Running route recommendation method, device and equipment

Abstract

The invention discloses a running route recommendation method, device and equipment, and the method comprises the steps: obtaining a basic route demand and multi-dimensional environment data corresponding to a route recommendation request in real time when the route recommendation request input by a user is received; calling a route planning model to perform route planning according to the basic route demand to obtain a plurality of candidate routes; according to the multi-dimensional environment data and the user portrait of the user, calculating the comprehensive score of each candidate route, selecting the candidate route with the highest comprehensive score, and optimizing the candidate route to obtain a target route; when a running start signal input by the user is received, physiological data and route warning information of the user are periodically collected; and generating change recommendation information of the target route according to the physiological data and / or the route warning information, and pushing the change recommendation information to wearable equipment of the user. Therefore, the candidate paths are optimized and screened in combination with the multi-dimensional environment data and the user portrait, and meanwhile, real-time change recommendation of the paths is performed in combination with the livelihood data and the path warning information in the running process.

Description

Technical Field

[0001] This invention relates to the field of route recommendation technology, and in particular to a method, apparatus, and device for recommending running routes. Background Technology

[0002] With increasing health awareness, running, as a low-barrier and easy-to-learn form of exercise, has gained widespread popularity. The development of mobile internet technology has spawned a large number of running and fitness apps, covering a broad user base from daily fitness to professional training, and promoting the digitalization and convenience of running. At the same time, the maturity of technologies such as artificial intelligence recommendation algorithms, Geographic Information Systems (GIS), and the Internet of Things (IoT) has provided technical support for the functional upgrades of running apps, and users' demands for personalized, health-oriented, safe, and social running experiences are becoming increasingly prominent.

[0003] Existing running and fitness apps primarily focus on recording basic data such as running trajectory, distance, and pace, as well as simple route planning. Their core functions are concentrated on the statistics and display of exercise data. However, with the diversification of the runner community and the increasing complexity of urban environments, runners are not only concerned with exercise results but also increasingly valuing the impact of the environment on health, the match between the route and their own abilities, safety during the run, and the interactive experience brought by sports-related social networking.

[0004] Although existing running apps have implemented basic exercise recording and route planning functions, in actual use, due to insufficient consideration of environmental factors, they may guide users to areas where the environment is not suitable or construction areas, failing to simultaneously take into account user health and exercise safety. Summary of the Invention

[0005] This invention provides a running route recommendation method, apparatus, and device, which solves the technical problem that existing running applications have achieved basic exercise recording and route planning functions, but in practical applications, due to insufficient consideration of environmental factors, they may guide users to areas where the environment is not suitable or construction areas, and cannot simultaneously take into account the user's health and exercise safety.

[0006] The first aspect of this invention provides a running route recommendation method, comprising:

[0007] When a user inputs a route recommendation request, the system obtains the basic route requirements and multi-dimensional environmental data corresponding to the route recommendation request in real time.

[0008] The route planning model is invoked to plan routes according to the basic route requirements, resulting in multiple candidate routes;

[0009] Based on the multi-dimensional environmental data and the user profile, calculate the comprehensive score of each candidate route, select the candidate route with the highest comprehensive score and optimize it to obtain the target route;

[0010] When the user inputs a signal to start running, the system periodically collects the user's physiological data and route warning information.

[0011] Based on the physiological data and / or the route warning information, generate change recommendation information for the target route and push it to the user's wearable device.

[0012] Optionally, the multi-dimensional environmental data includes air quality data, terrain data, roadside landscape data, road safety data, and weather data; the step of calculating a comprehensive score for each candidate route based on the multi-dimensional environmental data and the user's profile, selecting the candidate route with the highest comprehensive score, and optimizing it to obtain the target route includes:

[0013] Based on the user demand information within the basic route requirements, the route preference information within the user profile, and the air quality data, the weight values ​​of each rating dimension are adjusted to the target value using a preset weight adjustment rule to obtain the dimension weights.

[0014] A comprehensive score is calculated for each candidate route based on the weights of each dimension and the air quality data, terrain data, roadside landscape data, road safety data, and weather data.

[0015] The candidate route with the highest overall score is selected as the intermediate route;

[0016] The intermediate route is optimized according to preset optimization rules to obtain the target route.

[0017] Optionally, the step of calculating a comprehensive score for each candidate route based on the weights of each dimension and the air quality data, terrain data, roadside landscape data, road safety data, and weather data includes:

[0018] Based on the air quality data and the weather data, a quality scoring table is matched to determine the air quality score corresponding to each candidate route;

[0019] Based on the road surface type and slope of each road segment in the terrain data, match the road segment difficulty coefficient corresponding to each candidate route;

[0020] The landscape richness of each candidate route is determined by summing scores based on the landscape type and number of landscapes in the roadside landscape data.

[0021] Based on the road safety data, a safety score is matched for each candidate route;

[0022] Substitute the difficulty coefficient of each of the aforementioned road segments and the running level in the user profile into the preset difficulty matching degree calculation formula to generate the difficulty matching degree corresponding to each of the aforementioned candidate routes;

[0023] The air quality score, difficulty matching degree, landscape richness and safety score are weighted and superimposed using the weights of each dimension to calculate the comprehensive score of each candidate route.

[0024] Optionally, the step of optimizing the intermediate route according to a preset optimization rule to obtain the target route includes:

[0025] If the intermediate route passes through the marked area, the route planning model is invoked to replan the route segments corresponding to the marked area to obtain the first updated route that avoids the marked area.

