An electric wheelchair headrest adjustment method and device

Abstract

The present application provides a method and device for adjusting the headrest of an electric wheelchair, which relates to the technical field of wheelchair headrest adjustment. The key points of its technical solution are as follows: when it is detected that the user triggers the headrest adjustment instruction through a one-key operation, the current positioning data of the electric wheelchair, the head pressure data of the user, the headrest angle data, and the usage time data are obtained; the current usage scenario is identified according to the obtained data; the target position of the headrest is determined according to the identified current usage scenario and the pre-stored personalized parameters; the position difference between the current position and the target position of the headrest is calculated, and a headrest adjustment trajectory is generated based on this position difference; the headrest is controlled to perform an adjustment action according to the headrest adjustment trajectory, and at the same time, the force feedback and acceleration during the adjustment process are monitored. When an abnormality is detected, the adjustment is automatically stopped or the adjustment speed is reduced. The method and device for adjusting the headrest of an electric wheelchair provided by the present application have the advantages of simple operation and intelligent scenario recognition.

Description

Technical Field

[0001] This application relates to the technical field of wheelchair headrest adjustment. Specifically, it relates to an electric wheelchair headrest adjustment method and device. Background Art

[0002] In the existing electric wheelchair headrest adjustment technology, manual or simple electric adjustment methods are usually adopted. However, these methods are inconvenient for elderly users with limited hand mobility or mild cognitive impairment. For example, an elderly user who has been using an electric wheelchair for a long time needs to frequently adjust the headrest position in daily life. The user's family has set up multiple preset scenario modes for him, including eating, watching TV, and resting. However, due to the user's limited hand mobility, he is unable to precisely operate the complex control panel. In addition, the user also has mild cognitive impairment and has difficulty remembering the specific settings corresponding to different scenarios.

[0003] In this case, the user needs to quickly switch the headrest position between different scenarios, while also considering the emergency adjustment needs in case of sudden situations. More specifically, when the user switches from the eating mode to the watching TV mode, the headrest needs to be quickly and smoothly adjusted to a suitable angle and height while ensuring comfort. In addition, considering the user's safety, sudden shaking or pressure on the head needs to be avoided during the adjustment process.

[0004] The existing technology is difficult to meet these special needs and mainly has the following problems: complex operation, not suitable for users with limited hand mobility; lack of intelligent scenario recognition function and unable to automatically adapt to different usage scenarios; lack of guarantee for smoothness and safety during the adjustment process; unable to meet personalized needs and emergency adjustment requirements. These problems seriously affect the comfort and safety of electric wheelchair users, especially for those users with special needs, who urgently need a more intelligent, convenient and safe headrest adjustment method.

[0005] In view of the above problems, the existing technology urgently needs to be improved. Summary of the Invention

[0006] The purpose of this application is to provide an electric wheelchair headrest adjustment method and device, which have the advantages of simple operation, intelligent scenario recognition, smooth and safe adjustment, and meeting personalized needs.

[0007] This application provides an electric wheelchair headrest adjustment method, and the technical solution is as follows:

[0008] The method includes: when it is detected that the user triggers the headrest adjustment instruction through a one-key operation, obtaining the current positioning data of the electric wheelchair, the user's head pressure data, the headrest angle data, and the usage time data; identifying the current usage scenario according to the positioning data, the user's head pressure data, the headrest angle data, and the usage time data; determining the target position of the headrest according to the identified current usage scenario and the pre-stored personalized parameters; calculating the position difference between the current position and the target position of the headrest, and generating a headrest adjustment trajectory based on the position difference; controlling the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitoring the force feedback and acceleration during the adjustment process, and automatically stopping or reducing the adjustment speed when an abnormality is detected.

[0009] Further, the present application also proposes that the step of identifying the current usage scenario according to the positioning data, the user's head pressure data, the headrest angle data, and the usage time data includes: constructing a multi-dimensional feature space, and mapping the obtained positioning data, head pressure data, headrest angle data, and usage time data in the multi-dimensional feature space as feature vectors, where each dimension corresponds to a data type; analyzing the feature vectors to obtain a continuous scenario description value representing the current state of the user; dynamically determining the current usage scenario according to the continuous scenario description value, in combination with the stored user historical usage data and the physiologically state data collected in real time.

[0010] Further, the present application also proposes that the step of analyzing the feature vectors to obtain a continuous scenario description value representing the current state of the user includes: when it is detected that the user's state changes, analyzing the feature vectors to obtain an initial continuous scenario description value; performing time series correction on the initial continuous scenario description value based on the change trend of the historical continuous scenario description values; dynamically adjusting the weights of each feature according to the importance of different features in the current time period, where the features at least include one of the head pressure distribution, the headrest angle, and the usage time; optimizing the corrected continuous scenario description value in combination with the adjusted feature weights to obtain the final continuous scenario description value.

[0011] Further, the present application also proposes that the step of performing time series correction on the initial continuous scenario description value based on the change trend of the historical continuous scenario description values includes: obtaining a sequence of historical continuous scenario description values within a specified time window based on the change trend of the historical continuous scenario description values, and establishing a state transition probability matrix, where the state transition probability matrix reflects the conversion rule of the user between each preset scenario; calculating the confidence level of the initial continuous scenario description value according to the state transition probability matrix, and when the confidence level is lower than a preset threshold, correcting the initial continuous scenario description value based on the state transition probability matrix to obtain the corrected continuous scenario description value.

[0012] Furthermore, the present application also proposes that the step of optimizing the corrected continuous scene description value by combining the adjusted feature weights to obtain the final continuous scene description value includes: after obtaining the physiological state data of the user, calculating the physiological state matching degree of each preset scene based on the preset scene-physiological state correspondence relationship; using the physiological state matching degree as an additional feature and fusing it with the adjusted feature weights to obtain a comprehensive feature weight; and using the comprehensive feature weight to perform weighted averaging on the corrected continuous scene description value to obtain the final continuous scene description value.

[0013] Furthermore, the present application also proposes that the step of correcting the initial continuous scene description value based on the state transition probability matrix includes: after obtaining the historical scene transition data and the current physiological state data of the user, calculating the scene transition frequency of the user in different time periods based on the historical scene transition data, and dynamically adjusting a preset threshold according to the scene transition frequency and the current physiological state data; when the confidence level of the initial continuous scene description value is lower than the dynamically adjusted preset threshold, correcting the initial continuous scene description value based on the state transition probability matrix to obtain the corrected continuous scene description value.

[0014] Furthermore, the present application also proposes that the step of calculating the position difference between the current position and the target position of the headrest and generating a headrest adjustment trajectory based on the position difference includes: after obtaining the current position coordinates and the target position coordinates of the headrest, calculating a position difference vector based on the current position coordinates and the target position coordinates, and dynamically adjusting the maximum speed and maximum acceleration parameters in combination with the physiological state data of the user; according to the magnitude of the position difference vector and the adjusted maximum speed and maximum acceleration parameters, generating an S-shaped speed curve considering the neck bearing capacity of the user, where the S-shaped speed curve includes an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section; and based on the S-shaped speed curve and the position difference vector, calculating the intermediate position coordinate sequence during the adjustment of the headrest to form a smooth headrest adjustment trajectory adapted to the physiological characteristics of the user.

[0015] Furthermore, the present application also proposes that the step of generating an S-shaped speed curve considering the neck bearing capacity of the user according to the magnitude of the position difference vector and the adjusted maximum speed and maximum acceleration parameters includes: after obtaining the neck physiological parameters of the user, calculating the maximum safe acceleration of the user's neck based on the neck physiological parameters, and comparing it with the preset maximum acceleration parameter, and selecting the smaller value as the actual maximum acceleration parameter; according to the actual maximum acceleration parameter and the magnitude of the position difference vector, calculating the time and distance of the acceleration section and the deceleration section, and generating an S-shaped speed curve including an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section.

[0016] Further, the present application also proposes that the step of optimizing the corrected continuous scene description value by combining the adjusted feature weights to obtain the final continuous scene description value includes: obtaining the corrected continuous scene description value X = (x1, x2, ..., xn) and the adjusted feature weights W = (w1, w2, ..., wn), where xi represents the description value of the i-th preset scene, wi represents the weight of the i-th feature, n is the number of preset scenes, and real-time collecting the physiological state data P = (p1, p2, ..., pm) of the user, where pj represents the j-th physiological state parameter and m is the number of physiological state parameters; based on the pre-established scene-physiological state correspondence matrix S, calculating the physiological state matching degree M = S·P, where S is an n×m matrix, and fusing the feature weights and the physiological state matching degree to obtain the comprehensive feature weight W' = α·W + (1 - α)·normalize(M), where α is a balance parameter, 0 ≤ α ≤ 1, and normalize is a normalization function; calculating the final continuous scene description value Y = (y1, y2, ..., yn), where Y = softmax(β·X + (1 - β)·W'), and β is a balance parameter, 0 ≤ β ≤ 1.

