Carsickness identification method and device, computer equipment and storage medium

Abstract

The invention relates to a carsickness identification method and device, computer equipment and a storage medium, and relates to the technical field of vehicles, and the method in one embodiment comprises the steps: obtaining riding data of a target passenger identifier, and obtaining a riding signal feature corresponding to the target passenger identifier based on the riding data; comparing the riding signal feature with a baseline signal feature corresponding to the target passenger identifier to obtain a comparison result, and determining a carsickness identification result based on the comparison result; the baseline signal feature corresponding to the target passenger identifier is determined based on a first distribution feature and a second distribution feature of historical riding data corresponding to the target passenger identifier, and the first distribution feature is a distribution feature obtained by analyzing riding signal features of multiple pieces of historical riding data in a short first time window; the second distribution characteristics are distribution characteristics obtained by analyzing the riding signal characteristics of a plurality of pieces of historical riding data in a longer second time window. By adopting the scheme of the embodiment, the carsickness identification accuracy can be improved.

Description

Technical Field

[0001] This application relates to the field of vehicle interaction technology, and in particular to a motion sickness recognition method, device, computer equipment, storage medium, and computer program product. Background Technology

[0002] With the development of vehicle technology and the increase in vehicle ownership, driving or riding in vehicles has become an increasingly common travel need. While vehicles provide people with various intelligent and convenient experiences, people's demands for travel comfort are also rising. During a vehicle journey, some passengers may experience adverse symptoms such as sweating, dizziness, paleness, nausea, or even vomiting; these symptoms are collectively known as motion sickness. To improve motion sickness and enhance passenger comfort, it is necessary to identify whether passengers are experiencing motion sickness and the severity of their condition.

[0003] However, the accuracy of motion sickness recognition in related technologies is low, and there is a need to improve the accuracy of motion sickness recognition. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product that can accurately identify motion sickness in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a motion sickness detection method, the method comprising:

[0006] Acquire the passenger data of the target passenger identifier, and obtain the passenger signal characteristics corresponding to the target passenger identifier based on the passenger data;

[0007] The characteristics of the vehicle occupant signal are compared with the baseline signal characteristics corresponding to the target occupant identifier to obtain the comparison results, and the motion sickness identification result is determined based on the comparison results.

[0008] The baseline signal characteristics corresponding to the target occupant identifier are determined based on the first distribution characteristics and the second distribution characteristics of the historical travel data corresponding to the target occupant identifier. The first distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a first time window, and the second distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a second time window. The time length of the second time window is longer than the time length of the first time window, and the second time window includes the first time window.

[0009] In some embodiments, the determination of the baseline signal characteristics corresponding to the target occupant identifier includes:

[0010] Obtain the individual database corresponding to the target passenger identifier. The individual database includes multiple historical travel records corresponding to the target passenger identifier.

[0011] Multiple first-time historical travel data within the first time window of the target time are obtained from the individual database, and the travel signal characteristics of multiple first-time historical travel data are analyzed to obtain the first distribution characteristics;

[0012] Multiple second-historical travel data within the second time window of the target time are obtained from the individual database, and the travel signal characteristics of the multiple second-historical travel data are analyzed to obtain the second distribution characteristics;

[0013] Based on the first and second distribution features, the baseline signal features corresponding to the target occupant identifier are obtained.

[0014] In some embodiments, the ride signal characteristics of multiple first historical ride data are analyzed to obtain first distribution characteristics, including:

[0015] Based on the first forgetting factor corresponding to the first time window, the ride signal characteristics of multiple first historical ride data are calculated and processed to obtain the first distribution characteristics.

[0016] In some embodiments, the travel signal characteristics of multiple second historical travel data are analyzed to obtain second distribution characteristics, including:

[0017] Based on the second forgetting factor corresponding to the second time window, the ride signal characteristics of multiple second historical ride data are calculated and processed to obtain the second distribution characteristics.

[0018] In some embodiments, the ride signal characteristics of multiple first historical ride data are analyzed to obtain first distribution characteristics, including:

[0019] Based on the first forgetting factor corresponding to the first time window, the target travel signal features of the target historical travel data are used to update the historical first distribution features to obtain the first distribution features; the target historical travel data is the historical travel data closest to the target time among multiple first historical travel data corresponding to the first time window; the historical first distribution features are the travel signal features of the historical travel data of the first time window before the target historical travel data.

[0020] In some embodiments, the travel signal characteristics of multiple second historical travel data are analyzed to obtain second distribution characteristics, including:

[0021] Based on the second forgetting factor corresponding to the second time window, the target travel signal features of the target historical travel data are used to update the historical second distribution features to obtain the second distribution features; the target historical travel data is the historical travel data closest to the target time among multiple second historical travel data corresponding to the second time window; the historical second distribution features are the travel signal features of historical travel data in the second time window before the target historical travel data.

[0022] In some embodiments, the difference between 1 and the first forgetting factor is used as the weighting coefficient of the historical first distribution feature, and the first forgetting factor is used as the weighting coefficient of the target travel signal feature of the target historical travel data. The historical first distribution feature and the target travel signal feature are weighted and summed to obtain the first distribution feature.

[0023] In some embodiments, the difference between 1 and the second forgetting factor is used as the weighting coefficient of the historical second distribution feature, and the second forgetting factor is used as the weighting coefficient of the target travel signal feature of the target historical travel data. The historical second distribution feature and the target travel signal feature are weighted and summed to obtain the second distribution feature.

[0024] In some embodiments, baseline signal features corresponding to the target occupant identifier are obtained based on a first distribution feature and a second distribution feature, including:

[0025] Based on the first and second distribution characteristics, the adaptive weights are calculated and determined;

[0026] The first and second distribution features are weighted and summed based on adaptive weights to obtain the baseline signal features corresponding to the target occupant identifier.

[0027] In some embodiments, the fitness weights are calculated and determined based on a first distribution feature and a second distribution feature, including:

[0028] Based on the first and second distribution characteristics, the drift degree is calculated and determined.

[0029] The drift is processed by an exponential function to obtain the adaptive weights.

[0030] In some embodiments, the vehicle signal features are compared with the baseline signal features corresponding to the target occupant identifier to obtain a comparison result, and the motion sickness recognition result is determined based on the comparison result, including: calculating the feature difference between the vehicle signal features and the baseline signal features corresponding to the target occupant identifier; comparing the feature difference with a threshold corresponding to the vehicle signal features to obtain the motion sickness recognition result.

[0031] In some embodiments, the method further includes updating the threshold based on baseline signal features corresponding to the target occupant identifier.

[0032] In some embodiments, the baseline signal features include the baseline mean and the baseline variance; the threshold is updated based on the baseline signal features corresponding to the target occupant identifier, including:

[0033] The updated threshold is the sum of the product of the square root of the baseline variance and the threshold adjustment factor and the baseline mean.

[0034] In some embodiments, the method further includes: increasing the threshold adjustment coefficient when the motion sickness recognition results include motion sickness for a first preset number of consecutive times and no motion sickness confirmation feedback is received; and decreasing the threshold adjustment coefficient when the motion sickness recognition results include motion sickness for a second preset number of consecutive times and a motion sickness confirmation feedback is received.

[0035] Secondly, this application also provides a motion sickness detection device. The device includes:

[0036] The feature acquisition module is used to acquire the boarding data of the target occupant identifier and to acquire the boarding signal features corresponding to the target occupant identifier based on the boarding data;

[0037] The motion sickness confirmation module is used to compare the characteristics of the passenger signal with the baseline characteristics of the target passenger identifier, obtain the comparison result, and determine the motion sickness recognition result based on the comparison result. The baseline characteristics of the target passenger identifier are determined based on the first distribution characteristics and the second distribution characteristics of the historical passenger data corresponding to the target passenger identifier. The first distribution characteristics are obtained by analyzing the passenger signal characteristics of multiple historical passenger data within a first time window, and the second distribution characteristics are obtained by analyzing the passenger signal characteristics of multiple historical passenger data within a second time window. The time length of the second time window is longer than the time length of the first time window, and the second time window includes the first time window.

[0038] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the motion sickness recognition method in any of the embodiments described above.

[0039] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the motion sickness recognition method in any of the embodiments described above.

[0040] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the motion sickness recognition method in any of the embodiments described above.

