Artificial intelligence-based state recognition navigation recommendation method and elderly care terminal device

Abstract

The application relates to the field of old-age care equipment, and discloses a state recognition navigation recommendation method based on artificial intelligence and an old-age terminal device. The method comprises the following steps: acquiring terrain features, speed features, gait features and physiological features of a user in multiple walking trips when the user uses an intelligent walking stick, and acquiring map information and a walking destination of the user; wherein the terrain features, the speed features and the gait features are detected by a motion detection component of the intelligent walking stick; through a state recognition-navigation recommendation model, a navigation route and a walking guidance suggestion of the user in the next walking trip are determined according to the terrain features, the speed features, the gait features, the physiological features, the map information and the walking destination in the multiple walking trips; and the walking guidance suggestion comprises matters needing attention information when the user walks. The application can identify and detect the physical state of the old people and improve the nursing effect on the old people.

Description

Technical Field

[0001] This application relates to the field of elderly care equipment technology, and more specifically, to an artificial intelligence-based state recognition navigation recommendation method and elderly care terminal equipment. Background Technology

[0002] Currently, most smart cane-based elderly care terminal devices adopt a single-function design, mainly integrating obstacle detection, fall alarm, and simple positioning functions. This results in high coupling between functional modules and a high false alarm rate. Although some products introduce manual alarms or basic navigation, they lack multi-device collaboration capabilities and dynamic service expansion mechanisms. At the hardware level, they rely on a closed architecture, making it difficult to be compatible with new sensors or third-party devices. Furthermore, data processing is mostly limited to local threshold judgments, failing to fully utilize cloud computing and navigation services. Consequently, smart cane-based elderly care terminal devices have weak environmental adaptability and cannot meet the needs of the elderly in complex outdoor scenarios for accurate obstacle avoidance, intelligent navigation, and multi-dimensional collaborative response in emergency situations.

[0003] Patent CN104257048B (application number: CN201410460857.1) discloses an elderly assistance system based on a smart cane, employing a three-layer architecture: the bottom hardware layer integrates an accelerometer, GPS, ultrasonic module, and wireless communication device, allowing connection to peripherals such as smart bracelets and fabrics; the middle service layer encapsulates the data processing functions of heterogeneous hardware through a unified interface, providing standardized service interfaces; the application layer implements three core functions—ultrasonic obstacle avoidance (real-time ranging and acoustic vibration warning), one-click navigation (cloud-based path planning and voice guidance), and intelligent fall detection (three-axis acceleration analysis combined with audible and visual alarms and SMS notifications)—by calling the bottom-layer services. While the smart cane in patent CN104257048B can improve the safety and convenience of elderly people's outdoor activities, it cannot identify or detect the elderly person's physical condition, resulting in poor care effectiveness. Summary of the Invention

[0004] The purpose of this application is to provide an artificial intelligence-based state recognition navigation recommendation method and elderly care terminal device, which solves the technical problems of not being able to identify and detect the physical condition of the elderly and the poor care effect of the elderly, and achieves the technical effect of being able to identify and detect the physical condition of the elderly and improving the care effect of the elderly.

[0005] This application provides an artificial intelligence-based state recognition navigation recommendation method. The method includes: acquiring terrain features, speed features, gait features, and physiological features of a user using a smart cane over multiple walking distances, and acquiring map information and the user's walking destination; wherein, the terrain features, speed features, and gait features are detected by the motion detection component of the smart cane, and the physiological features include the user's health record and real-time physiological features detected by wearable devices; through a state recognition-navigation recommendation model, based on the terrain features, speed features, gait features, physiological features, map information, and walking destination over multiple walking distances, determining the user's navigation route and walking guidance suggestions for the next walking distance, the walking guidance suggestions including information on precautions for the user while walking.

[0006] In one possible implementation, a state recognition-navigation recommendation model is used to determine the navigation route and walking guidance suggestions for the user's next walking journey based on terrain features, speed features, gait features, physiological features, map information, and walking destination within multiple walking journeys. This includes: determining state recognition features based on terrain features, speed features, gait features, and physiological features within multiple walking journeys using a state recognition unit; determining multiple navigation routes based on map information and walking destination using a navigation recommendation unit; and determining a route recommendation index and walking guidance suggestions for each navigation route based on the state recognition features and multiple navigation routes using a route selection unit, and recommending multiple navigation routes to the user in descending order of route recommendation index.

[0007] In another possible implementation, the method further includes: using a state recognition unit to determine current state recognition features based on terrain features, speed features, gait features, and physiological features within the current walking distance; and using a route selection unit to determine alternative navigation routes corresponding to the current walking distance and walking guidance suggestions corresponding to the alternative navigation routes based on the current state recognition features and multiple navigation routes, wherein the alternative navigation routes are used to replace the navigation routes corresponding to the current walking distance.

[0008] In another possible implementation, the method further includes: when the user's gait characteristics meet preset gait characteristic conditions and the user's physiological characteristics meet preset physiological characteristic conditions, the state recognition unit determines multiple trip segmentation points corresponding to the current navigation route based on the terrain features, speed features, gait features, and physiological features within the current navigation route; obtains the terrain features, speed features, gait features, and physiological features corresponding to the walking distance between two adjacent trip segmentation points, and determines the segmentation state recognition features between two adjacent trip segmentation points based on the terrain features, speed features, gait features, and physiological features corresponding to the walking distance between two adjacent trip segmentation points; the segmentation navigation recommendation unit determines multiple segmentation navigation routes based on the current navigation route; and the segmentation route selection unit determines the route recommendation index and walking guidance suggestions corresponding to each segmentation navigation route based on the segmentation state recognition features and the multiple segmentation navigation routes, and recommends multiple segmentation navigation routes to the user in descending order of the route recommendation index.

[0009] In another possible implementation, the method further includes: using a segmented route selection unit, based on segmentation state identification features and multiple segmented navigation routes, determining rest time periods between multiple adjacent segmented navigation routes and rest guidance suggestions corresponding to each rest time period; when the user walks along the target segmented navigation route, recommending the target rest time period corresponding to the target segmented navigation route and the rest guidance suggestions corresponding to the target rest time period to the user.

[0010] In another possible implementation, the method further includes: acquiring the target segmentation state identification features of the user within the target segmentation navigation route, and acquiring the user's rest physiological features during the rest period after the target segmentation navigation route; based on the target segmentation state identification features, rest physiological features, and multiple segmentation navigation routes, determining the subsequent rest suggestions and subsequent guidance segmentation navigation routes corresponding to the target segmentation navigation route, wherein the subsequent rest suggestions are used to indicate the rest suggestions for the user after the target segmentation navigation route.

