Personalized recommendation method and system for riding vehicle and related equipment

Abstract

The invention provides a personalized recommendation method and system for a riding vehicle and related equipment, and the method comprises the steps: collecting the historical riding data of a user, and the historical riding data comprises the driving track data, the driving state data, the pedal pressure data and the user portrait data; analyzing the historical riding data to obtain road section slope feature data; constructing a riding recommendation model based on the historical riding data, the road section gradient feature data and a user preference / road condition multi-dimensional vehicle model recommendation algorithm; and recommending a riding vehicle for each riding of the user through the riding recommendation model. According to the vehicle recommendation method, during model construction, historical riding data of the user is combined, road section gradient feature data is also combined, user preferences and road condition multiple dimensions can be considered during vehicle recommendation, and the user has better vehicle recommendation experience.

Description

Technical Field

[0001] The present invention relates to the technical field of shared bicycles, and in particular to a personalized recommendation method, system, and related equipment for riding bicycles. Background Art

[0002] Currently, the primary recommendation method in the shared bike sector is to recommend the nearest bike. The main advantages of recommending the nearest bike are simplicity, simple distance determination, and improved user experience in finding a bike. However, its main drawbacks include: 1) failure to consider the user's historical riding habits, which may result in the recommended bike not meeting the user's needs; and 2) failure to consider the geographical environment in which the user may ride, such as uphill or downhill terrain, which may result in a poor user experience. Therefore, the existing technology lacks a method for multi-dimensional bike recommendation.

[0003] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0004] The present invention provides a method, system and related equipment for personalized recommendation of cycling vehicles. The main purpose of the present invention is to solve the technical problems mentioned in the background technology of the existing technology.

[0005] A first aspect of the present invention provides a personalized recommendation method for a cycling vehicle, comprising:

[0006] Collecting historical riding data of the user, including riding trajectory data, riding status data, pedal pressure data, and user profile data;

[0007] Analyzing the historical riding data to obtain road section slope characteristic data;

[0008] A cycling recommendation model is constructed based on the historical cycling data, the road section slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition;

[0009] The riding recommendation model is used to recommend a riding vehicle for each riding of the user.

[0010] In an optional implementation of the first aspect of the present invention, analyzing the historical riding data to obtain the road section slope characteristic data includes:

[0011] For each driving trajectory, construct a vehicle speed time variation curve and a pedal pressure time variation curve of the driving trajectory;

[0012] Performing time sequence alignment on the vehicle speed time variation curve and the pedal pressure time variation curve;

[0013] The velocity change rate and pressure change rate of each window are obtained in sequence along the time axis through the preset sliding window;

[0014] Matching the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic;

[0015] The road section slope characteristic data of each driving trajectory is obtained based on the matching result.

[0016] In an optional implementation of the first aspect of the present invention, matching the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic includes:

[0017] For each of the windows, performing a positive or negative judgment on the velocity change rate and the pressure change rate;

[0018] If the speed change rate is negative and the pressure change rate is positive, further determining whether the speed change rate is less than a preset first negative threshold and whether the pressure change rate is greater than a preset first positive threshold;

[0019] If the speed change rate is positive and the pressure change rate is negative, it is further determined whether the speed change rate is greater than a preset second positive threshold and whether the pressure change rate is less than a preset second negative threshold;

[0020] If the speed change rate is less than the first negative threshold and the pressure change rate is greater than the first positive threshold, the window is determined to be an uphill slope; if the speed change rate is greater than the second positive threshold and the pressure change rate is less than the second negative threshold, the window is determined to be a downhill slope; if the speed change rate and the pressure change rate are other situations, the window is determined to be a flat road.

[0021] In an optional implementation of the first aspect of the present invention, obtaining the road section slope characteristic data of each driving trajectory based on the matching result includes:

[0022] For each driving trajectory, adjacent windows with the same uphill and downhill attributes are selected along the time axis; if the adjacent windows are all uphill, the path segment corresponding to the adjacent windows is determined to be uphill; if the adjacent windows are all downhill, the path segment corresponding to the adjacent windows is determined to be downhill;

[0023] For the window that cannot have the same uphill and downhill attributes as the left and right adjacent windows, it is determined that the path segment corresponding to the window is a flat road;

[0024] All the path segments are combined with similar uphill and downhill attributes to obtain the road segment slope characteristic data of each driving trajectory.

