Navigation processing method and device based on artificial intelligence, computer equipment and medium

Abstract

The invention provides a navigation processing method and device based on artificial intelligence, computer equipment and a storage medium. The method comprises the following steps: after a driver starts a navigation application, acquiring preorder trajectory data of the driver based on the number of trajectory points and acquired current position information of the driver; acquiring target trajectory data corresponding to the driver when last navigation is finished; performing data integration on the preorder trajectory data and the target trajectory data to obtain first trajectory data; preprocessing the first trajectory data to obtain second trajectory data; performing feature extraction processing on the second trajectory data based on a feature extraction strategy to obtain a target feature set; performing positioning processing on the target feature set based on a sorting model, and outputting a corresponding target navigation path; and processing the target navigation path based on the path calculation service to obtain a corresponding target navigation result, and feeding back the target navigation result to the navigation application. Based on the use of the sorting model, more accurate and efficient navigation service can be provided.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based navigation processing method, device, computer equipment and storage medium. Background Art

[0002] In the existing path planning technology scenario, the path planning system usually relies on a single starting point and end point coordinates (i.e. single-point GPS coordinate data) to calculate the route to provide navigation services to users. This path planning method based on single-point GPS coordinates can provide users with relatively accurate route guidance under ideal conditions, that is, when the GPS signal is stable and there is no external interference.

[0003] However, in practical applications, due to the influence of various factors, the accuracy of single-point GPS coordinate data is often not guaranteed. First, GPS signals are easily interfered by various factors during propagation, such as natural factors such as the atmosphere and ionosphere, as well as obstructions from buildings and structures such as urban high-rise buildings, tunnels, and viaducts. These interferences and obstructions can cause GPS signal drift or signal loss, thereby causing deviations in single-point GPS coordinate data. Secondly, even when the GPS signal is stable, due to the complex and changeable actual driving environment of vehicles in the city, such as vehicles parked on the roadside, driving under viaducts, and traveling in tunnels, the actual position of the vehicle in these cases may also be significantly different from the position located by GPS. If this position deviation is directly used for path planning, it will cause a significant deviation between the planned route and the driver's actual driving route.

[0004] Due to the above reasons, the existing path planning methods based on single-point GPS coordinates often have the problem of low path planning quality in practical applications and cannot provide users with accurate and reliable navigation services. Summary of the invention

[0005] The main purpose of the present invention is to provide a navigation processing method, device, computer equipment and storage medium based on artificial intelligence, aiming to solve the technical problem that the existing path planning method based on single-point GPS coordinates has low path planning quality in practical applications and cannot provide users with accurate and reliable navigation services.

[0006] To achieve the above object, the present invention provides a navigation processing method based on artificial intelligence, the method comprising:

[0007] After the driver starts the navigation application, the current location information of the driver is obtained, and the previous trajectory data of the driver is collected based on a preset number of trajectory points and the current location information;

[0008] Acquire target trajectory data corresponding to the driver at the end of the last navigation;

[0009] Integrating the preceding trajectory data with the target trajectory data to obtain corresponding first trajectory data;

[0010] Preprocessing the first trajectory data to obtain corresponding second trajectory data;

[0011] Performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set;

[0012] Performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path;

[0013] The target navigation path is processed based on a preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application.

[0014] Optionally, before performing positioning processing on the target feature set based on a preset sorting model and outputting a corresponding target navigation path, the method further includes:

[0015] Obtaining pre-collected driver driving log data;

[0016] The driver's driving log data is labeled and processed based on the preset offline trajectory flow algorithm to obtain the corresponding target data;

[0017] Performing feature engineering on the target data to obtain corresponding sample data;

[0018] Call the preset initial model;

[0019] Training and optimizing the initial model based on the sample data to obtain a specified model that meets the preset construction requirements;

[0020] The specified model is used as the sorting model.

[0021] Optionally, the training and optimization processing of the initial model based on the sample data to obtain a specified model that meets preset construction requirements includes:

[0022] Dividing the sample data into a training set and a validation set based on a preset division ratio;

[0023] Using the training set to train the initial model to obtain a trained first model;

[0024] Based on a preset gain value evaluation index, using the validation set to perform a model evaluation on the first model to obtain a corresponding performance evaluation result;

[0025] Based on the performance evaluation result, the first model is optimized using a preset optimization strategy to obtain an optimized second model;

[0026] The second model is used as the designated model.