[0026] If the number of turns in the first updated route is greater than a preset threshold, the route planning model is invoked to replan the first updated route until the number of turns is lower than the threshold, thus obtaining a second updated route.

[0027] If the second updated route includes identical round-trip sections, then with the goal of increasing the amount of roadside landscape data, the route planning model is invoked to replan the second updated route and generate the target route.

[0028] Optionally, the step of generating change recommendation information for the target route based on the physiological data and / or the route warning information and pushing it to the user's wearable device includes:

[0029] When the physiological data matches the preset route change conditions, the corresponding updated route segment is retrieved according to the matching result, and the change recommendation information of the target route is generated;

[0030] When the warning area corresponding to the route warning information overlaps with the target route, calculate the distance to be moved between the user's current location and the warning area;

[0031] If the distance to be moved is greater than or equal to a preset distance threshold, the route planning model is invoked to replan the target route and generate change recommendation information to avoid the warning area.

[0032] If the distance to be moved is less than a preset distance threshold, the route planning model is invoked to generate change recommendation information from the current location to the nearest safe area;

[0033] The changed recommendation information is pushed to the user's wearable device for display.

[0034] Optionally, it also includes:

[0035] In response to user-inputted registration requests, collect basic information authorized by the user, route preference information, and historical running records;

[0036] The running ability assessment model is invoked to determine the user's corresponding running ability level based on the basic information and the historical running records;

[0037] A user profile is constructed using the basic information, the route preference information, and the running ability level.

[0038] Optionally, it also includes:

[0039] When the user inputs a running plan, the matching degree between the running plan and multiple existing running plans is calculated according to the plan content.

[0040] Send running invitation messages to target users whose matching score is higher than the matching threshold;

[0041] When the target user's response to the running invitation is received, route planning is performed according to the running plan and the target user's existing running plan to generate a common route;

[0042] In response to social requests from the user and the target user, a temporary chat room is created, real-time location is displayed, and / or a virtual group photo is generated.

[0043] Optionally, it also includes:

[0044] In response to the user's authorization command, the physiological data and running trajectory are stored in a local database;

[0045] The user profile is periodically updated based on the physiological data and the running trajectory.

[0046] The route planning model is periodically updated in response to the user's rating of the target route.

[0047] A second aspect of the present invention provides a running route recommendation device, comprising:

[0048] The data acquisition module is used to acquire the basic route requirements and multi-dimensional environmental data corresponding to the route recommendation request in real time when a user inputs a route recommendation request.

[0049] The route planning module is used to call the route planning model to plan routes according to the basic route requirements and obtain multiple candidate routes;

[0050] The route optimization module is used to calculate the comprehensive score of each candidate route based on the multi-dimensional environmental data and the user profile, select the candidate route with the highest comprehensive score and optimize it to obtain the target route;

[0051] The running data acquisition module is used to periodically collect the user's physiological data and route warning information when it receives the running start signal input by the user.

[0052] The real-time running route recommendation module is used to generate change recommendation information for the target route based on the physiological data and / or the route warning information, and push it to the user's wearable device.

[0053] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the running route recommendation method as described in any of the first aspects of the present invention.

[0054] As can be seen from the above technical solutions, the present invention has the following advantages:

[0055] Upon receiving a route recommendation request from a user, the system acquires the corresponding basic route requirements and multi-dimensional environmental data in real time. It then calls a route planning model to plan a route based on these requirements, generating multiple candidate routes. Based on the multi-dimensional environmental data and the user's profile, it calculates a comprehensive score for each candidate route, selects the highest-scoring route, and optimizes it to obtain the target route. When the user inputs a start signal, the system periodically collects the user's physiological data and route warning information. Based on the physiological data and / or route warning information, it generates route change recommendations and pushes them to the user's wearable device. This approach combines multi-dimensional environmental data and user profiles to optimize and filter candidate routes, while simultaneously providing real-time route change recommendations during the run based on physiological data and route warning information. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating the steps of a running route recommendation method provided in an embodiment of the present invention;

[0058] Figure 2This is a structural block diagram of a running route recommendation device provided in an embodiment of the present invention. Detailed Implementation

[0059] This invention provides a running route recommendation method, apparatus, and device to address the technical problem that existing running applications, while having implemented basic exercise recording and route planning functions, may guide users to unsuitable environments or construction areas due to insufficient consideration of environmental factors in practical applications, failing to simultaneously ensure user health and exercise safety.

[0060] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0061] Please see Figure 1 , Figure 1 A flowchart illustrating the steps of a running route recommendation method provided in an embodiment of the present invention.

[0062] This invention provides a running route recommendation method, comprising:

[0063] Step 101: When a route recommendation request is received from the user, the basic route requirements and multi-dimensional environmental data corresponding to the route recommendation request are obtained in real time.

[0064] A route recommendation request refers to a request-type message used to trigger the running route recommendation function. It can be implemented through virtual / physical buttons, template submission, or other information submission methods.

[0065] Basic route requirements refer to the core needs information that are explicitly or implicitly included when a user initiates a route recommendation request, including but not limited to the starting point, ending point, expected distance, and exercise goals (such as fat loss or training).

[0066] Multi-dimensional environmental data refers to various environmental data along the route that match the basic route requirements, including but not limited to air quality data, terrain data, roadside landscape data, road safety data, and weather data.

