An intelligent feedback regulation method for human-computer interaction experience optimization

Abstract

The application discloses a kind of intelligent feedback regulation and control method for man-machine interactive experience optimization, the present application relates to interactive regulation and control technical field, lack of the receiving and strategy adjustment mechanism of user feedback is solved, cannot form the technical problem of complete interactive closed loop, the present application is through the weighted calculation of theta wave power ratio and alpha wave invasion rate, through a large number of driving data statistics, establish personalized fatigue index judgment interval, solve the problem of different user eeg baseline difference, compared with single frequency band analysis, fatigue monitoring accuracy can be improved, cross validation is carried out in combination with other modal fatigue state information, avoid single signal misjudgment, secondly, real-time analysis road condition, automatically generate the optimization route of shorter time, and through multimodal feedback, guide user decision, support user to the "acceptance / rejection" feedback of route optimization, and dynamically adjust navigation strategy accordingly.

Description

Technical Field

[0001] This invention relates to the field of interactive control technology, specifically to an intelligent feedback control method for optimizing the human-computer interaction experience. Background Technology

[0002] With the development of intelligent driving technology, the importance of human-computer interaction in vehicle systems is becoming increasingly prominent.

[0003] According to patent application CN107357416A, a human-computer interaction device and interaction method are disclosed, including a data acquisition unit, an analysis unit, a processing unit, and an execution unit. The data acquisition unit is used to acquire the user's facial expressions, actions, voice, and / or information input through a controller. The analysis unit is used to analyze the above information. The processing unit is used to comprehensively process the results of the analysis module. The execution unit is used to display a virtual waiter and allow the virtual waiter to perform the operation to be performed by the user, and to feed back the operation execution result to the user.

[0004] Traditional in-vehicle interaction relies primarily on fixed navigation commands and basic driving data monitoring, but lacks the ability to perceive and dynamically respond to the driver's state in real time. While existing navigation systems can provide route planning, they struggle to personalize adjustments based on driver fatigue levels; fatigue monitoring is often based on a single modality, resulting in insufficient accuracy and an inability to achieve closed-loop feedback. Furthermore, traditional interaction methods suffer from lag in real-time response to complex road conditions and lack multimodal adaptive adjustment mechanisms, leading to limitations in the user experience. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent feedback control method for optimizing human-computer interaction experience, solving the problem of lacking a mechanism for receiving user feedback and adjusting strategies, thus failing to form a complete interactive loop.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent feedback control method for optimizing human-computer interaction experience, comprising the following steps: Based on vehicle driving information, determine whether to use navigation and generate navigation driving analysis signals or autonomous driving analysis signals; The autonomous driving analysis signal is analyzed, the driver's brainwaves are matched with the fatigue range reference standard to generate fatigue state information, and the fatigue index is calculated according to the formula. At the same time, fatigue judgment information is generated based on the index judgment interval. Compare the fatigue state information and fatigue judgment information to see if they are the same. If they are not the same, use the fatigue judgment information as the standard to generate fatigue monitoring results, and compare them with the early warning results to generate early warning information or normal monitoring signals. The navigation driving analysis signal is analyzed, and the vehicle's real-time positioning information and route navigation information are used to determine whether there is an optimized route, and generate an optimization feedback analysis signal or a normal positioning monitoring signal. The system analyzes the optimized feedback analysis signal, generates real-time monitoring information based on the feedback information, analyzes the normal positioning monitoring signal, determines the real-time distance based on the route navigation information, compares it with the warning distance, and generates navigation information and normal monitoring signal.

[0007] As a further aspect of the present invention, the specific method for generating the navigation driving analysis signal or autonomous driving analysis signal is as follows: The system acquires vehicle driving information and determines whether the vehicle is using navigation. If navigation is used, a navigation driving analysis signal is generated; otherwise, an autonomous driving analysis signal is generated.

[0008] As a further aspect of the present invention, the specific method for analyzing the autonomous driving analysis signal is as follows: By using the brainwave information of the driver within time t through auxiliary equipment, and matching the obtained brainwave information with the frequency range reference standard, the corresponding fatigue state information is generated. Get driver correspondence wave corresponding Power ratio and wavy Intrusion rate, according to the formula The fatigue index FI corresponding to the driver was calculated, where 0.6 and 0.4 are respectively... Power ratio and The weighting percentage corresponding to the intrusion rate is used to match the obtained fatigue index FI with the corresponding index judgment interval to generate fatigue judgment information.

