Route intelligent planning system and method based on intelligent cycling helmet

Abstract

The application discloses a route intelligent planning system and method based on an intelligent riding helmet, which is characterized in that: through the data time queue continuously collected by an IMU sensor, the original IMU data is converted through a low-dimensional structured coding of a full connection layer, and is input into a specially designed motion state nonlinear time sequence type integral coding network for processing, from which key features reflecting the real motion trajectory are extracted, and then a position offset compensation value is calculated through a decoder, so that the accurate correction of the current position of the user is finally realized. In this way, reliable position information can be provided in the case that the GPS signal is weak or missing, the accuracy and reliability of path planning are ensured, and the safety and experience of riding are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent planning, and more specifically, to a route intelligent planning system and method based on an intelligent riding helmet. BACKGROUND

[0002] In the rapid development of current intelligent transportation systems, riding safety and efficiency have gradually become the focus of attention. Although there are various riding assistance devices on the market, such as smart watches, smart phone applications, etc., they mainly provide navigation and route planning functions to help riders choose the best path. However, these devices often rely on external devices or mobile phones as the main information processing unit, which not only increases the user's burden of carrying additional equipment, but also may affect the use experience and safety due to improper placement of the device.

[0003] In addition, existing technical solutions still face challenges in terms of location accuracy. Although conventional riding navigation systems use GPS positioning technology, they do not fully consider the problem of positioning drift caused by multipath interference in urban canyons. Especially when riders are in complex scenes such as elevated bridge areas, dense building groups or tunnels, simply relying on GPS modules will form cumulative positional deviations, seriously affecting the reliability of subsequent path planning. Although some systems attempt to introduce inertial measurement units for auxiliary positioning, the conventional linear compensation model is difficult to effectively analyze the nonlinear attitude changes in riding motion, such as centripetal acceleration interference during sharp turns or random vibration noise on bumpy road sections, resulting in significant deviations between motion state estimation and true trajectory.

[0004] In addition, traditional route planning systems are mostly based on static data for calculation, i.e., only considering the shortest distance or estimated time between the starting point and the end point. At the same time, the traffic flow data updated at a fixed period in existing technologies cannot capture the instantaneous changes in road conditions caused by sudden accidents, resulting in the evaluation of the impact of real-time changing traffic conditions and weather factors on road passability remaining at the level of static threshold judgment.

[0005] In view of the above problems, an optimized route intelligent planning method based on an intelligent riding helmet is expected. SUMMARY

[0006] In order to solve the above technical problems, the present application is proposed. Embodiments of the present application provide a route intelligent planning system and method based on an intelligent riding helmet, which converts IMU raw data through a low-dimensional structured coding of a full connection layer by using a data time queue continuously collected by an IMU sensor, and inputs the converted data into a specially designed motion state nonlinear time series integral coding network for processing, extracts key features reflecting a real motion trajectory therefrom, calculates a position offset compensation value through a decoder, and finally realizes accurate correction of a current position of a user. In this way, reliable position information can be provided in the case of weak or missing GPS signals, ensuring the accuracy and reliability of path planning, and further improving the safety and experience of riding.

[0007] According to one aspect of the present application, a route intelligent planning method based on an intelligent riding helmet is provided, which includes:

[0008] collecting a time queue of real-time positions of a user and a time queue of IMU data by a GPS module and an IMU sensor deployed in the intelligent riding helmet;

[0009] receiving a time queue of traffic data and a time queue of weather data by an external data receiving module deployed in the intelligent riding helmet;

[0010] extracting a current position of the user from the time queue of real-time positions of the user;

[0011] correcting the current position of the user for position deviation based on the time queue of IMU data to obtain an optimized current position of the user;

[0012] inputting the optimized current position of the user, the time queue of traffic data, and the time queue of weather data into a road environment comprehensive state evaluation module to obtain a road environment comprehensive state evaluation result;

[0013] based on the road environment comprehensive state evaluation result and a set riding preference, and using a dynamic route optimization algorithm to adjust a riding route plan in real time.

[0014] According to another aspect of the present application, a route intelligent planning system based on an intelligent riding helmet is provided, which includes:

[0015] a data collection module for collecting a time queue of real-time positions of a user and a time queue of IMU data by a GPS module and an IMU sensor deployed in the intelligent riding helmet;

[0016] an external data receiving module for receiving a time queue of traffic data and a time queue of weather data by an external data receiving module deployed in the intelligent riding helmet;

[0017] a user current position extraction module configured to extract a user current position from a time queue of real-time positions of the user;

[0018] a position deviation correction module configured to correct the user current position based on a time queue of the IMU data to obtain an optimized user current position;

[0019] a road environment comprehensive state evaluation module configured to input the optimized user current position, the time queue of the traffic data, and the time queue of the weather data into the road environment comprehensive state evaluation module to obtain a road environment comprehensive state evaluation result;

[0020] a route real-time adjustment module configured to real-time adjust a cycling route plan based on the road environment comprehensive state evaluation result and a set cycling preference, and using a dynamic route optimization algorithm.

[0021] Compared with the prior art, the route intelligent planning system and method based on the intelligent cycling helmet provided by the present application can continuously collect data time queues by using an IMU sensor, convert IMU original data through a low-dimensional structured coding of a full connection layer, input the converted data into a specially designed motion state nonlinear time series integral coding network for processing, extract key features reflecting real motion trajectories therefrom, calculate a position offset compensation value through a decoder, and finally realize accurate correction of a user current position. In this way, reliable position information can be provided even in the case of weak or missing GPS signals, ensuring the accuracy and reliability of path planning and further improving the safety and experience of cycling. BRIEF DESCRIPTION OF DRAWINGS

[0022] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of embodiments of the present application, when taken in conjunction with the accompanying drawings. The drawings provided in the specification and the embodiments of the present application together serve to explain constituents of the present application and to provide a further understanding hereto, but do not limit the present application. In the drawings, like reference numerals refer to like elements or steps throughout.