[0041] The aforementioned motion sickness recognition method, device, computer equipment, storage medium, and computer program product, when performing motion sickness recognition, obtains the travel data of the target occupant identifier and, based on the travel data, obtains the corresponding travel signal characteristics, which are then compared with the baseline signal characteristics corresponding to the target occupant identifier. This achieves personalized motion sickness recognition based on the user corresponding to the target occupant identifier. Moreover, the baseline signal characteristics corresponding to the target occupant identifier are jointly determined based on the first distribution characteristics of historical travel data within a shorter first time window and the second distribution characteristics of historical travel data within a longer second time window. This simultaneously considers both the short-term and long-term changes in the historical travel data of the target occupant identifier, enabling the tracking of both short-term fluctuations and long-term stable trends of the user corresponding to the target occupant identifier. It achieves adaptive tracking of individual characteristics, improves the accuracy of the obtained baseline signal characteristics corresponding to the target occupant identifier, and further contributes to improving the accuracy of motion sickness recognition. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is an application environment diagram of the motion sickness recognition method in one embodiment;

[0044] Figure 2 This is a schematic diagram of the process for obtaining baseline signal characteristics in one embodiment;

[0045] Figure 3 This is a schematic diagram of the process for obtaining baseline signal characteristics in one embodiment;

[0046] Figure 4 This is a schematic diagram of the process for obtaining baseline signal characteristics in another embodiment;

[0047] Figure 5 This is a flowchart illustrating the process of obtaining baseline signal characteristics in another embodiment;

[0048] Figure 6 This is a flowchart illustrating the process of obtaining baseline signal characteristics in yet another embodiment;

[0049] Figure 7This is a schematic diagram of a process for obtaining baseline signal features based on a first distribution feature and a second distribution feature in one embodiment;

[0050] Figure 8 This is a structural block diagram of a motion sickness detection device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this application.

[0053] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0054] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. The term "multiple" as used in this application means two or more, unless otherwise explicitly defined. The terms "comprising" and "having," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusion. The term "and / or" is merely a description of the relationship between related objects, indicating one of the solutions or any combination of multiple solutions. Furthermore, the character " / " in this document generally indicates that the preceding and following related objects are in an "or" relationship.

[0055] It should be noted that all information and data involved in this application (including but not limited to data used for analysis, stored data, and displayed data) are information and data authorized by the user or fully authorized by all parties, and the acquisition, transmission, storage, use, and processing of related data comply with the relevant provisions of national laws and regulations. Users can refuse content pushed to them or can easily refuse content pushes. In the embodiments of this application, certain existing solutions in the industry, such as software, components, and models, may be mentioned. These should be considered exemplary, and their purpose is merely to illustrate the feasibility of implementing the technical solution of this application, but does not mean that the applicant has already used or necessarily used such a solution.

[0056] Currently, when identifying motion sickness in vehicle passengers, the method typically involves recognizing their physiological characteristics to determine whether they are experiencing motion sickness and the severity of their condition. This method usually compares the signal characteristics of different passengers with a pre-defined, identical baseline characteristic (also referred to as the baseline signal characteristic in this embodiment) and obtains the corresponding motion sickness identification result based on the comparison. However, in actual vehicle travel, different passengers exhibit individual differences. Using the same baseline signal characteristic as the judgment criterion for different passengers can easily lead to low accuracy. Furthermore, even for the same passenger, their individual characteristics can change over time. Therefore, using the same baseline signal characteristic as the criterion for determining whether someone is experiencing motion sickness for an extended period can also result in low accuracy.

[0057] Research has shown that while an individual passenger's characteristics generally tend to stabilize over a longer period, they can fluctuate over a shorter period. Therefore, combining long-term stable characteristics with short-term fluctuating characteristics to determine the passenger's baseline signal features can more accurately reflect their individual baseline signal characteristics. Based on this, motion sickness identification can be personalized and its accuracy further improved.

[0058] Accordingly, this application provides a motion sickness recognition method that can be applied to vehicles. This in-vehicle interaction method can be executed by the vehicle's controller, which can be a dedicated controller within the vehicle, a vehicle control unit (VCU), or a combination of the VCU and other controllers within the vehicle. This application does not impose any specific limitations on this method.

[0059] In one embodiment, such as Figure 1As shown, a motion sickness detection method is provided, including the following steps:

[0060] Step S101: Obtain the ride data of the target occupant identifier, and obtain the ride signal features corresponding to the target occupant identifier based on the ride data.

[0061] A target occupant identifier is a unique identifier corresponding to a vehicle occupant. The method for obtaining the target occupant identifier is not limited. For example, it can be obtained by taking an image of the occupant in the vehicle seat and identifying the corresponding user by recognizing their face in the image, then using that user's identifier as the target occupant identifier. Alternatively, the vehicle may be equipped with audio acquisition devices, fingerprint acquisition devices, etc., where voiceprint information identified from audio collected by the audio acquisition device, or fingerprint information collected by the fingerprint acquisition device, can uniquely identify a user, and that user's identifier can be used as the target occupant identifier. It is understood that in other embodiments, other methods can also be used to obtain the target occupant identifier, as long as it can uniquely identify a corresponding occupant. The specific type of the target occupant identifier is not limited, as long as it uniquely corresponds to a single user.

[0062] The types of signal features obtained from ride data are not limited. Some examples of ride signal features include, but are not limited to, facial expression features, posture features, heart rate features, and breathing movement features.

[0063] Facial expression features refer to characteristics related to an occupant's facial expressions. The methods for obtaining these features are not limited; for example, facial images of the target occupant can be captured using a camera (such as a high-resolution camera) (i.e., passenger data includes facial images), and facial expression recognition can be performed on the captured images to obtain facial expression features, such as raised eyebrows, frowning, and twitching of the corners of the mouth. The recognized facial expression features can include expression intensity and frequency.

[0064] Postural characteristics refer to information related to the posture of a target occupant. The methods for obtaining these characteristics are not limited. For example, images containing the occupant's head can be acquired using a camera (such as a high-resolution camera) (i.e., vehicle data includes head images), and posture characteristics can be obtained by analyzing these head images. Another example is capturing information about the relative posture changes between the occupant and the vehicle using signals from devices such as the seat IMU and vehicle posture sensors (i.e., vehicle data includes sensor signals from the seat IMU and vehicle posture sensors). The specific types of posture characteristics are not limited; for example, they can include head angular velocity and fluctuation amplitude. Postural characteristics can reflect the perceptual interference of vehicle dynamics changes on the occupant, thus serving as a basis for assessing the degree of motion sickness.

[0065] Heart rate characteristics refer to features related to the heart rate of a target occupant. There are no limitations on how heart rate characteristics can be obtained. For example, various non-invasive sensors, such as ECG straps and skin electrode patches, can be used to monitor and obtain corresponding signals in real time (i.e., passenger data includes signals collected by ECG straps and skin electrode patches), and heart rate characteristics can be obtained based on these signals. Another example is the acquisition of facial images of the target occupant using a camera (such as a high-resolution camera) (i.e., passenger data includes facial images), followed by analysis of the facial images using remote photoplethysmography to analyze changes in facial skin color difference, extract pulse wave signals from the video stream, and calculate heart rate and heart rate variability indicators, etc., but these methods are not limited to these.

[0066] Respiratory movement characteristics refer to features related to the breathing process of a target occupant. There are no limitations on the methods for obtaining respiratory movement characteristics. For example, the tiny rise and fall of the chest cavity caused by breathing can be detected by millimeter-wave radar built into the seat back (i.e., the occupant data includes radar signals from the millimeter-wave radar built into the seat back), and respiratory movement characteristics can be determined based on the signals captured by the millimeter-wave radar, but it is not limited to this.

[0067] Step S102: Compare the vehicle signal features with the baseline signal features corresponding to the target occupant identifier, obtain the comparison results, and determine the motion sickness recognition result based on the comparison results.

[0068] The baseline signal characteristics corresponding to the target occupant identifier are determined based on the first distribution characteristics and the second distribution characteristics of the historical travel data corresponding to the target occupant identifier. The first distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a first time window, and the second distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a second time window. The time length of the second time window is longer than that of the first time window, and the second time window includes the first time window.

[0069] Baseline signal features are a benchmark used to compare with the occupant's motion sickness features. By comparing the motion sickness features of the target occupant with the baseline signal features, the motion sickness identification result of the occupant corresponding to the target occupant can be determined.