[0011] In another possible implementation, the method further includes: using a state recognition unit to determine current state recognition features based on terrain features, speed features, gait features, and physiological features within the current walking journey; obtaining historical navigation routes that the user has previously followed and historical state recognition features corresponding to those historical navigation routes; using a navigation recommendation unit to determine multiple navigation routes based on map information and the walking destination; and using a route selection unit to determine alternative navigation routes and walking guidance suggestions corresponding to the current walking journey based on the current state recognition features, historical state recognition features, and multiple navigation routes, wherein the alternative navigation routes are used to replace the navigation routes corresponding to the current walking journey.

[0012] In another possible implementation, the method further includes: using a route selection unit to determine multiple trip split points corresponding to the current navigation route based on current state identification features and historical state identification features; using a segmented navigation recommendation unit to determine multiple segmented navigation routes based on the current navigation route, wherein the segmented navigation routes are alternative navigation routes between adjacent trip split points of the current navigation route; obtaining the current state identification features and historical state identification features corresponding to the walking distance between two adjacent trip split points, and using a segmented route selection unit to determine the route recommendation index and walking guidance suggestions corresponding to each segmented navigation route based on the current state identification features, historical state identification features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip split points, and recommending multiple segmented navigation routes to the user in descending order of route recommendation index.

[0013] In another possible implementation, the method further includes: using a route selection unit, determining rest periods between multiple adjacent segmented navigation routes and rest guidance suggestions corresponding to each rest period based on the current state recognition features, historical state recognition features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip segmentation points; when the user walks along the target segmented navigation route, recommending the target rest period corresponding to the target segmented navigation route and the rest guidance suggestions corresponding to the target rest period to the user.

[0014] This application also provides an elderly care terminal device, including a unit for executing the artificial intelligence-based state recognition navigation recommendation method as described in any of the preceding claims.

[0015] The beneficial effects of the embodiments in this application compared with the prior art are:

[0016] This application provides an AI-based state recognition navigation recommendation method. The method includes: acquiring terrain features, speed features, gait features, and physiological features of a user using a smart cane over multiple walking distances, and acquiring map information and the user's walking destination. The terrain features, speed features, and gait features are detected by the smart cane's motion detection component, and the physiological features include the user's health record and real-time physiological features detected by wearable devices. Using a state recognition-navigation recommendation model, based on the terrain features, speed features, gait features, physiological features, map information, and walking destination over multiple walking distances, the method determines the navigation route and walking guidance suggestions for the user's next walking distance. The walking guidance suggestions include information on precautions for the user while walking. This AI-based state recognition navigation recommendation method can identify the state of elderly users using canes, select the optimal navigation route from the navigation routes based on the state recognition results, and provide corresponding walking guidance suggestions for the elderly user's walking navigation route. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the first artificial intelligence-based state recognition navigation recommendation method provided in this application embodiment;

[0019] Figure 2 A schematic diagram illustrating the workflow of the first artificial intelligence-based state recognition navigation recommendation method provided in this application embodiment;

[0020] Figure 3 A flowchart illustrating the second AI-based state recognition navigation recommendation method provided in this application embodiment;

[0021] Figure 4 A flowchart illustrating the third AI-based state recognition navigation recommendation method provided in this application embodiment;

[0022] Figure 5 This is a schematic diagram of the logical structure of an elderly care terminal device provided in an embodiment of this application. Detailed Implementation

[0023] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0024] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0026] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0027] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0028] Existing smart canes can improve the safety and convenience of outdoor activities for the elderly, but they cannot identify and detect the elderly’s physical condition, resulting in poor care for the elderly.

[0029] Based on the above reasons, this application provides an AI-based state recognition navigation recommendation method. The method includes: acquiring terrain features, speed features, gait features, and physiological features of a user using a smart cane over multiple walking distances, and acquiring map information and the user's walking destination; wherein, terrain features, speed features, and gait features are detected by the motion detection component of the smart cane, and physiological features include the user's health record and real-time physiological features detected by wearable devices; through a state recognition-navigation recommendation model, based on the terrain features, speed features, gait features, physiological features, map information, and walking destination over multiple walking distances, determining the user's navigation route and walking guidance suggestions for the next walking distance, the walking guidance suggestions including precautions for the user while walking. The AI-based state recognition navigation recommendation method in this application can identify the state of elderly users using canes, select the optimal navigation route from the navigation routes based on the state recognition results, and provide corresponding walking guidance suggestions for the elderly user's walking navigation route.

[0030] In some scenarios, the AI-based state recognition navigation recommendation method of this application embodiment can be applied to elderly care terminal devices on smart canes used by elderly users. The elderly care terminal devices on smart canes can plan navigation routes based on the state of the elderly user and provide walking guidance suggestions based on the navigation routes.

[0031] The following describes in detail, with specific examples, an artificial intelligence-based state recognition navigation recommendation method provided in the embodiments of this application.

[0032] Figure 1 A flowchart illustrating the first artificial intelligence-based state recognition navigation recommendation method provided in this application embodiment is shown below. Figure 1 As shown, the above method includes S110 to S120, and S110 to S120 will be described in detail below.

[0033] S110. Acquire terrain features, speed features, gait features, and physiological features of the user during multiple walking distances while using the smart cane, and obtain map information and the user's walking destination. Among them, terrain features, speed features, and gait features are detected by the motion detection component of the smart cane, and physiological features include the user's health record and real-time physiological features detected by the wearable device.

[0034] In this implementation, the terrain features, speed features, and gait features of the user during multiple walking distances using the smart cane can be acquired first. Specifically, the motion detection components built into the smart cane can collect these features in real time. Terrain features can be obtained through pressure sensors and inertial measurement units at the bottom of the cane, which can identify, for example, road slope, ground material, and obstacle distribution. Speed ​​features can be calculated using accelerometers and gyroscopes to determine the user's walking speed and rhythm changes. Gait features can be dynamically captured by observing the user's movement patterns during the support and swing phases, such as stride uniformity and body balance.

[0035] It should be noted that multiple walking trips can be multiple preceding walking trips that the user has already taken before the current walking trip.

[0036] In this implementation, the physiological characteristics of the user during multiple walking distances while using the smart cane can also be acquired. These physiological characteristics include the user's health record and real-time physiological characteristics detected by the wearable device. The user's health record information, including past medical history, exercise capacity assessment data, and physical function indicators, can be pre-entered into the system or synchronously obtained from a medical institution's database. Real-time physiological characteristics are continuously monitored through the user's wearable smart device, such as parameters like heart rate, blood oxygen saturation, and muscle fatigue.

[0037] In this implementation, map information and the user's walking destination can also be obtained. The map information can be obtained through the positioning module and geographic information system connected to the smart cane, and combined with the walking destination actively entered by the user to form complete basic data for route planning.