[0023] Figure 1 a flowchart of the route intelligent planning method based on the intelligent cycling helmet according to an embodiment of the present application;

[0024] Figure 2 a data flow schematic diagram of the route intelligent planning method based on the intelligent cycling helmet according to an embodiment of the present application;

[0025] Figure 3 a flowchart of sub-step S4 of the route intelligent planning method based on the intelligent cycling helmet according to an embodiment of the present application;

[0026] Figure 4Flow chart for sub-step S42 of the method for route intelligent planning based on the intelligent cycling helmet according to the embodiments of the present application;

[0027] Figure 5 Block diagram of the system for route intelligent planning based on the intelligent cycling helmet according to the embodiments of the present application. DETAILED DESCRIPTION

[0028] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of embodiments of the present application, and thus are not to be used to limit the whole embodiments of the present application, and it should be understood that the present application is not limited to the described example embodiments.

[0029] As shown in the present application and claims, unless the context clearly indicates otherwise, the words "one", "an", "a", and / or "the" do not mean to specify a single number, but also include a plurality. Generally, the terms "comprising" and "including" only indicate including the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0030] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0031] Flow charts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in sequence. Instead, various steps can be processed in reverse order or simultaneously, as needed. Meanwhile, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0032] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of embodiments of the present application, and thus are not to be used to limit the whole embodiments of the present application, and it should be understood that the present application is not limited to the described example embodiments.

[0033] To break through the data island effect under the traditional split equipment architecture, rely on the high-precision sensor array integrated in the helmet body, establish a positioning compensation mechanism and environment state evaluation model with stronger time and space continuity in the riding scene. The core idea of the technical solution is to build a multi-modal data fusion decision system integrated in the riding helmet body, realize the technical leap of the riding navigation system in the three dimensions of positioning accuracy, environment perception and dynamic response through the cooperative action of embedded sensor network and intelligent algorithm. Specifically, considering the deficiencies of the prior art in position accuracy, real-time data processing and user experience, the technical solution adopts the combination of GPS module and IMU sensor to collect user real-time position and motion state information, thereby effectively overcoming the urban canyon effect problem caused by single dependence on GPS positioning, and through the introduction of low-dimensional structured coding based on full connection layer and nonlinear time series integral coding network, the accurate capture and compensation of the complex motion posture of the rider are realized. In addition, the time queue of traffic data and weather data is obtained by using the external data receiving module, and is combined with the road environment comprehensive state evaluation module, so that the system can dynamically adjust according to the real-time changing road conditions and riding preferences.

[0034] Specifically, in the technical solution of the present application, a route intelligent planning method based on an intelligent riding helmet is proposed. Figure 1 The flow chart of the route intelligent planning method based on the intelligent riding helmet according to the embodiment of the present application. Figure 2 The data flow schematic diagram of the route intelligent planning method based on the intelligent riding helmet according to the embodiment of the present application. As Figure 1 And Figure 2 As shown in the route intelligent planning method based on the intelligent riding helmet according to the embodiment of the present application, comprising the steps: S1, collecting the time queue of user real-time position and the time queue of IMU data through the GPS module and IMU sensor deployed in the intelligent riding helmet; S2, receiving the time queue of traffic data and the time queue of weather data by using the external data receiving module deployed in the intelligent riding helmet; S3, extracting the current position of the user from the time queue of user real-time position; S4, correcting the position deviation of the current position of the user based on the time queue of IMU data to obtain the optimized current position of the user; S5, inputting the optimized current position of the user, the time queue of traffic data and the time queue of weather data into the road environment comprehensive state evaluation module to obtain the road environment comprehensive state evaluation result; S6, based on the road environment comprehensive state evaluation result and the set riding preference, and using a dynamic route optimization algorithm to adjust the riding route planning in real time.

[0035] In particular, the S1 collects a time queue of user real-time position and a time queue of IMU data by means of a GPS module and an IMU sensor deployed in the smart cycling helmet. It should be understood that in urban environments, GPS signals are prone to interference, especially in areas with high-rise buildings, under tunnels or bridges, etc. This phenomenon is particularly evident. Due to multipath effects (i.e. GPS signals reflected from multiple directions to the receiver), positioning drift occurs, making the positioning result relying solely on GPS not accurate enough. In these cases, GPS can not be able to provide accurate enough position information to support effective route planning. While the IMU sensor can work independently of external radio signals, it can track the motion state of the device by measuring acceleration and angular velocity. Therefore, combining IMU data can complement the shortcomings of GPS in complex environments and ensure relatively accurate position estimation even in poor GPS signal conditions. Therefore, in the technical solution of the present application, based on the time queue of the IMU data, the position deviation of the current position of the user is corrected to obtain the optimized current position of the user. It is worth mentioning that the IMU (Inertial Measurement Unit) sensor is a device for measuring the three-axis attitude angle (or angular velocity) and acceleration of an object. In the technical solution of the present application, the IMU sensor can provide high-frequency acceleration and angular velocity data, which are collected and formed into a time queue, so that the system can capture very subtle changes in the rider's head and even the entire cycling process. This is very important for understanding the real-time motion state of the rider, especially in poor or unavailable GPS signal conditions (such as passing through tunnels, canyon effect between high-rise buildings, etc.), the continuous data provided by the IMU can be supplemented to help maintain the accuracy of position estimation.

[0036] In particular, S2 receives a time queue of traffic data and a time queue of weather data using an external data receiving module deployed in the smart cycling helmet. It should be understood that by obtaining a time queue of real-time traffic data, the system can dynamically adjust the cycling route to avoid congested sections or accident sites. Traditional route planning is often based on static map information and cannot timely reflect sudden situations occurring on the road, such as traffic accidents, road construction, etc. The introduction of real-time traffic data enables cyclists to choose the optimal path based on the latest road condition information, avoiding delays and potential safety risks. In addition, weather conditions have a direct impact on cycling safety. For example, rain can cause the road surface to be slippery, increasing the risk of falling; strong winds can affect the balance of cyclists, especially on viaducts or open areas. By receiving a time queue of weather data, the system can provide early warning of impending adverse weather and recommend that cyclists take appropriate preventive measures or adjust their schedules to ensure safety during cycling. Combining these external data with information collected by IMU sensors and GPS modules can provide a more comprehensive assessment of the overall state of the cycling environment. It is worth mentioning that the external data receiving module is connected to the Internet through wireless communication technology (such as Bluetooth, Wi-Fi, or cellular network), thereby enabling access to the latest traffic conditions and weather forecast information. Specifically, this module regularly downloads updated data streams from cloud servers, which include traffic flow, accident reports, road construction information, and detailed weather forecasts in the current area, such as rainfall, temperature changes, wind speed, and other factors that may affect cycling conditions.