[0027] Optionally, the acquiring of target trajectory data corresponding to the driver at the end of last navigation includes:

[0028] Acquire the first historical track data of the GPS point corresponding to the driver at the end of the last navigation;

[0029] Acquire the second historical track data of the GPS point after the route matching at the end of the last navigation corresponding to the driver;

[0030] Integrate the first historical trajectory data and the second historical trajectory data to obtain corresponding designated trajectory data;

[0031] The designated trajectory data is used as the target trajectory data.

[0032] Optionally, preprocessing the first trajectory data to obtain corresponding second trajectory data includes:

[0033] Performing noise removal processing on the first trajectory data to obtain corresponding first processed data;

[0034] Performing duplicate point removal processing on the first processed data to obtain corresponding second processed data;

[0035] Performing outlier filtering processing on the second processed data to obtain corresponding third processed data;

[0036] The third processed data is used as the second trajectory data.

[0037] Optionally, performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set includes:

[0038] Extracting trajectory features from the preceding trajectory data; wherein the trajectory features include trajectory continuity features, directionality features, and speed features;

[0039] Acquire first attribute information of a current point corresponding to the current position information;

[0040] Obtaining second attribute information of a preset candidate path;

[0041] Combining the trajectory feature, the first attribute information, and the second attribute information to obtain corresponding combined data;

[0042] The combined data is used as the target feature set.

[0043] Optionally, after the target navigation path is processed based on the preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application, the method further includes:

[0044] Determining whether feedback information corresponding to the target navigation result returned by the driver is received;

[0045] If yes, updating the target navigation result based on the feedback information to obtain an updated navigation result;

[0046] The updated navigation result is fed back to the navigation application.

[0047] In addition, to achieve the above-mentioned purpose, the present invention further provides a navigation processing device based on artificial intelligence, and the navigation processing device based on artificial intelligence includes:

[0048] A first processing module, configured to obtain the current location information of the driver after the driver starts the navigation application, and collect the previous trajectory data of the driver based on a preset number of trajectory points and the current location information;

[0049] A first acquisition module is used to acquire target trajectory data corresponding to the driver at the end of the last navigation;

[0050] An integration module, used for integrating the preceding trajectory data with the target trajectory data to obtain corresponding first trajectory data;

[0051] A preprocessing module, used for preprocessing the first trajectory data to obtain corresponding second trajectory data;

[0052] An extraction module, used to perform feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set;

[0053] A second processing module, used for performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path;

[0054] The third processing module is used to process the target navigation path based on a preset route calculation service to obtain a corresponding target navigation result, and feed the target navigation result back to the navigation application.

[0055] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0056] The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the artificial intelligence-based navigation processing methods proposed in the embodiments of the present application are implemented.

[0057] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0058] The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of any one of the artificial intelligence-based navigation processing methods proposed in the embodiments of the present application are implemented.

[0059] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0060] The present invention provides a navigation processing method, device, computer equipment and storage medium based on artificial intelligence. The method includes: after a driver starts a navigation application, firstly obtaining the current position information of the driver, and collecting the previous trajectory data of the driver based on a preset number of trajectory points; then obtaining the target trajectory data corresponding to the driver at the end of the last navigation; then integrating the previous trajectory data with the target trajectory data to obtain the corresponding first trajectory data; and pre-processing the first trajectory data to obtain the corresponding second trajectory data; subsequently performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set; further performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path; finally processing the target navigation path based on a preset route calculation service to obtain a corresponding target navigation result, and feeding back the target navigation result to the navigation application. After the driver starts the navigation application, the present invention obtains the driver's current location information, and collects the driver's previous trajectory data based on the preset number of trajectory points and the current location information, then obtains the target trajectory data corresponding to the driver at the end of the last navigation, and integrates the previous trajectory data with the target trajectory data to obtain the corresponding first trajectory data, then pre-processes the first trajectory data to obtain the corresponding second trajectory data, and performs feature extraction processing on the second trajectory data based on the use of a preset feature extraction strategy to obtain the corresponding target feature set, and then performs positioning processing on the target feature set based on the use of a preset sorting model, outputs the corresponding target navigation path, and subsequently calculates the target navigation path based on the preset route calculation service. The corresponding target navigation result is processed and the target navigation result is fed back to the navigation application. In this way, the present invention integrates the driver's previous trajectory data with the target trajectory data at the end of the driver's last navigation, and then extracts features from the integrated trajectory data to obtain the corresponding target feature set, which is then beneficial for the sorting model to perform intelligent analysis and processing on the target feature set, which can effectively improve the positioning accuracy, thereby accurately predicting the driver's actual position and driving direction and outputting a more accurate target navigation path, so that reliable starting point positioning can be provided for path planning in the future, ensuring the effectiveness and accuracy of the target navigation result generated based on the target navigation path, and is conducive to providing drivers with more accurate and efficient navigation services. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1is an exemplary system architecture diagram to which the present application may be applied;