[0017] Further, the present application also proposes an electric wheelchair headrest adjustment device, which includes: an acquisition module, configured to obtain the current positioning data of the electric wheelchair, the head pressure data of the user, the headrest angle data, and the usage time data when detecting that the user triggers a headrest adjustment instruction through a one-key operation; an identification module, configured to identify the current usage scene according to the positioning data, the head pressure data of the user, the headrest angle data, and the usage time data; a determination module, configured to determine the target position of the headrest according to the identified current usage scene and the pre-stored personalized parameters; a generation module, configured to calculate the position difference between the current position and the target position of the headrest, and generate a headrest adjustment trajectory based on the position difference; a control module, configured to control the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitor the force feedback and acceleration during the adjustment process, and automatically stop or reduce the adjustment speed when detecting an abnormality.

[0018] As can be seen from the above, an electric wheelchair headrest adjustment method and device provided by the present application simplify the operation process through one-key operation and intelligent scene recognition, and are suitable for users with limited hand mobility; through multi-dimensional data analysis and dynamic adjustment, intelligent scene recognition and personalized adjustment are realized; through generating a smooth adjustment trajectory and real-time monitoring, the smoothness and safety of the adjustment process are ensured, and it has the advantages of simple operation, intelligent scene recognition, smooth and safe adjustment, and meeting personalized needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1A flowchart of a method for adjusting the headrest of an electric wheelchair provided by this application.

[0020] Figure 2 A schematic structural diagram of a device for adjusting the headrest of an electric wheelchair provided by this application.

[0021] In the figure: 210, acquisition module; 220, recognition module; 230, determination module; 240, generation module; 250, control module. Specific embodiments

[0022] Next, the technical solutions in this application will be clearly and completely described in conjunction with the accompanying drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Usually, the components of this application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of this application that is required to be protected, but only represents the selected embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0023] It should be noted that: Similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0024] In actual applications, the electric wheelchair headrest adjustment system faces technical problems such as low intelligence, insufficient safety and comfort. Specifically, existing systems usually adopt manual or simple electric adjustment methods, which are inconvenient for elderly users with limited hand mobility or mild cognitive impairment. The system lacks an intelligent scene recognition function and cannot automatically adapt to different usage scenarios, resulting in users needing to frequently manually adjust the headrest position. In addition, the adjustment process lacks guarantees of smoothness and safety, which may cause discomfort or potential safety hazards to users. These problems seriously affect the usability of the system, the level of intelligence, and the ability to ensure user comfort and safety.

[0025] In response to this, referring to Figure 1 , this application proposes a method for adjusting the headrest of an electric wheelchair, and this method includes:

[0026] S110. When it is detected that the user triggers the headrest adjustment instruction through a one-key operation, obtain the current positioning data of the electric wheelchair, the head pressure data of the user, the headrest angle data, and the usage time data;

[0027] S120. Identify the current usage scenario based on the positioning data, the user's head pressure data, the headrest angle data, and the usage time data;

[0028] S130. Determine the target position of the headrest according to the identified current usage scenario and the pre-stored personalized parameters;

[0029] S140. Calculate the position difference between the current position and the target position of the headrest, and generate a headrest adjustment trajectory based on this position difference;

[0030] S150. Control the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitor the force feedback and acceleration during the adjustment process. When an abnormality is detected, automatically stop or reduce the adjustment speed.

[0031] Among them, the one-key operation refers to a way in which the user can trigger the headrest adjustment instruction through a simple single operation. Specifically, technologies such as touch buttons, voice commands, or gesture recognition can be used to implement it. This simplified operation method reduces the user's operation difficulty and is especially suitable for users with limited hand mobility.

[0032] Among them, the positioning data refers to the information of the current position of the electric wheelchair. Specifically, GPS, indoor positioning systems, or other positioning technologies can be used to obtain it. These data help the system judge the environment where the user is located, so as to more accurately identify the current usage scenario.

[0033] Among them, the head pressure data refers to the pressure distribution information generated when the user's head contacts the headrest. Specifically, a pressure sensor array can be used to collect it. These data reflect the posture and support needs of the user's head and are crucial for judging the user's comfort and adjustment needs.

[0034] Among them, the headrest angle data refers to the current tilt angle of the headrest. Specifically, an angle sensor or an encoder can be used to measure it. These data directly reflect the current state of the headrest and are an important reference basis for adjusting the position of the headrest.

[0035] Among them, the usage time data refers to the continuous usage time of the user in the current state. Specifically, it can be recorded through a timer or a timestamp. These data help the system judge whether the user needs to adjust the headrest to avoid discomfort caused by maintaining the same posture for a long time.

[0036] Among them, scenario recognition refers to automatically judging the user's current usage situation based on the acquired multi-dimensional data. Specifically, preset algorithms, machine learning algorithms such as decision trees or neural networks can be used to implement it. This intelligent recognition technology enables the system to automatically adapt to different usage scenarios without the user manually selecting a mode.

[0037] Among them, the personalized parameters refer to the headrest adjustment preferences preset for specific users, which can specifically include information such as the ideal headrest angles and heights in different scenarios. These parameters enable the system to make more precise adjustments according to the individual needs of users.

[0038] Among them, the headrest adjustment trajectory refers to the path of the headrest moving from the current position to the target position, which can specifically be generated by using curve interpolation or trajectory planning algorithms. This smooth adjustment trajectory ensures the comfort and safety of the adjustment process.

[0039] Among them, the force feedback refers to the detected resistance or pressure changes during the headrest adjustment process, which can specifically be measured by force sensors. These data help the system to timely detect abnormal situations during the adjustment process.

[0040] Among them, the acceleration refers to the rate of change of the headrest's movement speed during the adjustment process, which can specifically be measured by acceleration sensors. These data are used to monitor the smoothness of the adjustment process and prevent sudden jolts from causing discomfort to the user.

[0041] The core innovation of this application lies in the proposal to realize the automatic judgment of the user's usage status through data collection and intelligent scenario recognition technologies. Combining the pre-stored personalized parameters, the system can automatically determine the most suitable headrest target position for the user. During the adjustment process, by generating a smooth adjustment trajectory and real-time monitoring of force feedback and acceleration, the safety and comfort of the adjustment are ensured.

[0042] The working principle of this application can be divided into the following key steps:

[0043] First of all, when the user triggers the headrest adjustment instruction through a one-key operation, the system will start the data collection process, including obtaining the current positioning data of the electric wheelchair, the user's head pressure data, the headrest angle data, and the usage time data.

[0044] Next, these multi-dimensional data are input into the pre-trained scenario recognition algorithm. This algorithm can be based on machine learning technologies such as decision trees or neural networks, and can automatically judge the current usage scenario according to the input data features, such as whether it is the eating, watching TV, or resting state.

[0045] After identifying the current scenario, the system will combine the pre-stored personalized parameters to determine the target position of the headrest. These personalized parameters can include information such as the headrest angles and heights preferred by the user in different scenarios, and these information can be obtained through the user's usage history or manual settings, that is, different scenario templates can be preset.

[0046] The system will then calculate the position difference between the current position and the target position of the headrest. Based on this position difference, the system will generate a smooth adjustment trajectory to ensure the smoothness of the headrest movement process.

[0047] When performing the adjustment action, the system will control the movement of the headrest according to the generated trajectory. At the same time, the force sensor and the acceleration sensor will continuously monitor the force feedback and acceleration changes during the adjustment process. If an abnormal situation is detected, such as a sudden increase in force feedback or the acceleration exceeding the preset threshold, the system will immediately react and automatically stop or reduce the adjustment speed to ensure the safety and comfort of the user.