[0067] In this embodiment of the invention, the method can be implemented through an APP or terminal device. When a user registers or logs in, the system obtains authorization for information collection and connects to map software or other similar software to acquire multi-dimensional environmental data. Taking the APP implementation as an example, when a user inputs a running route recommendation request through the APP, the system first extracts the basic route requirements, including the user-specified starting point (or automatically obtains the current location), expected running distance, preset exercise goals, and other core information. Simultaneously, by calling third-party data interfaces, connecting to virtual city data sources, and integrating map point-of-interest data, the system collects multi-dimensional environmental data in real time, including air quality, terrain features, road safety conditions, weather conditions, and landscape resources in the area and its surroundings, ensuring that the data covers key environmental influencing factors throughout the running process.

[0068] In addition, it can predict basic route requirements based on users' historical running records, such as recommending running distances and starting points that users have frequently chosen recently by default; it adds a noise pollution data dimension, avoiding noisy sections such as main roads and construction areas through the urban noise monitoring interface; and it allows users to authorize smart home devices to work together, such as using indoor air quality sensors to determine whether to prioritize recommending indoor running routes.

[0069] In another example of the invention, the method further includes the following steps:

[0070] In response to user-inputted registration requests, collect basic information authorized by the user, route preference information, and historical running records;

[0071] The running ability assessment model is invoked to determine the user's corresponding running ability level based on basic information and historical running records;

[0072] User profiles are built using basic information, route preference information, and running ability levels.

[0073] In this embodiment of the invention, after a user enters a registration request, the APP or terminal responds and provides a user authorization page. After obtaining the user's authorization, a user basic information input page is displayed, allowing the user to enter basic information such as gender, age, height, and weight, set running goals (weight loss, improved physical fitness, marathon training, etc.), and route preference information such as "likes park greenways," "prefers flat routes," and "likes challenges." After each run, the user can rate and tag the route.

[0074] After the user submits their input, the system associates this basic information and route preference information with the user's account. It also interfaces with major sports platforms to retrieve historical running records or allow users to upload custom data, once authorized by the user. After collecting and initially cleaning the three types of information, a pre-defined running ability assessment model is invoked. Using the collected basic information and historical running records as input, the model analyzes multi-dimensional sports-related indicators to derive a running ability level that matches the user's actual athletic performance. Specifically, the assessment dimensions may include, but are not limited to, aerobic capacity level (beginner / intermediate / advanced / professional), endurance index (based on the longest historical running distance and completion time), hill climbing ability (based on the relationship between gradient and pace on historical routes), and recovery ability (based on the time interval and performance between two consecutive runs).

[0075] Finally, the basic information, route preference information, and running ability level obtained from the assessment are structurally integrated, multi-dimensional feature tags are added through tagging, and potential sports demand features are supplemented by data correlation analysis to build a comprehensive user profile that can accurately reflect the user's sports characteristics, preferences, and ability level.

[0076] Step 102: Call the route planning model to plan routes according to the basic route requirements and obtain multiple candidate routes;

[0077] After obtaining the basic route requirements, the algorithm is parsed to extract the user-defined starting point (current location or specified location) and expected distance (e.g., 5 kilometers). Taking the user's basic route requirements as the core constraint, and combining regional road network data, a graph theory algorithm incorporating multi-objective optimization logic is adopted. Based on traditional path planning, factors such as road connectivity and environmental adaptability are considered to generate multiple candidate routes with different directions and covering diverse environmental characteristics. This satisfies the basic requirements such as starting point and distance, and provides sufficient screening samples for subsequent comprehensive scoring.

[0078] It should be noted that the route planning model can integrate graph theory algorithms and multi-objective constraint logic to generate multiple candidate routes that meet the basic requirements.

[0079] In addition, when generating candidate routes, a closed-loop route priority generation logic can be added to prioritize recommending routes with similar start and finish points, thereby improving the convenience of running for users; and by combining real-time public event information (such as marathon events and park activities), temporary densely populated sections can be avoided to optimize the feasibility of candidate routes.

[0080] Step 103: Based on multi-dimensional environmental data and user profiles, calculate the comprehensive score of each candidate route, select the candidate route with the highest comprehensive score and optimize it to obtain the target route;

[0081] User profiles refer to comprehensive user characteristic models built based on basic user information, historical running data, ability assessment results, and personal preference tags.

[0082] In this embodiment, after obtaining multiple candidate routes, the user profile is retrieved to obtain basic physical information, running ability level, endurance index, route preference and other features. Each candidate route is comprehensively scored according to a preset scoring system from dimensions such as environmental health, difficulty matching, landscape experience and safety. The candidate route with the highest score is selected, and then optimization is completed by automatically avoiding dangerous areas, optimizing the turning frequency at intersections and enriching the landscape diversity of the round trip route, so as to finally determine the target route.

[0083] In addition, a temporary demand adaptation mechanism can be added, dynamically adding special scoring factors for temporary needs such as running with children or pets.

[0084] In one example of the present invention, the multi-dimensional environmental data includes air quality data, terrain data, roadside landscape data, road safety data, and weather data; step 103 may include the following sub-steps S11-S14:

[0085] S11. Based on the user demand information within the basic route requirements, the route preference information within the user profile, and the air quality data, and combined with the preset weight adjustment rules, adjust the weight values ​​of each rating dimension to the target value to obtain the dimension weights.

[0086] User demand information includes, but is not limited to, running starting point, expected distance such as 5 kilometers, and exercise goals (such as weight loss, improving physical fitness, marathon training, etc.). Route preference information refers to the user profile's description of preferences for running route type, environmental characteristics, etc. (such as preference for park greenways, flat sections, etc.).