[0063] Figure 2 is a flow chart of a navigation processing method based on artificial intelligence provided by an embodiment of the present invention;

[0064] Figure 3 is a structural schematic diagram of an embodiment of an artificial intelligence-based navigation processing device according to the present application;

[0065] Figure 4 This is a basic structural block diagram of the computer device in this embodiment. DETAILED DESCRIPTION

[0066] The artificial intelligence-based navigation processing method provided by the embodiment of the present invention is applied to the artificial intelligence-based navigation processing device. Unless otherwise defined, all technical and scientific terms used in this document have the same meaning as those generally understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0067] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0068] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0069] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0070] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social online platform software, etc.

[0071] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0072] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0073] It should be noted that the artificial intelligence-based navigation processing method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based navigation processing device is generally arranged in the server / terminal device.

[0074] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0075] Existing positioning methods usually rely on single-point GPS coordinate data for position calculation. This method is simple and direct, but because it cannot combine rich external information for comprehensive judgment, the positioning accuracy and robustness in complex scenarios are often greatly limited. Single-point GPS positioning is mainly based on the real-time satellite signals received by the device, ignoring factors such as the target's motion trajectory, road network structure, navigation context information, etc. This single positioning mode is susceptible to the following limitations:

[0076] 1) Binding problem at complex intersections: Single-point GPS coordinate data is only a point value, and it is difficult to accurately distinguish multiple roads that are close or parallel. Especially at complex intersections, overpasses, and high-speed nodes with multi-layer structures, it is easy to make binding errors, that is, binding the device location to the wrong road, causing the navigation system to generate an incorrect path.

[0077] 2) Insufficient direction recognition capability in dynamic scenarios: Single-point GPS coordinates cannot fully utilize the target’s motion trajectory direction information. Therefore, in vehicle start-up, deceleration or parking scenarios, it is difficult to accurately determine the device’s location and direction of movement, which may cause deviations in route planning.

[0078] 3) Decreased accuracy in parking lots and signal-blocked areas: Parking lots and underground spaces are subject to severe signal interference, and single-point GPS data can have significant deviations in accuracy. Even if the device is located outdoors, when there are many obstacles such as buildings (such as the urban canyon effect) or when the vehicle is stationary, a single real-time coordinate is unlikely to provide a stable positioning result.

[0079] 4) The lack of contextual information leads to limited navigation effect: Previous positioning methods cannot make full use of contextual information and historical records during the navigation process, such as the last navigation destination, path location, etc., and lack the ability to analyze and apply historical navigation results, resulting in positioning being difficult to adapt to the needs of long-term or segmented navigation.

[0080] 5) The planned route takes a detour or U-turn: The error of single-point GPS positioning is often in the range of several meters to more than ten meters. This error accumulation in a complex road network often causes the device to be mistakenly believed to be located on the wrong section of the road. The navigation system then generates a detour, U-turn instructions, or even cross-road planning results, seriously affecting the user experience.

[0081] To solve the above problems, the present invention fundamentally improves the positioning method. It combines the position and direction information of the previous trajectory to perform a global analysis of the movement characteristics of the target device during the navigation process, and further integrates the road network data to correct the current possible position of the device, and more intelligently identifies the matching relationship between the device and the road network structure. In addition, the addition of the last navigation end information enables the system to effectively memorize the important data at the end of navigation, and use it as the initial reference point to make up for the shortcomings of single-point coordinate calculation.

[0082] In the technical implementation, the present invention uses a sorting model to dynamically process and analyze the priority and credibility of multiple data sources to ensure the accuracy and flexibility of device location judgment. This method brings the following significant advantages:

[0083] Eliminate road binding errors: For scenes such as intersections and highway intersections, combine road level and direction information to reduce the probability of incorrect road binding.