[0038] It should be noted that map information can specifically include map details such as slope and steps, and then navigation routes can be planned based on this information.

[0039] S120. Using a state recognition-navigation recommendation model, based on terrain features, speed features, gait features, physiological features, map information, and walking destination within multiple walking journeys, determine the navigation route and walking guidance suggestions for the user's next walking journey. The walking guidance suggestions include information on precautions for the user while walking.

[0040] In this implementation, a state recognition-navigation recommendation model can be used to determine the user's navigation route and walking guidance suggestions for the next walking trip based on terrain features, speed features, gait features, physiological features, map information, and walking destination within multiple walking trips. The walking guidance suggestions include information on precautions for the user while walking, thus realizing the planning of the user's navigation route and walking guidance suggestions.

[0041] For example, the state recognition-navigation recommendation model can be implemented using an attention-based deep learning model, specifically trained using terrain features, speed features, gait features, physiological features, map information, walking destinations, sample navigation routes, and sample walking guidance suggestions within multiple sample walking journeys.

[0042] After data collection is completed, the state recognition-navigation recommendation model can perform fusion analysis on multi-dimensional features. For example, when it detects that the user's current gait features show a decline in lower limb stability and that the real-time heart rate data exceeds the safety threshold set in the health record, the state recognition-navigation recommendation model can dynamically adjust the navigation route to avoid challenging terrain by combining steep slope information identified in the terrain features.

[0043] In this implementation, the process of generating walking guidance suggestions can take into account the user's long-term health data and real-time status. For example, for users with cardiovascular disease, flat paths are prioritized when planning routes, and precautions such as controlling walking speed are suggested in the suggestions.

[0044] The beneficial effect of the above implementation method is that by integrating users' personalized health data, real-time physiological status and multi-dimensional movement characteristics, it can provide accurate navigation route planning and walking safety guidance for elderly users with different physical conditions, significantly improving the safety and adaptability of outdoor activities for special groups.

[0045] The beneficial effect of the above implementation method is that, through adaptive suggestions formed by intelligent learning of users' long-term behavioral patterns, users' walking habits and exercise capabilities can be gradually optimized while ensuring basic navigation functions, thereby improving the user experience.

[0046] In some implementations, in S120 above, the state recognition-navigation recommendation model determines the navigation route and walking guidance suggestions for the user in the next walking trip based on the terrain features, speed features, gait features, physiological features, map information and walking destination within multiple walking trips, including S121 to S122. S121 to S122 will be explained in detail below.

[0047] S121. The state recognition unit determines state recognition features based on terrain features, speed features, gait features, and physiological features within multiple walking distances. The navigation recommendation unit determines multiple navigation routes based on map information and the walking destination.

[0048] Figure 2 A schematic diagram of the workflow of the first artificial intelligence-based state recognition navigation recommendation method provided in the embodiments of this application is shown below. Figure 2As shown in this implementation, during the process of a user using a smart cane for walking navigation, the state recognition unit can determine the state recognition features based on the terrain features, speed features, gait features and physiological features within multiple walking journeys. Then, the state recognition unit can perform multi-dimensional feature extraction and comprehensive analysis on historical journey data.

[0049] Specifically, the state recognition unit can fuse data such as road slope, obstacle density and ground friction parameters in terrain features, average movement speed and acceleration change rate in speed features, stride fluctuation coefficient and body balance index in gait features, and heart rate variability and muscle fatigue in physiological features to obtain state recognition features.

[0050] For example, when the difference in support time between the left and right feet in a user's recent gait characteristics exceeds a preset threshold, the imbalance of lower limb strength can be judged by combining the real-time heart rate rise trend, thereby generating state recognition features that characterize the user's exercise ability.

[0051] In this implementation, a navigation recommendation unit can also determine multiple navigation routes based on map information and walking destination. Specifically, a geographic information system can be used to perform topological parsing on the map information and generate multiple candidate routes in combination with the user's walking destination.

[0052] For example, in a scenario where a destination needs to reach a park two kilometers away, three route options can be planned: a riverside walkway, a shortcut through a commercial street, and internal community roads. Each route is marked with an estimated length, a terrain complexity score, and information on surrounding facilities. The process of generating candidate routes can comprehensively consider path efficiency, the distribution of public facilities, and environmental safety factors. For example, routes with benches and emergency call devices along the way are given priority.

[0053] S122. Through the route selection unit, based on the status recognition features and multiple navigation routes, determine the route recommendation index and walking guidance suggestions corresponding to each navigation route, and recommend multiple navigation routes to the user in descending order of the route recommendation index.

[0054] like Figure 2 As shown, in this implementation, the route selection unit can determine the route recommendation index and walking guidance suggestions for each navigation route based on the status recognition features and multiple navigation routes. The unit then recommends multiple navigation routes to the user in descending order of the route recommendation index. The user can then select the navigation route from the multiple navigation routes with the recommendation index from high to low.

[0055] For example, the route selection unit can dynamically match status recognition features with candidate routes. For instance, if the system detects knee strain in a user, it can automatically lower the recommendation index for routes containing steps or slopes, while adding positive scoring weights to routes with a high proportion of flat surfaces. Walking guidance suggestions can be generated by incorporating real-time physiological fluctuations; for example, if a user's blood oxygen saturation remains below the baseline value in their health record, a prompt such as "It is recommended to pause and take deep breaths every 200 meters" can be inserted into the recommended routes. Finally, the system can prioritize routes according to their recommendation index and sequentially push the optimal routes and their corresponding suggestions via the smart cane's voice module or vibration feedback.

[0056] The beneficial effects of the above implementation method are that, by integrating users' real-time physiological status, long-term exercise capacity data, and environmental information, it is possible to provide personalized navigation solutions that balance safety and efficiency for special groups. The beneficial effects are also reflected in the dynamic route scoring mechanism, which effectively balances the contradiction between objective environmental factors and subjective health conditions in route planning by quantifying users' physical condition characteristics into a calculable recommendation index.

[0057] The beneficial effect of the above implementation method is that the collaborative output mode of walking guidance suggestions and route recommendations can continuously provide adaptive interventions during navigation, helping users optimize walking posture and adjust exercise intensity, thereby forming a healthy and sustainable outdoor activity support system.

[0058] In some implementations, the above method also includes S130 to S140, which are described in detail below.

[0059] S130. Through the state recognition unit, determine the current state recognition features based on the terrain features, speed features, gait features and physiological features within the current walking distance.

[0060] In this implementation, when a user uses a smart cane for walking navigation, and the user is walking along a predetermined navigation route, the state recognition unit can determine the current state recognition features based on the terrain features, speed features, gait features, and physiological features within the current walking distance. This enables dynamic analysis of real-time data of the current journey through the state recognition unit, and further allows for the planning of the navigation route based on the current state recognition features within the current walking distance.