[0025] In an optional implementation of the first aspect of the present invention, sequentially acquiring the velocity change rate and the pressure change rate of each window along the time axis through a preset sliding window includes:

[0026] For each of the windows, obtaining the maximum speed value, the minimum speed value, the maximum pedal pressure value, and the minimum pedal pressure value within the window time;

[0027] The speed change rate is obtained by subtracting the first time from the last time from the maximum speed value and the minimum speed value, and then dividing the difference by the window time;

[0028] The pressure change rate is obtained by subtracting the earlier time from the later time to calculate the difference between the maximum pedaling pressure value and the minimum pedaling pressure value, and then dividing the difference by the window time.

[0029] In an optional implementation of the first aspect of the present invention, the cycling recommendation model constructed based on the historical cycling data, the road section slope characteristic data, and the user preference / road condition multi-dimensional vehicle type recommendation algorithm includes:

[0030] Draw a historical riding trajectory on a preset map based on the historical riding data;

[0031] Adding the road section slope characteristic data to the historical riding trajectory to obtain a road section characteristic map;

[0032] Classifying all riding vehicles to obtain a vehicle classification library of the riding vehicles;

[0033] Establishing associations between various types of riding vehicles in the vehicle classification library and different user portraits and riding scenarios under different road conditions;

[0034] A cycling recommendation model is constructed based on the road section feature map, the vehicle type classification library and the association relationship.

[0035] In an optional implementation of the first aspect of the present invention, recommending a cycling vehicle for each ride of the user using the cycling recommendation model includes:

[0036] Receive the user's input of the riding start point and riding end point;

[0037] Planning a path between the cycling starting point and the cycling end point using the shortest path;

[0038] Matching the planned path with the road section feature map;

[0039] Determining a road segment overlap ratio between the planned path and the road segment feature map;

[0040] If the road segment overlap ratio is less than a preset threshold, recommending a vehicle type based on the preference of the user profile data;

[0041] If the road section overlap ratio is greater than the preset threshold, the uphill and downhill ratios of the overlapping road sections are further obtained, and vehicle type recommendations are made based on the uphill and downhill ratios.

[0042] A second aspect of the present invention provides a personalized recommendation system for riding vehicles, the personalized recommendation system for riding vehicles comprising:

[0043] A data collection module is used to collect the user's historical riding data, wherein the historical riding data includes driving trajectory data, driving status data, pedal pressure data, and user portrait data;

[0044] A data analysis module is used to analyze the historical riding data to obtain road slope characteristic data;

[0045] A model building module is used to build a cycling recommendation model based on the historical riding data, the road slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition;

[0046] The vehicle recommendation module is used to recommend a riding vehicle for each ride of the user through the riding recommendation model.

[0047] A third aspect of the present invention provides a terminal device, comprising: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line;

[0048] The at least one processor calls the instructions in the memory to enable the terminal device to execute the personalized recommendation method for a bicycle as described in any one of the first aspects of the present invention.

[0049] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for personalized recommendation of a riding vehicle as described in any one of the first aspects of the present invention is implemented.

[0050] Beneficial effects: The present invention provides a personalized recommendation method, system and related equipment for cycling vehicles, the method comprising collecting historical cycling data of a user, wherein the historical cycling data comprises driving trajectory data, driving status data, pedal pressure data and user portrait data; analyzing the historical cycling data to obtain road slope characteristic data; constructing a cycling recommendation model based on the historical cycling data, the road slope characteristic data and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition; and recommending a cycling vehicle for each ride of the user through the cycling recommendation model. The vehicle recommendation method of the present invention combines both the user's historical cycling data and the road slope characteristic data when constructing the model, and can take into account both user preference and multi-dimensional road condition when recommending vehicles, so that users have a better vehicle recommendation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of an embodiment of the main method steps of a personalized recommendation method for a bicycle to be ridden according to the present invention;

[0052] Figure 2 A schematic diagram of an embodiment of a personalized recommendation system for cycling vehicles according to the present invention;

[0053] Figure 3 The figure is a schematic diagram of an embodiment of a terminal device of the present invention. DETAILED DESCRIPTION

[0054] The terms "first," "second," "third," "fourth," and the like (if any) in the description and claims of the present invention and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0055] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 A first aspect of the present invention provides a personalized recommendation method for a cycling vehicle, comprising:

[0056] S100. Collect the user's historical riding data, the historical riding data includes driving trajectory data, driving status data, pedal pressure data and user portrait data; in the present invention, the trajectory data includes each historical riding trajectory of the user recorded by the system, the driving status data includes the speed change data of each riding trajectory; the pedal pressure data includes the pedal pressure change data of each riding trajectory; the user portrait data includes data such as occupation, residence / workplace and preferences. Speed ​​data monitoring can obtain the user's position changes in real time through the GPS module installed on the vehicle, and calculate the user's speed changes in real time. Pedal pressure monitoring can use the pressure sensor installed on the bicycle, and the system can record the pressure changes applied by the user on the pedals during riding in real time.