[0084] Improve positioning accuracy: Constrain the current coordinates through trajectory and direction information, greatly reducing positioning errors in outdoor scenes.

[0085] Improve robustness: In parking lots, complex overpasses and other conditions, the stability and reliability of positioning results can be improved by integrating multiple data sources.

[0086] Optimize route planning: Based on context analysis and road network correction, avoid problems such as detours and U-turns in route planning.

[0087] Continue to refer Figure 2 , shows a flow chart of an embodiment of a navigation processing method based on artificial intelligence proposed in the present application. The embodiment of the present application can acquire and process relevant data based on artificial intelligence technology.

[0088] The artificial intelligence-based navigation processing method provided by the embodiment of the present invention comprises the following steps:

[0089] S210, after the driver starts the navigation application, the current location information of the driver is obtained, and the previous trajectory data of the driver is collected based on a preset number of trajectory points and the current location information.

[0090] In this step, the execution subject of the present invention may specifically be a mobile terminal installed with a navigation application. Specifically, when the driver starts the navigation application, the driver's current location information can be collected in real time through positioning technologies such as GPS, Wi-Fi, and base stations. The collected current location information may include longitude and latitude, speed, positioning accuracy, angle, positioning source, and time information. Among them, the collected current location information can be cached in real time according to the preset number of trajectory points to obtain the driver's previous trajectory data. There is no specific limitation on the selection of the numerical value of the above-mentioned number of trajectory points, which can be set according to actual business needs, for example, it can be set to 25. Then, the longitude and latitude, speed, positioning accuracy, angle, positioning source, and time information of the driver's first 25 trajectory points can be cached in real time and used as the above-mentioned previous trajectory data, and the previous trajectory data can be used to provide a reference for the current positioning.

[0091] The application scenarios of the present invention include: dense road network: suitable for scenes with dense traffic network such as viaducts, main and auxiliary roads, which helps to accurately select the best driving path.

[0092] Intersections: At complex intersections, the sorting model can effectively identify the correct direction of travel of vehicles and optimize crossing strategies.

[0093] Static vehicle: Suitable for positioning stationary vehicles. It estimates the most likely position of the vehicle through historical trajectory information to ensure accurate positioning.

[0094] S220, obtaining target trajectory data corresponding to the driver at the end of the last navigation.

[0095] In this step, the specific implementation process of obtaining the target trajectory data corresponding to the driver at the end of the last navigation will be further described in detail in the subsequent specific embodiments of the present invention, and will not be elaborated on here.

[0096] S230: Integrate the preceding trajectory data and the target trajectory data to obtain corresponding first trajectory data.

[0097] In this step, the corresponding designated integrated data can be obtained by integrating the preceding trajectory data and the target trajectory data, and the designated integrated data can be used as the first trajectory data. The first trajectory data is a trajectory data set including the preceding trajectory data and the target trajectory data.

[0098] S240: Preprocess the first trajectory data to obtain corresponding second trajectory data.

[0099] In this step, the specific implementation process of preprocessing the first trajectory data to obtain the corresponding second trajectory data will be further described in detail in subsequent specific embodiments of the present invention, and will not be elaborated on here.

[0100] S250: Perform feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set.

[0101] In this step, the starting point can be located and matched by calling the binding service, and the second trajectory data can be feature extracted and processed by the binding service to extract important features that affect positioning and navigation. Specifically, the specific implementation process of performing feature extraction and processing on the second trajectory data based on the preset feature extraction strategy to obtain the corresponding target feature set will be further described in detail in the subsequent specific embodiments of the present invention, and will not be elaborated on here.

[0102] S260, performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path.

[0103] In this step, the model input of the above-mentioned sorting model is the target feature set, and the input is the score of each candidate path. Then, based on the score, the optimal path and direction can be further screened out as the best estimate of the current position. The optimal path (i.e., the target navigation path) has a high score and is consistent with the driver's driving history and current state. The construction process of the above-mentioned sorting model will be further described in detail in the subsequent specific embodiments of the present invention, and will not be elaborated on here.

[0104] S270: Process the target navigation path based on a preset route calculation service to obtain a corresponding target navigation result, and feed the target navigation result back to the navigation application.