[0087] In this embodiment, the weights of each rating dimension can be adjusted in real time based on user needs, route preferences, and air quality data to obtain the dimensional weights for different rating dimensions. Specifically, if a user marks "want an easy run today," the weight of difficulty matching is reduced, while the weight of landscape experience is increased; if the user's goal is "marathon training," the weight of difficulty matching is increased; when air quality is poor, the weight of environmental health is automatically increased, and routes with better air quality are recommended first.

[0088] S12. Calculate the comprehensive score for each candidate route based on the weights of each dimension and air quality data, terrain data, roadside landscape data, road safety data, and weather data.

[0089] Optionally, S12 may include the following sub-steps:

[0090] By matching air quality data and weather data with a quality scoring table, the air quality score corresponding to each candidate route is determined.

[0091] Based on the road surface type and slope of each road segment in the terrain data, match the road difficulty coefficient corresponding to each candidate route;

[0092] The landscape richness of each candidate route is determined by summing scores based on the landscape type and quantity in the roadside landscape data.

[0093] Each candidate route is matched with a safety score based on road safety data.

[0094] The difficulty coefficients of each route and the running level in the user profile are substituted into the preset difficulty matching formula to generate the difficulty matching degree for each candidate route;

[0095] The air quality score, difficulty matching degree, landscape richness and safety score are weighted and superimposed using the weights of each dimension to calculate the comprehensive score of each candidate route.

[0096] In this embodiment, air quality data can be obtained by calling third-party APIs (such as the China National Environmental Monitoring Centre and Hefeng Weather) to obtain real-time AQI (Air Quality Index), PM2.5, PM10, ozone, and other data for the user's location and surrounding areas, with accuracy down to the street level. An air quality scoring system is established based on this data: AQI 0-50 (Excellent): 9-10 points; AQI 51-100 (Good): 7-8 points; AQI 101-150 (Lightly Polluted): 4-6 points; AQI 151 and above (Moderate and above Pollution): 0-3 points. Terrain data is collected by accessing a high-precision map API to obtain slope data and road surface types such as asphalt, synthetic running track, dirt road, and cobblestone road for each road segment. Each road segment is then graded by slope: flat (0-3%): difficulty coefficient 1.0; gentle slope (3-5%): difficulty coefficient 1.2; medium slope (5-8%): difficulty coefficient 1.5; steep slope (>8%): difficulty coefficient 2.0, etc. Landscape data is obtained by statistically analyzing surrounding parks, lakes, rivers, historical buildings, and other scenic resources based on map points of interest and other data. For example, +2 points are awarded for every kilometer of natural landscape such as parks / lakes, +1 point for every kilometer of historical buildings / landmarks, and 0 points for areas primarily consisting of buildings or industrial zones. Road safety data collection can be achieved by connecting to the city traffic management department's API or crawling construction notices, marking construction sections and their expected completion times. Based on publicly available security incident data and user reports, security risk areas can be marked, real-time traffic flow data can be obtained to avoid densely trafficked sections, and the distribution of streetlights can be marked, prioritizing well-lit sections for night runs. Weather data is collected by obtaining real-time weather, temperature, humidity, wind speed, and precipitation probability data to calculate a weather suitability score.

[0097] For example, the following dimensions are used to calculate the comprehensive score for each candidate route: environmental health (40% weight): air quality score × 0.4; difficulty matching (25% weight): the degree to which the overall difficulty of the route matches the user's ability; landscape experience (20% weight): landscape richness score; and safety (15% weight): no construction, good public security, and traffic safety.

[0098] Its comprehensive scoring formula is as follows:

[0099]

[0100] in, For comprehensive scoring, Rate the air quality. For difficulty matching, For landscape richness, Rate the safety.

[0101] The formula for calculating the difficulty matching score is:

[0102]

[0103] This ensures that the difficulty of the route matches the user's ability, neither too easy nor too difficult.

[0104] S13. Select the candidate route with the highest overall score as the intermediate route;

[0105] S14. Optimize the intermediate route according to the preset optimization rules to obtain the target route.

[0106] Furthermore, S14 may include the following sub-steps:

[0107] If the intermediate route passes through the marked area, the route planning model is called to replan the road segments corresponding to the marked area to obtain the first updated route that avoids the marked area.

[0108] If the number of turns in the first updated route exceeds a preset threshold, the route planning model is invoked to replan the first updated route until the number of turns is lower than the threshold, thus obtaining the second updated route.

[0109] If the second updated route includes the exact same round-trip sections, then with the goal of increasing the amount of landscape data along the route, the route planning model is invoked to replan the second updated route and generate the target route.

[0110] In this embodiment of the invention, the selected intermediate routes are first checked for path verification. If an intermediate route is detected to pass through a preset marked area, the route planning model is immediately invoked to replan the corresponding route segments with the core objective of avoiding the marked area, generating a first updated route that does not pass through the marked area. Subsequently, the number of turns of the first updated route is counted and compared with a preset threshold. If the number of turns exceeds the threshold, the route planning model is continuously invoked to iteratively replan the first updated route, optimizing intersection selection and route orientation, until the number of turns is lower than the threshold, resulting in a second updated route with smooth driving. Finally, it is checked whether the second updated route contains completely identical round-trip segments. If such a situation exists, the route planning model is invoked to redesign the round-trip segments with the core objective of increasing the amount of roadside landscape data, prioritizing roads that pass through different landscape resources, and generating a target route that balances safety, smoothness, and landscape experience.

