the underwater platform navigation recorded to judge whether the feedback is correct; If the feedback is correct, then dividing the t2 by the length of the specific value of the navigation state data to obtain the average feedback time for each number in the specific value.

6. The method for evaluating the attention of underwater platform navigation drivers according to claim 3, characterized in that, Making a comprehensive determination and giving an evaluation result for the average response duration ta1, the mean value ta3 of the average feedback time, and the feedback correct rate P includes: If P > Pv, and ta1 < tav1 & ta3 < tav3, then the driver's attention is concentrated; If P < Pv, then the driver's attention is not concentrated; If P > Pv, and tav1 < ta1 || tav3 < ta3, then the driver's attention is not concentrated; Among them, Pv is the set feedback correct rate threshold, tav1 is the set average response duration threshold, and tav3 is the set mean value threshold of the average feedback time.

7. The method for evaluating the attention of underwater platform navigation drivers according to claim 2, characterized in that, If the operating frequency is medium frequency, then combine the two evaluation results for comprehensive determination and output an alarm, including: If the evaluation result of the face fatigue monitoring is that the driver is fatigued and the evaluation result of the voice response monitoring is that the driver's attention is not concentrated, output a first-level alarm; If the evaluation result of the face fatigue monitoring is that the driver is fatigued and the evaluation result of the voice response monitoring is that the driver's attention is concentrated, output a second-level alarm; If the evaluation result of the face fatigue monitoring is that the driver is not fatigued and the evaluation result of the voice response monitoring is that the driver's attention is not concentrated, output a third-level alarm; Otherwise, no alarm is given.

8. The method for evaluating the attention of underwater platform navigation drivers according to claim 2, characterized in that, If the operating frequency is low frequency, then combine the two evaluation results for comprehensive determination and output an alarm, including: If the evaluation result of the face fatigue monitoring is that the driver is fatigued and the evaluation result of the voice response monitoring is that the driver's attention is not concentrated, output a first-level alarm; If the evaluation result of the face fatigue monitoring is that the driver is not fatigued and the evaluation result of the voice response monitoring is that the driver's attention is not concentrated, output a second-level alarm; If the evaluation result of the face fatigue monitoring is that the driver is fatigued and the evaluation result of the voice response monitoring is that the driver's attention is concentrated, output a third-level alarm; Otherwise, no alarm is given.

9. An attention assessment system for underwater platform navigation drivers, the system being deployed on an underwater platform, characterized in that, It includes a comprehensive evaluation computer, a camera, a voice output module, and a voice input module; The comprehensive evaluation computer is communicatively connected to the underwater platform navigation console for obtaining the navigation status data of the underwater platform and the operation instructions input by the driver; The comprehensive evaluation computer is communicatively connected to the camera for obtaining the facial video information of the driver and performing face fatigue monitoring; The comprehensive evaluation computer is communicatively connected to the voice output module for outputting voice broadcast instructions for the driver to execute; The comprehensive evaluation computer is communicatively connected to the voice input module for obtaining the voice information feedback by the driver; The comprehensive evaluation computer internally executes the method for evaluating the attention of the underwater platform navigation driver as described in any one of claims 1-8 for evaluating the attention of the navigation driver and giving an appropriate alarm.

10. The underwater platform navigation driver attention evaluation system according to claim 9, characterized in that, The voice broadcast instruction is to randomly broadcast the name corresponding to a certain navigation status data concerned by the driver, and the list of names corresponding to all the navigation status data concerned by the driver is stored in the comprehensive evaluation computer in the form of a configuration table; Among them, the name list includes depth, depth rate, height, height rate, heading, trim, roll, remaining power, and navigation position.

Images