[0105] In this step, the route calculation service is a service that has the function of generating corresponding navigation results according to the navigation path. The target navigation path can be transmitted to the route calculation service to generate the final target navigation result. The target navigation result includes information such as the optimal path (i.e., the target navigation path), estimated time and distance. In addition, the generated target navigation result is fed back to the navigation application for the driver's reference and use. By displaying the target navigation path, estimated time and distance and other information in the navigation application, a driver-friendly navigation experience can be intelligently provided.

[0106] In an embodiment of the present invention, after a driver starts a navigation application, the current location information of the driver is first obtained, and the previous trajectory data of the driver is collected based on a preset number of trajectory points; then the target trajectory data corresponding to the driver at the end of the last navigation is obtained; then the previous trajectory data and the target trajectory data are integrated to obtain corresponding first trajectory data; and the first trajectory data is preprocessed to obtain corresponding second trajectory data; subsequently, feature extraction processing is performed on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set; further positioning processing is performed on the target feature set based on a preset sorting model, and a corresponding target navigation path is output; finally, the target navigation path is processed based on a preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application. After the driver starts the navigation application, the present invention obtains the driver's current location information, and collects the driver's previous trajectory data based on the preset number of trajectory points and the current location information, then obtains the target trajectory data corresponding to the driver at the end of the last navigation, and integrates the previous trajectory data with the target trajectory data to obtain the corresponding first trajectory data, then pre-processes the first trajectory data to obtain the corresponding second trajectory data, and performs feature extraction processing on the second trajectory data based on the use of a preset feature extraction strategy to obtain the corresponding target feature set, and then performs positioning processing on the target feature set based on the use of a preset sorting model, outputs the corresponding target navigation path, and subsequently calculates the target navigation path based on the preset route calculation service. The corresponding target navigation result is processed and the target navigation result is fed back to the navigation application. In this way, the present invention integrates the driver's previous trajectory data with the target trajectory data at the end of the driver's last navigation, and then extracts features from the integrated trajectory data to obtain the corresponding target feature set, which is then beneficial for the sorting model to perform intelligent analysis and processing on the target feature set, which can effectively improve the positioning accuracy, thereby accurately predicting the driver's actual position and driving direction and outputting a more accurate target navigation path, so that reliable starting point positioning can be provided for path planning in the future, ensuring the effectiveness and accuracy of the target navigation result generated based on the target navigation path, and is conducive to providing drivers with more accurate and efficient navigation services.

[0107] Optionally, before performing positioning processing on the target feature set based on a preset sorting model and outputting a corresponding target navigation path, the method further includes:

[0108] Obtain pre-collected driver driving log data.

[0109] In this step, the above-mentioned driver driving log data can be obtained by integrating the driving history data of all drivers collected in advance within the preset time period, including GPS tracks, speeds, directions, etc., and the road network data collected in advance, including road types, direction angles, lengths, etc., and the order data collected in advance, including the starting point, end point, time, etc. There is no specific limitation on the time selection of the above-mentioned preset time period, which can be set according to actual business needs, for example, within the past year. The collected driver driving log data can also be cleaned, including removing duplicate, invalid or abnormal data, filling in missing values, or making reasonable inferences based on the context, and standardizing the data, such as unifying the timestamp format and speed units, so as to improve the data quality of the driver driving log data.

[0110] The driver's driving log data is labeled and processed based on the preset offline trajectory flow algorithm to obtain the corresponding target data.

[0111] In this step, the offline trajectory flow algorithm can be used to calculate the corresponding binding route truth value (i.e., the actual driving route) according to the complete order trajectory of the order data. The binding route truth value is used as the label of the corresponding driver driving log data for supervised learning.

[0112] Feature engineering is performed on the target data to obtain corresponding sample data.

[0113] In this step, the process of feature engineering the target data includes feature extraction, feature selection, and feature processing. Specifically, feature extraction includes: extracting features related to candidate paths and directions from the original data, such as the position, speed, direction, etc. of the current point; extracting features related to candidate paths from the road network data, such as road type, direction angle, length, etc. Feature selection includes: using statistical methods or machine learning algorithms (such as correlation analysis, chi-square test, recursive feature elimination, etc.) to select features that have a significant impact on model performance; and removing redundant or noise features to reduce model complexity. Feature processing includes: normalizing, discretizing, encoding, and other processing of features to improve the learning efficiency and performance of the model; and transforming nonlinear features, such as polynomial features, logarithmic features, etc., to capture the complex relationship between features.