[0061] For example, the state recognition unit can integrate newly collected terrain features, speed features, gait features and physiological features during the current walking journey. For example, it can detect data such as increased road surface slipperiness, sudden decrease in user walking speed, irregular fluctuations in stride length and abnormal increase in heart rate in real time. The state recognition unit can generate a dynamic feature set that reflects the user's current physical state and environmental changes, so that the real-time updated state recognition features can capture sudden changes in physical load and external environmental interference that are ignored by traditional navigation systems.

[0062] S140. Through the route selection unit, based on the current state identification features and multiple navigation routes, determine the alternative navigation route corresponding to the current walking trip and the walking guidance suggestions corresponding to the alternative navigation route. The alternative navigation route is used to replace the navigation route corresponding to the current walking trip.

[0063] In this implementation, in order to optimize the navigation route, the route selection unit can determine the alternative navigation route and the walking guidance suggestion corresponding to the current walking trip based on the current state recognition features and multiple navigation routes. This enables real-time matching of dynamically generated state recognition features with multiple pre-generated navigation routes, and the alternative navigation route is used to replace the navigation route corresponding to the current walking trip.

[0064] For example, when the route selection unit detects a feature in the user's current gait where the pressure distribution between the left and right feet is unbalanced and exceeds a safety threshold, the route selection unit can immediately perform a risk assessment on the current navigation route that includes steps or slopes, and quickly retrieve alternative navigation routes from the map database.

[0065] For example, the process of selecting alternative navigation routes can combine the current state identification features to determine the user's remaining physical strength estimate, and based on the user's remaining physical strength estimate and the distribution of surrounding facilities, for example, prioritize routes with barrier-free access and close to medical points as alternatives. Meanwhile, the corresponding walking guidance suggestions can include specific operational guidelines such as "reduce the step frequency to 90 steps per minute" or "use a cane for support for 3 seconds every 50 meters".

[0066] For example, when a user encounters a temporary obstacle while passing through a construction section, the smart cane can provide vibration feedback to indicate a route change, and detour instructions can be displayed on the cane's screen simultaneously.

[0067] For example, when a wearable device detects an abnormally high user body temperature, the alternative navigation route recommendation process can automatically incorporate weights for tree cover and drinking water point distribution, while simultaneously generating an accompanying suggestion to "replenish water every 5 minutes".

[0068] As an optimization, the process of selecting alternative navigation routes can also be optimized by combining historical travel data, such as automatically blocking road segments that have caused muscle soreness in users.

[0069] The beneficial effect of the above implementation method is that by collecting and analyzing dynamic feature data in real time, it can quickly generate alternative navigation schemes that ensure safety when the user's physical condition or external environment changes suddenly, effectively improving the emergency response capability of the pedestrian navigation system.

[0070] The beneficial effects of the above-mentioned implementation method are that, through the deep coupling of alternative routes and real-time health data, the risk of sports injuries can be significantly reduced by instantly matching the user's current physiological limits with environmental challenges. Furthermore, the coordinated output of dynamically adjusted walking guidance suggestions and alternative routes can maintain the stability of the user's exercise rhythm during path changes, forming a continuously adaptive navigation support system.

[0071] Figure 3 A flowchart illustrating the second AI-based state recognition navigation recommendation method provided in this application embodiment is shown below. Figure 3 As shown, the above method also includes S210 to S230, which will be explained in detail below.

[0072] S210. When the user's gait characteristics meet the preset gait characteristic conditions and the user's physiological characteristics meet the preset physiological characteristic conditions, the state recognition unit determines multiple trip split points corresponding to the current navigation route based on the terrain features, speed features, gait features and physiological characteristics within the current navigation route.

[0073] In this implementation, when the user's gait characteristics meet the preset gait characteristic conditions and the user's physiological characteristics meet the preset physiological characteristic conditions, it means that the user's gait characteristics and physiological characteristics during walking meet the conditions for navigation route optimization, which can trigger the intelligent segmentation mechanism of the navigation route.

[0074] For example, preset gait characteristic conditions may include consecutive occurrences of shortened stride length and a difference in the time it takes for the left and right feet to touch the ground exceeding a balance threshold in the user's gait characteristics.

[0075] For example, preset physiological characteristics may include a significant decrease in heart rate variability detected by the wearable device.

[0076] When segmenting navigation routes, the state recognition unit can automatically analyze the frequency of terrain undulations, road surface material change nodes, and public facility distribution characteristics within the current navigation route, thereby determining key location points suitable for route segmentation. Specifically, it can determine multiple trip segmentation points corresponding to the current navigation route based on terrain features, speed features, gait features, and physiological features within the current navigation route.

[0077] For example, the generation of these travel segmentation points can take into account the inflection point of user physical exertion and the area of ​​sudden change in environmental risk. For instance, flexible segmentation points can be set every 300 meters in long-distance trails, or mandatory segmentation points can be automatically generated when a group of steps is detected ahead.

[0078] S220: Obtain the terrain features, speed features, gait features, and physiological features corresponding to the walking distance between two adjacent trip split points. Based on these features, determine the segmentation state recognition features between the two adjacent trip split points. Through the segmentation navigation recommendation unit, determine multiple segmented navigation routes based on the current navigation route.

[0079] In this implementation, after obtaining multiple trip split points corresponding to the current navigation route, the terrain features, speed features, gait features, and physiological features corresponding to the walking trip between two adjacent trip split points can be obtained. The terrain features, speed features, gait features, and physiological features corresponding to the walking trip between two adjacent trip split points represent the terrain features and user status when walking between the two trip split points. Then, based on the terrain features, speed features, gait features, and physiological features corresponding to the walking trip between two adjacent trip split points, the segmentation status recognition features between two adjacent trip split points can be determined.

[0080] For example, after obtaining multiple trip segmentation points, the system can extract features from the sub-trips between adjacent segmentation points. For instance, within the segmentation zone between a commercial area and a park, data such as the road surface smoothness score, the user's average walking speed fluctuation curve, foot pressure distribution characteristics during the support phase, and blood oxygen saturation change trends can be extracted. These data are then used to form segmentation state recognition features reflecting the user's adaptability to that road segment through a feature fusion algorithm. These segmentation features can accurately characterize the user's physical response patterns under different environmental conditions, such as the increased muscle tension observed on gravel road sections.

[0081] It should be noted that when determining the segmentation state recognition features between two adjacent trip segmentation points based on the terrain features, speed features, gait features, and physiological features corresponding to the walking trips between two adjacent trip segmentation points, the number of walking trips between two adjacent trip segmentation points can be one or more. This implementation does not limit the number of walking trips between two adjacent trip segmentation points.