[0009] As a further aspect of the present invention, the specific method for generating early warning information or normal monitoring signals is as follows: The obtained fatigue state information is matched with the fatigue judgment information to determine whether they are the same. If they are the same, the corresponding fatigue monitoring result is generated. Otherwise, if they are different, the fatigue judgment is used as the standard to generate the fatigue monitoring result. The fatigue detection results are compared with the corresponding warning results. If the fatigue detection results are greater than the warning results, a fatigue warning signal is generated; otherwise, a normal monitoring signal is generated.

[0010] As a further aspect of the present invention, the specific method for analyzing the navigation driving analysis signal is as follows: The system acquires real-time vehicle location information and performs route monitoring and analysis based on route navigation information to determine whether there is an optimized route. An optimized route is defined as a route that, compared to the currently planned route, allows the vehicle to reach its destination in a shorter time. If such a route exists, an optimization feedback analysis signal is generated; otherwise, a normal location monitoring signal is generated.

[0011] As a further aspect of the present invention, the specific method for analyzing the optimized feedback analysis signal is as follows: The system obtains the corresponding optimized route and displays it to the driver. It also obtains the driver's feedback information, including whether the driver accepts the signal or not. If the driver accepts the signal, the system modifies the current route navigation information based on the optimized route and generates real-time monitoring information. If the driver does not accept the signal, the system maintains the current route navigation information.

[0012] As a further aspect of the present invention, the specific method for analyzing the normal positioning monitoring signal is as follows: Based on the vehicle's real-time location information and route navigation information, a comprehensive judgment is made to obtain the corresponding non-straight driving information in the route navigation information. This information is labeled as i, and i = 1, 2, ..., j, where j represents the number of non-straight driving information. Then, the real-time distance between the non-straight driving information i and the vehicle is determined based on the vehicle's real-time location information. The obtained real-time distance is then compared with the warning distance. If the real-time distance is less than the warning distance, corresponding navigation information is generated, and the driver is reminded based on the navigation information. If the real-time distance is greater than the warning distance, a normal monitoring signal is generated.

[0013] This invention provides an intelligent feedback control method for optimizing the human-computer interaction experience. Compared with existing technologies, it has the following advantages: This invention improves fatigue detection accuracy by weighting the theta wave power ratio and alpha wave intrusion rate compared to single-band analysis. It establishes personalized fatigue index judgment intervals through the statistical analysis of a large amount of driving data, solving the problem of differences in EEG baselines among different users. At the same time, it combines fatigue state information from other modalities for cross-validation to avoid misjudgment based on a single signal. Furthermore, it analyzes road conditions in real time, automatically generates optimized routes with shorter response times, and guides user decision-making through multimodal feedback. It supports user feedback on route optimization ("accept / reject") and dynamically adjusts navigation strategies accordingly. Attached Figure Description

[0014] Figure 1 This is a diagram illustrating the steps and methods of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 This application provides an intelligent feedback control method for optimizing human-computer interaction experience, which specifically includes the following steps: Step S1: The system acquires the vehicle's driving information in real time, including but not limited to vehicle speed, driving direction, latitude and longitude. Based on this, it determines whether the vehicle has activated the navigation function. Specifically, this is determined by detecting whether the vehicle has set a destination and whether it receives and executes navigation commands (such as turn prompts, distance prompts, etc.). If the system determines that the vehicle is using navigation, it will generate a navigation driving analysis signal. Conversely, if the vehicle is not using navigation, the system will generate an autonomous driving analysis signal. The system will process the different signals generated separately.