[0057] S200: Analyze the historical riding data to obtain road slope characteristic data. In step S200, the analysis primarily utilizes the driving trajectory data, driving status data, and pedal pressure data from the historical riding data. The driving trajectory data can record the locations the user has traveled on a map, and the speed data from the driving status data can not only reflect the user's riding status but also provide a preliminary assessment of terrain changes. For example, a sudden drop in speed may be due to an uphill slope. Of course, changes in pedal pressure are directly related to the user's required riding effort and can effectively assist in assessing changes in the riding environment.

[0058] In an optional implementation of step S200 of the present invention, analyzing the historical riding data to obtain the road section slope characteristic data includes:

[0059] S201. For each driving trajectory, construct a time-varying curve of vehicle speed and a time-varying curve of pedal pressure for the driving trajectory. Before using the driving state data and pedal pressure data to construct the curves, the present invention may also preprocess the data. This preprocessing includes data smoothing. Because sensor data may be affected by various factors (such as uneven roads and signal interference), generating noise, the system needs to smooth the collected speed and pressure data. This processing method includes moving average, etc., to eliminate noise and ensure data accuracy.

[0060] S202. Time-series alignment is performed on the vehicle speed time variation curve and the pedal pressure time variation curve. In the present invention, the speed data and the pressure data need to be aligned on the time axis to ensure that the speed and pressure at the same time point are synchronized. This step is crucial because it directly affects the subsequent judgment of uphill and downhill.

[0061] S203. Obtain the speed change rate and pressure change rate of each window in sequence along the time axis through a preset sliding window; in the present invention, the window length can be selected as a sliding window of 5-10 seconds to balance real-time performance and stability; in an optional implementation of the first aspect of the present invention, obtaining the speed change rate and pressure change rate of each window in sequence along the time axis through a preset sliding window includes: for each of the windows, obtaining the maximum speed value, minimum speed value, maximum pedal pressure value and minimum pedal pressure value within the window time; subtracting the time from the maximum speed value and the minimum speed value by the time in the rear and dividing the result by the window time to obtain the speed change rate; subtracting the time from the maximum pedal pressure value and the minimum pedal pressure value by the time in the rear and dividing the result by the window time to obtain the pressure change rate. Taking the speed change rate as an example, the calculation process is to first determine which of the maximum speed value and the minimum speed value comes first and which comes later. If the maximum speed value comes later in time sequence and the minimum speed value comes earlier in time sequence, the speed change rate is calculated as (the maximum speed value time - the minimum speed value) / the window time, otherwise it is calculated as (the minimum speed value time - the maximum speed value) / the window time. The pressure change rate is calculated in the same way. The present invention obtains the change rate by subtracting the minimum value from the maximum value in the time window, which can make the obtained change more significant, thereby reducing the influence of the instrument accuracy error.

[0062] S204. Match the speed change rate and the pressure change rate of each window with the preset uphill and downhill judgment logic; in the present invention, the logic of judging uphill and downhill based on the speed change rate and the pressure change rate can be simply understood as: if the speed drops significantly and the pressure rises significantly, it is an uphill slope; if the speed rises significantly and the pressure drops significantly, it is a downhill slope. If the above conditions are not met, it is considered to be in a flat road state.

[0063] In an optional embodiment of the first aspect of the present invention, matching the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic includes: for each window, performing a positive or negative judgment on the speed change rate and the pressure change rate; if the speed change rate is negative and the pressure change rate is positive, further judging whether the speed change rate is less than a preset first negative threshold and whether the pressure change rate is greater than a preset first positive threshold; if the speed change rate is positive and the pressure change rate is negative, further judging whether the speed change rate is greater than a preset second positive threshold and whether the pressure change rate is less than a preset second negative threshold; if the speed change rate is less than the first negative threshold and the pressure change rate is greater than the first positive threshold, determining that the window is uphill; if the speed change rate is greater than the second positive threshold and the pressure change rate is less than the second negative threshold, determining that the window is downhill; if the speed change rate and the pressure change rate are other situations, determining that the window is a flat road.