[0070] The motion sickness recognition method described in the above embodiment, when performing motion sickness recognition, obtains the travel data of the target occupant identifier and, based on the travel data, obtains the corresponding travel signal characteristics, which are then compared with the baseline signal characteristics corresponding to the target occupant identifier. This achieves personalized motion sickness recognition based on the user corresponding to the target occupant identifier. Moreover, the baseline signal characteristics corresponding to the target occupant identifier are determined jointly based on the first distribution characteristics of each historical travel data within a shorter first time window and the second distribution characteristics of each historical travel data within a longer second time window. This simultaneously considers the changes in the historical travel data of the target occupant identifier in both the short and long term, enabling the tracking of both short-term fluctuations and long-term stable trends of the user corresponding to the target occupant identifier. It achieves adaptive tracking of individual characteristics, improves the accuracy of the obtained baseline signal characteristics corresponding to the target occupant identifier, and further contributes to improving the accuracy of motion sickness recognition.

[0071] As mentioned above, the baseline signal characteristics corresponding to the target occupant identifier can be obtained based on historical travel data in the individual database corresponding to the target occupant identifier. Accordingly, in some embodiments, reference is made to... Figure 2 As shown, the methods for determining the baseline signal characteristics corresponding to the target occupant identifier include:

[0072] Step S201: Obtain the individual database corresponding to the target passenger identifier. The individual database includes multiple historical travel data corresponding to the target passenger identifier.

[0073] The individual database corresponding to the target occupant identifier is a database used to store relevant data from the target occupant's historical travel history. The individual database contains multiple data entries, each containing historical travel data corresponding to the target occupant identifier. In some embodiments, each data entry in the individual database may also include travel signal features obtained by feature extraction from the historical travel data of that data entry.

[0074] Furthermore, in some embodiments, each data entry in the individual database may also include motion sickness identification results for that historical travel data.

[0075] In other embodiments, the individual database may include an individual motion sickness database and an individual non-motion sickness database: wherein the individual motion sickness database contains historical travel data in which motion sickness is identified, or contains historical travel data in which motion sickness is identified and user confirmation has been received; the individual non-motion sickness database contains historical travel data in which motion sickness is identified and no motion sickness has been identified, or contains historical travel data in which motion sickness is identified and user confirmation of no motion sickness has been received.

[0076] Step S202: Obtain multiple first historical travel data within the first time window of the target time from the individual database, and analyze the travel signal characteristics of the multiple first historical travel data to obtain the first distribution characteristics.

[0077] The length of the first time window is the time window within which the fluctuation of the occupant corresponding to the target occupant identifier can be assessed and determined. The specific length of the first time window is not limited, as long as the fluctuation of the occupant corresponding to the target occupant identifier can be assessed within this time window.

[0078] The target time is the time used to determine the first time window. There are no restrictions on how the target time is determined. For example, in some examples, it can be the time when the process of determining the baseline signal characteristics corresponding to the target occupant identifier begins, or the time when the update of the baseline signal characteristics corresponding to the target occupant identifier is triggered, but it is not limited to these.

[0079] Step S203: Obtain multiple second historical travel data within the second time window of the target time from the individual database, and analyze the travel signal characteristics of the multiple second historical travel data to obtain the second distribution characteristics.

[0080] The length of the second time window is the time window that can be used to assess and determine the long-term stability characteristics of the occupant corresponding to the target occupant identifier. The specific length of the second time window is not limited, as long as it can reflect the stability of the occupant corresponding to the target occupant identifier within the relatively long second time window. That is, the state of the occupant corresponding to the target occupant identifier can usually reflect the stability characteristics within the length of the second time window.

[0081] Step S204: Based on the first distribution feature and the second distribution feature, obtain the baseline signal feature corresponding to the target occupant identifier.

[0082] The motion sickness recognition method based on this embodiment determines motion sickness recognition by comparing the obtained travel signal features corresponding to the target occupant identifier with the baseline signal features corresponding to the target occupant identifier. This achieves personalized motion sickness recognition based on the user corresponding to the target occupant identifier. Furthermore, the baseline signal features corresponding to the target occupant identifier are jointly determined based on the first distribution features of historical travel data within a shorter first time window and the second distribution features of historical travel data within a longer second time window in the individual database of the target occupant identifier. This simultaneously considers both the short-term and long-term changes in the historical travel data of the target occupant identifier, enabling the tracking of both short-term fluctuations and long-term stable trends of the user corresponding to the target occupant identifier. This allows for adaptive tracking of individual characteristics, improves the accuracy of the obtained baseline signal features corresponding to the target occupant identifier, and further enhances the accuracy of motion sickness recognition.

[0083] The specific method for determining the first distribution feature based on multiple first historical travel data is not limited; in some embodiments, reference is made to... Figure 3 As shown, the step S202 above, which determines the first distribution feature based on multiple first historical travel data, includes:

[0084] Step S2021: Based on the first forgetting factor corresponding to the first time window, calculate and process the ride signal features of multiple first historical ride data to obtain the first distribution features.

[0085] The first forgetting factor is a factor that characterizes the forgetting of historical travel data during the process of obtaining the first distribution feature that reflects the volatility over a short period of time. It reflects the parameter of attenuation of historical travel data. By setting the first forgetting factor, more attention can be paid to the latest data in the process of determining the first distribution feature, so as to improve the parameter tracking ability and improve the accuracy of the obtained first distribution feature.

[0086] The range of the first forgetting factor is usually greater than 0 and less than 1. The specific value of the first forgetting factor is not limited. In some specific examples, the first forgetting factor can be set to 0.2, but it is not limited to this.

[0087] Based on the first forgetting factor corresponding to the first time window, the travel signal characteristics of multiple first historical travel data are calculated and processed to obtain the first distribution characteristics. The specific method for obtaining the first distribution characteristics is not limited; some examples may include:

[0088] Based on the distance between each first historical travel data point and the target time, and the first forgetting factor, the weight coefficient of each first historical travel data point is determined. For example, the exponential function of the first forgetting factor can be used as the weight coefficient of each first historical travel data point based on its distance from the target time. For example, in the relevant example, the weight of the first historical travel data point closest to the target time is the first forgetting factor, and the weight of the first historical travel data point second closest to the target time is a value determined based on the square of the first forgetting factor. However, it is not limited to this; as long as the distance from the target time is greater, the determined weight coefficient should be smaller.

[0089] Based on the weighting coefficients of each first historical travel data, the first historical travel data features of each first historical travel data are adjusted to obtain the adjusted first historical travel data features.

[0090] The adjusted first historical travel data features are processed by calculating the first distribution feature to obtain the first distribution feature. For example, in some examples, the first distribution feature may be obtained by calculating the mean or variance of the adjusted first historical travel data features.

[0091] Accordingly, when determining the first distribution feature based on multiple first historical travel data, the travel signal features of multiple first historical travel data are calculated and processed based on the first forgetting factor corresponding to the first time window. Thus, the first distribution feature is obtained on the basis of the set first forgetting factor, so that while considering the short-term fluctuations of the user corresponding to the target passenger identifier, the influence of the most recent historical travel data is considered as much as possible, which can further improve the accuracy of the short-term fluctuations reflected by the obtained first distribution feature.

[0092] In other embodiments, reference is made to... Figure 4 As shown, step S202 above involves analyzing the ride signal characteristics of multiple first historical ride data to obtain first distribution characteristics, including:

[0093] Step S2022: Based on the first forgetting factor corresponding to the first time window and the target travel signal features of the target historical travel data, update the historical first distribution features to obtain the first distribution features; the target historical travel data is the historical travel data closest to the target time among multiple first historical travel data corresponding to the first time window; the historical first distribution features are the travel signal features based on the travel data of the historical travel data of the first time window before the target historical travel data.

[0094] The first historical distribution feature refers to the first distribution feature determined based on historical travel data prior to the target historical travel data. In other words, it is the first distribution feature determined based on existing historical travel data before obtaining the target historical travel data. Therefore, after obtaining the target historical travel data, updating the first historical distribution feature based on the first forgetting factor and the target historical travel data can avoid the problem of large computational load and high resource consumption when performing full calculations on the full dataset within a new first time window based on the target historical travel data. This can improve the efficiency of obtaining and updating the first distribution feature.