[0017] Step S2: Analyze the generated autonomous driving analysis signal. This involves using the driver's brainwave information within time t from an auxiliary device (specifically, an in-vehicle EEG headband). The corresponding brainwaves are acquired and analyzed for their corresponding frequency bands. This brainwave information is then matched with a frequency range reference standard to generate corresponding fatigue state information. The frequency range reference standard represents the frequency range corresponding to different brainwave frequency bands and the corresponding fatigue state. Specifically, this is manifested as follows: The frequency range corresponding to the wave is 8-13Hz. The fatigue state is manifested as the brain shifting from alertness to relaxation, which may indicate fatigue. The frequency range corresponding to the wave is 4-7Hz. The fatigue state is manifested as: decreased cortical arousal and slowed nerve activity. The frequency range corresponding to the wave is 0.5-3Hz. The fatigue state is manifested as the brain approaching a sleep state and cognitive function collapsing. The frequency range corresponding to the wave is 14-30Hz. The fatigue state is manifested as: difficulty in maintaining attention and depletion of neural resources. Simultaneously calculate the corresponding drivers within time t. The power proportion of the wave, denoted as Power ratio, specifically referring to the power ratio in EEG signals The proportion of wave power (frequency 4-7Hz) in the overall EEG signal power was calculated. The wave intrusion rate, denoted as Intrusion rate, specifically refers to the rate at which intrusion should not occur. Waves (frequency 8-13Hz, normally dominant when eyes are closed and relaxed) in a conscious, open-eye state. The proportion of wave anomalies that appear and occupy the area, then according to the formula The fatigue index FI corresponding to the driver was calculated, where 0.6 and 0.4 are respectively... Power ratio and The weight ratio corresponding to the intrusion rate is used to match the obtained fatigue index FI with the corresponding index judgment interval. The index judgment interval is obtained through a large amount of data statistics, and the specific value is set according to the actual situation to generate fatigue judgment information. The obtained fatigue state information is matched with the fatigue judgment information to determine whether they are the same. If they are the same, the corresponding fatigue monitoring result is generated. Otherwise, if they are different, the fatigue judgment is used as the standard to generate the fatigue monitoring result.

[0018] The system will calculate, in real time, the characteristic parameters of relevant waves in the driver's electroencephalogram (EEG) signal within a specific time interval t. Specifically, it will calculate... The proportion of the power of the wave (frequency range of 4-7Hz) in the total EEG signal power is denoted as Power ratio. Simultaneously, calculate... The proportion of waves (normally dominant when eyes are closed and relaxed, with a frequency range of 8-13Hz) that abnormally appear and occupy a position in a conscious, open-eye state is denoted as [the percentage of waves that would normally dominate when eyes are closed and relaxed]. Intrusion rate.

[0019] Based on the above two parameters, the system follows the formula To calculate the driver's fatigue index FI, where 0.6 and 0.4 are respectively... Power ratio and The weighting percentage corresponding to the intrusion rate; Suppose that the electroencephalogram (EEG) signals of a driver are monitored over a period of 3 minutes (i.e., time t is 3 minutes). Calculations show that... The wave power ratio is 0.3. The intrusion rate is 0.2. The driver's fatigue index, calculated using the formula, is 0.26. It is known that the fatigue index range established through extensive data statistics is: 0-0.2 for mild fatigue, 0.2-0.4 for moderate fatigue, and 0.4-1 for severe fatigue. Therefore, the driver's fatigue level is determined to be moderate.

[0020] Through statistical analysis of extensive EEG data from real-world driving scenarios, different fatigue index judgment intervals were established. For example, the fatigue index range was divided into mild fatigue, moderate fatigue, and severe fatigue intervals. The system matches the calculated fatigue index FI with these judgment intervals to generate corresponding fatigue judgment information. This fatigue judgment information obtained through EEG signal analysis is then compared with fatigue state information acquired through other methods. If they match, the corresponding fatigue detection result is directly generated; if they do not match, the fatigue judgment information based on EEG signal analysis is used to generate the final fatigue detection result.

[0021] The obtained fatigue detection results are compared with the corresponding warning results. The warning results here specifically represent the fatigue warning level of the corresponding fatigue detection results. If the fatigue detection results are greater than the warning results, a fatigue warning signal is generated and a corresponding warning message is issued. The warning message is given through an alarm sound. If the fatigue detection results are less than the warning results, a normal monitoring signal is generated.