[0037] In particular, S3 extracts the current location of the user from the time queue of the user's real-time location. The time queue of the user's real-time location contains the latitude and longitude coordinates of the user at different time points, forming the basis for the user's movement trajectory. Based on accurate current location information, the system can more effectively perform path planning algorithms.

[0038] In particular, S4 performs position deviation correction on the current location of the user based on the time queue of IMU data to obtain an optimized current location of the user. Specifically, in one specific example of the present application, as shown in Figure 3 S4 includes: S41, structurally encoding each IMU data in the time queue of IMU data to obtain a time queue of IMU low-dimensional structured encoding vectors; S42, inputting the time queue of IMU low-dimensional structured encoding vectors into a motion state nonlinear time series quasi-integral encoding network to obtain a motion state time series quasi-integral cumulative encoding vector; S43, determining a position offset compensation value based on the motion state time series quasi-integral cumulative encoding vector; S44, performing position deviation correction on the current location of the user based on the position offset compensation value to obtain the optimized current location of the user.

[0039] Specifically, the S41, each IMU data in the time queue of the IMU data is structured and coded to obtain a time queue of IMU low-dimensional structured coding vectors. It should be understood that in a complex urban riding scene, due to the intermittent shielding of viaducts, dense building reflection interference, and multi-source noise coupling caused by sudden braking or snake-shaped obstacle avoidance, the time queue of the original IMU data often contains high-dimensional redundant features and nonlinear motion noise. When the traditional linear compensation model directly differentiates the original six-axis inertial data, it is difficult to separate the coupling effects of motion state essential features and sensor random drift. In order to overcome the above problems, the original high-dimensional IMU data is converted into a low-dimensional structured coding vector which is more representative and easy to process. In the technical solution of the present application, each IMU data in the time queue of the IMU data is low-dimensional structured coding based on a full connection layer to obtain a time queue of IMU low-dimensional structured coding vectors. By applying the low-dimensional structured coding technology of the full connection layer, the key features that best reflect the true motion state of the rider can be extracted from a large number of IMU sensor readings. This not only simplifies the subsequent data processing process, but also improves the effectiveness of data feature extraction, so that the system can more accurately understand the motion pattern of the rider, and even in the case of poor GPS signal, it can also maintain high positioning accuracy. In particular, the coding layer adopts a weight sharing mechanism across the sensor axis system, orthogonally projects the three-axis time series data of the accelerometer, gyroscope and magnetometer in the time-space dimension, and selects the intrinsic motion vector strongly related to the riding motion pattern through a nonlinear activation function. Specifically, the IMU six-dimensional original data (three-axis acceleration + three-axis angular velocity) of each timestamp is mapped to a three-dimensional latent space, which retains the core motion features of the riding posture while filtering the dimensional noise caused by the inherent zero bias of the sensor and environmental magnetic interference. In this way, not only the accuracy of position positioning is improved, but also more reliable data support is provided for road environment comprehensive state evaluation and dynamic route optimization.

[0040] Specifically, the S42 inputs the time queue of the IMU low-dimensional structured encoding vector into a motion state nonlinear time series type integral encoding network to obtain a motion state time series type integral accumulation encoding vector. It should be understood that the multi-modal inertial characteristics of the cycling motion have the characteristics of coexistence of long-range dependence and local mutation, and the traditional linear compensation model is difficult to accurately capture the complex nonlinear posture changes in the cycling process, such as centripetal acceleration interference in sudden turns, rapid acceleration or deceleration, and random vibration noise in bumpy road sections, so that the traditional position compensation model based on linear integral or simple recursive network is difficult to accurately analyze the real motion trajectory. In order to capture the global motion trend and instantaneous posture change in the cycling process at the same time, the key features that best reflect the real motion state of the cyclist are extracted from the time queue of the IMU low-dimensional structured encoding vector, and in the technical solution of the present application, the time queue of the IMU low-dimensional structured encoding vector is input into a motion state nonlinear time series type integral encoding network to obtain a motion state time series type integral accumulation encoding vector. That is, through the multi-stage encoding mode from coarse granularity to fine granularity and from global to local, not only the overall motion mode can be captured, but also the saliency of node individualized features can be enhanced in detail, so as to realize multi-level and multi-scale complex feature integration. In this process, first, the main mode of the cycling motion (such as the overall acceleration trend, the angular velocity change period) is extracted through the information kernel coarse-grained aggregation network, and the global motion representation in the cycling process is constructed; then the local abnormal motion segment (such as sudden acceleration drop when emergency braking, sudden turning when avoiding obstacles) is dynamically identified through the kernel aggregation compensation factor, and the compensation weight is explicitly regulated by using the gating function, so that the encoding vector can not only represent the smooth displacement in continuous cycling, but also highlight the instantaneous influence of sudden action. Among them, the coarse-grained aggregation encoding describes the macro displacement trend of the cyclist along the planned path, and the fine-grained compensation aggregation accurately captures the micro trajectory deviation caused by environmental interference (such as crosswind disturbance) or active obstacle avoidance. In this way, the real position deviation of the cyclist can be captured, which helps to realize more accurate position deviation correction, thereby improving the reliability and user experience of the entire route planning system. Specifically, in one specific example of the present application, as shown in FIG. 6, the time queue of the IMU low-dimensional structured encoding vector is input into the motion state nonlinear time series type integral encoding network to obtain the motion state time series type integral accumulation encoding vector. The motion state nonlinear time series type integral encoding network includes a coarse-grained aggregation encoding network and a fine-grained compensation aggregation encoding network. The coarse-grained aggregation encoding network is used to extract the main mode of the cycling motion (such as the overall acceleration trend, the angular velocity change period), and the global motion representation in the cycling process is constructed. The fine-grained compensation aggregation encoding network is used to dynamically identify the local abnormal motion segment (such as sudden acceleration drop when emergency braking, sudden turning when avoiding obstacles), and the compensation weight is explicitly regulated by using the gating function, so that the encoding vector can not only represent the smooth displacement in continuous cycling, but also highlight the instantaneous influence of sudden action. Figure 4As shown, the S42 comprises: S421, inputting the time queue of the IMU low-dimensional structured encoding vector into an information kernel coarse-grained aggregation network to obtain an IMU coarse-grained aggregation encoding vector; S422, determining an IMU kernel aggregation compensation weight of each IMU low-dimensional structured encoding vector in the time queue of the IMU low-dimensional structured encoding vector based on the IMU coarse-grained aggregation encoding vector to obtain a set of IMU kernel aggregation compensation weight factors; S423, inputting the set of IMU kernel aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector into a node fine-grained dynamic compensation aggregation network to obtain an IMU fine-grained compensation aggregation encoding vector; and S424, inputting the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector into a residual unit to obtain the motion state time series class integral accumulation encoding vector.