[0082] By segmenting the navigation recommendation unit, multiple segmented navigation routes are determined based on the current navigation route. These multiple segmented navigation routes are alternative routes corresponding to each sub-trip of the current navigation route. This allows the segmented navigation recommendation unit to generate multiple alternative solutions for each sub-trip while preserving the overall framework of the current navigation route.

[0083] For example, for a 500-meter sub-trip that includes steep slopes, two types of split navigation routes can be planned: one that bypasses the barrier-free passage and the other that maintains the original path but inserts auxiliary support guidance. The generation of each alternative route will take into account the terrain difficulty unique to that sub-trip and the user's current physiological load capacity. For example, for users who are sensitive to knee joint pressure, an alternative solution to avoid steps will be automatically generated.

[0084] S230. Through the segmented route selection unit, based on the segmentation status identification features and multiple segmented navigation routes, determine the route recommendation index and walking guidance suggestions corresponding to each segmented navigation route, and recommend multiple segmented navigation routes to the user in descending order of the route recommendation index.

[0085] In this implementation, the segmented route selection unit can further determine the route recommendation index and walking guidance suggestions corresponding to each segmented navigation route based on the segmentation status recognition features and multiple segmented navigation routes. The unit then recommends multiple segmented navigation routes to the user in descending order of the route recommendation index, thereby enabling route recommendation for multiple segmented navigation routes.

[0086] For example, the segmented route selection unit can use a dynamic evaluation mechanism to calculate the matching degree between segmented features and alternative routes. For instance, when it is detected that the user's Achilles tendon load is continuously exceeding the limit in a certain sub-journey, the system can assign a higher recommendation index to the alternative route with a soft plastic track, and at the same time generate a targeted suggestion of "using a small stride and high frequency of movement".

[0087] For example, the ranking process of the recommendation results can incorporate multi-dimensional weights, such as prioritizing the display of the top three options with the greatest reduction in physiological burden, and displaying comparative data on the advantages and disadvantages of each option through a cane touch screen in pagination.

[0088] The beneficial effect of the above implementation method is that by intelligently segmenting long-distance navigation routes and implementing segmented optimization, it can significantly improve the exercise sustainability of elderly people with special physical conditions in complex environments and effectively prevent health risks caused by excessive fatigue.

[0089] The aforementioned implementation method also offers the advantage of providing more scenario-adaptive navigation decision support by capturing the user's differentiated state in different environmental segments through refined processing of segmented features. Furthermore, the deep integration of dynamic segmentation mechanisms and real-time physiological monitoring enables flexible adjustments to micro-paths while maintaining navigation continuity, forming a multi-layered navigation protection system that balances efficiency and safety.

[0090] In some implementations, in S210 above, the state recognition unit determines multiple trip segmentation points corresponding to the current navigation route based on terrain features, speed features, gait features and physiological features within the current navigation route, including S211 to S212. S211 to S212 will be explained in detail below.

[0091] S211. Obtain road surface images corresponding to the current navigation route through a camera, and determine the pedestrian flow corresponding to the current navigation route based on the road surface images.

[0092] In this implementation, the existing smart cane can also be equipped with a camera to acquire real-time road images. Therefore, when recognizing the walking status of elderly users, the camera can be used to acquire the road image corresponding to the current navigation route. Based on the road image, the pedestrian flow corresponding to the current navigation route can be determined. The pedestrian flow characterizes the degree of influence of the pedestrian flow in the current navigation route on the road image. The smaller the pedestrian flow, the smaller the influence on the road image of the current navigation route.

[0093] For example, when determining the pedestrian flow corresponding to the current navigation route based on the road surface image, the number of people identified per minute in the road surface image can be used as the pedestrian flow corresponding to the current navigation route.

[0094] S212. When the pedestrian flow is less than the preset pedestrian flow, the status recognition unit determines multiple travel segmentation points corresponding to the current navigation route based on the road surface image corresponding to the current navigation route and the speed characteristics, gait characteristics and physiological characteristics within the current navigation route.

[0095] In this implementation, when the pedestrian flow is less than the preset pedestrian flow, it indicates that the pedestrian flow corresponding to the current navigation route is small and will not have a significant impact on the obstacle recognition in the road image of the current navigation route. At this time, the state recognition unit can extract the obstacle status in the road image corresponding to the current navigation route based on the speed characteristics, gait characteristics, and physiological characteristics of the elderly user in the current navigation route. Specifically, the state recognition unit can extract the obstacle status in the road image corresponding to the current navigation route and dynamically plan multiple travel segmentation points corresponding to the current navigation route based on the obstacle status in the road image and the speed characteristics, gait characteristics, and physiological characteristics of the elderly user in the current navigation route. This improves the reasonable planning of the elderly walking route and compensates for the situation where obstacles may temporarily appear in fixed terrain features. It realizes intelligent analysis and early warning of walking routes based on the surrounding environment, thereby improving the user experience of elderly users.

[0096] For example, the state recognition unit can be trained using an image dataset containing obstacle annotations, and the state recognition unit can be used to identify the state of obstacles in road images.

[0097] The beneficial effect of the above implementation method is that, when there is less pedestrian traffic, dynamic planning is performed on multiple travel segmentation points corresponding to the current navigation route based on the obstacle status in the road image and the speed characteristics, gait characteristics and physiological characteristics of elderly users within the current navigation route, thereby improving the accuracy of dynamic planning of walking routes for elderly users.

[0098] In some implementations, the above method also includes S240 to S250, which will be described in detail below.

[0099] S240. Through the segmented route selection unit, based on the segmentation status identification features and multiple segmented navigation routes, determine the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period.

[0100] In this implementation, to improve the protection of elderly users while walking, the segmented route selection unit can determine the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period based on the segmentation status recognition features and multiple segmented navigation routes.

[0101] For example, the physiological load of a sub-trip can be dynamically assessed by a segmented route selection unit. The segmented route selection unit can intelligently calculate the appropriate rest interval between adjacent segmented navigation routes based on segmentation state recognition features related to muscle fatigue accumulation curves and heart rate recovery rates, combined with the corresponding terrain complexity score.

[0102] For example, between two consecutive split navigation routes containing steep slopes, the system can automatically generate a rest suggestion including a 5-minute break between the two split navigation routes containing steep slopes, based on the user's knee flexion and extension angle change trend and Achilles tendon pressure peak data, and set rest nodes in the terrain transition area.

[0103] For example, the system can intelligently allocate rest programs of different intensities based on the predicted difficulty of the subsequent route.