[0111] Furthermore, priority grading is introduced during the verification of marked areas. Avoidance strategies are adopted for high-risk marked areas (such as sudden construction or security incident sites), while the avoidance method can be flexibly adjusted based on route smoothness for low-impact areas (such as temporary densely populated areas). The frequency threshold can be dynamically adapted based on user profiles, such as setting lower turning thresholds for beginner and senior runners, while retaining challenging and complex route options for professional runners. When replanning the second updated route, the intersection turning types are optimized simultaneously, prioritizing the avoidance of intersections with continuous sharp turns and poor visibility to improve running safety. When optimizing the round-trip route, in addition to adding landscape data, the road surface type and landscape type are matched in combination with user preference tags (such as liking rubber tracks or natural landscapes), while controlling the total distance deviation within the preset range to avoid deviating from the user's expected running distance. A new route smoothness check has been added, prioritizing the selection of road sections with smooth surfaces during replanning to further improve the overall experience of the target route.

[0112] Step 104: When the user inputs a running start signal, periodically collect the user's physiological data and route warning information;

[0113] Physiological data refers to key indicators reflecting a user's physical condition during running, including but not limited to pace (real-time speed, average pace, and target pace comparison), heart rate (real-time heart rate, heart rate zone determination), cadence, stride length, cumulative distance, and cumulative time. This data can be obtained through wearable devices such as the user's mobile phone sensors, smart bracelets, or smartwatches.

[0114] Route warning information refers to real-time risk information related to the target route that may affect safety or the running experience, such as construction, road damage, street light malfunctions, stray dogs, or traffic.

[0115] In this embodiment of the invention, after receiving a running start signal sent by the user through an APP or wearable device, a periodic data collection process is initiated. The user's physiological data such as pace, heart rate, and cadence are collected in real time through the mobile phone sensor or the connected smart wearable device. At the same time, a continuous connection with the server is maintained to periodically obtain real-time roadside construction information, traffic accidents, security risks, weather changes, and other route warning information along the route, ensuring the real-time and continuous nature of data collection.

[0116] In addition, the collection frequency can be dynamically adjusted according to the user's real-time exercise intensity, and the interval can be shortened when the exercise intensity is high to improve the response speed; a new user active reporting channel has been added, allowing real-time uploading of abnormal road conditions such as road damage and stray dog ​​presence as supplementary warning information; and it can be linked with urban intelligent traffic sensors, road lighting equipment and other public facilities to obtain more accurate road condition and environmental warning data.

[0117] Step 105: Based on physiological data and / or route warning information, generate route change recommendation information and push it to the user's wearable device.

[0118] Change recommendation information refers to suggestions generated based on the user's real-time physiological status or route warning information, used to adjust the target route, including adjusting direction and alternative routes.

[0119] In this embodiment, after collecting physiological data and route warning information, the data is analyzed in real time. If changes in the user's physiological state, such as abnormal heart rate or pace deviating from the target range, are detected, or if warning information such as newly added dangerous areas or sudden weather changes appear on the route ahead, the target route is replanned according to the physiological data and / or route warning information to generate corresponding route detour plans, distance adjustment suggestions, rest point recommendations, and other change recommendation information. This information is then pushed to the user's wearable device through voice reminders, vibration alerts, and other means to ensure that the user receives and responds in a timely manner.

[0120] In addition, multiple change options can be provided and marked with options such as the fastest detour and the gentlest route for users to choose from; the push notification method can be automatically matched according to the user's historical preferences, such as prioritizing voice reminders if the user prefers voice; if the user is in a running invitation state, the new route and the reason for the change will be pushed to fellow runners at the same time after the route is changed, ensuring the coordination of the team running.

[0121] In one example of the present invention, step 105 may include the following sub-steps:

[0122] When physiological data matches preset route change conditions, the corresponding updated route segment is retrieved according to the matching result, and the change recommendation information of the target route is generated.

[0123] When the warning area corresponding to the route warning information overlaps with the target route, calculate the distance to be moved between the user's current location and the warning area;

[0124] If the distance to be moved is greater than or equal to the preset distance threshold, the route planning model is invoked to replan the target route and generate change recommendation information to avoid the warning area.

[0125] If the distance to be moved is less than the preset distance threshold, the route planning model is invoked to generate change recommendation information from the current location to the nearest safe area;

[0126] The updated recommendation information is pushed to the user's wearable device for display.

[0127] Route change conditions refer to the physiological data triggering criteria preset by the system to determine whether the target route needs to be adjusted. These include, but are not limited to, heart rate that is consistently higher than a certain threshold and pace that differs from historical pace by more than a certain threshold.

[0128] In this embodiment of the invention, user physiological data collected from mobile phone sensors or connected smart wearable devices is received and compared in real time with preset route change conditions. When the physiological data meets the change conditions, the system combines the user's current location, the distribution of the target route segments, and surrounding road resources to retrieve updated route segments suitable for the user's physical condition, generating change recommendation information that includes route adjustment direction and advantages of alternative route segments. For example, if the heart rate is consistently higher than a certain threshold, the system recommends nearby moderate road segments or rest points as updated route segments; if the pace is significantly lower than historical levels, the system prompts the user to shorten the route.

[0129] Simultaneously, the system monitors route warning information in real time. In the event of sudden severe weather, it plans the route to the nearest shelter or terminates the target route. It can also extract the warning area and determine its spatial overlap with the target route. If overlap exists, it calculates the distance between the user's current location and the warning area using geographic information system technology. If this distance is greater than or equal to a preset distance threshold, it immediately invokes the route planning model to replan the route while avoiding the warning area, taking into account the core requirements of the original route and generating change recommendation information, prompting "Road construction ahead, replanning your route." If the distance is less than the threshold, it invokes the route planning model to prompt the user to slow down or stop the target route, replanning the target route to the nearest safe area and generating short-distance change recommendation information from the current location to the nearest safe area. Finally, it pushes various change recommendation information to the user's wearable device via wireless communication, displaying it with a clear visual interface and vibration alerts to ensure timely access for the user.