[0095] Therefore, when determining the first distribution feature based on multiple first historical travel data, the first historical distribution feature is updated by combining the most recently obtained target historical travel data and the first forgetting factor. This allows the existing first historical distribution feature to be updated only based on the latest historical travel data (i.e., the target historical travel data) and the first forgetting factor after obtaining new historical travel data, without having to recalculate the entire dataset based on multiple historical travel data. This improves the efficiency of obtaining the first distribution feature.

[0096] Based on the target historical travel data and the first forgetting factor, the historical first distribution feature is updated. The specific method of obtaining the first distribution feature is not limited. In some specific examples, the difference between 1 and the first forgetting factor can be used as the weight coefficient of the historical first distribution feature, and the first forgetting factor can be used as the weight coefficient of the target travel signal feature of the target historical travel data. The historical first distribution feature and the target travel signal feature of the target historical travel data are weighted and summed to obtain the first distribution feature.

[0097] Therefore, the method of updating the historical primary distribution characteristics to obtain the primary distribution characteristics can be expressed by the formula:

[0098]

[0099] in, This represents the first distribution characteristic. This indicates the characteristics of the first historical distribution. Indicates the first forgetting factor. This represents the target travel signal characteristics of the target's historical travel data. Indicates the sample sequence number.

[0100] The number and type of specific features of the first distribution feature are not limited. In the relevant embodiments of this application, taking the distribution feature as an example that includes two features, the first distribution feature may include a first sub-feature of category one and a first sub-feature of category two. In a specific example, category one may be the mean and category two may be the variance. Then, the first sub-feature of category one may be the first mean and the first sub-feature of category two may be the first variance, but it is not limited to these.

[0101] Taking the first sub-feature of category 1 as the first mean and the first sub-feature of category 2 as the first variance as an example, the method of updating the historical first distribution feature to obtain the first distribution feature can be expressed by the formula:

[0102]

[0103]

[0104] in, This represents the first mean. Indicates the first forgetting factor. This represents the highest historical average. This represents the target travel signal characteristics of the target's historical travel data. Indicates the sample sequence number; Indicates the first variance. This represents the first variance in history.

[0105] Therefore, when updating the historical first distribution feature based on the target historical travel data and the first forgetting factor, the first forgetting factor is used as the weighting coefficient of the most recent target historical travel data. On this basis, the historical first distribution feature and the target travel signal feature of the target historical travel data are weighted and summed. This reflects that when considering short-term fluctuations, the first distribution feature obtained by forgetting the target historical travel data with the first forgetting factor can improve the accuracy of the obtained first distribution feature.

[0106] The specific method for analyzing the travel signal characteristics of multiple second historical travel data to obtain the second distribution characteristics is not limited. In some embodiments, reference is made to... Figure 5 As shown, step S203 involves analyzing the ride signal characteristics of multiple second historical ride data to obtain second distribution characteristics, including:

[0107] Step S2031: Based on the second forgetting factor corresponding to the second time window, calculate and process the ride signal features of multiple second historical ride data to obtain the second distribution features.

[0108] The second forgetting factor is a factor that characterizes the forgetting of historical travel data during the process of obtaining the second distribution feature that reflects stability over a longer period of time. It reflects the parameter of attenuation of historical travel data. By setting the second forgetting factor, more attention can be paid to the latest data in the process of determining the second distribution feature, so as to improve the parameter tracking ability and improve the accuracy of the obtained second distribution feature.

[0109] The range of the second forgetting factor is usually greater than 0 and less than 1. The specific value of the second forgetting factor is not limited. In some specific examples, the first forgetting factor can be set to 0.03, but it is not limited to this.

[0110] Based on the second forgetting factor corresponding to the second time window, the travel signal characteristics of multiple second historical travel data are calculated and processed to obtain the second distribution characteristics. The specific method for obtaining the second distribution characteristics is not limited; some examples may include:

[0111] Based on the distance between each second historical travel data point and the target time, and the second forgetting factor, the weight coefficient of each second historical travel data point is determined. For example, the exponential function of the second forgetting factor can be used as the weight coefficient of each second historical travel data point based on its distance from the target time. For instance, in the relevant example, the weight of the second historical travel data point closest to the target time (i.e., the target historical travel data) is the second forgetting factor, and the weight of the second historical travel data point second closest to the target time is a value determined based on the square of the second forgetting factor. However, this is not limited to this; as long as the distance from the target time is greater, the determined weight coefficient should be smaller.

[0112] Based on the weighting coefficients of each second historical travel data, the second historical travel data features of each second historical travel data are adjusted to obtain the adjusted second historical travel data features.

[0113] The adjusted second historical travel data features are then processed to calculate the second distribution feature, thus obtaining the second distribution feature. For example, in some examples, the second distribution feature might be obtained by calculating the mean or variance of the adjusted second historical travel data features.

[0114] Accordingly, when determining the second distribution feature based on multiple second historical travel data, the calculation and processing of multiple second historical travel data is based on the second forgetting factor corresponding to the second time window. Thus, the second distribution feature is obtained on the basis of the set second forgetting factor. This allows for consideration of the long-term stable trend of the user corresponding to the target passenger identifier while taking into account the influence of more recent historical travel data, thereby further improving the accuracy of the long-term stable trend feature reflected by the obtained second distribution feature.

[0115] In other embodiments, reference is made to... Figure 6 As shown, step S203 involves analyzing the ride signal characteristics of multiple second historical ride data to obtain second distribution characteristics, including:

[0116] Step S2032: Based on the second forgetting factor corresponding to the second time window and the target travel signal features of the target historical travel data, update the historical second distribution features to obtain the second distribution features; the target historical travel data is the historical travel data closest to the target time among multiple second historical travel data corresponding to the second time window; the historical second distribution features are the travel signal features of the historical travel data of the second time window before the target historical travel data.

[0117] The historical second distribution feature refers to the second distribution feature determined based on historical travel data prior to the target historical travel data. In other words, it refers to the second distribution feature determined based on existing historical travel data before obtaining the target historical travel data. Therefore, after obtaining the target historical travel data, the historical second distribution feature can be updated based on the second forgetting factor and the target historical travel data. This avoids the problem of large computational load and high resource consumption when performing full calculations on the full dataset within a new second time window based on the target historical travel data, and can improve the efficiency of obtaining and updating the second distribution feature.

[0118] Therefore, when determining the second distribution feature based on multiple second historical travel data, the historical second distribution feature is updated by combining the latest target historical travel data and the second forgetting factor. This allows the existing historical second distribution feature to be updated only based on the latest historical travel data and the second forgetting factor after obtaining new historical travel data, without having to recalculate the entire dataset based on multiple historical travel data, which helps to improve the efficiency of obtaining the second distribution feature.

[0119] Based on the target historical travel data and the second forgetting factor, the historical second distribution feature is updated. The specific method of obtaining the second distribution feature is not limited. In some specific examples, the difference between 1 and the second forgetting factor is used as the weight coefficient of the historical second distribution feature, and the second forgetting factor is used as the weight coefficient of the target travel signal feature of the target historical travel data. The historical second distribution feature and the target travel signal feature of the target historical travel data are weighted and summed to obtain the second distribution feature.

[0120] Therefore, the method for updating the historical secondary distribution characteristics to obtain the secondary distribution characteristics can be expressed by the formula:

[0121]

[0122] in, This represents the characteristics of the second distribution. This indicates the characteristics of the second historical distribution. Indicates the second forgetting factor. This represents the target travel signal characteristics of the target's historical travel data. .

[0123] The number and type of specific features of the second distribution feature are not limited. In the relevant embodiments of this application, taking the distribution feature as an example that includes two features, the second distribution feature may include a second sub-feature of category one and a second sub-feature of category two. In a specific example, category one may be the mean and category two may be the variance. Then, the second sub-feature of category one may be the second mean and the second sub-feature of category two may be the second variance, but it is not limited to these.

[0124] Taking the second sub-feature of category 1 as the second mean and the second sub-feature of category 2 as the second variance as an example, the method of updating the historical second distribution feature to obtain the second distribution feature can be expressed by the formula:

[0125]

[0126]

[0127] in, This represents the second mean. Indicates the second forgetting factor. This represents the second-highest historical average. This represents the target travel signal characteristics of the target's historical travel data. Indicates the sample sequence number; Indicates the second variance. This represents the second variance in history.