[0104] S250. When a user walks along a target segmented navigation route, the system recommends the target rest time period corresponding to the target segmented navigation route and rest guidance suggestions corresponding to the target rest time period.

[0105] In this implementation, when a user is walking along a selected segmented route, the system recommends a target rest period corresponding to the target segmented navigation route and rest guidance suggestions corresponding to the target rest period. The recommendation system can interactively implement accompanying reminders to remind the user to rest during the walk.

[0106] The beneficial effects of the above implementation method are that by intelligently matching physiological recovery needs with route characteristics, sports injuries can be effectively prevented and the sustainability of walking can be improved; in terms of the personalized generation mechanism of rest suggestions, by integrating user health records and environmental variable data, recovery plans with preventive medical value can be formed.

[0107] The beneficial effects of the above implementation method are that the dynamically adjusted layout of rest nodes and the real-time feedback guidance can optimize the user's physical fitness allocation strategy while ensuring navigation continuity, and build a complete closed loop for sports and health management.

[0108] In some implementations, the above method also includes S260 to S270, which will be described in detail below.

[0109] S260. Obtain the target segmentation status recognition features of the user within the target segmentation navigation route, and obtain the rest physiological features of the user during the rest period after the target segmentation navigation route.

[0110] In this implementation, in order to further and more accurately monitor the user's state, the target segmentation state recognition features of the user within the target segmentation navigation route can be obtained, and the rest physiological features of the user during the rest period after the target segmentation navigation route can be obtained, so as to further adjust the rest process after the user completes the target segmentation navigation route.

[0111] In this implementation, the sensor array built into the smart cane can continuously collect the user's motion state data in a specific sub-journey. The process of acquiring target segmentation state recognition features can be achieved by recording abnormal data of wrist swing amplitude through the six-axis inertial sensor of the cane handle, and by combining the force point offset trend captured by the insole pressure sensor to form a comprehensive feature vector characterizing the user's motion load on that section of road.

[0112] For example, during rest periods, wearable devices can continuously monitor resting physiological characteristics such as heart rate recovery curves and blood oxygen saturation recovery gradients, for instance, by using the photoplethysmography (PPG) sensor of a smart bracelet to obtain venous return efficiency indicators.

[0113] S270. Based on the target segmentation state identification features, rest physiological features, and multiple segmentation navigation routes, determine the subsequent rest suggestions and subsequent guidance segmentation navigation routes corresponding to the target segmentation navigation route. The subsequent rest suggestions are used to indicate to the user the rest suggestions after the target segmentation navigation route.

[0114] In this implementation, the subsequent rest suggestions and subsequent guidance segmentation navigation routes corresponding to the target segmentation navigation route can be determined based on the target segmentation state identification features, rest physiological features, and multiple segmentation navigation routes. The subsequent rest suggestions are used to indicate to the user the rest suggestions after the target segmentation navigation route, so as to guide the user's rest process according to the subsequent rest suggestions.

[0115] When generating subsequent rest suggestions, the system can compare and analyze the cumulative fatigue-related features in the target segmentation state recognition features with the recovery efficacy-related features in the rest physiological features to generate subsequent rest suggestions.

[0116] For example, when it is detected that the rate of decrease in Achilles tendon temperature during the resting period is lower than the baseline value in the health record after a sub-stroke with excessive calf muscle load, an enhanced recovery suggestion of "applying Achilles tendon ice for 10 minutes" can be generated.

[0117] It should be noted that the optimization process of subsequent navigation schemes in this implementation is time-dependent. For example, after a user completes a sub-trip containing continuous steps, the system can dynamically adjust the allowable slope threshold for sloping sections in subsequent routes based on physiological characteristics monitored during rest periods.

[0118] The beneficial effect of the above implementation method is that by establishing a dynamic feedback mechanism between trip characteristics and recovery effects, a health guidance plan with temporal continuity can be formed, which significantly improves the precision of physical condition management during long-distance navigation.

[0119] The beneficial effect of the above implementation method is that, in the closed-loop health management of exercise and recovery, by feeding back real-time physiological recovery data to subsequent route decisions, an intelligent protection system for preventing sports injuries can be built.

[0120] Figure 4 A flowchart illustrating the third AI-based state recognition navigation recommendation method provided in this application embodiment is shown below. Figure 4 As shown, the above method also includes S310 to S320, which will be described in detail below.

[0121] S310. The state recognition unit determines the current state recognition features based on terrain features, speed features, gait features, and physiological features within the current walking distance. It also acquires the user's historical navigation routes and corresponding historical state recognition features for the current walking distance. Finally, the navigation recommendation unit determines multiple navigation routes based on map information and the walking destination.

[0122] In this implementation, the current state recognition feature can be determined by the state recognition unit based on the terrain features, speed features, gait features and physiological features within the current walking journey. The current state recognition feature represents the user's various environmental features and walking status in the current journey. At the same time, the historical navigation routes that the user has followed and the historical state recognition features corresponding to the historical navigation routes can be obtained, and the current walking journey can be optimized based on the historical state recognition features.

[0123] For example, during the process of a user using a smart cane for navigation, the real-time data of the current journey can be fused and analyzed in multiple dimensions by the state recognition unit. The state recognition unit can simultaneously integrate parameters such as road slope angle and obstacle distribution density in terrain features, instantaneous acceleration and gait frequency fluctuation coefficient in speed features, foot lift-off height and trunk swing amplitude in gait features, and skin conductance response and breathing rhythm in physiological features to form a feature set that reflects the user's real-time movement state as the current state recognition feature.

[0124] For example, the system can retrieve the user's historical navigation records in the area, such as gait stability coefficients and heart rate recovery curves recorded on the same route three months ago, to provide a benchmark for current decision-making.

[0125] In this implementation, a navigation recommendation unit can also determine multiple navigation routes based on map information and walking destination, and then optimize the selection of navigation routes among the multiple navigation routes.

[0126] S320. Through the route selection unit, based on the current state identification features, historical state identification features, and multiple navigation routes, determine the alternative navigation route corresponding to the current walking trip and the walking guidance suggestions corresponding to the alternative navigation route. The alternative navigation route is used to replace the navigation route corresponding to the current walking trip.

[0127] In this implementation, after obtaining the current state identification features and the historical state identification features, the route selection unit can determine the alternative navigation route corresponding to the current walking trip and the walking guidance suggestions corresponding to the alternative navigation route based on the current state identification features, the historical state identification features and multiple navigation routes. The alternative navigation route is used to replace the navigation route corresponding to the current walking trip.