[0130] In addition, you can set up a smart voice reminder system in the APP and configure voice broadcast rules to trigger voice reminders in the following situations: broadcasting the current pace, cumulative time, and heart rate after completing 1 kilometer; reminding you "too fast / too slow, adjustment recommended" when the pace deviates from the target by ±10%; reminding you "too high / too low heart rate, adjust intensity" when the heart rate exceeds the target range; reminding you "uphill / downhill ahead, adjust pace" 50 meters in advance when approaching a slope; reminding you "intersection / heavy traffic ahead, be careful" when approaching a dangerous section; and reminding you to drink water every 15-20 minutes (adjusted according to temperature and exercise intensity).

[0131] In another example of the invention, the method further includes the following steps:

[0132] When a running plan is received from the user, the matching degree between the running plan and multiple existing running plans is calculated based on the content of the running plan.

[0133] Send running invitations to target users whose match rate is higher than the matching threshold;

[0134] When a confirmation message is received from the target user in response to the running invitation, a common route is generated based on the running plan and the target user's existing running plan.

[0135] Responding to social requests from users and target users, it can create temporary chat rooms, display real-time locations, and / or generate virtual group photos.

[0136] A running plan refers to exercise arrangements posted by users that include core information such as running time, starting point, expected distance, and pace requirements.

[0137] Matching degree refers to the degree of fit calculated by comparing the running plan with other users' existing running plans based on key factors of the running plan (such as time, location, pace, distance, social preferences, etc.).

[0138] A temporary chat room refers to an instant messaging scenario set up for both parties to invite each other to a run, used for pre-run communication and in-run interaction.

[0139] Virtual group photos refer to composite group photos generated using AI technology, based on the running trajectories and image data of both users, without requiring them to be in the same frame. This is done with the authorization of both users.

[0140] In this embodiment, upon receiving a running plan uploaded by any user, the running plan is parsed to extract information such as time, starting point, expected distance, pace, whether reservations are open, and social preferences. The app or system then compares this information with corresponding elements from multiple existing running plans and calculates the matching degree between each existing running plan and the new running plan using a preset algorithm. The calculation formula is as follows:

[0141]

[0142] in, For matching degree, The ability matching degree can be obtained by similarity calculation based on pace and expected distance. The distance between the starting points is 1 if it is less than 2 kilometers, otherwise it is 0. The time matching score is 1 if the time difference is less than 30 minutes, otherwise it is 0. For social preferences, such as gender requirements and age group, a match is assigned a value of 1, and a non-match is assigned a value of 0.

[0143] Users with a match score higher than the system's preset matching threshold are identified as target users, and running invitations containing the initiator's running plan's core information and matching highlights are sent to them. Upon receiving a confirmation response from the target user, the system integrates the common needs and personalized preferences of the initiator's running plan and the target user's existing running plan, calls the route planning model to design a comprehensive route, and generates a shared route that takes into account both parties' distance, pace, and route preferences. Subsequently, in response to social requests initiated by both parties, a temporary chat room is automatically set up for pre-run or real-time communication. During the run, the real-time location of each other is displayed through map visualization to facilitate meeting and interaction. After the run, a virtual group photo is generated based on both parties' running tracks and exercise data to enrich the social experience.

[0144] Furthermore, real-time environmental data can be integrated into shared route planning, prioritizing routes with high environmental health and safety, while reserving flexible adjustment points to allow for minor route tweaks during the running invitation process. Temporary chat rooms now include features such as reminders for sports aid stations and synchronized weather alerts, enhancing usability. Additionally, a running invitation credit system can be established, using user credit scores based on mutual feedback to lower matching priority for users who frequently cancel, thus ensuring a higher success rate for running invitations.

[0145] In another example of the invention, the method further includes the following steps:

[0146] In response to the user's authorized instructions, physiological data and running trajectory are stored in a local database;

[0147] User profiles are updated periodically based on physiological data and running routes;

[0148] The route planning model is updated periodically in response to user ratings of the target route.

[0149] In this embodiment, after receiving authorization from the user, the physiological data collected during the run and the recorded running trajectory are securely stored in encrypted form in the user's local database, such as an SQLite database, to ensure data privacy and security. Simultaneously, the physiological data and running trajectory are extracted from the local database at preset intervals, and the core features in the user profile, such as running ability level, endurance index, and route preference tags, are updated based on data change trends to maintain consistency between the user profile and the actual exercise status. Furthermore, user ratings of the target route are continuously collected, and this feedback data is summarized and analyzed at fixed intervals. The data features are then used to optimize key parameters in the route planning model, such as the weighting of rating dimensions and the route generation logic, improving the accuracy and adaptability of the model's recommended routes.

[0150] Furthermore, with user authorization via the cloud, running tracks and physiological data stored in the local database can be periodically encrypted and synchronized to the cloud server. Simultaneously, during the training of the route planning model, user ratings and feedback on the route can be collected periodically to analyze the discrepancy between the user's actual performance and predictions. A / B testing logic is incorporated into route planning model updates to simultaneously verify the effectiveness of different optimization strategies, prioritizing adjustments that improve user route satisfaction. After each run, users are invited to rate the route (1-5 stars) and add tags, collecting feedback on the frequency and content of voice reminders to ensure accurate updates to the route planning model.

[0151] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0152] The running route recommendation device provided in the embodiments of the present invention is described below. The running route recommendation device described below can be referred to in correspondence with the running route recommendation method described above.