[0128] Therefore, when updating the historical second distribution feature based on the target historical travel data and the second forgetting factor, the second forgetting factor is used as the weighting coefficient of the most recent target historical travel data. On this basis, the historical second distribution feature and the target travel signal feature of the target historical travel data are weighted and summed. This reflects that, under the condition of considering the long-term stable trend, the second distribution feature when forgetting the target historical travel data with the second forgetting factor can improve the accuracy of the obtained second distribution feature.

[0129] refer to Figure 7 As shown, in some embodiments, obtaining the baseline signal features corresponding to the target occupant identifier based on the first distribution feature and the second distribution feature in step S203 above may include:

[0130] Step S701: Calculate and determine the adaptive weights based on the first distribution characteristics and the second distribution characteristics.

[0131] The adaptive weight is used to determine and balance the degree of adaptation between the first distribution feature corresponding to a shorter time window (first time window) and the second distribution feature corresponding to a longer time window (second time window). It reflects the adaptation between the volatility feature in the shorter time and the stability feature in the longer time.

[0132] Step S702: Perform weighted summation on the first distribution feature and the second distribution feature based on adaptive weights to obtain the baseline signal feature corresponding to the target occupant identifier.

[0133] Based on the adaptive weights, the first distribution feature and the second distribution feature can be weighted and summed to obtain the baseline signal feature corresponding to the target member identifier.

[0134] Taking a distribution feature that includes the mean and variance as an example, the calculated baseline signal features can include the baseline mean and the baseline variance. In this case, the first distribution feature corresponding to the first time window includes the first mean and the first variance, and the second distribution feature corresponding to the second time window includes the second mean and the second variance, which can be expressed by the formula:

[0135]

[0136]

[0137] in, Indicates the baseline mean. Indicates the adaptive weights, This represents the second mean corresponding to the second time window. This represents the first mean value corresponding to the first time window. Indicates the baseline variance. This represents the second variance corresponding to the second time window. This represents the first variance corresponding to the first time window.

[0138] Accordingly, when obtaining baseline signal features based on the first distribution feature and the second distribution feature, the adaptive weights are calculated based on the first distribution feature and the second distribution feature, and then the first distribution feature and the second distribution feature are weighted and summed based on the adaptive weights to obtain the corresponding baseline signal features. Since the calculated adaptive weights can reflect the degree of adaptation between the short-term fluctuations and long-term stable trends of the features of the user corresponding to the target occupant identifier, the accuracy of the obtained baseline signal features can be further improved.

[0139] The method for obtaining the adaptive weights is not limited. In a specific example, the adaptive weights are calculated and determined based on the first distribution feature and the second distribution feature, which may include:

[0140] Based on the first and second distribution characteristics, the drift degree is calculated and determined.

[0141] The drift is processed by an exponential function to obtain the adaptive weights.

[0142] The drift degree reflects the first distribution characteristic of short-term volatility over a shorter time period (first time window), and the degree of drift relative to the stability characteristic over a longer time period (second time window). Based on the obtained drift degree, further processing can be performed to obtain adaptive weights.

[0143] Taking distribution characteristics including mean and variance as an example, in a specific example, the drift can be the ratio of the absolute difference between the first mean corresponding to the first time window and the second mean corresponding to the second time window to the square root of the second variance corresponding to the second time window. This drift can be expressed by the formula:

[0144] ,in .

[0145] in, Indicates the degree of drift. This represents the first mean. This represents the second mean. This represents the second variance.

[0146] Based on the obtained drift, the drift can be further processed by an exponential function to obtain the fitness weights. In a specific example, the method for determining the fitness weights based on the drift can be expressed by the formula:

[0147] .

[0148] in, Indicates the adaptive weights, It is an exponential function. Indicates the degree of drift.

[0149] Accordingly, when calculating and determining the adaptive weights based on the first and second distribution features, they are obtained on the basis of the drift degree calculated based on the first and second distribution features. Since the drift degree reflects the degree of drift between the short-term fluctuations and long-term stable trends of the user characteristics corresponding to the target occupant identifier, obtaining the adaptive weights based on the calculated drift degree can effectively take into account the influence of the drift degree of short-term fluctuations and long-term stable trends. Calculating the adaptive weights accordingly can further improve the accuracy of the baseline signal features obtained based on the adaptive weights.

[0150] In some embodiments, the above-mentioned comparison of the occupant signal features with the baseline signal features corresponding to the target occupant identifier to obtain a comparison result, and the determination of the motion sickness recognition result based on the comparison result, may include:

[0151] Calculate the feature difference between the passenger signal features and the baseline signal features corresponding to the target occupant identifier;

[0152] The motion sickness identification result is obtained by comparing the feature difference with the threshold corresponding to the characteristics of the vehicle signal.

[0153] It is understandable that the different signal types in the motion sickness signal features correspond to different baseline signal features. Therefore, the features of different signal types can be compared with the corresponding baseline signal features to obtain their respective comparison results. These results are then compared with their corresponding thresholds, and a comprehensive assessment of the motion sickness recognition results is obtained based on the comparisons between each signal type's features and the thresholds.

[0154] Therefore, when obtaining motion sickness recognition results, the difference between the multimodal riding signal features and the baseline signal features corresponding to the target occupant identification is calculated, and the difference is compared with the corresponding threshold, which is simple and convenient.

[0155] In cases where the motion sickness signal features include features of different signal types, the motion sickness signal features of different signal types can be compared with the baseline signal features corresponding to that signal type to obtain the change in the signal features of that signal type. Based on the threshold range in which the change in the signal features is located (i.e., the range determined by the threshold), the sub-motion sickness level corresponding to that signal type can be determined. Based on the sub-motion sickness levels corresponding to each signal type, the final motion sickness level can be obtained by combining the results. In this case, the motion sickness identification result includes the final motion sickness level.

[0156] Taking the aforementioned multimodal vehicle-related signal features, including facial expression features, posture features, heart rate features, and respiratory movement features, as an example, the method for obtaining motion sickness recognition results can be as follows.

[0157] For facial expression features, the identified features (such as facial expressions, differences in facial skin tone, and eyelid closure) are compared with the baseline signal features of the corresponding facial expression features to determine the amount of change in the facial expression features. Based on the amount of change, a first motion sickness level is obtained based on the facial expression features. For example, if no nausea / vomiting-related expressions are detected, the amount of change in facial skin tone is within the normal range, and the amount of change in eyelid closure is within the normal range, the first motion sickness level is determined to be level 0; if a slight nausea expression is detected, or either the amount of change in facial skin tone or the amount of change in eyelid closure deviates slightly from the normal range, the first motion sickness level is determined to be level 1; if a nausea expression or near-vomiting action is detected, or either the amount of change in facial skin tone or the amount of change in eyelid closure deviates significantly from the normal range, the first motion sickness level is determined to be level 2, but this is not limited to these cases.

[0158] For posture features, the identified head posture features are compared with the baseline signal features of the corresponding head posture (head posture features in a normal sitting posture) to obtain a second motion sickness level based on head posture. For example, the current head angular velocity and the angle difference between the current head angle and the normal sitting reference angle are obtained; the second motion sickness level is determined based on the threshold range of the angular velocity and angle difference. In a specific example, if the angular velocity is less than or equal to the first angular velocity threshold and the angle difference is less than or equal to the first angle difference threshold, the second motion sickness level is determined to be mild disturbance (e.g., level 0); if the angular velocity is greater than the first angular velocity threshold and less than or equal to the second angular velocity threshold, or the angle difference is greater than the first angle difference threshold and less than or equal to the second angle difference threshold, the second motion sickness level is determined to be moderate disturbance (e.g., level 1); if the angular velocity is greater than the second angular velocity threshold, or the angle difference is greater than the second angle difference threshold, the second motion sickness level is determined to be severe disturbance (e.g., level 2), but this is not limited to these cases.