[0128] It should be noted that the route selection unit can optimize dynamic decision-making by cross-referencing the differences between current and historical state characteristics. For example, when it detects that a user's current breathing rate has increased by 20% compared to historical data for the same period, the system can assign priority weights to alternative routes that include gentle slopes and generate targeted suggestions such as "using the zigzag uphill technique." The introduction of historical state characteristics can also help identify trends in user movement patterns. For instance, by comparing the stride decay curves over six months, the system can automatically avoid ditch sections requiring large strides when recommending routes.

[0129] For example, when an increase in a user's anxiety index is detected during the morning rush hour, the system can refer to the user's preference for low-traffic routes during the same historical time period and prioritize recommending quieter routes through community gardens. Simultaneously, the recommendation strategy can be optimized by incorporating environmental variable data from historical travel patterns; for instance, historical routes that have caused users to slip and fall during rainy weather can be permanently downgraded.

[0130] The beneficial effect of the above implementation method is that by integrating real-time physiological data with long-term exercise pattern characteristics, it is possible to provide personalized navigation solutions that balance safety and habit continuity for patients with chronic diseases and users in the recovery period.

[0131] The beneficial effects of the above implementation method are also that, through in-depth mining of historical data and the establishment of time series models of user status changes, exercise risks can be predicted in advance and preventive route adjustments can be implemented. At the same time, the introduction of a dynamic comparison mechanism enables the navigation system to have continuous learning capabilities, automatically optimizing recommendation strategies as user physical functions and exercise preferences change, forming an intelligent navigation companion system that evolves along with the user's health status.

[0132] In some implementations, the above method also includes S330 to S340, which will be described in detail below.

[0133] S330. The route selection unit determines multiple trip split points corresponding to the current navigation route based on the current state identification features and historical state identification features. The segmented navigation recommendation unit determines multiple segmented navigation routes based on the current navigation route. The segmented navigation routes are alternative navigation routes between adjacent trip split points of the current navigation route.

[0134] In this implementation, the route selection unit can determine multiple trip segmentation points corresponding to the current navigation route based on the current state identification features and the historical state identification features. These multiple trip segmentation points are used to segment and evaluate the current navigation route, thereby enabling the segmentation and optimization of the current trip based on the comparison between the historical state and the current state.

[0135] For example, the layout of trip split points can take into account the dynamic balance between environmental characteristics and user exercise capabilities. For instance, when a cobblestone road section that has historically caused a heart rate increase exceeding a preset range is detected ahead, a trip split point can be set 200 meters in advance.

[0136] For example, when the route selection unit performs dynamic segmentation processing on the navigation route, it can intelligently determine the route segmentation points suitable for the user's current physical fitness by combining the current and historical state identification features of the user that are currently detected.

[0137] By using the segmented navigation recommendation unit, multiple segmented navigation routes are determined based on the current navigation route. These segmented navigation routes are alternative navigation routes between adjacent trip split points of the current navigation route. This allows for the replacement of navigation routes between adjacent trip split points of the current navigation route based on the segmented navigation routes, thereby optimizing the navigation routes between adjacent trip split points of the current navigation route.

[0138] For example, when generating segmented navigation routes, the segmented navigation recommendation unit can perform micro-path optimization based on the framework of the original navigation route. For instance, in a 1.2-kilometer-long riverside walkway, the system can generate three options for each 200-meter segment: a passageway inside the riverside railing, a tree-lined walkway, and a detour route around the viewing platform.

[0139] S340. Obtain the current state identification features and historical state identification features corresponding to the walking distance between two adjacent trip split points. Through the segmented route selection unit, based on the current state identification features, historical state identification features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip split points, determine the route recommendation index and walking guidance suggestions corresponding to each segmented navigation route, and recommend multiple segmented navigation routes to the user in descending order of route recommendation index.

[0140] In this implementation, the current state identification features and historical state identification features corresponding to the walking distance between two adjacent trip split points can be obtained. Through the segmented route selection unit, based on the current state identification features, historical state identification features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip split points, the route recommendation index and walking guidance suggestions corresponding to each segmented navigation route are determined. Multiple segmented navigation routes are recommended to the user in descending order of the route recommendation index, thereby achieving the purpose of determining multiple segmented navigation routes for the current navigation route and achieving the purpose of segmented optimization of the current navigation route.

[0141] When determining the route recommendation index between adjacent dividing points, the system can integrate features of physiological data from multiple periods. For example, for historical records of respiratory rhythm disorders, the system can reduce the route recommendation index corresponding to the historical records of respiratory rhythm disorders and insert a "use the two-step breathing method" prompt when recommending uphill routes.

[0142] For example, after a user completes the first 100-meter segment of the route, the system can dynamically adjust the recommended order of subsequent segmented routes based on the current and historical status identification features of the completed segments.

[0143] The beneficial effect of the above implementation method is that by implementing dynamic micro-segmentation of navigation routes, it is possible to provide patients with chronic pain with a progressive exercise program that is precisely matched to their physical tolerance, effectively preventing the cumulative effect of sports injuries.

[0144] The beneficial effect of the above implementation method is that, in optimizing the spatiotemporal correlation of segmented routes, a dynamic comparison mechanism between the current state and historical performance can be established to achieve refined control of motion load. The deep coupling between the segmented recommendation system and multi-dimensional feature data can provide users with path selection strategies that balance motion safety and experience optimization while ensuring the overall navigation objective.

[0145] In some implementations, the above method also includes S350 to S360, which will be described in detail below.

[0146] S350: Through the route selection unit, based on the current state recognition features, historical state recognition features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent route segmentation points, determine the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period.

[0147] In this implementation, in order to improve the rest experience for elderly users, the route selection unit can determine the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period based on the current state recognition features, historical state recognition features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip segmentation points.

[0148] In this implementation, when executing a segmented navigation route, the physiological load of the sub-trip can be dynamically evaluated by the segmented route selection unit. Specifically, by combining the current state identification features, historical state identification features, and multiple segmented navigation routes, suitable rest nodes can be intelligently calculated.

[0149] For example, the route selection unit can intelligently calculate suitable rest stops based on the gait symmetry index between adjacent split points and the real-time terrain complexity score. For instance, if the system detects that a user's heel impact force exceeds 15% of the historical average for 5 consecutive minutes in a certain segment of the route, it can set up a buffer rest area 50 meters before the end of that segment.

[0150] As an optimization, the suggestions can be dynamically adjusted based on real-time environmental variables. For example, in hot weather, suggestions for salt replenishment and sun protection reminders can be automatically inserted, or when a sudden increase in air humidity is detected, anti-slip footwear inspection steps can be recommended.

[0151] S360. When a user walks along a target segmented navigation route, the system recommends the target rest time period corresponding to the target segmented navigation route and rest guidance suggestions corresponding to the target rest time period.

[0152] In this implementation, when a user walks along a target segmented navigation route, the system can recommend the target rest time period and rest guidance suggestions corresponding to the target rest time period, thereby pushing the user's rest time period and rest guidance suggestions to improve the user experience.