[0114] Recall the preset initial model.

[0115] In this step, the initial model may specifically adopt the lambdaMART model, which performs well in processing large-scale data, handling nonlinear relationships, and capturing feature interactions.

[0116] The initial model is trained and optimized based on the sample data to obtain a specified model that meets the preset construction requirements.

[0117] In this step, the specific implementation process of training and optimizing the initial model based on the sample data to obtain a specified model that meets the preset construction requirements will be further described in detail in the subsequent specific embodiments of the present invention and will not be elaborated on here.

[0118] The specified model is used as the sorting model.

[0119] In this step, after the ranking model is built, the trained ranking model can be deployed to an online environment for real-time processing of the driver's driving data and road network data, and the ranking model can be used to optimize the candidate paths and directions, and the results can be fed back to the driver or navigation system. The performance of the model can be monitored regularly to ensure that the model can continue to accurately identify the optimal path and direction. In addition, the model can be regularly updated and optimized based on new data and user feedback. New features, algorithms or technologies can also be introduced to improve the accuracy and robustness of the model.

[0120] In an embodiment of the present invention, the driver's driving log data collected in advance is obtained; then the driver's driving log data is annotated based on a preset offline trajectory flow algorithm to obtain corresponding target data; then the target data is feature engineered to obtain corresponding sample data; subsequently the preset initial model is called; the initial model is further trained and optimized based on the sample data to obtain a designated model that meets the preset construction requirements; and the designated model is used as the sorting model. The present invention annotates the driver's driving log data collected in advance based on the use of a preset offline trajectory flow algorithm to obtain corresponding target data, then feature engineers the target data to obtain corresponding sample data, and subsequently trains and optimizes the initial model based on the sample data, thereby achieving efficient and accurate sorting models that meet the construction requirements, effectively improving the construction efficiency of the sorting model, and ensuring the model effect of the sorting model obtained.

[0121] Optionally, the training and optimization processing of the initial model based on the sample data to obtain a specified model that meets preset construction requirements includes:

[0122] The sample data is divided into a training set and a validation set based on a preset division ratio.

[0123] In this step, there is no specific limitation on the numerical value selection of the above division ratio, which can be determined according to actual business needs, for example, 7:3 can be selected.

[0124] The initial model is trained using the training set to obtain a trained first model.

[0125] In this step, the initial model can be trained using a labeled training set, and appropriate model parameters such as learning rate, number of iterations, number and depth of trees can be set. Then, the performance of the model can be evaluated using methods such as cross-validation, and the parameters can be adjusted to optimize the model, thereby obtaining a trained first model.

[0126] Based on the preset gain value evaluation index, the first model is evaluated using the validation set to obtain a corresponding performance evaluation result.

[0127] In this step, the gain value evaluation index specifically adopts NDCG (Normalized Discounted Cumulative Gain), which is an index that comprehensively considers the relationship between the model ranking result and the real sequence. The first model is evaluated by using the above validation set, and the NDCG value between the model ranking result and the real sequence is calculated as the above performance evaluation result, and is used to evaluate the accuracy of the model.

[0128] Based on the performance evaluation result, the first model is optimized using a preset optimization strategy to obtain an optimized second model.

[0129] In this step, the above optimization strategy may specifically include trying different feature combinations, feature transformations, and model parameter settings to find the best model configuration. The advantages and differences of the model can be analyzed based on the performance evaluation results on the validation set, and then the above optimization strategy can be used to perform corresponding model optimization on the above first model, so as to obtain a second model whose accuracy and robustness meet the requirements. And the second model can accurately identify the optimal path and direction.

[0130] The second model is used as the designated model.

[0131] In an embodiment of the present invention, the sample data is divided into a training set and a validation set based on a preset division ratio; the initial model is then trained using the training set to obtain a trained first model; then, based on a preset gain value evaluation index, the first model is evaluated using the validation set to obtain a corresponding performance evaluation result; subsequently, based on the performance evaluation result, the first model is optimized using a preset optimization strategy to obtain an optimized second model; and the second model is used as the designated model. The present invention divides the sample data into a training set and a validation set based on a preset division ratio, and then the initial model is trained using the training set to obtain a trained first model; then, based on a gain value evaluation index, the validation set and the optimization strategy are used to optimize the first model to obtain an optimized second model and use it as the final designated model, thereby constructing an efficient and accurate sorting model, effectively improving the efficiency of sorting model construction, and ensuring the model effect of the sorting model obtained.