[0041] More specifically, the S421 inputs the time queue of the IMU low-dimensional structured encoding vector into an information kernel coarse-grained aggregation network to obtain an IMU coarse-grained aggregation encoding vector. Since complex environments such as urban canyons and viaducts can cause GPS signal multipath interference to form cumulative position deviation, and the cycling motion contains nonlinear attitude changes such as sharp turns and bumpy roads, it is difficult for traditional linear compensation models to accurately capture these dynamic characteristics. At this time, although the accelerometer and gyroscope data collected by the IMU sensor can provide high-frequency motion state information, the original signals have high-frequency noise interference and nonlinear coupling problems, and direct use will cause the motion state estimation error to be amplified. In the technical solution of the present application, the time queue of the IMU low-dimensional structured encoding vector is input into an information kernel coarse-grained aggregation network to obtain an IMU coarse-grained aggregation encoding vector. Here, the global feature compression of the IMU low-dimensional encoding is performed through the information kernel coarse-grained aggregation network, which essentially reconstructs the similarity structure between nodes in the implicit high-dimensional space through the nonlinear mapping mechanism to summarize the global features of the time queue of the IMU low-dimensional structured encoding vector, so as to extract the main mode that best represents the true motion state of the cyclist. In this way, the overall motion trend of the cyclist in the entire time sequence can be effectively captured, which is crucial for understanding the macro movement mode of the cyclist. Specifically, in the embodiments of the present application, the time queue of the IMU low-dimensional structured encoding vector is input into the information kernel coarse-grained aggregation network according to the following information kernel coarse-grained aggregation formula to obtain the IMU coarse-grained aggregation encoding vector; wherein the information kernel coarse-grained aggregation formula is:

[0042]

[0043] wherein, the time queue of the IMU low-dimensional structured encoding vector, These are the 1st, 2nd, and 3rd times in the time queue of the IMU low-dimensional structured encoded vector, respectively. The and the first Each IMU low-dimensional structured encoded vector and This represents taking the maximum and minimum values ​​in the vector. Kernel factors representing global feature information for function, The kernel weights are global feature information. The number of vectors in the time queue of the IMU low-dimensional structured encoded vectors. This represents the coarse-grained convergent coding vector of the IMU.

[0044] More specifically, in step S422, based on the IMU coarse-grained convergent coding vector, the IMU kernel convergence compensation weights of each IMU low-dimensional structured coding vector in the time queue of the IMU low-dimensional structured coding vector are determined to obtain a set of IMU kernel convergence compensation weight factors. In the embodiments of this application, firstly, the kernel convergence compensation factor of each IMU low-dimensional structured coding vector in the time queue of the IMU low-dimensional structured coding vector relative to the IMU coarse-grained convergent coding vector is calculated to obtain a set of IMU kernel convergence compensation factors; it should be understood that although the information kernel coarse-grained convergent network can effectively capture the overall movement trend of the cyclist throughout the entire time series, this compressed aggregation process of global features may dilute or even completely lose the personalized information of some important nodes. For example, during nonlinear attitude changes such as sharp turns or rapid acceleration and deceleration, local detailed features are crucial for accurate position deviation correction. If these details are ignored or lost, it will lead to a significant deviation between the position estimation and the actual trajectory. Therefore, the kernel convergence compensation factor of each IMU low-dimensional structured encoded vector in the time queue of the IMU low-dimensional structured encoded vector is calculated relative to the IMU coarse-grained converged encoded vector. This extracts key features reflecting local details from the time queue of the IMU low-dimensional structured encoded vector, generating a set of IMU kernel convergence compensation factors. In other words, by calculating the kernel convergence compensation factor, the correction rules for each node (i.e., the IMU data at each time point) relative to the global features are dynamically adjusted, ensuring that the personalized information of important nodes is preserved and providing strong support for subsequent fine-grained compensation modeling.