[0153] Please see Figure 2 This invention provides a running route recommendation device, comprising:

[0154] The data acquisition module 201 is used to acquire the basic route requirements and multi-dimensional environmental data corresponding to the route recommendation request in real time when a route recommendation request is received from the user.

[0155] The route planning module 202 is used to call the route planning model to plan routes according to the basic route requirements and obtain multiple candidate routes.

[0156] The route optimization module 203 is used to calculate the comprehensive score of each candidate route based on multi-dimensional environmental data and user profiles, select the candidate route with the highest comprehensive score and optimize it to obtain the target route;

[0157] The running data acquisition module 204 is used to periodically collect the user's physiological data and route warning information when it receives a running start signal input by the user.

[0158] The real-time running route recommendation module 205 is used to generate change recommendation information for the target route based on physiological data and / or route warning information and push it to the user's wearable device.

[0159] Optionally, the multi-dimensional environmental data includes air quality data, terrain data, roadside landscape data, road safety data, and weather data; the route optimization module 203 includes:

[0160] The dimension weight determination submodule is used to adjust the weight values ​​of each rating dimension to the target value based on user demand information within the basic route requirements, route preference information within the user profile, and air quality data, combined with preset weight adjustment rules, to obtain the dimension weights.

[0161] The comprehensive score calculation submodule is used to calculate the comprehensive score of each candidate route according to the weight of each dimension and air quality data, terrain data, roadside landscape data, road safety data, and weather data.

[0162] The intermediate route selection submodule is used to select the candidate route with the highest overall score as the intermediate route;

[0163] The route optimization submodule is used to optimize intermediate routes according to preset optimization rules to obtain the target route.

[0164] Optionally, the comprehensive score calculation submodule is specifically used for:

[0165] By matching air quality data and weather data with a quality scoring table, the air quality score corresponding to each candidate route is determined.

[0166] Based on the road surface type and slope of each road segment in the terrain data, match the road difficulty coefficient corresponding to each candidate route;

[0167] The landscape richness of each candidate route is determined by summing scores based on the landscape type and quantity in the roadside landscape data.

[0168] Each candidate route is matched with a safety score based on road safety data.

[0169] The difficulty coefficients of each route and the running level in the user profile are substituted into the preset difficulty matching formula to generate the difficulty matching degree for each candidate route;

[0170] The air quality score, difficulty matching degree, landscape richness and safety score are weighted and superimposed using the weights of each dimension to calculate the comprehensive score of each candidate route.

[0171] Optionally, the route optimization submodule is specifically used for:

[0172] If the intermediate route passes through the marked area, the route planning model is called to replan the road segments corresponding to the marked area to obtain the first updated route that avoids the marked area.

[0173] If the number of turns in the first updated route exceeds a preset threshold, the route planning model is invoked to replan the first updated route until the number of turns is lower than the threshold, thus obtaining the second updated route.

[0174] If the second updated route includes the exact same round-trip sections, then with the goal of increasing the amount of landscape data along the route, the route planning model is invoked to replan the second updated route and generate the target route.

[0175] Optionally, the real-time running track recommendation module 205 is specifically used for:

[0176] When physiological data matches preset route change conditions, the corresponding updated route segment is retrieved according to the matching result, and the change recommendation information of the target route is generated.

[0177] When the warning area corresponding to the route warning information overlaps with the target route, calculate the distance to be moved between the user's current location and the warning area;

[0178] If the distance to be moved is greater than or equal to the preset distance threshold, the route planning model is invoked to replan the target route and generate change recommendation information to avoid the warning area.

[0179] If the distance to be moved is less than the preset distance threshold, the route planning model is invoked to generate change recommendation information from the current location to the nearest safe area;

[0180] The updated recommendation information is pushed to the user's wearable device for display.

[0181] Optionally, the device also includes a user profile building module, specifically used for:

[0182] In response to user-inputted registration requests, collect basic information authorized by the user, route preference information, and historical running records;

[0183] The running ability assessment model is invoked to determine the user's corresponding running ability level based on basic information and historical running records;

[0184] User profiles are built using basic information, route preference information, and running ability levels.

[0185] Optionally, the device also includes a running invitation module, specifically used for:

[0186] When a running plan is received from the user, the matching degree between the running plan and multiple existing running plans is calculated based on the content of the running plan.

[0187] Send running invitations to target users whose match rate is higher than the matching threshold;

[0188] When a confirmation message is received from the target user in response to the running invitation, a common route is generated based on the running plan and the target user's existing running plan.

[0189] Responding to social requests from users and target users, it can create temporary chat rooms, display real-time locations, and / or generate virtual group photos.

[0190] Optionally, the device also includes a data storage and update module, specifically used for:

[0191] In response to the user's authorized instructions, physiological data and running trajectory are stored in a local database;

[0192] User profiles are updated periodically based on physiological data and running routes;

[0193] The route planning model is updated periodically in response to user ratings of the target route.

[0194] This invention provides an electronic device, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the running route recommendation method as described in any embodiment of this invention.

[0195] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0196] In the several embodiments provided by this invention, 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 is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules 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 modules may be electrical, mechanical, or other forms.

[0197] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0198] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0199] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 the present invention.