[0159] Based on heart rate characteristics, a third motion sickness level can be obtained. In a specific example, the difference between the current heart rate characteristic and the individual baseline heart rate characteristic corresponding to the target occupant's identifier can be calculated. The difference between the ratio of low-frequency to high-frequency variability in the current heart rate and the high-low frequency ratio of the baseline can also be calculated. Based on the intervals in which the heart rate difference and the high-low frequency ratio difference fall, the third motion sickness level is determined. For example, if the heart rate difference is less than or equal to a first heart rate difference threshold, and the high-low frequency ratio difference is less than or equal to a first ratio difference threshold, the third motion sickness level is determined to be mild (e.g., level 0). If the heart rate difference is greater than the first heart rate difference threshold and less than or equal to a second heart rate difference threshold, or the high-low frequency ratio difference is greater than the first ratio threshold and less than or equal to the second ratio threshold, the third motion sickness level is determined to be moderate (e.g., level 1). If the heart rate difference is greater than the second heart rate difference threshold, or the high-low frequency ratio difference is greater than the second ratio threshold, the third motion sickness level is determined to be severe (e.g., level 2).

[0160] Based on respiratory movement characteristics, a fourth motion sickness level can be obtained. In a specific example, the difference between the current respiratory rate and the baseline respiratory rate corresponding to the target occupant's identifier can be calculated, along with the rate of change of the current respiratory amplitude relative to the baseline. The fourth motion sickness level is determined based on the intervals within which the respiratory rate difference and the rate of change of respiratory amplitude fall. For instance, if the respiratory rate difference is less than or equal to the minimum first frequency difference threshold, and the rate of change of respiratory amplitude is less than or equal to the minimum first rate of change threshold, the fourth motion sickness level is determined to be mild (e.g., level 0). If the respiratory rate difference is greater than the minimum first frequency difference threshold and less than or equal to the second frequency difference threshold, or if the rate of change of respiratory amplitude is greater than the minimum first rate of change threshold and less than or equal to the second rate of change threshold, the fourth motion sickness level is determined to be moderate (e.g., level 1). If the respiratory rate difference is greater than the second frequency difference threshold, or if the rate of change of respiratory amplitude is greater than the second rate of change threshold, the fourth motion sickness level is determined to be severe (e.g., level 2).

[0161] After obtaining the first, second, third, and fourth motion sickness levels, the highest level among these four levels can be used as the final motion sickness level.

[0162] As described above, in the process of determining the level of motion sickness, the difference between the obtained signal features and the corresponding baseline signal features is compared with a corresponding threshold. Therefore, in the process of updating the baseline signal features, the thresholds can be further adjusted. Accordingly, in some embodiments, the method of this application further includes:

[0163] The threshold is updated based on the baseline signal characteristics corresponding to the target occupant identifier.

[0164] The method of adjusting the threshold based on the baseline signal characteristics corresponding to the target occupant identifier is not limited. Taking the above distribution characteristics including mean and variance as an example, the baseline signal characteristics include baseline mean and baseline variance. The adjusted threshold can be obtained based on the baseline mean, baseline variance, and threshold adjustment coefficient. Specifically, the product of the square root of the baseline variance and the threshold adjustment coefficient, plus the sum of the baseline mean, can be used as the updated threshold.

[0165] Taking the above-mentioned three levels of motion sickness as an example, for each signal characteristic, two corresponding thresholds can be set, and the adjusted thresholds can be expressed as:

[0166]

[0167]

[0168]

[0169] in, This represents a smaller threshold, such as the first angle difference threshold or the first heart rate difference threshold mentioned above. Indicates the baseline mean. Represents the square root of the baseline variance. Indicates the baseline variance. This indicates a larger threshold, such as the second angle difference threshold or the second heart rate difference threshold mentioned above. Indicates a smaller threshold The corresponding first threshold adjustment coefficient, Indicates a larger threshold The corresponding second threshold adjustment coefficient.

[0170] Accordingly, based on the first distribution characteristics of the shorter time window corresponding to the target occupant identifier and the second distribution characteristics of the longer time window, the threshold used to obtain the motion sickness identification result can be adjusted, which can realize the dynamic update of the threshold when comparing motion sickness, and can help to further improve the accuracy of the obtained motion sickness identification result.

[0171] In some embodiments, the method further includes:

[0172] If the motion sickness detection results include motion sickness for the first preset number of consecutive times, and no motion sickness confirmation feedback is received, the threshold adjustment coefficient is increased.

[0173] Therefore, if the motion sickness recognition result includes motion sickness for the first preset number of consecutive times, and no motion sickness confirmation feedback is received, it indicates that the motion sickness recognition result including motion sickness may be incorrect. This may be due to the threshold setting being too small. Therefore, the threshold adjustment coefficient can be increased to increase the threshold, which will help improve the accuracy of motion sickness recognition.

[0174] In some embodiments, the method further includes:

[0175] If motion sickness is detected in the second consecutive preset number of motion sickness detections and a motion sickness confirmation feedback is received, the threshold adjustment coefficient is reduced.

[0176] Therefore, if the motion sickness recognition result includes motion sickness in the second consecutive preset number of times and a motion sickness confirmation feedback is received, it indicates that the motion sickness recognition result including motion sickness is accurate. In order to improve the accuracy of motion sickness recognition and to help alleviate motion sickness and improve comfort when adjusting the driving process based on motion sickness recognition, the threshold adjustment coefficient can be reduced to reduce the threshold, which helps to improve the accuracy of motion sickness recognition and improve driving comfort.

[0177] The specific number of times the first preset number of times and the second preset number of times are not limited. For example, in some examples, the first preset number of times may include 3 times and the second preset number of times may include 2 times, etc., but it is not limited to this.

[0178] Based on the examples described above, the following specific examples will be used for illustration. The motion sickness recognition method provided in this application embodiment can realize a motion sickness recognition system based on multimodal features of in-vehicle cameras and non-contact sensors. By establishing a database of individual passengers and introducing a subjective feedback mechanism, dynamic and accurate recognition of motion sickness can be achieved. Specific examples are as follows.

[0179] After passengers are seated in the vehicle, or during vehicle operation, passengers can be identified to obtain passenger identification identifiers, thereby obtaining target passenger identifiers. Specifically, if the passenger is determined to be a new passenger based on the target passenger identifier (e.g., no matching passenger identifier is found for the passenger's facial image) or an old passenger who has not ridden the vehicle for a long time, the process of modeling the identified passenger is triggered.

[0180] In the process of modeling occupants, multimodal features of occupants can be collected after confirming that the occupants are not experiencing motion sickness or physical discomfort. The collected multimodal features are then used as the baseline signal features of the occupants, and the baseline signal features are bound to the target occupant identifier and stored.

[0181] If the target passenger identifier is neither a new passenger nor an old passenger who has not ridden for a long time, or if the target passenger identifier is a new passenger or an old passenger who has not ridden for a long time, but a baseline signal feature has been established and stored for them, then motion sickness can be identified based on the baseline signal feature corresponding to the passenger during the passenger's subsequent ride.

[0182] Specifically, during vehicle driving, passenger riding data can be collected and processed to obtain riding signal features corresponding to the target occupant identifier. The signal features of each signal type in the riding signal features are compared with the baseline signal features corresponding to the signal type to obtain the signal difference. The signal difference is then compared with the thresholds corresponding to the signal type to determine the interval in which the signal difference falls. Based on the interval in which the signal difference falls, the motion sickness level corresponding to each signal type is determined. The highest level of motion sickness level corresponding to each signal type is taken as the final motion sickness recognition result.

[0183] If the final motion sickness identification result indicates that the user is experiencing motion sickness, a prompt message can be output for confirmation, such as a voice prompt confirmation message: "Hello, are you feeling unwell or experiencing motion sickness?".

[0184] If positive feedback from the user is detected, such as the user replying with a voice message "yes" or "I am carsick", the carickness status is marked, and the obtained original travel data is stored as incremental data for the carickness sample library. The individual database corresponding to the target passenger identifier is then updated based on this incremental data.

[0185] If negative feedback from the user is detected, such as the user replying "no" via voice, the original travel data obtained will be stored as incremental data for the non-motion sickness sample database, and the individual database corresponding to the target passenger identifier will be updated.

[0186] When updating the individual database corresponding to the target occupant identifier, an online incremental learning approach can be adopted. Through a dual exponential sliding window mechanism, short-term physiological state fluctuations and long-term stable trends can be tracked separately. Based on the drift degree calculated in real time, the baseline signal characteristics and thresholds can be dynamically fused and adjusted, as follows.