[0048] Specifically, in some specific embodiments, the relationships between the positioning data, the user's head pressure data, the headrest angle data, the usage time data and the current usage scenario are as follows:

[0049] Scene Location data Head pressure data Headrest angle data Usage time data Possible headrest adjustments Eating Kitchen / Dining room Low headrest pressure, head tilted forward Angle 0 - 15° Shorter time Adjust the headrest to a low angle to avoid interfering with eating Watching TV Living room Uniform headrest pressure Angle 30 - 45° Longer time Adjust the headrest to a comfortable angle and provide slow fine-tuning Taking a nap Bedroom or living room Higher headrest pressure, evenly distributed Angle 60 - 90° Time gradually increases Automatically adjust the headrest angle to a relaxing angle to prevent muscle tension Reading Living room / Study Headrest pressure concentrated at the lower part Angle 15 - 30° Moderate The headrest tilts slightly forward to provide good support Sleeping Bedroom Stable and high headrest pressure Angle 75 - 90° No change for a long time The headrest enters a long-term static mode to avoid frequent adjustments

[0050] In this way, the present application realizes an intelligent, safe and comfortable electric wheelchair headrest adjustment method, effectively solving the problems such as complex operation, lack of intelligent scenario recognition and lack of safety guarantee in the adjustment process in the traditional method.

[0051] In some of the above embodiments of the present application, steps for identifying the current usage scenario based on the positioning data, the user's head pressure data, the headrest angle data and the usage time data are proposed to determine the current usage scenario. However, in this process, how to achieve more accurate, continuous and personalized scenario recognition for the user during the adjustment of the electric wheelchair headrest to adapt to the transition state between different scenarios of the user and improve the accuracy and comfort of the headrest adjustment has become a problem.

[0052] In response to this, the present application further proposes that the steps for identifying the current usage scenario based on the positioning data, the user's head pressure data, the headrest angle data and the usage time data include: constructing a multi-dimensional feature space, and mapping the obtained positioning data, head pressure data, headrest angle data and usage time data in the multi-dimensional feature space as feature vectors, where each dimension corresponds to a data type respectively; analyzing the feature vectors to obtain a continuous scenario description value representing the current state of the user; and dynamically determining the current usage scenario according to the continuous scenario description value in combination with the stored user historical usage data and the physiologically state data collected in real time.

[0053] This application realizes the unified processing and analysis of multi-dimensional data by constructing a multi-dimensional feature space and mapping various data types into feature vectors. This method improves the data processing efficiency and provides a more comprehensive information basis for subsequent scene recognition. The analysis of the feature vectors yields continuous scene description values, and this representation method of continuous values can better reflect the subtle changes in the user's state compared to discrete scene classification, improving the accuracy and sensitivity of scene recognition. Finally, by combining the user's historical usage data and real-time physiological state data, the current usage scene is dynamically determined. This method takes into account the user's personalized characteristics and real-time state, making the scene recognition result more accurate and adaptable.

[0054] The technical solution of this application effectively solves the problems of low data processing efficiency and low scene recognition accuracy through the comprehensive analysis of multi-dimensional data and dynamic scene recognition. By introducing continuous scene description values and considering the user's historical data and real-time physiological state, this application can more accurately capture the dynamic changes in the user's state and provide a more accurate and adaptable scene recognition basis for subsequent headrest adjustment. This method not only improves the accuracy of scene recognition but also enhances the system's adaptability to the user's personalized needs, thus realizing a more intelligent and accurate headrest adjustment.

[0055] In this application, constructing a multi-dimensional feature space is a key step. This feature space can be realized in various ways. For example, dimensionality reduction techniques such as principal component analysis (PCA) or t-SNE can be used to construct a low-dimensional feature space, or the original data dimensions can be directly used to construct a high-dimensional feature space. Each dimension corresponds to a data type. For example, location data can include longitude and latitude coordinates, head pressure data can include a pressure distribution matrix, headrest angle data can include pitch angle and roll angle, and usage time data can include the current timestamp and the duration of continuous use.

[0056] When mapping the acquired data into the feature space, different normalization or standardization methods can be adopted to ensure the comparability of different types of data in the feature space. For example, Z-score standardization or Min-Max scaling can be used to process data with different dimensions.

[0057] Various machine learning or deep learning algorithms can be used for the analysis of feature vectors. For example, classification algorithms such as support vector machine (SVM) or random forest can be used to obtain discrete scene classification results, and then they can be converted into continuous scene description values through the softmax function. Or, a neural network model such as a multi-layer perceptron (MLP) or a long short-term memory network (LSTM) can be directly used to learn the mapping relationship from feature vectors to continuous scene description values.

[0058] The introduction of continuous scenario description values enables this application to capture subtle changes in the user's state. For example, if the user gradually transitions from a resting state to a state of watching TV, the continuous scenario description values can smoothly reflect this process, rather than suddenly jumping from one discrete state to another. This continuity makes the headrest adjustment smoother and more natural, enhancing the user's comfort.

[0059] Combining the user's historical usage data and real-time physiological state data to dynamically determine the current usage scenario further improves the accuracy and personalization of scenario recognition. Historical usage data can reflect the user's habits and preferences, while real-time physiological state data (such as heart rate, breathing rate, etc.) can reflect the user's immediate needs. The combination of these two types of data makes scenario recognition more comprehensive and accurate.

[0060] In the process of solving the problems of accurate, continuous, and personalized scenario recognition, the technical solution of this application realizes the unified processing of multi-source heterogeneous data through the construction of a multi-dimensional feature space, improving the data utilization efficiency. The introduction of continuous scenario description values enables the system to capture subtle changes in the user's state and adapt to the transitional states between different scenarios. The dynamic scenario determination mechanism that combines historical data and real-time physiological state data further improves the accuracy and personalization of scenario recognition.

[0061] Compared with traditional discrete scenario classification methods, this method can better adapt to the dynamic changes of the user's state. For example, when the user gradually transitions from a resting state to a state of watching TV, the headrest can be smoothly adjusted according to the continuously changing scenario description values, rather than suddenly jumping from one position to another. This not only enhances the user's comfort but also reduces the discomfort or safety risks that may be caused by sudden adjustments.

[0062] In addition, the method of this application achieves a higher degree of personalization by considering the user's historical usage data and real-time physiological state. For example, the system can learn that a certain user is more inclined to use a certain scenario mode during a specific time period, or requires a specific headrest adjustment in a certain physiological state. This personalized scenario recognition and adjustment greatly improve the adaptability of the system and the user experience.

[0063] As a preferred implementation manner, this application can be realized through the following steps:

[0064] First, construct a 5-dimensional feature space, corresponding to location data (longitude and latitude), head pressure data (pressure center coordinates), headrest angle data (pitch angle), and usage time data (current timestamp) respectively.

[0065] Then, preprocess and standardize the acquired data. For example, convert the latitude and longitude into offsets relative to the user's frequently used location, convert the head pressure data into offsets relative to the center of the headrest, convert the headrest angle data into angles relative to the horizontal position, and convert the usage time data into the percentage of the time of day.

[0066] Next, use a pre-trained deep neural network model to analyze the standardized feature vectors and output a continuous scene description value. This value can be a number between 0 and 1, where 0 represents a fully resting state, 1 represents a fully active state, and intermediate values represent transitional states.

[0067] Finally, combine the user's historical usage data (such as usage patterns at similar times and locations in the past week) and real-time physiological state data (such as heart rate and respiratory rate) to fine-tune the continuous scene description value. For example, if the historical data shows that the user is usually in a resting state at this time, but the real-time heart rate data shows that the user may be in an active state, the system will appropriately increase the scene description value.

[0068] In this way, the present application realizes accurate, continuous, and personalized scene recognition, provides a reliable basis for subsequent headrest adjustment, and thus improves the accuracy and comfort of the headrest adjustment of the electric wheelchair.

[0069] In some of the above embodiments of the present application, it is proposed to analyze the feature vectors to obtain a continuous scene description value representing the user's current state for identifying the current usage scene. However, there may be the following problems in this process: First, the initial continuous scene description value may not be accurate enough to fully reflect the dynamic changes of the user's state; Second, the importance of different features may change at different time periods and needs to be dynamically adjusted; Finally, simply relying on feature vector analysis may ignore the user's real-time physiological state information, resulting in the scene recognition result being incomplete and inaccurate.

[0070] In response to this, the present application further proposes that when it is detected that the user's state changes, analyze the feature vectors to obtain an initial continuous scene description value; perform temporal correction on the initial continuous scene description value based on the change trend of the historical continuous scene description values; dynamically adjust the weights of each feature according to the importance of different features in the current time period, where the features at least include one of the head pressure distribution, headrest angle, and usage time; and optimize the corrected continuous scene description value by combining the adjusted feature weights to obtain the final continuous scene description value.