[0064] S205. Obtain road segment slope characteristic data for each driving trajectory based on the matching results. In the present invention, since the time window is typically short, the uphill and downhill results of the time window are affected not only by the slope of the road but also by other road conditions (e.g., uneven roads or obstacles). Therefore, the present invention requires that the road segment corresponding to multiple windows must meet the uphill condition before determining that the road segment is an uphill segment. The same principle applies to downhill determination.

[0065] In an optional embodiment of the first aspect of the present invention, obtaining the road section slope characteristic data for each driving trajectory based on the matching results includes: for each driving trajectory, filtering out adjacent windows (the number can be more than 2) with the same uphill and downhill attributes along the time axis; if the adjacent windows are all uphill, then determining that the path segment corresponding to the adjacent windows is uphill; if the adjacent windows are all downhill, then determining that the path segment corresponding to the adjacent windows is downhill; for the windows that cannot have the same uphill and downhill attributes as the left and right adjacent windows, then determining that the path segment corresponding to the window is flat; merging all the path segments with the same uphill and downhill attributes to obtain the road section slope characteristic data for each driving trajectory. In the present invention, the above-mentioned screening process mainly utilizes the principle that the state of the vehicle will not suddenly change when going uphill and downhill, that is, the vehicle's uphill and downhill states will have a certain degree of continuity. The advantages of this technical solution of the present invention are that it only relies on vehicle speed and pressure data, is low cost, and performs uphill and downhill judgment by sliding windows, which can significantly improve response speed and accuracy.

[0066] In another uphill / downhill judgment idea of ​​the present invention, the logic of uphill judgment can also be that the system analyzes the user's speed and pressure data. When the speed decreases and lasts for a period of time, and the pedal pressure increases significantly, the system determines that the user is on an uphill section. This recognition process takes advantage of the fact that the user needs to exert greater force when going uphill, while the speed naturally decreases. The logic of downhill judgment is the opposite. When the speed increases and there is no obvious change in the pedal pressure, the system determines that the user is on a downhill section. At this time, the user accelerates under the action of gravity without applying too much force.

[0067] S300. A cycling recommendation model is constructed based on the historical cycling data, the road section slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm based on user preferences / road conditions. In an optional implementation of the first aspect of the present invention, the cycling recommendation model is constructed based on the historical cycling data, the road section slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm based on user preferences / road conditions, including: drawing a historical cycling trajectory on a preset map based on the historical cycling data; adding the road section slope characteristic data to the historical cycling trajectory to obtain a road section characteristic map; classifying all cycling vehicles to obtain a vehicle model classification library of the cycling vehicles; establishing an association relationship between each type of cycling vehicle in the vehicle model classification library and different user portraits and different road condition cycling scenarios; and constructing a cycling recommendation model based on the road section characteristic map, the vehicle model classification library, and the association relationship.

[0068] In the present invention, when performing user preference analysis, a matrix decomposition module can be used to decompose user-vehicle interaction data (such as the frequency and duration of user use of various vehicles) into potential vectors to extract the user's long-term preferences. For short-term user interaction data, a temporal neural network (such as LSTM or Transformer) can be used to analyze the user's recent riding sequence to capture dynamic interests, and combined with a graph attention network, the correlation between users using different models can be captured; when constructing the association between vehicles and riding scenarios with different road conditions, the bicycle attribute database can be further subdivided into frame type (road bike / mountain bike / touring bike), tire width, transmission system and weight, etc., and annotated with applicable scenario labels (such as road bikes are suitable for flat road racing, touring bikes are suitable for long-distance load carrying).