Claims

1. A running course recommendation method characterized by comprising: The method comprises the following steps: when receiving a route recommendation request input by a user, acquiring in real time basic route requirements and multi-dimensional environment data corresponding to the route recommendation request; calling a route planning model to plan a route according to the basic route requirements, to obtain a plurality of candidate routes; calculating a comprehensive score of each candidate route according to the multi-dimensional environment data and a user portrait of the user, selecting a candidate route with the highest comprehensive score, and optimizing the candidate route to obtain a target route; when receiving a running start signal input by the user, periodically collecting physiological data and route warning information of the user; generating change recommendation information of the target route according to the physiological data and / or the route warning information, and pushing the change recommendation information to a wearable device of the user.

2. The running course recommendation method according to claim 1, characterized by, The multi-dimensional environment data comprises air quality data, terrain data, landscape data along a route, road safety data, and weather data; the step of calculating a comprehensive score of each candidate route according to the multi-dimensional environment data and a user portrait of the user, selecting a candidate route with the highest comprehensive score, and optimizing the candidate route to obtain a target route comprises the following steps: adjusting a weight value of each score dimension to a target value according to user demand information in the basic route requirements, route preference information in the user portrait, and the air quality data, and combining a preset weight adjustment rule to obtain a dimension weight; calculating a comprehensive score of each candidate route according to the dimension weight and the air quality data, the terrain data, the landscape data along the route, the road safety data, and weather data; selecting a candidate route with the highest comprehensive score as an intermediate route; optimizing the intermediate route according to a preset optimization rule to obtain a target route.

3. The running course recommendation method according to claim 2, characterized by, The step of calculating a comprehensive score of each candidate route according to the dimension weight and the air quality data, the terrain data, the landscape data along the route, the road safety data, and weather data comprises the following steps: determining an air quality score corresponding to each candidate route according to the air quality data and a quality score table; determining a road segment difficulty coefficient corresponding to each candidate route according to a road surface type and a slope of each road segment in the terrain data; determining landscape richness of each candidate route by score superposition according to a landscape type and a landscape quantity in the landscape data along the route; determining a safety score corresponding to each candidate route based on the road safety data; generating a difficulty matching degree corresponding to each candidate route by substituting each road segment difficulty coefficient and a running level in the user portrait into a preset difficulty matching degree calculation formula; weighting and superimposing the air quality score, the difficulty matching degree, the landscape richness, and the safety score according to each dimension weight to calculate a comprehensive score of each candidate route.

4. The running course recommendation method according to claim 2, characterized by, The step of optimizing the intermediate route according to a preset optimization rule to obtain a target route comprises the following steps: if the intermediate route passes through a marked area, calling the route planning model to re-plan a passing road segment corresponding to the marked area to obtain a first updated route that avoids the marked area; If the number of turns of the first updated route is greater than a preset number threshold, the route planning model is called to re-plan the first updated route until the number of turns is less than the number threshold, to obtain a second updated route; If the second updated route includes the same return route segment, the route planning model is called to re-plan the second updated route to generate a target route, with the goal of increasing the amount of data of the along-the-way landscape data.

5. The running course recommendation method according to claim 1, characterized by, The step of generating the change recommendation information of the target route according to the physiological data and / or the route warning information and pushing the change recommendation information to the wearable device of the user includes: When the physiological data matches a preset route change condition, the corresponding updated route segment is retrieved according to the matching result to generate the change recommendation information of the target route; When the warning area corresponding to the route warning information overlaps with the target route, the distance to be moved between the current position of the user and the warning area is calculated; If the distance to be moved is greater than or equal to a preset distance threshold, the route planning model is called to re-plan the target route to generate change recommendation information to avoid the warning area; If the distance to be moved is less than the preset distance threshold, the route planning model is called to generate change recommendation information from the current position to the nearest safe area; The change recommendation information is pushed to the wearable device of the user for display.

6. The running course recommendation method according to claim 1, characterized by, Further comprising: In response to a registration request input by the user, basic information, route preference information and historical running records authorized by the user are collected; A running ability evaluation model is called to determine the running ability level of the user according to the basic information and the historical running records; The basic information, the route preference information and the running ability level are used to construct a user portrait.

7. The running course recommendation method according to claim 1, characterized by, Further comprising: When the running plan input by the user is received, the matching degree between the running plan and a plurality of existing running plans is calculated according to the plan content of the running plan; Running invitation information is sent to a target user whose matching degree is higher than a matching threshold; When confirmation information returned by the target user in response to the running invitation information is received, a common route is generated according to the running plan and the existing running plan of the target user; In response to a social request of the user and the target user, a temporary chat room is established, real-time positions are displayed and / or virtual photos are generated.

8. The running course recommendation method according to claim 1, characterized by, Further comprising: In response to an authorization instruction of the user, the physiological data and the running trajectory are stored in a local database; The user portrait is periodically updated according to the physiological data and the running trajectory; In response to score information of the target route input by the user, the route planning model is periodically updated.

9. A running course recommendation device characterized by comprising: Comprising: A data acquisition module is configured to, when a route recommendation request input by a user is received, acquire basic route requirements and multi-dimensional environmental data corresponding to the route recommendation request in real time; A route planning module is configured to call a route planning model to plan a route according to the basic route requirements, to obtain a plurality of candidate routes; a route optimization module, configured to calculate a comprehensive score of each of the candidate routes according to the multi-dimensional environment data and the user portrait of the user, select a candidate route with the highest comprehensive score, and optimize the candidate route to obtain a target route; a running data collection module, configured to periodically collect physiological data of the user and route alert information when a running start signal input by the user is received; a real-time route recommendation module, configured to generate change recommendation information of the target route according to the physiological data and / or the route alert information, and push the change recommendation information to a wearable device of the user.

10. An electronic device, comprising: a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to make the processor perform the steps of the running route recommendation method according to any one of claims 1-8.

Images