[0071] The technical solution of this application effectively improves the accuracy and adaptability of scene recognition by introducing mechanisms such as timing correction, dynamic weight adjustment, and optimization of continuous scene description values. Specifically, the timing correction mechanism takes into account the change trend of historical continuous scene description values, avoids sudden changes in scene recognition results, and improves the stability of the system. The dynamic weight adjustment mechanism adjusts the weights according to the importance of each feature in different time periods, enabling the system to better adapt to the changing needs of users at different times of the day. The use of continuous scene description values can more precisely capture the transition state of users between different scenes, realizing smoother and more comfortable headrest adjustment.

[0072] Furthermore, the technical solution of this application focuses on parameters directly related to the adjustment of the electric wheelchair headrest, such as head pressure distribution, headrest angle, and usage time, ensuring a high correlation between the scene recognition result and the actual headrest adjustment requirements. This targeted parameter selection and optimization strategy enables the system to more accurately and timely capture the state changes of users during the use of the electric wheelchair, providing a more reliable basis for subsequent headrest adjustment.

[0073] In specific implementation, the acquisition of the initial continuous scene description value can be achieved in various ways. For example, a pre-trained neural network model can be used, with the feature vector as the input, and a multi-dimensional continuous scene description value is output. Another method is to use a clustering algorithm to map the feature vector into a predefined scene space, obtaining the probability distribution of each preset scene as the initial continuous scene description value.

[0074] The timing correction step can adopt the sliding window method combined with the exponentially weighted moving average (EWMA) algorithm. Specifically, a fixed-size queue of historical continuous scene description values can be maintained. When a new initial continuous scene description value is generated, it is subjected to EWMA calculation with the historical values in the queue to obtain the corrected continuous scene description value. This method takes into account both historical trends and can timely reflect the changes in the current state.

[0075] The dynamic weight adjustment can be implemented based on time periods and user behavior patterns. For example, a feature importance matrix for different time periods can be predefined, and combined with the user's historical usage data, the weights of each feature are dynamically adjusted. In addition, a reinforcement learning algorithm can be introduced to optimize the feature weight allocation strategy by continuously learning the user's feedback.

[0076] The optimization process of the final continuous scene description value can adopt weighted average or more complex fusion algorithms. For example, a soft voting mechanism can be used to fuse the corrected continuous scene description value and the scene probability distribution calculated based on feature weights to obtain the final continuous scene description value. This method not only retains the stability of timing correction but also incorporates the flexibility of dynamic weight adjustment.

[0077] In a specific embodiment, it is assumed that an electric wheelchair is equipped with a pressure sensor array, an angle sensor, and a timer, which are respectively used to collect head pressure distribution, headrest angle, and usage time data. The system collects data every 5 seconds and constructs a feature vector.

[0078] First, a pre-trained convolutional neural network model is used to process the feature vector to obtain an initial 5-dimensional continuous scene description value, which respectively corresponds to five preset scenes: rest, eating, watching TV, reading, and chatting. Then, the system maintains a sliding window containing the data of the most recent 10 minutes and uses the EWMA algorithm (α = 0.7) to perform temporal correction on the initial value.

[0079] Next, the system adjusts the feature weights according to the current time period (for example, the breakfast time period from 8:00 to 9:00). During this time period, the weights of the features related to eating (such as the head forward tilt angle) will be increased. Suppose the adjusted weights are: head pressure distribution 0.4, headrest angle 0.3, usage time 0.3.

[0080] Finally, the system combines the corrected continuous scene description value and the adjusted feature weights and uses the weighted average method to calculate the final continuous scene description value. For example, if the corrected value is [0.1, 0.6, 0.2, 0.05, 0.05], it may be adjusted to [0.08, 0.65, 0.18, 0.04, 0.05] after considering the feature weights, which more accurately reflects the state that the user may be in the eating scene.

[0081] Through this method, the technical solution of the present application can more accurately and timely capture the state changes of the user during the use of the electric wheelchair. For special user groups who have difficulty accurately expressing their needs, this improvement significantly improves the accuracy and comfort of headrest adjustment. For example, when the user switches from eating to watching TV, the system can smoothly adjust the headrest position, avoiding sudden changes that cause discomfort to the user. In addition, due to considering time factors and historical trends, the system can better predict and adapt to the changes in the user's needs, reducing unnecessary adjustment times and improving the user experience.

[0082] In some of the above embodiments of the present application, a step of performing temporal correction on the initial continuous scene description value based on the change trend of the historical continuous scene description value is proposed to improve the accuracy of scene recognition. However, in this process, there may be a situation where the initial continuous scene description value does not match the actual scene, resulting in inaccurate correction results. In addition, simply relying on historical data for correction may not be able to timely reflect sudden changes in the user's state, affecting the real-time performance and adaptability of scene recognition.

[0083] In response to this, the present application further proposes to obtain a sequence of historical consecutive scenario description values within a specified time window based on the change trend of historical consecutive scenario description values, and establish a state transition probability matrix, which reflects the conversion rules of users between various preset scenarios; calculate the confidence level of the initial consecutive scenario description value according to the state transition probability matrix, and when the confidence level is lower than the preset threshold, correct the initial consecutive scenario description value based on the state transition probability matrix to obtain the corrected consecutive scenario description value.

[0084] The present application collects historical scenario description values with time correlation by setting an appropriate time window, providing a data basis for subsequent analysis. This method is particularly suitable for elderly users with fixed daily routines. By establishing a state transition probability matrix, the conversion rules of elderly users between daily scenarios such as eating, watching TV, and resting are modeled. This method can capture the regularity of the behavior patterns of elderly users and improve the accuracy of scenario recognition.

[0085] By calculating the confidence level of the initial consecutive scenario description value and comparing it with a dynamically adjusted preset threshold, possible incorrect scenario judgments are identified. The setting of the dynamic threshold takes into account the degree of cognitive impairment of elderly users, enabling the system to better adapt to the characteristics of different users. When the confidence level of the recognition result is low, the state transition probability matrix is used for correction to make the scenario recognition result more in line with the actual usage habits of elderly users. This mechanism can effectively prevent misjudgments caused by sudden changes in sensor data or environmental interference.

[0086] The technical solution of the present application can be implemented in the following manner: First, based on the change trend of historical consecutive scenario description values, an appropriate time window is set, such as the most recent 24 hours or the most recent week. Within this time window, the system collects and stores the sequence of historical consecutive scenario description values of the user. These data can be stored in a database in the form of timestamps and corresponding scenario description values.

[0087] Next, based on the collected historical data, a state transition probability matrix is established. Assuming that the preset scenarios include n scenarios such as eating, watching TV, and resting, then the state transition probability matrix will be an n×n matrix. Each element P(i,j) in the matrix represents the probability of transitioning from scenario i to scenario j. This probability can be calculated by counting the frequency of scenario transitions in the historical data. For example, if in the historical data, the number of times the user transitions from the eating scenario to the watching TV scenario accounts for 30% of all the times the user transitions out of the eating scenario, then P(eating, watching TV)=0.3.

[0088] After obtaining the state transition probability matrix, the system can use this matrix to calculate the confidence of the initial continuous scene description value. A possible calculation method is to use the forward algorithm. Specifically, assuming that the initial continuous scene description value at the current moment is S_t and the scene at the previous moment is S_(t-1), then the confidence of S_t can be expressed as:

[0089] Confidence(S_t) = P(S_t | S_(t-1)) * Confidence(S_(t-1));

[0090] where P(S_t | S_(t-1)) can be directly obtained from the state transition probability matrix.

[0091] The system also needs to set a preset threshold for determining whether the initial continuous scene description value is credible. This threshold can be dynamically adjusted according to the characteristics of the user. For example, for users with mild cognitive impairment, a higher threshold, such as 0.8, can be set; while for users with severe cognitive impairment, a lower threshold, such as 0.6, can be set.

[0092] When the calculated confidence is lower than the preset threshold, the system will activate the correction mechanism. The correction process can use the maximum likelihood estimation method. Specifically, the system will calculate the most likely scene sequence under the current state transition probability matrix. This can be achieved through the Viterbi algorithm. The algorithm will consider the historical scene sequence and the currently observed data, find the most likely scene sequence, and use the last scene in this sequence to replace the initial continuous scene description value.

[0093] In this way, the technical solution of the present application can effectively improve the accuracy and stability of scene recognition, and reduce the impact of misjudgment on the user experience of elderly users. Especially for elderly users with limited hand mobility and mild cognitive impairment, this scene recognition method based on historical data and behavior patterns can better meet their needs, thus providing a more reliable basis for the precise adjustment of the headrest and improving the comfort and safety of using an electric wheelchair.

[0094] As a preferred implementation manner, the technical solution of the present application can be implemented in an intelligent electric wheelchair system. The system includes a central processing unit, multiple sensors (such as pressure sensors, angle sensors, GPS positioning modules, etc.), a data storage unit, and a headrest adjustment actuator.