[0045] In particular, for the deviation compensation between the node feature vector and the coarse-grained aggregation encoding vector, in one preferred example, the deviation correction requirement between the local feature and the global representation is dynamically quantified by a game-based counterfactual framework. Due to the cumulative drift of GPS signals easily disturbed by multipath interference in complex environments such as urban canyons and viaducts, and the nonlinear attitude changes such as sharp turns and bumpy roads in cycling movements, the traditional global feature aggregation may over-compress the local motion details (such as the inertial disturbance caused by arm swing or the effective motion information in short vibration noise). At this time, the calculation of the kernel aggregation compensation factor maps the vector norm difference between each node feature vector and the coarse-grained encoding vector to the decision point loss of the strategy action, and uses the normalized counterfactual regret value to measure the potential benefits of the local features not retained in global modeling. This compensation mechanism not only captures the conflict between centripetal acceleration disturbance and global motion trend during sharp turns, but also dynamically adjusts the compensation weight under the explicit regulation of the gating function, so that the model can not only maintain the consistency of the overall motion direction (such as the angular velocity trend of continuous turning) in viaduct areas and other scenes, but also retain the local significance of instantaneous obstacle avoidance actions (such as the IMU mutation signal during emergency lane changing). By regarding the compensation factor as the potential benefit bias of "not taking action", the model can adaptively balance the feature conflict between global and local under the information kernel compression assumption. Specifically, in the embodiments of the present application, the specific steps of calculating the kernel aggregation compensation factor of each IMU low-dimensional structured encoding vector in the time queue of the IMU low-dimensional structured encoding vector relative to the IMU coarse-grained aggregation encoding vector are as follows: modulating the IMU low-dimensional structured encoding vector and the IMU coarse-grained aggregation encoding vector to obtain an IMU low-dimensional structured encoding modulation vector and an IMU coarse-grained aggregation encoding modulation vector; calculating the compensation information between the IMU low-dimensional structured encoding modulation vector and the IMU coarse-grained aggregation encoding modulation vector to obtain an IMU kernel aggregation compensation weight vector; based on the IMU low-dimensional structured encoding vector and the IMU coarse-grained aggregation encoding vector, calculating an IMU kernel aggregation deviation compensation factor; based on the IMU kernel aggregation compensation weight vector and the IMU kernel aggregation deviation compensation factor, calculating a kernel aggregation compensation factor to obtain the IMU kernel aggregation compensation factor.The specific process for calculating the IMU kernel convergence bias compensation factor based on the IMU low-dimensional structured coding vector and the IMU coarse-grained convergence coding vector is as follows: In response to the fact that the L2 norm of the IMU low-dimensional structured coding vector is less than the L2 norm of the IMU coarse-grained convergence coding vector, the L2 norm of the IMU low-dimensional structured coding vector is calculated, divided by the L2 norm of the IMU coarse-grained convergence coding vector, and then added to a constant, and the base-2 logarithmic function value is calculated; In response to the fact that the L2 norm of the IMU low-dimensional structured coding vector is greater than or equal to the L2 norm of the IMU coarse-grained convergence coding vector, the L2 norm of the IMU low-dimensional structured coding vector is calculated, divided by the L2 norm of the IMU coarse-grained convergence coding vector.

[0046] More specifically, in this embodiment, the kernel convergence compensation factor of each IMU low-dimensional structured coding vector in the time queue of the IMU low-dimensional structured coding vector relative to the IMU coarse-grained converged coding vector is calculated using the following kernel convergence compensation calculation formula to obtain a set of IMU kernel convergence compensation factors; wherein, the kernel convergence compensation calculation formula is:

[0047]

[0048] in, For point convolutional layer processing, This is the first weight matrix. This is the second weight matrix. for function, The time queue of the IMU low-dimensional structured encoding vector Each IMU low-dimensional structured coded modulation vector For IMU coarse-grained convergent-coded modulation vectors, For positional subtraction, To take the absolute value, For the IMU core convergence compensation weight vector, The L2 norm of a vector. It is the value of a logarithmic function with base 2. This represents the IMU core convergence bias compensation factor. and These represent the compensation weight matrix and the compensation bias vector, respectively. To compensate for the modulation vector, for The corresponding IMU core convergence compensation factor.

[0049] Further, the set of IMU kernel aggregation compensation factors is explicitly modeled based on a gating function to obtain a set of IMU kernel aggregation compensation weight factors. Due to the easy multipath interference of GPS signals in complex environments such as urban canyons and viaducts, and the nonlinear attitude changes such as sharp turns and bumpy roads in cycling movements, the original IMU data will contain high-frequency noise and nonlinear coupling errors. Traditional global feature compression may over-restrict local motion details. At this time, the kernel aggregation compensation factor is explicitly modeled by a gating function, which can parameterize the action intensity of the compensation factor and dynamically select the local information that is crucial to the global feature correction under nonlinear constraints, so that the system can filter out only locally significant features while appropriately suppressing irrelevant or redundant information. In this way, the influence of each node on the global feature can be more accurately described, ensuring accurate position information even in complex dynamic environments. That is, the effective intensity of the compensation factor is adjusted flexibly according to the current motion state, and the kernel aggregation compensation weight factor obtained by explicit modeling of the gating function can dynamically adjust the correction intensity of local features on global encoding, so that the model can retain key local information of centripetal acceleration during sharp turns and suppress the interference of redundant vibration noise on the overall motion direction in viaduct areas and other scenarios. In this way, the model balances the continuity of the global motion direction and the instantaneous response of the local obstacle avoidance action in viaduct areas and other scenarios. Specifically, in the embodiments of the present application, the set of IMU kernel aggregation compensation factors is explicitly modeled based on a gating function to obtain a set of IMU kernel aggregation compensation weight factors according to the following compensation formula:

[0050]

[0051] wherein, is a preset threshold, is an explicit modeling operation of compensation, is the corresponding IMU kernel aggregation compensation weight factor.

[0052] More specifically, the S423 inputs the set of IMU core aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector into a node fine-grained dynamic compensation aggregation network to obtain an IMU fine-grained compensation aggregation encoding vector. That is, by performing node fine-grained dynamic compensation aggregation analysis on the set of IMU core aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector, the key features that can best reflect the actual motion state of the rider are extracted from the set of IMU core aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector. In this way, the model can dynamically optimize the cooperative expression of global and local features according to the motion state (such as the riding speed and the steering amplitude) and the environmental complexity (such as the road density and the obstacle distribution) in the real-time scene, so as to improve the accuracy of position deviation correction. Specifically, in the embodiments of the present application, the set of IMU core aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector are input into a node fine-grained dynamic compensation aggregation network to obtain an IMU fine-grained compensation aggregation encoding vector according to the following fine-grained dynamic compensation aggregation formula:

[0053]

[0054] wherein, represents the IMU fine-grained compensation aggregation encoding vector.