[0187] The short-term window (the aforementioned first time window) is used to quickly respond to changes in physiological state during the most recent ride, and it is equipped with a corresponding first forgetting factor. First forgetting factor In a specific example, the value could be 0.2; the long-term window (the second time window mentioned above) is used to maintain a stable historical state and avoid baseline deviation due to occasional anomalies, and it has a corresponding second forgetting factor. The second forgetting factor In this specific example, the value could be 0.03. The short-term and long-term windows can be adjusted using adaptive weights. The data is then fused to form the final adaptive baseline signal features and decision threshold:

[0188] Taking the baseline signal characteristics, including mean and variance, as an example, the update process for the baseline signal characteristics can be as follows.

[0189] Based on the short-term window (the aforementioned first time window), the recursive method for updating the first distribution feature of the short-term window (the aforementioned first time window) can be as follows:

[0190]

[0191]

[0192] in, This represents the first mean. Indicates the first forgetting factor. This represents the highest historical average. This refers to the target historical travel data (i.e., the most recently identified historical travel data). Indicates the sample sequence number; Indicates the first variance. This represents the first variance in history.

[0193] Based on the long-term window (the second time window mentioned above), the recursive method for updating the second distribution feature of the long-term window (the second time window mentioned above) can be as follows:

[0194]

[0195]

[0196] in, This represents the second mean. Indicates the second forgetting factor. This represents the second-highest historical average. This refers to the target historical travel data (i.e., the most recently identified historical travel data). Indicates the sample sequence number; Indicates the second variance. This represents the second variance in history.

[0197] Define drift as ,in Then the fitness weight is .

[0198] Adaptive weights The closer it is to 1, the more significant the current drift and the greater the contribution of the short-term baseline.

[0199] In obtaining adaptive weights Based on this, it can be based on adaptive weights The first distribution feature of the short-term window (the first time window mentioned above) and the second distribution feature of the long-term window (the second time window mentioned above) are adaptively fused, which can be expressed by the formula:

[0200]

[0201]

[0202] in, This represents the baseline mean in the baseline signal characteristics. Indicates the adaptive weights, This represents the baseline variance in the baseline signal characteristics. This represents the second mean. This represents the first mean. Indicates the second variance. This represents the first variance.

[0203] The threshold can also be adjusted. Specifically, taking the example of two corresponding thresholds for signal characteristics, the adjusted threshold can be expressed as:

[0204]

[0205]

[0206]

[0207] in, This represents a smaller threshold, such as the first angle difference threshold or the first heart rate difference threshold mentioned above. Indicates the baseline mean. Represents the square root of the baseline variance. Indicates the baseline variance. This indicates a larger threshold, such as the second angle difference threshold or the second heart rate difference threshold mentioned above. Indicates a smaller threshold The corresponding first threshold adjustment coefficient, Indicates a larger threshold The corresponding second threshold adjustment coefficient.

[0208] Among them, under the condition that the threshold adjustment coefficient is satisfied, the first threshold adjustment coefficient can be adjusted. Second threshold adjustment coefficient Dynamic adjustments are made to dynamically adjust the sensitivity of motion sickness detection.

[0209] For example, in some cases, if the changes in the multimodal signal characteristics of the passenger corresponding to the target occupant identifier gradually weaken after multiple consecutive rides following motion sickness, that is, if the passenger corresponding to the target occupant identifier shows increased adaptability to the vehicle after multiple rides (such as a gradual weakening of changes in physiological indicators after motion sickness), then it can be confirmed that the threshold adjustment coefficient adjustment condition is met.

[0210] When adjusting the threshold adjustment coefficient, a specific example could be as follows:

[0211] If no motion sickness confirmation feedback is received for the first preset number of times (e.g., N ≥ 3), and the current multimodal signal features fall within a preset range, for example, in If the proportion of signal characteristics exceeds 80%, the trigger threshold is relaxed, that is, the threshold adjustment coefficient is increased, for example, by increasing the first threshold adjustment coefficient. Second threshold adjustment coefficient Taking an initial value of k1=1 and k2=2 as an example, the first threshold adjustment coefficient after threshold relaxation is... Second threshold adjustment coefficient They can be 1.2 and 2.4 respectively, i.e., k1=1.2, k2=2.4;

[0212] If at least two confirmed motion sickness events occur consecutively (for example, two times), and the number of abnormal samples exceeds the limit... If the range is exceeded, threshold tightening is triggered, i.e., the threshold adjustment coefficient is reduced, for example, by decreasing the first threshold adjustment coefficient. Second threshold adjustment coefficient Taking an initial value of k1=1 and k2=2 as an example, the reduced first threshold adjustment coefficient is... Second threshold adjustment coefficient They can be 0.8 and 1.6 respectively, i.e., k1=0.8 and k2=1.6.

[0213] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0214] Based on the same inventive concept, this application also provides a motion sickness recognition device for implementing the motion sickness recognition method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of the one or more motion sickness recognition device embodiments provided below can be found in the limitations of the motion sickness recognition method described above, and will not be repeated here.

[0215] In one embodiment, such as Figure 8 As shown, a motion sickness detection device is provided, including: a feature acquisition module 801 and a motion sickness confirmation module 802, wherein:

[0216] The feature acquisition module 801 is used to acquire the boarding data of the target occupant identifier and acquire the boarding signal features corresponding to the target occupant identifier based on the boarding data;

[0217] The motion sickness confirmation module 802 is used to compare the ride signal features with the baseline signal features corresponding to the target occupant identifier, obtain the comparison result, and determine the motion sickness recognition result based on the comparison result. The baseline signal features corresponding to the target occupant identifier are determined based on the first distribution feature and the second distribution feature of the historical ride data corresponding to the target occupant identifier. The first distribution feature is the distribution feature obtained by analyzing the ride signal features of multiple historical ride data within a first time window, and the second distribution feature is the distribution feature obtained by analyzing the ride signal features of multiple historical ride data within a second time window. The time length of the second time window is longer than the time length of the first time window, and the second time window includes the first time window.

[0218] In some embodiments, the apparatus further includes:

[0219] The baseline feature determination module is used to acquire an individual database corresponding to the target occupant identifier, the individual database including multiple historical travel data corresponding to the target occupant identifier; acquire multiple first historical travel data within a first time window of the target time from the individual database, and analyze the travel signal characteristics of the multiple first historical travel data to obtain a first distribution feature; acquire multiple second historical travel data within a second time window of the target time from the individual database, and analyze the travel signal characteristics of the multiple second historical travel data to obtain a second distribution feature; based on the first distribution feature and the second distribution feature, obtain the baseline signal feature corresponding to the target occupant identifier.

[0220] In some embodiments, the baseline feature determination module is used to calculate and process the ride signal features of multiple first historical ride data based on a first forgetting factor corresponding to a first time window, so as to obtain a first distribution feature.

[0221] In some embodiments, the baseline feature determination module is used to calculate and process the ride signal features of multiple second historical ride data based on a second forgetting factor corresponding to the second time window, so as to obtain a second distribution feature.

[0222] In some embodiments, the baseline feature determination module is used to update the historical first distribution feature based on the first forgetting factor corresponding to the first time window and the target ride signal feature of the target historical ride data to obtain the first distribution feature; the target historical ride data is the historical ride data closest to the target time among multiple first historical ride data corresponding to the first time window; the historical first distribution feature is the ride signal feature based on the ride signal feature of the historical ride data of the first time window before the target historical ride data.

[0223] In some embodiments, the baseline feature determination module is used to update the historical second distribution feature based on the second forgetting factor corresponding to the second time window and the target travel signal feature of the target historical travel data to obtain the second distribution feature; the target historical travel data is the historical travel data closest to the target time among multiple second historical travel data corresponding to the second time window; the historical second distribution feature is the travel signal feature based on the travel signal feature of the historical travel data of the second time window before the target historical travel data.

[0224] In some embodiments, the baseline feature determination module is used to take the difference between 1 and the first forgetting factor as the weighting coefficient of the historical first distribution feature, take the first forgetting factor as the weighting coefficient of the target travel signal feature of the target historical travel data, and perform a weighted summation of the historical first distribution feature and the target travel signal feature to obtain the first distribution feature.

[0225] In some embodiments, the baseline feature determination module is used to take the difference between 1 and the second forgetting factor as the weighting coefficient of the historical second distribution feature, take the second forgetting factor as the weighting coefficient of the target travel signal feature of the target historical travel data, and perform a weighted summation of the historical second distribution feature and the target travel signal feature to obtain the second distribution feature.