[0095] In practical applications, the system can set a 24-hour sliding time window. Every 5 minutes, the system records the user's scenario description value and stores this data in the data storage unit. Assume that 5 scenarios are preset: eating, watching TV, resting, reading, and chatting. Then the state transition probability matrix will be a 5×5 matrix.

[0096] The system continuously updates this matrix. For example, if within the most recent 24 hours, the user switches from the eating scenario to the watching TV scenario 20 times, and the total number of times the user switches out of the eating scenario is 50 times, then P(eating, watching TV) = 20 / 50 = 0.4.

[0097] When the user triggers a headrest adjustment instruction, the system first obtains an initial continuous scenario description value based on the current sensor data. Assume that this value indicates that the user is most likely watching TV at present (with a probability of 0.7). The system calculates the confidence of this initial continuous scenario description value. If the user was eating at the previous moment, then the confidence might be:

[0098] Confidence = 0.7 * P(watching TV|eating) = 0.7 * 0.4 = 0.28;

[0099] Assume that the preset threshold set by the system for this elderly user with mild cognitive impairment is 0.6. Since 0.28 < 0.6, the system activates the correction mechanism. Through the Viterbi algorithm, the system may find that considering the historical data and the current observation, the most likely state sequence of the user is: eating -> resting -> resting. Therefore, the system corrects the initial "watching TV" judgment to "resting".

[0100] This correction mechanism can effectively prevent misjudgments caused by sudden interferences, improving the accuracy and stability of scenario recognition. At the same time, since the system continuously updates the state transition probability matrix, it can also adapt to the slow changes in the user's behavior pattern and maintain long-term recognition accuracy. This can significantly improve the usage experience and comfort for elderly users who need to use an electric wheelchair for a long time.

[0101] In some of the above embodiments of the present application, steps are proposed to optimize the corrected continuous scenario description value by combining the adjusted feature weights to obtain the final continuous scenario description value for optimizing the continuous scenario description value. However, in this process, only considering the feature weights may not fully reflect the user's real-time physiological state, resulting in insufficient accuracy of scenario recognition. In addition, a single adjustment of feature weights may not be able to adapt to the complex and changeable usage environment, affecting the precision and personalization of headrest adjustment.

[0102] In response to this, the present application further proposes that after obtaining the physiological state data of the user, calculating the physiological state matching degree of each preset scenario based on the preset scenario-physiological state correspondence relationship; using the physiological state matching degree as an additional feature and fusing it with the adjusted feature weights to obtain comprehensive feature weights; and using the comprehensive feature weights to perform weighted averaging on the corrected continuous scenario description values to obtain the final continuous scenario description values.

[0103] The technical solution of the present application realizes the integration of multi-dimensional information and improves the accuracy and personalization of scenario recognition by introducing physiological state data and the scenario-physiological state correspondence relationship. Specifically, the solution first obtains the physiological state data of the user, such as heart rate, respiratory rate, and body temperature. These data can reflect the real-time physiological state of the user and provide additional personalized information for scenario recognition.

[0104] Next, based on the pre-established scenario-physiological state correspondence relationship, calculate the physiological state matching degree of each preset scenario. This step takes into account the typical physiological characteristics that the user may exhibit in different scenarios, enhancing the pertinence of scenario recognition. For example, the eating scenario may be associated with specific heart rate and breathing patterns, while the resting scenario may correspond to different physiological indicators.

[0105] Use the calculated physiological state matching degree as a new feature and fuse it with the previously adjusted feature weights to obtain comprehensive feature weights. This fusion method realizes the integration of multi-dimensional information and makes scenario recognition more comprehensive and accurate. The fusion process can adopt weighted averaging or other more complex algorithms to ensure a reasonable balance between physiological state information and other feature information.

[0106] Finally, use the obtained comprehensive feature weights to perform weighted averaging on the corrected continuous scenario description values to obtain the final continuous scenario description values. This step comprehensively considers multiple information sources and improves the accuracy and reliability of scenario description.

[0107] The technical solution of the present application significantly improves the accuracy and personalization of scenario recognition by introducing physiological state data and the scenario-physiological state correspondence relationship. This method is particularly suitable for elderly users who have difficulty accurately expressing their needs. The system can assist in judging the current scenario through changes in their physiological states. For example, for an elderly user with mild cognitive impairment, even if he cannot clearly express his needs, the system can more accurately judge whether the user is in the eating, watching TV, or resting state by monitoring changes in physiological indicators such as his heart rate and respiratory rate, combined with the current time and environmental information.

[0108] In specific implementation, machine learning algorithms can be used to establish and optimize the correspondence between scenarios and physiological states. For example, algorithms such as Support Vector Machine (SVM) or Random Forest can be used to train a model based on a large amount of user data to establish the mapping relationship between different scenarios and physiological states. This model can be updated regularly to adapt to the long-term changes in the user's physiological state.

[0109] When calculating the physiological state matching degree, methods such as cosine similarity or Euclidean distance can be used to compare the similarity between the physiological state vector of the current user and the standard physiological state vector of the preset scenario. For example, assume that the system presets three scenarios: eating, watching TV, and resting, and each scenario has a corresponding standard physiological state vector. After obtaining the real-time physiological state data of the user, calculate the similarity between this data and the three standard vectors to obtain the matching degree of each scenario.

[0110] When fusing the feature weights and the physiological state matching degree, an adaptive weight algorithm can be used. For example, based on the user's historical data, the weights of the feature weights and the physiological state matching degree in the final decision can be dynamically adjusted. If it is found that the change in a user's physiological state is highly correlated with the change in the scenario, the system will increase the weight of the physiological state matching degree; conversely, if it is found that the change in the physiological state is not obvious, the system will correspondingly reduce its weight.

[0111] In this way, the technical solution of the present application can more comprehensively capture the state information of elderly users during the use of electric wheelchairs. This not only improves the accuracy of scenario recognition but also enhances the system's adaptability to individual differences of users. Finally, this optimized scenario description value can provide a more reliable basis for the precise adjustment of the electric wheelchair headrest, thereby improving the comfort and safety of elderly users when using electric wheelchairs in daily scenarios such as eating, watching TV, and resting.

[0112] Through this implementation method, the system can flexibly adjust the weights of physiological state information and other feature information in the final decision, so as to adapt to the individual differences of different users and the special needs of different scenarios. For example, for a user with heart disease, the system may increase the weight of the heart rate parameter to more sensitively capture possible abnormal conditions. For a user with sleep disorders, the system may pay more attention to the breathing frequency and body movement to more accurately identify the resting state.

[0113] This method not only improves the accuracy of scenario recognition but also enhances the adaptability and personalization of the system. By comprehensively considering multiple information sources, the system can better understand the actual needs of users, thereby providing more precise headrest adjustment services. This is of particularly important significance for elderly users with limited hand mobility and possible cognitive impairments, and can significantly improve their comfort and safety when using electric wheelchairs.

[0114] In some of the above embodiments of the present application, a step of correcting the initial continuous scene description value based on the state transition probability matrix is proposed to improve the accuracy of scene recognition. However, in this process, the fixity of the preset threshold may lead to insufficient adaptability of the system to changes in the user's behavior pattern. Especially when the user's physiological state changes, it may affect the sensitivity and accuracy of scene recognition.

[0115] In response to this, the present application further proposes that after obtaining the user's historical scene transition data and current physiological state data, calculate the scene transition frequency of the user in different time periods based on the historical scene transition data, and dynamically adjust the preset threshold according to the scene transition frequency and the current physiological state data; when the confidence level of the initial continuous scene description value is lower than the dynamically adjusted preset threshold, correct the initial continuous scene description value based on the state transition probability matrix to obtain the corrected continuous scene description value.

[0116] The technical solution of the present application improves the adaptability and accuracy of scene recognition by introducing a dynamic threshold adjustment mechanism. Specifically, the system first obtains the user's historical scene transition data and current physiological state data. The historical scene transition data may include the user's scene switching records in the past period (such as one week or one month), and the current physiological state data may include indicators such as heart rate, blood pressure, and body temperature.

[0117] Based on the historical scene transition data, the system calculates the scene transition frequency of the user in different time periods. This step can be achieved through statistical analysis methods, such as using time series analysis or sliding window techniques to calculate the number of scene transitions in each time period (such as every hour or every half day). The calculation of the scene transition frequency takes into account the time correlation of the user's behavior and can reflect the user's activity patterns at different times of the day.