[0069] S400. Recommending a cycling vehicle for each ride of the user through the cycling recommendation model. In an optional implementation of the first aspect of the present invention, recommending a cycling vehicle for each ride of the user through the cycling recommendation model includes: receiving a cycling start point and a cycling end point input by the user; planning a path for the cycling start point and the cycling end point using the shortest path; matching the planned path with the road section feature map; determining the road section overlap ratio between the planned path and the road section feature map; if the road section overlap ratio is less than a preset threshold, recommending a vehicle type based on the preference of the user portrait data (this implementation method is that the user's historical cycling data is too little and the path cannot match enough road section information, in which case the vehicle recommendation is mainly based on the user's preference); if the road section overlap ratio is greater than the preset threshold, further obtaining the uphill and downhill ratio of the overlapping road section, and recommending a vehicle type based on the uphill and downhill ratio. After identifying uphill and downhill sections, the system calculates the specific distance of each section. Based on the ratio of uphill to downhill sections in the user's current cycling route, the system determines whether the user will find the section strenuous. For example, if the user's current cycling section has a long uphill section, the system will recommend a bike with better climbing performance. These bikes are typically equipped with a lower gear ratio or electric power assist, which can reduce the difficulty of riding uphill. If the proportion of downhill or flat sections is large, the system will recommend a stable bike. Stable bikes are lightweight, do not require additional climbing ability, are suitable for downhill or flat riding, and offer better safety.

[0070] In the present invention, after obtaining the recommended vehicle type, the present invention will obtain information about surrounding rental points, and then obtain vehicle information of each rental point, and screen out rental points with the recommended vehicle type. After the user selects a rental point and plans a walking route, the map information including the user's current location and the rental point location will first be converted into a node adjacency matrix, and then a road network between the user's current location and the rental point location will be constructed based on the node adjacency matrix; Dijkstra will be used to search for feasible paths in the road network to obtain a navigation path.

[0071] See also Figure 2 A second aspect of the present invention provides a personalized recommendation system for riding vehicles, the personalized recommendation system for riding vehicles comprising:

[0072] A data collection module 10 is used to collect historical riding data of the user, wherein the historical riding data includes riding trajectory data, riding status data, pedal pressure data, and user portrait data;

[0073] A data analysis module 20 is used to analyze the historical riding data to obtain road slope characteristic data;

[0074] A model building module 30 is configured to build a cycling recommendation model based on the historical cycling data, the road slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition;

[0075] The vehicle recommendation module 40 is configured to recommend a vehicle for each ride of the user using the riding recommendation model.

[0076] In an optional implementation of the second aspect of the present invention, the data analysis module includes:

[0077] A curve construction unit, configured to construct, for each driving trajectory, a vehicle speed time variation curve and a pedal pressure time variation curve of the driving trajectory;

[0078] a timing alignment unit, configured to perform timing alignment on the vehicle speed time variation curve and the pedal pressure time variation curve;

[0079] A change rate calculation unit is used to obtain the velocity change rate and pressure change rate of each window in sequence along the time axis through a preset sliding window;

[0080] a logic judgment unit, configured to match the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic;

[0081] The road section slope acquisition unit is used to obtain the road section slope characteristic data of each driving trajectory based on the matching result.

[0082] In an optional implementation of the second aspect of the present invention, the logic judgment unit includes:

[0083] a positive / negative judgment subunit, configured to perform a positive / negative judgment on the velocity change rate and the pressure change rate for each of the windows;

[0084] a first sub-situation judgment sub-unit, configured to, if the speed change rate is negative and the pressure change rate is positive, further judge whether the speed change rate is less than a preset first negative threshold and whether the pressure change rate is greater than a preset first positive threshold;

[0085] a second sub-case judgment sub-unit, configured to, if the speed change rate is positive and the pressure change rate is negative, further judge whether the speed change rate is greater than a preset second positive threshold and whether the pressure change rate is less than a preset second negative threshold;

[0086] The window uphill and downhill attribute determination subunit is used to determine that the window is uphill if the speed change rate is less than the first negative threshold and the pressure change rate is greater than the first positive threshold; if the speed change rate is greater than the second positive threshold and the pressure change rate is less than the second negative threshold, then determine that the window is downhill; if the speed change rate and the pressure change rate are other situations, then determine that the window is a flat road.

[0087] In an optional implementation of the second aspect of the present invention, the road section slope acquisition unit includes:

[0088] The road segment uphill and downhill attribute judgment subunit is configured to, for each driving trajectory, filter adjacent windows along the time axis that have the same uphill and downhill attributes; if the adjacent windows are both uphill, then the path segment corresponding to the adjacent windows is determined to be uphill; if the adjacent windows are both downhill, then the path segment corresponding to the adjacent windows is determined to be downhill; and for windows that cannot have the same uphill and downhill attributes as the adjacent windows on the left and right, then the path segment corresponding to the window is determined to be flat.

[0089] The same-attribute road segment merging subunit is used to merge all the path segments with the same uphill and downhill attributes to obtain the road segment slope characteristic data of each driving trajectory.