[0022] Step S3: The system performs in-depth processing on the acquired navigation driving analysis signals. First, it accurately acquires the vehicle's real-time location information and, combined with the current route navigation information, continuously monitors and analyzes the driving route. This analysis comprehensively considers factors such as real-time road conditions (e.g., traffic congestion, accident-prone sections), road speed limits, and historical travel times for different road sections to determine if a better route exists. An optimized route primarily refers to a route that, compared to the currently planned route, allows the vehicle to reach its destination in a shorter time.

[0023] If the analysis determines that an optimized route exists, the system will generate an optimization feedback analysis signal; otherwise, it will generate a normal positioning monitoring signal. The system will perform targeted analysis on each of these two different signals.

[0024] When the system processes the generated optimization feedback analysis signal, it accurately obtains the corresponding optimized route and presents it to the driver in a clear and intuitive way (such as highlighting it on the in-vehicle navigation screen or providing voice prompts). At the same time, the system obtains the driver's feedback information in real time, which is mainly divided into accepting the signal and not accepting the signal.

[0025] If the system receives an acceptance signal from the driver, it will modify the current route navigation information in real time based on the optimized route and generate real-time monitoring information to continuously track the vehicle's driving situation on the new route, ensuring the accuracy and timeliness of navigation. If the system receives a non-acceptance signal, it will maintain the current route navigation information and continue to provide navigation guidance to the driver according to the original route.

[0026] For example, a driver is traveling from the city center to a suburban destination, and the vehicle is driving normally after activating navigation. During the journey, the system detects severe congestion ahead based on real-time location information, with an estimated long travel time. After comprehensively analyzing surrounding road information and real-time traffic conditions, the system determines that there is an optimized route that avoids the congested section, allowing the vehicle to reach its destination faster. Therefore, the system generates an optimization feedback analysis signal.

[0027] After acquiring the optimized route, the system displays the new route on the in-vehicle navigation screen with a highlighted green line and prompts the driver with a voice message: "Congestion detected ahead. A faster route has been found for you. Would you like to switch?" The driver interacts via voice, and the system receives and recognizes the corresponding voice, generating an "accept" message and issuing an acceptance signal. At this point, the system uses this optimized route as the standard, modifies the current route navigation information, generates real-time monitoring information, continuously monitors the vehicle's driving situation on the new route, and makes timely adjustments based on changes in road conditions.

[0028] Step S4: The system performs in-depth analysis of the generated normal positioning monitoring signals. Based on accurate real-time vehicle positioning information and closely combined with route navigation information, a comprehensive judgment is made. During this process, the system extracts relevant content related to non-straight-line driving from the route navigation information. This non-straight-line driving information covers the route information corresponding to situations such as turning, lane changing, or U-turns during vehicle operation. For ease of differentiation and management, the system sequentially labels these non-straight-line driving information entries as i, where i = 1, 2, ..., j, and j represents the total number of non-straight-line driving information entries.

[0029] Next, the system will determine the real-time distance between each non-linear driving information i and the vehicle's current position based on the vehicle's real-time location information. Specifically, it will start with the non-linear driving information labeled i=1 and analyze them sequentially. Afterward, the system will compare the obtained real-time distance with the preset warning distance.

[0030] If the real-time distance is less than the warning distance, it indicates that the vehicle is approaching a non-straight-line section of road. The system will generate corresponding navigation information and, based on this information, will clearly and explicitly remind the driver of the upcoming driving maneuver (e.g., "Prepare to turn right in 50 meters ahead"). If the real-time distance is greater than the warning distance, it means that the vehicle is still some distance from the non-straight-line section of road. The system will continue to generate normal monitoring signals and continue to routinely monitor the vehicle's driving status and route information.

[0031] Suppose a driver is driving along a pre-set navigation route. At a certain moment, the system analyzes the normal positioning monitoring signals. By combining the vehicle's real-time positioning information and route navigation information, it detects three instances of non-straight driving on the current route, labeled i=1 (left turn at the first intersection ahead), i=2 (lane change required after driving a certain distance), and i=3 (U-turn after passing one more intersection), i.e., j=3.

[0032] The system first takes the non-straight-line driving information (left turn at the first intersection ahead) marked i=1 as the target and determines the real-time distance between it and the vehicle to be 100 meters based on the vehicle's real-time positioning information. Assuming the preset warning distance is 80 meters, since 100 meters is greater than 80 meters, the system continues to generate normal monitoring signals and continuously monitors the vehicle's driving situation.