[0118] Next, the system dynamically adjusts the preset threshold according to the calculated scene transition frequency and the current physiological state data. This process can be achieved by designing a weight function that takes the scene transition frequency and physiological state data as inputs and outputs an adjustment coefficient. For example, when the scene transition frequency is high, the system may lower the threshold to increase the sensitivity to scene changes; when the physiological state indicators of the user show fatigue or discomfort, the system may increase the threshold to reduce misjudgment.

[0119] After dynamically adjusting the threshold, the system compares the confidence level of the initial continuous scene description value with the new threshold. When the confidence level of the initial continuous scene description value is lower than the dynamically adjusted preset threshold, the system corrects the initial continuous scene description value based on the state transition probability matrix. This correction process can be achieved through matrix operations, multiplying the initial continuous scene description value by the state transition probability matrix to obtain the corrected scene description value.

[0120] This combined method of dynamic threshold adjustment and scene description value correction has multiple advantages. First, it improves the system's adaptability to changes in the user's behavior pattern. By considering the scene transition frequency in different time periods, the system can better capture the regular changes in the user's daily activities. Second, taking into account the current physiological state data enables the system to adjust the recognition sensitivity according to the user's real-time condition, which is particularly important for elderly users or users with volatile physical conditions. Finally, by combining dynamic threshold adjustment with correction based on the probability matrix, the system can effectively reduce misjudgments caused by environmental interference or sensor fluctuations while maintaining sensitivity.

[0121] In some of the above embodiments of the present application, steps of calculating the position difference between the current position and the target position of the headrest and generating a headrest adjustment trajectory based on the position difference are proposed to achieve the adjustment of the headrest. However, in this process, how to ensure the smoothness and safety of the adjustment process while considering the user's physiological characteristics and comfort has become an urgent problem to be solved. Especially for elderly users with limited hand mobility and mild cognitive impairment, how to avoid sudden head shaking or pressure while ensuring the adjustment effect has become a key technical challenge.

[0122] In response to this, the present application further proposes that after obtaining the current position coordinates and target position coordinates of the headrest, calculating a position difference vector based on the current position coordinates and target position coordinates, and dynamically adjusting the maximum speed and maximum acceleration parameters in combination with the user's physiological state data; generating an S-shaped speed curve considering the user's neck bearing capacity according to the magnitude of the position difference vector and the adjusted maximum speed and maximum acceleration parameters, where the S-shaped speed curve includes an initial slow acceleration section, a middle constant speed section, and a final slow deceleration section; calculating a sequence of intermediate position coordinates of the headrest during the adjustment process based on the S-shaped speed curve and the position difference vector to form a smooth headrest adjustment trajectory adapted to the user's physiological characteristics.

[0123] The technical solution of the present application provides a safe and comfortable headrest adjustment experience for elderly users with limited hand mobility and mild cognitive impairment through dynamic parameter adjustment, segmented speed control, smooth trajectory generation, and a safety guarantee mechanism. This method is particularly suitable for daily scenarios such as eating, watching TV, and resting. The system can generate the most suitable adjustment trajectory according to the user's real-time state, ensuring both the accuracy of the adjustment and avoiding unnecessary stimulation or harm to the user.

[0124] Specifically, the technical solution of the present application includes the following key features:

[0125] First, the dynamic parameter adjustment mechanism dynamically adjusts the maximum speed and maximum acceleration parameters by acquiring the user's physiological state data in real time. This can be achieved in a variety of ways, such as using wearable devices to collect the user's heart rate, blood pressure and other physiological data, or using the pressure sensor built into the headrest to detect the user's head posture and neck muscle tension. Based on this data, the system can use fuzzy logic or machine learning algorithms to determine the most suitable speed and acceleration parameters.

[0126] Secondly, the generation of the S-shaped speed curve takes into account the tolerance of the user's neck. In specific implementation, the entire adjustment process can be divided into three stages: the initial slow acceleration stage, the intermediate uniform speed stage, and the final slow deceleration stage. The time and distance of each stage can be dynamically calculated based on the size of the position difference vector and the user's physiological state. For example, for users with fragile necks, the system can extend the time of the acceleration and deceleration stages and shorten the time of the uniform speed stage to provide a smoother adjustment experience.

[0127] Finally, the calculation of the intermediate position coordinate sequence is based on the S-shaped velocity curve and the position difference vector. This can be achieved by numerical integration method, integrating the velocity curve in the time dimension to obtain the position of the headrest at each time point. In order to improve the computational efficiency, the Runge-Kutta method with adaptive step size can be used, using smaller step size at key points (such as acceleration change points) and larger step size in the uniform speed section.

[0128] Dynamic parameter adjustment directly affects the generation of the S-shaped speed curve, which in turn determines the calculation results of the intermediate position coordinate sequence. This design enables the entire adjustment process to accurately adapt to the user's real-time status and needs.

[0129] Through this design, the technical solution of the present application can effectively solve the problems existing in the prior art. Compared with simple linear adjustment, the S-shaped speed curve provides a smoother acceleration and deceleration process, significantly reducing the impact on the user's neck. The dynamic parameter adjustment mechanism ensures that the adjustment process can adapt to the changes in the user's physiological state in real time, further improving safety and comfort.

[0130] In some of the above embodiments of the present application, it is proposed to generate an S-shaped velocity curve that takes into account the user's neck bearing capacity according to the size of the position difference vector and the adjusted maximum velocity and maximum acceleration parameters to achieve smooth adjustment of the headrest. However, in this process, there is still the problem of how to more accurately consider the physiological characteristics of the user's neck and how to protect the user's neck safety to the greatest extent while ensuring the adjustment efficiency. Especially for some users with fragile necks, how to avoid possible discomfort or injury during the headrest adjustment process while achieving effective position adjustment is a technical problem that needs to be solved urgently.

[0131] In response to this, the present application further proposes that after obtaining the neck physiological parameters of the user, calculating the maximum safe acceleration of the user's neck based on the neck physiological parameters, comparing it with a preset maximum acceleration parameter, and selecting the smaller value as the actual maximum acceleration parameter; calculating the time and distance of the acceleration section and the deceleration section according to the actual maximum acceleration parameter and the magnitude of the position difference vector, and generating an S-shaped speed curve including an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section.

[0132] The technical solution of the present application calculates a personalized maximum safe acceleration by obtaining the neck physiological parameters of the user, takes into account the differences among elderly users, and ensures that the subsequent generated speed curve will not exceed the bearing capacity of the user's neck. This method not only considers the safety of the user, but also can adapt to the physiological characteristics of different users, providing a more personalized adjustment experience.

[0133] Specifically, the present application first obtains the neck physiological parameters of the user. These parameters may include factors such as neck muscle strength, cervical spine flexibility, age, etc. Based on these parameters, the system can use a preset algorithm or query a pre-established corresponding relationship table to calculate the maximum safe acceleration of the user's neck. For example, for a 65-year-old user with medium neck muscle strength and good cervical spine flexibility, the system may calculate that their maximum safe acceleration is 0.5m / s².

[0134] Furthermore, the system compares the calculated maximum safe acceleration with a preset maximum acceleration parameter. The preset maximum acceleration parameter may be set based on the general population, for example, 0.8m / s². The system will select the smaller of these two values as the actual maximum acceleration parameter. In this example, the system will select 0.5m / s² as the actual maximum acceleration parameter. This dynamic adjustment mechanism ensures that while the system guarantees the adjustment efficiency, it always puts the user's safety first.

[0135] According to the actual maximum acceleration parameter and the magnitude of the position difference vector, the system will accurately calculate the time and distance of the acceleration section and the deceleration section. For example, if the total distance that the headrest needs to move is 30cm, the system may calculate that each of the acceleration section and the deceleration section requires 5cm of distance and 1 second of time, and the intermediate constant speed section requires 20cm of distance and 2 seconds of time. This precise planning ensures the smoothness and controllability of the headrest adjustment process.

[0136] Based on the above calculations, the system generates an S-shaped speed curve including an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section. This curve design minimizes the impact on the neck of elderly users and provides a more comfortable adjustment experience.

[0137] Through this fine speed control, the technical solution of this application can provide a safe, comfortable and personalized headrest adjustment experience for elderly users with limited hand mobility and mild cognitive impairment. Especially in daily scenarios such as eating, watching TV, and resting, the system can generate the most suitable adjustment speed curve according to the individual characteristics of the user, ensuring both the accuracy and efficiency of the adjustment, and avoiding unnecessary irritation or harm to the user's neck.