[0055] More specifically, the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector are input into a residual unit to obtain the motion state time-series type integral accumulation encoding vector. It should be understood that although the information kernel coarse-grained aggregation network can effectively capture the overall motion trend of the rider and correct the global features by calculating the kernel aggregation compensation factor, and the fine-grained dynamic compensation aggregation network further refines the local detailed features, but if the two parts of information are directly combined, some key details may be lost or unnecessary noise may be introduced. Therefore, in the technical solution of the present application, the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector are input into a residual unit to extract the key features that best reflect the actual motion state of the rider from the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector, and generate an optimized motion state time-series type integral accumulation encoding vector. In addition, by inputting the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector into the residual unit, the system can ensure feature transmission while avoiding the gradient vanishing problem, so that the model is easier to optimize, and the significance and robustness of the encoding result are further improved. Through such fine adjustment, the intelligent riding helmet can provide reliable location service and optimal riding route recommendation for the rider in various complex environments. Specifically, in the embodiments of the present application, the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector are input into a residual unit to obtain the motion state time-series type integral accumulation encoding vector according to the following residual calculation formula; wherein the residual calculation formula is:

[0056]

[0057] wherein, and is a residual weight coefficient, represents the motion state time-series type integral accumulation encoding vector.

[0058] Specifically, the S43 determines a position offset compensation value based on the motion state time series integral accumulation encoding vector. That is, in the technical solution of the present application, the motion state time series integral accumulation encoding vector is input into a decoder-based position compensation estimation module to obtain a position offset compensation value. In this process, the decoder extracts the key features that best reflect the actual motion state of the rider from the motion state time series integral accumulation encoding vector, generating an accurate position offset compensation value. The position offset compensation value is calculated based on the actual motion trajectory of the rider, which can effectively compensate for the GPS positioning error and the noise in the IMU data, ensuring the accuracy and reliability of the position information. Further, the position offset compensation value is used to correct the position deviation of the user's current position to obtain the optimized user's current position. By correcting the position deviation of the user's current position based on the position offset compensation value, the system can significantly improve the accuracy of position estimation, especially in complex urban environments. This means that in various complex urban riding scenarios, such as viaduct areas or tunnels, the rider can obtain continuous and stable position information, greatly improving the safety and comfort of riding.

[0059] Specifically, the S44 corrects the position deviation of the user's current position based on the position offset compensation value to obtain the optimized user's current position. That is, the calculated position offset compensation value is combined with the preliminary position estimate provided by the current GPS. In this way, the errors caused by GPS positioning can be effectively corrected, especially the positioning drift problem caused by the multipath effect commonly seen in complex environments such as urban canyons, high-rise buildings, or tunnels. Finally, after such position deviation correction, the system can obtain a more accurate user's current position, i.e., the optimized user's current position.

[0060] In particular, the S5 inputs the optimized user current location, the time queue of traffic data, and the time queue of weather data into a road environment comprehensive state assessment module to obtain a road environment comprehensive state assessment result. The road environment comprehensive state assessment module is not just a simple data processing unit, but an intelligent analysis platform that integrates multiple algorithms and technologies. It provides a comprehensive, dynamic, and personalized road environment assessment result by fusing multi-source data, recommends the best path for cyclists, and ensures their safety and comfort. Advanced data analysis techniques such as machine learning algorithms, time series analysis, and pattern recognition are used inside the module to deeply analyze the data. Specifically, the road environment comprehensive state assessment module will conduct a detailed analysis of the traffic conditions near the optimized user current location, such as identifying whether there are congested road sections or accident sites, and calculating the travel efficiency of each possible path based on this information. In addition, the module will also consider the impact of weather factors on cycling safety, for example, if the road ahead is about to encounter heavy rain, it may recommend that cyclists choose other safer routes. In this process, the road environment comprehensive state assessment module not only simply aggregates and displays data, but also generates a comprehensive road environment state assessment result by considering all relevant factors. This includes but is not limited to recommended best cycling routes, estimated arrival time, and potential risk warnings.

[0061] In particular, the S6 adjusts the cycling route planning in real time based on the road environment comprehensive state evaluation result and the set cycling preference, and uses a dynamic route optimization algorithm. That is, after obtaining the road environment comprehensive state evaluation result, the system combines this information with the user's cycling preference. The cycling preference may include but is not limited to preference for scenic routes, desire to choose the shortest path, or avoidance of busy road sections, etc. These preference settings are usually set by the user in advance in the system or automatically generated according to historical cycling data. By understanding the user's personal preferences, the system can better meet their specific needs and provide more personalized services. Therefore, in the technical solution of the present application, the cycling route planning is adjusted in real time based on the road environment comprehensive state evaluation result and the set cycling preference, and a dynamic route optimization algorithm is used. The dynamic route optimization algorithm uses a series of advanced calculation methods to process these complex data, including but not limited to heuristic search algorithm, genetic algorithm, simulated annealing algorithm, etc. These algorithms can find the optimal solution among a large number of feasible paths, taking into account the balance of time cost, distance, safety, etc. More importantly, it can be dynamically adjusted according to the real-time updated data, which means that as the road conditions change during cycling, the system can recalculate the optimal route in real time. Then, the system will feed back the optimized cycling route to the user. This is usually presented in the form of navigation instructions to guide the cyclist to follow the recommended path. In this way, not only the safety and efficiency of cycling are improved, but also a more intelligent and comfortable cycling experience is provided for the user.

[0062] In summary, the route intelligent planning method based on the intelligent cycling helmet according to the embodiments of the present application is illustrated, which converts the IMU raw data through the low-dimensional structured coding of the full connection layer by using the data time queue continuously collected by the IMU sensor, and inputs it into the specially designed motion state nonlinear time series integral coding network for processing, extracts the key features that can reflect the real motion trajectory, and then calculates the position offset compensation value through the decoder to finally realize the accurate correction of the user's current position. In this way, reliable position information can be provided even in the case of weak or missing GPS signal, ensuring the accuracy and reliability of path planning, and thus improving the safety and experience of cycling.

[0063] Further, a route intelligent planning system based on an intelligent cycling helmet is also provided.

[0064] Figure 5 A block diagram of the route intelligent planning system based on the intelligent cycling helmet according to the embodiments of the present application is shown in FIG. 1. As shown in the figure, Figure 5As shown, the route intelligent planning system 300 based on the intelligent cycling helmet according to the embodiment of the present application comprises: a data collection module 310, configured to collect a time queue of user real-time positions and a time queue of IMU data by deploying a GPS module and an IMU sensor in the intelligent cycling helmet; an external data receiving module 320, configured to receive a time queue of traffic data and a time queue of weather data by deploying an external data receiving module in the intelligent cycling helmet; a user current position extraction module 330, configured to extract a user current position from the time queue of user real-time positions; a position deviation correction module 340, configured to correct the user current position based on the time queue of IMU data to obtain an optimized user current position; a road environment comprehensive state evaluation module 350, configured to input the optimized user current position, the time queue of traffic data and the time queue of weather data into the road environment comprehensive state evaluation module to obtain a road environment comprehensive state evaluation result; and a route real-time adjustment module 360, configured to adjust a cycling route plan in real time based on the road environment comprehensive state evaluation result and a set cycling preference, and using a dynamic route optimization algorithm.