[0033] As the vehicle continues to travel, when the system detects that the non-straight-line driving information i=1 is 70 meters away from the vehicle in real time, since 70 meters is less than 80 meters, the system immediately generates navigation information and prompts the driver through the in-vehicle navigation voice prompt: "Prepare to turn left in 70 meters ahead." At the same time, the left turn direction is indicated by an arrow on the screen, and the steering wheel vibrates to remind the driver to prepare in advance.

[0034] The data in the above formulas are all calculated using numerical values, without substituting the units of the parameters. In addition, the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0035] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart feedback control method for optimizing human-computer interaction experience, characterized in that, include: The system acquires vehicle driving information and determines whether the vehicle is using navigation. If it is, a navigation driving analysis signal is generated; otherwise, an autonomous driving analysis signal is generated. The autonomous driving analysis signal is analyzed, and the driver's brainwaves are matched with fatigue range reference standards to generate fatigue state information. The fatigue index is calculated according to a formula, and fatigue judgment information is generated based on the index judgment interval. The analysis of autonomous driving signals includes: using the driver's EEG information within time t via auxiliary equipment, matching the obtained EEG information with a frequency range reference standard to generate corresponding fatigue state information; and acquiring the driver's corresponding... wave corresponding Power ratio and wavy Intrusion rate, according to the formula The fatigue index FI corresponding to the driver was calculated, where 0.6 and 0.4 are respectively... Power ratio and The weight ratio corresponding to the intrusion rate is used to match the obtained fatigue index FI with the corresponding index judgment interval to generate fatigue judgment information. Compare the fatigue state information and fatigue judgment information to see if they are the same. If they are not the same, use the fatigue judgment information as the standard to generate fatigue monitoring results, and compare them with the early warning results to generate early warning information or normal monitoring signals. Generating early warning information or normal monitoring signals includes: matching the obtained fatigue state information with the fatigue judgment information to determine whether the two are the same. If they are the same, the corresponding fatigue monitoring result is generated; otherwise, if they are different, the fatigue judgment is used as the standard to generate the fatigue monitoring result. The obtained fatigue detection result is compared with the corresponding early warning result. If the fatigue detection result is greater than the early warning result, a fatigue early warning signal is generated; otherwise, a normal monitoring signal is generated. The navigation driving analysis signal is analyzed, and the vehicle's real-time positioning information and route navigation information are used to determine whether there is an optimized route, and generate an optimization feedback analysis signal or a normal positioning monitoring signal. The system analyzes the optimized feedback analysis signal, generates real-time monitoring information based on the feedback information, analyzes the normal positioning monitoring signal, determines the real-time distance based on the route navigation information, compares it with the warning distance, and generates navigation information and normal monitoring signal. The analysis of navigation driving analysis signals includes: acquiring real-time vehicle positioning information, and performing route monitoring analysis based on route navigation information to determine whether there is an optimized route. An optimized route mainly refers to a route that can make the vehicle reach its destination in a shorter time compared to the current planned route. If such a route exists, an optimization feedback analysis signal is generated; otherwise, a normal positioning monitoring signal is generated. The analysis of the optimization feedback signal includes: obtaining the corresponding optimized route and displaying it to the driver, while obtaining the driver's corresponding feedback information, including whether the signal is accepted or not. For the accepted signal, the current route navigation information is modified based on the optimized route, and real-time monitoring information is generated. For the unacceptable signal, the current route navigation information is maintained.

2. The intelligent feedback control method for optimizing human-computer interaction experience according to claim 1, characterized in that, The specific method for analyzing normal positioning monitoring signals is as follows: Based on the vehicle's real-time location information and route navigation information, a comprehensive judgment is made to obtain the corresponding non-straight driving information in the route navigation information. This information is labeled as i, and i = 1, 2, ..., j, where j represents the number of non-straight driving information. Then, the real-time distance between the non-straight driving information i and the vehicle is determined based on the vehicle's real-time location information. The obtained real-time distance is then compared with the warning distance. If the real-time distance is less than the warning distance, corresponding navigation information is generated, and the driver is reminded based on the navigation information. If the real-time distance is greater than the warning distance, a normal monitoring signal is generated.

Images