[0138] As a preferred implementation, the system can further dynamically adjust the S-shaped speed curve by combining real-time monitored force feedback and acceleration data. For example, if abnormal tension in the user's neck muscles is detected during the adjustment process, the system can immediately reduce the current acceleration or speed, or even pause the adjustment process. This real-time feedback mechanism further improves the safety and comfort of the adjustment process.

[0139] Another possible implementation is that the system can gradually optimize and adjust the calculation model of the maximum safe acceleration according to the user's usage habits and preferences. For example, if the system finds that a certain user can tolerate a higher acceleration without discomfort during multiple adjustment processes, then the system can appropriately increase the maximum safe acceleration threshold for this user, thereby improving the adjustment efficiency while ensuring safety.

[0140] By precisely considering the physiological characteristics of the user's neck, dynamically adjusting the maximum acceleration parameter, and generating a personalized S-shaped speed curve, the technical solution of this application effectively solves the problem of how to maximize the protection of the user's neck safety while ensuring the adjustment efficiency. This method not only improves the safety and comfort of the headrest adjustment, but also can adapt to the individual differences of different users, providing a more considerate usage experience for elderly users. Through this refined control, the technical solution of this application can significantly reduce the neck discomfort or harm that may be caused by improper headrest adjustment during daily use, thereby improving the comfort and safety of elderly users using electric wheelchairs.

[0141] In some of the above embodiments of this application, steps are proposed to optimize the corrected continuous scene description value by combining the adjusted feature weights to obtain the final continuous scene description value for optimizing the continuous scene description value. However, in this process, only considering the feature weights may not fully reflect the user's real-time physiological state, resulting in insufficient accuracy of scene recognition. In addition, simple weighted averaging may not be able to effectively handle the complex relationships between features, affecting the accuracy of the final scene description value.

[0142] In response to this, the present application further proposes to obtain the corrected continuous scene description values X = (x1, x2, ..., xn) and the adjusted feature weights W = (w1, w2, ..., wn), where xi represents the description value of the i-th preset scene, wi represents the weight of the i-th feature, n is the number of preset scenes, and the physiological state data P = (p1, p2, ..., pm) of the user is collected in real time, where pj represents the j-th physiological state parameter and m is the number of physiological state parameters; based on the pre-established scene-physiological state correspondence matrix S, the physiological state matching degree M = S·P is calculated, where S is an n×m matrix, and the feature weights and the physiological state matching degree are fused to obtain the comprehensive feature weight W' = α·W + (1 - α)·normalize(M), where α is a balance parameter, 0 ≤ α ≤ 1, and normalize is a normalization function; the final continuous scene description values Y = (y1, y2, ..., yn) are calculated, where Y = softmax(β·X + (1 - β)·W'), and β is a balance parameter, 0 ≤ β ≤ 1.

[0143] The technical solution of the present application introduces multiple key technical features, including collecting the physiological state data of the user in real time, calculating the physiological state matching degree through the scene-physiological state correspondence matrix, using the balance parameter α to fuse the feature weights and the physiological state matching degree to obtain the comprehensive feature weight, and introducing the softmax function and the balance parameter β to calculate the final continuous scene description values. These technical features work together to effectively solve the problems existing in the previous solutions.

[0144] By introducing the physiological state data and the scene-physiological state correspondence matrix, the system can more comprehensively consider the real-time state of the user and improve the accuracy of scene recognition. The combination of the balance parameter and the softmax function enables the system to more flexibly handle the relationship between features and improves the accuracy of the final scene description values.

[0145] The technical solution of the present application has the following innovations: directly incorporating the physiological state data of the user into the scene recognition process to achieve more personalized and dynamic scene recognition; efficiently fusing multiple data sources through matrix operations and normalization processing to improve the calculation efficiency; introducing multiple balance parameters to make the system more adaptable and robust and capable of being optimized according to different users and scenes.

[0146] In the technical solution of the present application, first of all, collecting the physiological state data P = (p1, p2, ..., pm) of the user in real time is the basis of the entire solution. These data can be obtained through various sensors, such as heart rate sensors, blood pressure sensors, body temperature sensors, etc. The types and quantities of the collected data can be adjusted according to actual needs to adapt to different application scenarios.

[0147] The scene - physiological state correspondence matrix S is pre - established, which reflects the association between different scenes and the user's physiological state. This matrix can be obtained by training through machine learning algorithms based on a large amount of historical data, or constructed by an expert system according to medical knowledge and experience. Each row of matrix S corresponds to a preset scene, each column corresponds to a physiological state parameter, and the matrix element represents the contribution degree of a certain physiological state parameter to a certain scene.

[0148] Calculating the physiological state matching degree M = S·P is achieved through matrix multiplication. This step matches the user's real - time physiological state with the preset scenes to obtain the matching degree of each scene. The higher the matching degree, the more the user's current physiological state conforms to the scene.

[0149] Fusing the feature weight and the physiological state matching degree to get the comprehensive feature weight W' = α·W+(1 - α)·normalize(M) is one of the key steps in this solution. Here, the balance parameter α is introduced to adjust the relative importance of the feature weight W and the physiological state matching degree M. The normalize function is used to normalize M so that it is on the same order of magnitude as W. The value of α can be dynamically adjusted according to the actual application scenario. For example, when the user's physiological state fluctuates greatly, the value of α can be increased to enhance the influence of the physiological state matching degree.

[0150] Finally, calculating the final continuous scene description value Y = softmax(β·X+(1 - β)·W') introduces the softmax function and another balance parameter β. The use of the softmax function ensures that the output Y is a probability distribution, and the sum of all elements is 1, which is convenient for subsequent processing and decision - making. The β parameter is used to balance the influence of the corrected continuous scene description value X and the comprehensive feature weight W'.

[0151] There are close associations and interactions among these features. For example, the physiological state data P is transformed into the physiological state matching degree M through the scene - physiological state correspondence matrix S, and then combined with the feature weight W to form the comprehensive feature weight W'. This multi - level data fusion enables the system to capture the user's state and scene features more comprehensively. At the same time, by introducing the balance parameters α and β, the system has the ability to dynamically adjust the influence of various factors and can adaptively adjust the strategy according to different situations.

[0152] In practical applications, the parameter range can be set according to specific requirements. For example, α and β can be adjusted between 0.3 and 0.7 to balance the influence of different factors. The dimension of matrix S can be determined according to the number of preset scenes and the number of monitored physiological parameters. For example, a 10×5 matrix can represent 10 preset scenes and 5 physiological parameters.

[0153] When the technical solution of this application solves the problem of optimizing continuous scene description values, by collecting the physiological state data of users in real time, the system can timely capture the changes in the user's state. These data are combined with the pre-established scene-physiological state correspondence matrix to calculate the physiological state matching degree, so as to directly incorporate the user's real-time physiological state into the scene recognition process.

[0154] Next, the system obtains the comprehensive feature weight by fusing the feature weight and the physiological state matching degree. The key to this step is the introduction of the balance parameter α, which enables the system to dynamically adjust the relative importance of the feature weight and the physiological state matching degree according to different situations. For example, when the user is in a stable state, they may be more inclined to rely on the historical feature weight; while when the user's state changes significantly, the system can consider the real-time physiological state matching degree more.

[0155] When calculating the final continuous scene description value, the system further introduces the softmax function and the balance parameter β. The use of the softmax function ensures that the output result is a probability distribution, which not only makes the result easier to interpret but also provides a better basis for subsequent decisions. The introduction of the balance parameter β allows the system to find the best balance point between the corrected continuous scene description value and the comprehensive feature weight.

[0156] Compared with simple weighted averaging, this solution can better handle the complex relationships between features, improving the accuracy and stability of scene recognition. At the same time, by introducing multiple adjustable parameters, the system has stronger adaptability and flexibility, and can be adaptively adjusted according to the needs of different users, different environments, and different time periods.

[0157] Compared with the previous methods, the technical solution of this application enables the system to more accurately reflect the user's current situation by introducing real-time physiological state data, improving the accuracy of scene recognition. The introduced balance parameters α and β enable the system to dynamically adjust the weights of different factors according to the actual situation, enhancing the adaptability of the system. Using the softmax function to process the final output enables the system to better handle the non-linear relationships between features. Through matrix operations and normalization processing, efficient data fusion and processing are achieved. The system can be optimized according to the characteristics of different users and scenarios, providing a more personalized headrest adjustment service.