[0153] The beneficial effect of the above implementation method is that by establishing a precise matching mechanism between historical exercise load, current exercise load and recovery strategy, sports injuries can be effectively prevented and the sustainability of long-distance navigation can be improved.

[0154] The beneficial effects of the above implementation method are also that, in terms of optimizing the spatiotemporal continuity of rest guidance, by deeply integrating real-time physiological data with environmental variables, a personalized recovery plan with preventive health intervention value can be formed. This can optimize the user's physical fitness management strategy while ensuring navigation efficiency, and build a complete closed-loop system for sports health monitoring.

[0155] This application also provides an elderly care terminal device, including a unit for executing the artificial intelligence-based state recognition navigation recommendation method as described in any of the preceding claims.

[0156] Figure 5 This is a schematic diagram of the logical structure of an elderly care terminal device provided in an embodiment of this application, such as... Figure 5 As shown, the elderly care terminal device 1 in this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to send and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above-described method. The beneficial effects of this embodiment have been described in the above-described method and will not be repeated here.

[0157] It should be noted that this elderly care terminal device can be installed on a smart cane as an accessory to improve the user experience of the smart cane equipped with this elderly care terminal device.

[0158] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0159] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0160] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0161] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0162] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0163] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0164] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0165] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A state recognition navigation recommendation method based on artificial intelligence, characterized in that, The method includes: The system acquires terrain features, speed features, gait features, and physiological features of the user during multiple walking distances while using the smart cane, and obtains map information and the user's walking destination. The terrain features, speed features, and gait features are detected by the motion detection component of the smart cane, and the physiological features include the user's health record and real-time physiological features detected by the wearable device. Using a state recognition-navigation recommendation model, based on terrain features, speed features, gait features, physiological features, map information, and walking destinations within multiple walking journeys, the system determines the navigation route and walking guidance suggestions for the user's next walking journey. The walking guidance suggestions include information on precautions for the user while walking. The method further includes: When a user's gait characteristics meet preset gait characteristic conditions and the user's physiological characteristics meet preset physiological characteristic conditions, the state recognition unit determines multiple trip split points corresponding to the current navigation route based on the terrain features, speed features, gait features, and physiological features within the current navigation route. The terrain features, speed features, gait features, and physiological features corresponding to the walking distance between two adjacent trip split points are obtained. Based on the terrain features, speed features, gait features, and physiological features corresponding to the walking distance between two adjacent trip split points, the segmentation state recognition features between two adjacent trip split points are determined. Through the segmentation navigation recommendation unit, multiple segmentation navigation routes are determined based on the current navigation route. The segmented route selection unit determines the route recommendation index and walking guidance suggestions for each segmented navigation route based on the segmentation status recognition features and multiple segmented navigation routes. It then recommends multiple segmented navigation routes to the user in descending order of the route recommendation index.

2. The method as described in claim 1, characterized in that, Using a state recognition-navigation recommendation model, based on terrain features, speed characteristics, gait characteristics, physiological characteristics, map information, and walking destinations across multiple walking journeys, the system determines navigation routes and walking guidance suggestions for the user's next walking journey, including: The state recognition unit determines state recognition features based on terrain features, speed features, gait features, and physiological features within multiple walking distances; the navigation recommendation unit determines multiple navigation routes based on map information and walking destinations. The route selection unit determines the route recommendation index and walking guidance suggestions for each navigation route based on status recognition features and multiple navigation routes, and recommends multiple navigation routes to the user in descending order of the route recommendation index.

3. The method as described in claim 2, characterized in that, The method further includes: The current state recognition features are determined by the state recognition unit based on the terrain features, speed features, gait features and physiological features within the current walking distance. The route selection unit identifies alternative navigation routes and walking guidance suggestions corresponding to the current walking distance based on the current status identification features and multiple navigation routes. The alternative navigation routes are used to replace the navigation routes corresponding to the current walking distance.

4. The method as described in claim 3, characterized in that, The method further includes: By using the segmented route selection unit, based on the segmentation status identification features and multiple segmented navigation routes, the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period are determined. As users walk along the target segmented navigation route, the system recommends the target rest time period corresponding to the target segmented navigation route and rest guidance suggestions for the target rest time period.

5. The method as described in claim 4, characterized in that, The method further includes: Obtain the target segmentation status recognition features of the user within the target segmentation navigation route, and obtain the user's rest physiological features during the rest period after the target segmentation navigation route; Based on the target segmentation state identification features, rest physiological features, and multiple segmentation navigation routes, subsequent rest suggestions and subsequent guidance segmentation navigation routes corresponding to the target segmentation navigation route are determined. The subsequent rest suggestions are used to indicate to the user what rest to do after the target segmentation navigation route.

6. The method as described in claim 5, characterized in that, The method further includes: The current state recognition feature is determined by the state recognition unit based on the terrain features, speed features, gait features and physiological features within the current walking route; the historical navigation routes that the user has followed and the historical state recognition features corresponding to the historical navigation routes are obtained; and multiple navigation routes are determined by the navigation recommendation unit based on map information and walking destination. The route selection unit determines the alternative navigation route and walking guidance suggestions corresponding to the current walking trip based on the current state recognition features, historical state recognition features, and multiple navigation routes. The alternative navigation route is used to replace the navigation route corresponding to the current walking trip.

7. The method as described in claim 6, characterized in that, The method further includes: The route selection unit determines multiple trip split points corresponding to the current navigation route based on the current state recognition features and historical state recognition features; the segmented navigation recommendation unit determines multiple segmented navigation routes based on the current navigation route, and the segmented navigation routes are alternative navigation routes between the navigation routes adjacent to the current navigation route. The system acquires the current state identification features and historical state identification features corresponding to the walking distance between two adjacent trip split points. Through the segmented route selection unit, based on the current state identification features, historical state identification features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip split points, it determines the route recommendation index and walking guidance suggestions for each segmented navigation route, and recommends multiple segmented navigation routes to the user in descending order of route recommendation index.

8. The method as described in claim 7, characterized in that, The method further includes: By using the route selection unit, based on the current state recognition features, historical state recognition features, and multiple segmented navigation routes corresponding to the walking distance between two adjacent trip segmentation points, the rest time periods between multiple adjacent segmented navigation routes and the rest guidance suggestions corresponding to each rest time period are determined. As users walk along the target segmented navigation route, the system recommends the target rest time period corresponding to the target segmented navigation route and rest guidance suggestions for the target rest time period.

9. A terminal device for elderly care, characterized in that, It includes a unit for performing the state recognition navigation recommendation method based on any one of claims 1 to 8.

Images