[0090] In an optional implementation of the second aspect of the present invention, the change rate calculation unit includes:

[0091] a maximum and minimum value acquisition subunit, configured to acquire, for each window, the maximum speed value, the minimum speed value, the maximum pedaling pressure value, and the minimum pedaling pressure value within the window time;

[0092] A speed change rate calculation subunit is configured to calculate the difference between the maximum speed value and the minimum speed value by subtracting the first time from the last time, and then divide the difference by the window time to obtain the speed change rate;

[0093] The pressure change rate calculation subunit is used to calculate the difference between the maximum pedaling pressure value and the minimum pedaling pressure value by subtracting the first time from the last time, and then divide the difference by the window time to obtain the pressure change rate.

[0094] In an optional embodiment of the second aspect of the present invention, the model building module includes:

[0095] A historical riding trajectory drawing unit, configured to draw a historical riding trajectory on a preset map based on the historical riding data;

[0096] a road section feature map acquisition unit, configured to add the road section slope feature data to the historical riding trajectory to obtain a road section feature map;

[0097] A vehicle type classification library building unit is used to classify all riding vehicles and obtain a vehicle type classification library of the riding vehicles;

[0098] an association establishing unit, configured to establish associations between the various types of riding vehicles in the vehicle classification library and different user portraits and riding scenarios under different road conditions;

[0099] A model building unit is used to build a cycling recommendation model based on the road section feature map, the vehicle type classification library and the association relationship.

[0100] In an optional implementation of the second aspect of the present invention, the vehicle recommendation module includes:

[0101] An input receiving unit, configured to receive a user input of a riding start point and a riding end point;

[0102] A path planning unit, configured to plan a path between the cycling starting point and the cycling end point using the shortest path;

[0103] A road segment coincidence matching unit, configured to match the planned path with the road segment feature map;

[0104] a road section overlap ratio obtaining unit, configured to determine a road section overlap ratio between the planned path and the road section feature map;

[0105] a preference recommendation unit, configured to recommend a vehicle type based on the preference of the user profile data if the road segment overlap ratio is less than a preset threshold;

[0106] The road condition recommendation unit is used to further obtain the uphill and downhill ratio of the overlapping road section if the road section overlap ratio is greater than the preset threshold, and recommend the vehicle type based on the uphill and downhill ratio.

[0107] Figure 3 : is a structural diagram of a terminal device provided by an embodiment of the present invention. The terminal device may have relatively large differences due to different configurations or performances, and may include one or more processors 50 (central processing units, CPU) (for example, one or more processors) and memory 60, and one or more storage media 70 for storing application programs or data (for example, one or more mass storage devices). Among them, the memory and storage medium can be temporary storage or permanent storage. The program stored in the storage medium may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the terminal device. Furthermore, the processor can be configured to communicate with the storage medium and execute a series of instruction operations in the storage medium on the terminal device.

[0108] The terminal device of the present invention may further include one or more power supplies 80, one or more wired or wireless network interfaces 90, one or more input and output interfaces 100, and / or one or more operating systems, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 3 The terminal device structure shown does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0109] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the steps of the personalized recommendation system for cycling vehicles.

[0110] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system or system or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0111] If the integrated unit is implemented in the form of 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, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., various media that can store program code.

[0112] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A personalized recommendation method for a cycling vehicle, characterized in that: include: Collecting historical riding data of the user, including riding trajectory data, riding status data, pedal pressure data, and user profile data; Analyzing the historical riding data to obtain road section slope characteristic data; A cycling recommendation model is constructed based on the historical cycling data, the road section slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition; The riding recommendation model is used to recommend a riding vehicle for each riding of the user.

2. The personalized recommendation method for riding vehicles according to claim 1, characterized in that: Analyzing the historical riding data to obtain the road section slope characteristic data includes: For each driving trajectory, construct a vehicle speed time variation curve and a pedal pressure time variation curve of the driving trajectory; Performing time sequence alignment on the vehicle speed time variation curve and the pedal pressure time variation curve; The velocity change rate and pressure change rate of each window are obtained in sequence along the time axis through the preset sliding window; Matching the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic; The road section slope characteristic data of each driving trajectory is obtained based on the matching result.