[0065] As described above, the route intelligent planning system 300 based on the intelligent cycling helmet according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with the route intelligent planning algorithm based on the intelligent cycling helmet, and the like. In a possible implementation manner, the route intelligent planning system 300 based on the intelligent cycling helmet according to the embodiment of the present application can be integrated into a wireless terminal as a software module and / or a hardware module. For example, the route intelligent planning system 300 based on the intelligent cycling helmet can be a software module in an operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the route intelligent planning system 300 based on the intelligent cycling helmet can also be one of many hardware modules of the wireless terminal.

[0066] Alternatively, in another example, the route intelligent planning system 300 based on the intelligent cycling helmet and the wireless terminal can also be separate devices, and the route intelligent planning system 300 based on the intelligent cycling helmet can be connected to the wireless terminal through a wired and / or wireless network, and transmit interactive information in an agreed data format.

[0067] Having described above several embodiments of the disclosure, any modifications and variations that fall within the scope of the described embodiments are also intended to be within the scope of the disclosure. As will be apparent to those skilled in the art, some modifications and variations to the embodiments described above can be practiced while staying within the scope and spirit of the described embodiments. The foregoing description of the described embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the described embodiments to the precise form disclosed. Many modifications and variations are possible in light of the above teachings. It is intended that the disclosed embodiments be limited only by the claims.