[0158] In the second aspect, referring to Figure 2 , this application further proposes an electric wheelchair headrest adjustment device, which includes:

[0159] An acquisition module 210, configured to obtain the current positioning data of the electric wheelchair, the head pressure data of the user, the headrest angle data, and the usage time data when detecting that the user triggers a headrest adjustment instruction through a one-key operation;

[0160] An identification module 220, configured to identify a current usage scenario according to positioning data, the user's head pressure data, headrest angle data, and usage time data;

[0161] A determination module 230, configured to determine a target position of the headrest according to the identified current usage scenario and pre-stored personalized parameters;

[0162] A generation module 240, configured to calculate a position difference between a current position and the target position of the headrest, and generate a headrest adjustment trajectory based on the position difference;

[0163] A control module 250, configured to control the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitor force feedback and acceleration during the adjustment process, and automatically stop or reduce the adjustment speed when an abnormality is detected.

[0164] Through one-key operation and intelligent scenario recognition, the operation process is simplified, which is suitable for users with limited hand mobility; through multi-dimensional data analysis and dynamic adjustment, intelligent scenario recognition and personalized adjustment are achieved; through generating a smooth adjustment trajectory and real-time monitoring, the smoothness and safety of the adjustment process are guaranteed, and it has the advantages of simple operation, intelligent scenario recognition, smooth and safe adjustment, and meeting personalized needs.

[0165] In addition, in some preferred embodiments, an electric wheelchair headrest adjustment device proposed by the present application can execute any one of the steps in the above method.

[0166] The above are only embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for adjusting the headrest of an electric wheelchair, characterized in that, The method includes: When it is detected that the user triggers a headrest adjustment instruction through a one-key operation, obtaining the current positioning data of the electric wheelchair, the user's head pressure data, the headrest angle data, and the usage time data; Identifying the current usage scenario according to the positioning data, the user's head pressure data, the headrest angle data, and the usage time data; Determining the target position of the headrest according to the identified current usage scenario and the pre-stored personalized parameters; Calculating the position difference between the current position and the target position of the headrest, and generating a headrest adjustment trajectory based on the position difference; Controlling the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitoring the force feedback and acceleration during the adjustment process, and automatically stopping or reducing the adjustment speed when an abnormality is detected; The step of identifying the current usage scenario according to the positioning data, the user's head pressure data, the headrest angle data, and the usage time data includes: Constructing a multi-dimensional feature space, and mapping the obtained positioning data, head pressure data, headrest angle data, and usage time data in the multi-dimensional feature space as feature vectors, where each dimension corresponds to a data type respectively; Analyzing the feature vectors to obtain a continuous scenario description value representing the current state of the user; Dynamically determining the current usage scenario according to the continuous scenario description value, in combination with the stored user historical usage data and the physiological state data collected in real time.

2. The method for adjusting the headrest of an electric wheelchair according to claim 1, wherein, The step of analyzing the feature vectors to obtain a continuous scenario description value representing the current state of the user includes: When it is detected that the user state changes, analyzing the feature vectors to obtain an initial continuous scenario description value; Performing time series correction on the initial continuous scenario description value based on the change trend of the historical continuous scenario description value; Dynamically adjusting the weights of each feature according to the importance of different features in the current time period, where the features at least include one of head pressure distribution, headrest angle, and usage time; Combining the adjusted feature weights to optimize the corrected continuous scenario description value to obtain the final continuous scenario description value.

3. The method for adjusting the headrest of an electric wheelchair according to claim 2, wherein The step of performing time series correction on the initial continuous scenario description value based on the change trend of the historical continuous scenario description value includes: Obtaining a sequence of historical continuous scenario description values within a specified time window based on the change trend of the historical continuous scenario description value, and establishing a state transition probability matrix, where the state transition probability matrix reflects the conversion rule of the user between each preset scenario; Calculating the confidence level of the initial continuous scenario description value according to the state transition probability matrix, and when the confidence level is lower than a preset threshold, correcting the initial continuous scenario description value based on the state transition probability matrix to obtain a corrected continuous scenario description value.

4. The electric wheelchair headrest adjustment method according to claim 2, characterized in that, The step of combining the adjusted feature weights to optimize the corrected continuous scenario description value to obtain the final continuous scenario description value includes: After obtaining the physiological state data of the user, calculating the physiological state matching degree of each preset scenario based on the preset scenario-physiological state correspondence; Taking the physiological state matching degree as an additional feature and fusing it with the adjusted feature weights to obtain a comprehensive feature weight; The weighted average of the corrected continuous scene description values is calculated using the comprehensive feature weights to obtain the final continuous scene description value.

5. The electric wheelchair headrest adjustment method according to claim 3, wherein The step of correcting the initial continuous scene description value based on the state transition probability matrix includes: After obtaining the user's historical scene transition data and current physiological state data, calculate the scene transition frequencies of the user in different time periods based on the historical scene transition data, and dynamically adjust the preset threshold according to the scene transition frequencies and the current physiological state data; When the confidence level of the initial continuous scene description value is lower than the preset threshold dynamically adjusted, correct the initial continuous scene description value based on the state transition probability matrix to obtain the corrected continuous scene description value.

6. The method for adjusting the headrest of an electric wheelchair according to claim 1, characterized in that The step of calculating the position difference between the current position and the target position of the headrest and generating a headrest adjustment trajectory based on the position difference includes: After obtaining the current position coordinates and target position coordinates of the headrest, calculate the position difference vector based on the current position coordinates and target position coordinates, and dynamically adjust the maximum speed and maximum acceleration parameters in combination with the user's physiological state data; According to the magnitude of the position difference vector and the adjusted maximum speed and maximum acceleration parameters, generate an S-shaped speed curve considering the user's neck bearing capacity, where the S-shaped speed curve includes an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section; Based on the S-shaped speed curve and the position difference vector, calculate the intermediate position coordinate sequence during the adjustment of the headrest to form a smooth headrest adjustment trajectory adapted to the user's physiological characteristics.

7. A method for adjusting a headrest of an electric wheelchair according to claim 6, characterized in that, The step of generating an S-shaped speed curve considering the user's neck bearing capacity according to the magnitude of the position difference vector and the adjusted maximum speed and maximum acceleration parameters includes: After obtaining the neck physiological parameters of the user, calculate the maximum safe acceleration of the user's neck based on the neck physiological parameters, and compare it with the preset maximum acceleration parameter, and select the smaller value as the actual maximum acceleration parameter; According to the actual maximum acceleration parameter and the magnitude of the position difference vector, calculate the time and distance of the acceleration section and the deceleration section, and generate an S-shaped speed curve including an initial slow acceleration section, an intermediate constant speed section, and a final slow deceleration section.

8. A method for adjusting the headrest of an electric wheelchair according to claim 2, characterized in that, The step of optimizing the corrected continuous scene description value in combination with the adjusted feature weights to obtain the final continuous scene description value includes: Obtain the corrected continuous scene description value X=(x1, x2, ..., xn) and the adjusted feature weights W=(w1, w2,..., wn), where xi represents the description value of the i-th preset scene, wi represents the weight of the i-th feature, n is the number of preset scenes, and collect the user's physiological state data P=(p1, p2, ..., pm) in real time, where pj represents the j-th physiological state parameter, and m is the number of physiological state parameters; Based on the pre-established scenario-physiological state correspondence matrix S, calculate the physiological state matching degree M = S·P, where S is an n×m matrix, and fuse the feature weight and the physiological state matching degree to obtain the comprehensive feature weight W' = α·W + (1 - α)·normalize(M), where α is a balance parameter, 0 ≤ α ≤ 1, and normalize is a normalization function; Calculate the final continuous scenario description value Y = (y1, y2,..., yn), where Y = softmax(β·X + (1 - β)·W'), and β is a balance parameter, 0 ≤ β ≤ 1.

9. An electric wheelchair headrest adjustment device for performing the method according to any one of claims 1 to 8, characterized in that, The device includes: An acquisition module, configured to obtain the current positioning data of the electric wheelchair, the user's head pressure data, the headrest angle data, and the usage time data when it is detected that the user triggers a headrest adjustment instruction through a one-key operation; An identification module, configured to identify the current usage scenario according to the positioning data, the user's head pressure data, the headrest angle data, and the usage time data; A determination module, configured to determine the target position of the headrest according to the identified current usage scenario and the pre-stored personalized parameters; A generation module, configured to calculate the position difference between the current position and the target position of the headrest, and generate a headrest adjustment trajectory based on the position difference; A control module, configured to control the headrest to perform an adjustment action according to the headrest adjustment trajectory, and simultaneously monitor the force feedback and acceleration during the adjustment process, and automatically stop or reduce the adjustment speed when an abnormality is detected.

Images