3. The personalized recommendation method for riding vehicles according to claim 2, characterized in that: The matching of the speed change rate and the pressure change rate of each window with a preset uphill and downhill judgment logic includes: For each of the windows, performing a positive or negative judgment on the velocity change rate and the pressure change rate; If the speed change rate is negative and the pressure change rate is positive, further determining whether the speed change rate is less than a preset first negative threshold and whether the pressure change rate is greater than a preset first positive threshold; If the speed change rate is positive and the pressure change rate is negative, it is further determined whether the speed change rate is greater than a preset second positive threshold and whether the pressure change rate is less than a preset second negative threshold; If the speed change rate is less than the first negative threshold and the pressure change rate is greater than the first positive threshold, the window is determined to be an uphill slope; if the speed change rate is greater than the second positive threshold and the pressure change rate is less than the second negative threshold, the window is determined to be a downhill slope; if the speed change rate and the pressure change rate are other situations, the window is determined to be a flat road.

4. The personalized recommendation method for riding vehicles according to claim 2, characterized in that: The road section slope characteristic data of each driving trajectory obtained based on the matching result includes: For each driving trajectory, adjacent windows with the same uphill and downhill attributes are selected along the time axis; if the adjacent windows are all uphill, the path segment corresponding to the adjacent windows is determined to be uphill; if the adjacent windows are all downhill, the path segment corresponding to the adjacent windows is determined to be downhill; For the window that cannot have the same uphill and downhill attributes as the left and right adjacent windows, it is determined that the path segment corresponding to the window is a flat road; All the path segments are combined with similar uphill and downhill attributes to obtain the road segment slope characteristic data of each driving trajectory.

5. The personalized recommendation method for riding vehicles according to claim 2, characterized in that: The method of sequentially obtaining the velocity change rate and pressure change rate of each window along the time axis through the preset sliding window includes: For each of the windows, obtaining the maximum speed value, the minimum speed value, the maximum pedal pressure value, and the minimum pedal pressure value within the window time; The speed change rate is obtained by subtracting the first time from the last time from the maximum speed value and the minimum speed value, and then dividing the difference by the window time; The pressure change rate is obtained by subtracting the earlier time from the later time to calculate the difference between the maximum pedaling pressure value and the minimum pedaling pressure value, and then dividing the difference by the window time.

6. The personalized recommendation method for riding vehicles according to claim 1, characterized in that: The cycling recommendation model constructed based on the historical cycling data, the road section slope characteristic data, and the user preference / road condition multi-dimensional vehicle type recommendation algorithm includes: Draw a historical riding trajectory on a preset map based on the historical riding data; Adding the road section slope characteristic data to the historical riding trajectory to obtain a road section characteristic map; Classifying all riding vehicles to obtain a vehicle classification library of the riding vehicles; Establishing associations between various types of riding vehicles in the vehicle classification library and different user portraits and riding scenarios under different road conditions; A cycling recommendation model is constructed based on the road section feature map, the vehicle type classification library and the association relationship.

7. The personalized recommendation method for riding a vehicle according to claim 6, characterized in that: The recommending a cycling vehicle for each riding of the user by using the cycling recommendation model includes: Receive the user's input of the riding start point and riding end point; Planning a path between the cycling starting point and the cycling end point using the shortest path; Matching the planned path with the road section feature map; Determining a road segment overlap ratio between the planned path and the road segment feature map; If the road segment overlap ratio is less than a preset threshold, recommending a vehicle type based on the preference of the user profile data; If the road section overlap ratio is greater than the preset threshold, the uphill and downhill ratios of the overlapping road sections are further obtained, and vehicle type recommendations are made based on the uphill and downhill ratios.

8. A personalized recommendation system for cycling vehicles, characterized in that: The personalized recommendation system for cycling vehicles includes: A data collection module is used to collect the user's historical riding data, wherein the historical riding data includes driving trajectory data, driving status data, pedal pressure data, and user portrait data; A data analysis module is used to analyze the historical riding data to obtain road slope characteristic data; A model building module is used to build a cycling recommendation model based on the historical riding data, the road slope characteristic data, and a multi-dimensional vehicle model recommendation algorithm of user preference / road condition; The vehicle recommendation module is used to recommend a riding vehicle for each ride of the user through the riding recommendation model.

9. A terminal device, characterized in that: The terminal device includes: a memory and at least one processor, wherein the memory stores instructions, and the memory and the at least one processor are interconnected via a line; The at least one processor calls the instructions in the memory to enable the terminal device to execute the personalized recommendation method for a cycling vehicle according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the personalized recommendation method for a riding vehicle according to any one of claims 1 to 7 is implemented.

Images