[0226] In some embodiments, the baseline feature determination module is used to calculate and determine adaptive weights based on first distribution features and second distribution features; and to perform weighted summation on the first distribution features and second distribution features based on the adaptive weights to obtain the baseline signal features corresponding to the target occupant identifier.

[0227] In some embodiments, the baseline feature determination module is used to calculate and determine the drift degree based on the first distribution feature and the second distribution feature; and to process the drift degree with an exponential function to obtain the adaptive weight.

[0228] In some embodiments, the motion sickness confirmation module 802 is used to calculate the feature difference between the ride signal feature and the baseline signal feature corresponding to the target occupant identifier; and compare the feature difference with the threshold corresponding to the ride signal feature to obtain the motion sickness recognition result.

[0229] In some embodiments, the baseline feature determination module is used to update the threshold based on the baseline signal features corresponding to the target occupant identifier.

[0230] In some embodiments, the baseline signal features include the baseline mean and the baseline variance; the baseline feature determination module is used to sum the product of the square root of the baseline variance and the threshold adjustment coefficient with the baseline mean as the updated threshold.

[0231] In some embodiments, the baseline feature determination module is used to increase the threshold adjustment coefficient when the motion sickness identification results include motion sickness for a first preset number of consecutive times and no motion sickness confirmation feedback information is received.

[0232] In some embodiments, the baseline feature determination module is used to reduce the threshold adjustment coefficient when the motion sickness identification results for a second consecutive preset number of times include motion sickness and motion sickness confirmation feedback information is received.

[0233] In an exemplary embodiment, a computer device is provided. This computer device can be a terminal, and its internal structure may include a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of this computer device provides computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of this computer device is used for exchanging information between the processor and external devices. The communication interface of this computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a motion sickness recognition method. The display unit of this computer device is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0234] Those skilled in the art will understand that the structure of the computer device described above is only a description of a part of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. A specific computer device may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements.

[0235] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the above embodiments.

[0236] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0237] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0238] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0239] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0240] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A motion sickness detection method, characterized in that, The method includes: Acquire the passenger data of the target passenger identifier, and obtain the passenger signal features corresponding to the target passenger identifier based on the passenger data; The vehicle occupant signal features are compared with the baseline signal features corresponding to the target occupant identifier to obtain a comparison result, and the motion sickness identification result is determined based on the comparison result. The baseline signal characteristics corresponding to the target occupant identifier are determined based on the first distribution characteristics and the second distribution characteristics of the historical travel data corresponding to the target occupant identifier. The first distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a first time window, and the second distribution characteristics are obtained by analyzing the travel signal characteristics of multiple historical travel data within a second time window. The time length of the second time window is greater than the time length of the first time window, and the second time window includes the first time window.

2. The method according to claim 1, characterized in that, The methods for determining the baseline signal features corresponding to the target occupant identifier include: Obtain the individual database corresponding to the target passenger identifier, wherein the individual database includes multiple historical travel data corresponding to the target passenger identifier; Multiple first historical travel data within the first time window of the target time are obtained from the individual database, and the travel signal characteristics of the multiple first historical travel data are analyzed to obtain the first distribution characteristics; Multiple second historical travel data within the second time window of the target time are obtained from the individual database, and the travel signal characteristics of the multiple second historical travel data are analyzed to obtain the second distribution characteristics; Based on the first distribution feature and the second distribution feature, the baseline signal feature corresponding to the target occupant identifier is obtained.

3. The method according to claim 2, characterized in that, Includes at least one of the following: First item: The step of analyzing the ride signal characteristics of the multiple first historical ride data to obtain the first distribution characteristics includes: Based on the first forgetting factor corresponding to the first time window, the ride signal features of the multiple first historical ride data are calculated and processed to obtain the first distribution feature; Second item: The step of analyzing the ride signal characteristics of the multiple second historical ride data to obtain the second distribution characteristics includes: Based on the second forgetting factor corresponding to the second time window, the ride signal characteristics of the multiple second historical ride data are calculated and processed to obtain the second distribution characteristics.

4. The method according to claim 2, characterized in that, Includes at least one of the following: First item: The step of analyzing the ride signal characteristics of the multiple first historical ride data to obtain the first distribution characteristics includes: Based on the first forgetting factor corresponding to the first time window, the target travel signal features of the target historical travel data are used to update the historical first distribution features to obtain the first distribution features; the target historical travel data is the historical travel data closest to the target time among the multiple first historical travel data corresponding to the first time window; the historical first distribution features are the travel signal features of historical travel data in the first time window before the target historical travel data; Second item: The step of analyzing the ride signal characteristics of the multiple second historical ride data to obtain the second distribution characteristics includes: Based on the second forgetting factor corresponding to the second time window, the target travel signal features of the target historical travel data are used to update the historical second distribution features to obtain the second distribution features; the target historical travel data is the historical travel data closest to the target time among the multiple second historical travel data corresponding to the second time window; the historical second distribution features are the travel signal features of historical travel data in the second time window before the target historical travel data.

5. The method according to claim 4, characterized in that, Includes at least one of the following: First item: The difference between 1 and the first forgetting factor is used as the weighting coefficient of the historical first distribution feature, and the first forgetting factor is used as the weighting coefficient of the target travel signal feature of the target historical travel data. The historical first distribution feature and the target travel signal feature are weighted and summed to obtain the first distribution feature. Second item: The difference between 1 and the second forgetting factor is used as the weighting coefficient of the historical second distribution feature, and the second forgetting factor is used as the weighting coefficient of the target travel signal feature of the target historical travel data. The historical second distribution feature and the target travel signal feature are weighted and summed to obtain the second distribution feature.

6. The method according to claim 2, characterized in that, The step of obtaining the baseline signal features corresponding to the target occupant identifier based on the first distribution feature and the second distribution feature includes: Based on the first distribution characteristics and the second distribution characteristics, the adaptive weights are calculated and determined; The first distribution feature and the second distribution feature are weighted and summed based on the adaptive weights to obtain the baseline signal features corresponding to the target occupant identifier.

7. The method according to claim 6, characterized in that, The step of calculating and determining the adaptive weights based on the first distribution feature and the second distribution feature includes: Based on the first distribution feature and the second distribution feature, the drift degree is calculated and determined; The adaptive weights are obtained by applying an exponential function to the drift degree.

8. The method according to any one of claims 1 to 7, characterized in that, The step of comparing the occupant signal features with the baseline signal features corresponding to the target occupant identifier, obtaining a comparison result, and determining the motion sickness recognition result based on the comparison result includes: Calculate the feature difference between the occupant signal feature and the baseline signal feature corresponding to the target occupant identifier; The motion sickness recognition result is obtained by comparing the feature difference with the threshold corresponding to the ride signal feature.

9. The method according to claim 8, characterized in that, The method further includes: The threshold is updated based on the baseline signal characteristics corresponding to the target occupant identifier.

10. The method according to claim 9, characterized in that, The baseline signal features include the baseline mean and baseline variance; updating the threshold based on the baseline signal features corresponding to the target occupant identifier includes: The updated threshold is the sum of the square root of the baseline variance and the threshold adjustment factor, and the baseline mean.

11. The method according to claim 10, characterized in that, The method further includes: If the motion sickness detection result includes motion sickness for the first preset number of consecutive times, and no motion sickness confirmation feedback is received, the threshold adjustment coefficient is increased. If motion sickness is detected in a second preset number of consecutive motion sickness detections and a motion sickness confirmation feedback is received, the threshold adjustment coefficient is reduced.

12. A motion sickness detection device, characterized in that, The device includes: The feature acquisition module is used to acquire the ride data of the target occupant identifier and acquire the ride signal feature corresponding to the target occupant identifier based on the ride data; The motion sickness confirmation module is used to compare the travel signal features with the baseline signal features corresponding to the target occupant identifier, obtain a comparison result, and determine a motion sickness recognition result based on the comparison result. The baseline signal features corresponding to the target occupant identifier are determined based on a first distribution feature and a second distribution feature of the historical travel data corresponding to the target occupant identifier. The first distribution feature is a distribution feature obtained by analyzing the travel signal features of multiple historical travel data within a first time window, and the second distribution feature is a distribution feature obtained by analyzing the travel signal features of multiple historical travel data within a second time window. The duration of the second time window is longer than the duration of the first time window, and the second time window includes the first time window.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 11.

15. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 11.

Images