Claims

1. A route intelligent planning method based on an intelligent cycling helmet, characterized in that, The method comprises the following steps: Collecting a time queue of user real-time positions and a time queue of IMU data through a GPS module and an IMU sensor deployed in a smart cycling helmet; Receiving a time queue of traffic data and a time queue of weather data through an external data receiving module deployed in the smart cycling helmet; Extracting a user current position from the time queue of user real-time positions; Based on the time queue of IMU data, performing position deviation correction on the user current position to obtain an optimized user current position, comprising: Performing low-dimensional structured coding on each IMU data in the time queue of IMU data based on a full connection layer to obtain a time queue of IMU low-dimensional structured coding vectors; Inputting the time queue of IMU low-dimensional structured coding vectors into an information kernel coarse-grained aggregation network to obtain an IMU coarse-grained aggregation coding vector, and the specific calculation formula is: in, This represents the time queue of the IMU low-dimensional structured encoding vector. These are the 1st, 2nd, and 3rd times in the time queue of the IMU low-dimensional structured encoded vector, respectively. The and the first Each IMU low-dimensional structured encoded vector and This represents taking the maximum and minimum values ​​in the vector. Kernel factors representing global feature information for function, The kernel weights are global feature information. The number of vectors in the time queue of the IMU low-dimensional structured encoded vectors. This indicates the coarse-grained convergence encoding direction of the IMU; Based on the IMU coarse-grained aggregation coding vector, determining the IMU kernel aggregation compensation weight of each IMU low-dimensional structured coding vector in the time queue of IMU low-dimensional structured coding vectors to obtain a set of IMU kernel aggregation compensation weight factors, and the specific calculation formula is: wherein, is a point convolution layer processing, is a first weight matrix, is a second weight matrix, is a function, is the th IMU low-dimensional structured encoding modulation vector in the time queue of the IMU low-dimensional structured encoding vectors, is the th IMU low-dimensional structured encoding modulation vector, is an IMU coarse-grained aggregation encoding modulation vector, is a position-wise subtraction, is an absolute value taking, is an IMU kernel aggregation compensation weight vector, denotes the two-norm of a vector, is a logarithm function value with base 2, denotes an IMU kernel aggregation bias compensation factor, and denote a compensation weight matrix and a compensation bias vector, respectively, is a compensation modulation vector, is a corresponding IMU kernel aggregation compensation factor, is a preset threshold, is a compensation explicit modeling operation, is a corresponding IMU kernel aggregation compensation weight factor; Inputting the set of IMU kernel aggregation compensation weight factors, the IMU coarse-grained aggregation coding vector and the time queue of IMU low-dimensional structured coding vectors into a node fine-grained dynamic compensation aggregation network to obtain an IMU fine-grained compensation aggregation coding vector, and the specific calculation formula is: wherein, denotes the IMU fine-grained compensation aggregated encoding vector; Inputting the IMU coarse-grained aggregation coding vector and the IMU fine-grained compensation aggregation coding vector into a residual unit to obtain a motion state time series type integral accumulation coding vector, and the specific calculation formula is: wherein, and are residual weight coefficients, denotes the motion state time series integral cumulative encoding vector; Based on the motion state time series type integral accumulation coding vector, determining a position offset compensation value; Based on the position offset compensation value, performing position deviation correction on the user current position to obtain the optimized user current position; Inputting the optimized user current position, the time queue of traffic data and the time queue of weather data into a road environment comprehensive state evaluation module to obtain a road environment comprehensive state evaluation result; Based on the road environment comprehensive state evaluation result and a set riding preference, using a dynamic route optimization algorithm to adjust a cycling route plan in real time. 2.The smart ride based on a smart ride helmet route intelligent planning method according to claim 1, characterized in that, Based on the IMU coarse-grained aggregation coding vector, determining the IMU kernel aggregation compensation weight of each IMU low-dimensional structured coding vector in the time queue of IMU low-dimensional structured coding vectors to obtain a set of IMU kernel aggregation compensation weight factors, comprising: Calculating the kernel aggregation compensation factor of each IMU low-dimensional structured coding vector in the time queue of IMU low-dimensional structured coding vectors relative to the IMU coarse-grained aggregation coding vector to obtain a set of IMU kernel aggregation compensation factors; Performing compensation explicit modeling based on a gating function on the set of IMU kernel aggregation compensation factors to obtain the set of IMU kernel aggregation compensation weight factors. 3.The smart ride based on a smart ride helmet route intelligent planning method according to claim 2, characterized in that, calculating each IMU low-dimensional structured encoding vector in the time queue of the IMU low-dimensional structured encoding vector relative to a core convergence compensation factor of the IMU coarse-grained convergence encoding vector to obtain a set of IMU core convergence compensation factors, comprising: modulating the IMU low-dimensional structured encoding vector and the IMU coarse-grained convergence encoding vector to obtain an IMU low-dimensional structured encoding modulation vector and an IMU coarse-grained convergence encoding modulation vector; calculating compensation information between the IMU low-dimensional structured encoding modulation vector and the IMU coarse-grained convergence encoding modulation vector to obtain an IMU core convergence compensation weight vector; calculating an IMU core convergence bias compensation factor based on the IMU low-dimensional structured encoding vector and the IMU coarse-grained convergence encoding vector; calculating a core convergence compensation factor based on the IMU core convergence compensation weight vector and the IMU core convergence bias compensation factor to obtain the IMU core convergence compensation factor. 4.The smart ride based on a smart ride helmet route intelligent planning method according to claim 3, characterized in that, calculating an IMU core convergence bias compensation factor based on the IMU low-dimensional structured encoding vector and the IMU coarse-grained convergence encoding vector, comprising: in response to a two-norm of the IMU low-dimensional structured encoding vector being less than a two-norm of the IMU coarse-grained convergence encoding vector, calculating a two-norm of the IMU low-dimensional structured encoding vector divided by a two-norm of the IMU coarse-grained convergence encoding vector and then adding a constant one, and calculating a value of a logarithmic function with 2 as a base of the obtained number; in response to a two-norm of the IMU low-dimensional structured encoding vector being greater than or equal to a two-norm of the IMU coarse-grained convergence encoding vector, calculating a two-norm of the IMU low-dimensional structured encoding vector divided by a two-norm of the IMU coarse-grained convergence encoding vector. 5.The smart ride based on a smart ride helmet route intelligent planning method according to claim 4, characterized in that, determining a position offset compensation value based on the motion state time series quasi-integral cumulative encoding vector, comprising: inputting the motion state time series quasi-integral cumulative encoding vector into a decoder-based position compensation estimation module to obtain a position offset compensation value. 6.A route intelligent planning system based on an intelligent cycling helmet, characterized in that, comprising: a data acquisition module configured to acquire a time queue of user real-time positions and a time queue of IMU data through a GPS module and an IMU sensor deployed in a smart cycling helmet; an external data receiving module configured to receive a time queue of traffic data and a time queue of weather data by using an external data receiving module deployed in the smart cycling helmet; a user current position extraction module configured to extract a user current position from the time queue of user real-time positions; a position bias correction module configured to perform position bias correction on the user current position based on the time queue of IMU data to obtain an optimized user current position, comprising: performing low-dimensional structured encoding on each IMU data in the time queue of IMU data based on a full connection layer to obtain a time queue of IMU low-dimensional structured encoding vectors; inputting the time queue of IMU low-dimensional structured encoding vectors into an information core coarse-grained convergence network to obtain an IMU coarse-grained convergence encoding vector, and the specific calculation formula is: in, This represents the time queue of the IMU low-dimensional structured encoding vector. These are the 1st, 2nd, and 3rd times in the time queue of the IMU low-dimensional structured encoded vector, respectively. The and the first Each IMU low-dimensional structured encoded vector and This represents taking the maximum and minimum values ​​in the vector. Kernel factors representing global feature information for function, The kernel weights are global feature information. The number of vectors in the time queue of the IMU low-dimensional structured encoded vectors. This indicates the coarse-grained convergence encoding direction of the IMU; Based on the IMU coarse-grained aggregation encoding vector, determine the IMU core aggregation compensation weight of each IMU low-dimensional structured encoding vector in the time queue of the IMU low-dimensional structured encoding vector to obtain a set of IMU core aggregation compensation weight factors, and the specific calculation formula is: wherein, is a point convolution layer processing, is a first weight matrix, is a second weight matrix, is a function, is the th IMU low-dimensional structured encoding modulation vector in the time queue of the IMU low-dimensional structured encoding vectors, is the th IMU low-dimensional structured encoding modulation vector in the time queue of the IMU low-dimensional structured encoding vectors, is an IMU coarse-grained aggregation encoding modulation vector, is a position-wise subtraction, is an absolute value taking, is an IMU kernel aggregation compensation weight vector, denotes the two-norm of a vector, is a logarithm function value with base 2, denotes an IMU kernel aggregation bias compensation factor, and denote a compensation weight matrix and a compensation bias vector, respectively, is a compensation modulation vector, is a corresponding IMU kernel aggregation compensation factor, is a preset threshold, is a compensation explicit modeling operation, is a corresponding IMU kernel aggregation compensation weight factor; Input the set of IMU core aggregation compensation weight factors, the IMU coarse-grained aggregation encoding vector and the time queue of the IMU low-dimensional structured encoding vector into the node fine-grained dynamic compensation aggregation network to obtain an IMU fine-grained compensation aggregation encoding vector, and the specific calculation formula is: wherein, denotes the IMU fine-grained compensation aggregated encoding vector; Input the IMU coarse-grained aggregation encoding vector and the IMU fine-grained compensation aggregation encoding vector into the residual unit to obtain a motion state time series type integral accumulation encoding vector, and the specific calculation formula is: wherein, and are residual weight coefficients, denotes the motion state time series integral accumulated encoding vector; Based on the motion state time series type integral accumulation encoding vector, determine the position offset compensation value; Based on the position offset compensation value, correct the position deviation of the current position of the user to obtain the optimized current position of the user; The road environment comprehensive state evaluation module is used for inputting the optimized current position of the user, the time queue of the traffic data and the time queue of the weather data into the road environment comprehensive state evaluation module to obtain a road environment comprehensive state evaluation result; The route real-time adjustment module is used for adjusting the riding route planning in real time based on the road environment comprehensive state evaluation result and the set riding preference, and using a dynamic route optimization algorithm.

Images