[0132] Optionally, the acquiring of target trajectory data corresponding to the driver at the end of last navigation includes:

[0133] The first historical track data of the GPS point corresponding to the driver at the end of the last navigation is obtained.

[0134] In this step, when the driver finishes the last navigation, the position, direction, speed, and positioning accuracy information of the original GPS point at the end of the last navigation, that is, the first historical trajectory data, will be cached.

[0135] The second historical track data of the GPS point after the route matching at the end of the last navigation corresponding to the driver is obtained.

[0136] In this step, at the same time, when the driver's last navigation ends, the GPS point information after route matching, that is, the above-mentioned second historical trajectory data, will be saved. Among them, the above-mentioned GPS point information is obtained by matching the real-time GPS point with the road on the map, including information such as the matched position, direction, speed and positioning accuracy. This matching process is usually to improve positioning accuracy and reduce noise interference, because the original GPS point may deviate from the actual road due to equipment accuracy, environmental factors and other reasons. Specifically, the location information at the end of the last navigation can be used as the starting point or reference point for subsequent navigation or location tracking. The GPS point information after route matching can help the system understand the driver's driving path and current position more accurately. In the subsequent navigation process, the system can use this information to plan the optimal path, provide real-time navigation prompts, etc.

[0137] The first historical trajectory data and the second historical trajectory data are integrated to obtain corresponding designated trajectory data.

[0138] In this step, the designated trajectory data includes the first historical trajectory data and the second historical trajectory data.

[0139] The designated trajectory data is used as the target trajectory data.

[0140] In an embodiment of the present invention, by acquiring the first historical trajectory data of the GPS point at the end of the last navigation corresponding to the driver; and acquiring the second historical trajectory data of the GPS point after route matching at the end of the last navigation corresponding to the driver; then integrating the first historical trajectory data with the second historical trajectory data to obtain the corresponding designated trajectory data; and subsequently using the designated trajectory data as the target trajectory data. The present invention acquires the first historical trajectory data of the GPS point at the end of the last navigation corresponding to the driver, and acquires the second historical trajectory data of the GPS point after route matching at the end of the last navigation corresponding to the driver, and then integrating the first historical trajectory data with the second historical trajectory data, so that the target trajectory data at the end of the last navigation corresponding to the driver can be automatically and accurately obtained, thereby ensuring the data accuracy of the obtained target trajectory data.

[0141] Optionally, preprocessing the first trajectory data to obtain corresponding second trajectory data includes:

[0142] The first trajectory data is subjected to noise removal processing to obtain corresponding first processed data.

[0143] In this step, the first trajectory data may be smoothed by using an algorithm such as a Kalman filter to reduce noise, thereby obtaining corresponding first processed data.

[0144] The first processed data is subjected to duplicate point removal processing to obtain corresponding second processed data.

[0145] In this step, the distance and time difference between adjacent points can be calculated from the first processed data, and points that are less than 2 meters away from the previous point can be removed to avoid repeated recording, thereby obtaining the corresponding second processed data.

[0146] The second processed data is subjected to outlier filtering processing to obtain corresponding third processed data.

[0147] In this step, outliers can be filtered from the second processed data according to the confidence and positioning accuracy to ensure the accuracy and reliability of the third processed data. If the network positioning point has no speed and direction and is easy to deviate from the current track, the network positioning point is filtered.

[0148] The third processed data is used as the second trajectory data.

[0149] In an embodiment of the present invention, the first trajectory data is subjected to noise removal processing to obtain corresponding first processed data; then the first processed data is subjected to duplicate point removal processing to obtain corresponding second processed data; then the second processed data is subjected to outlier filtering processing to obtain corresponding third processed data; and the third processed data is subsequently used as the second trajectory data. The present invention performs noise removal processing, duplicate point removal processing, and outlier filtering processing on the first trajectory data, thereby achieving efficient and accurate preprocessing of the first trajectory data, and effectively ensuring the accuracy and reliability of the generated second trajectory data.

[0150] Optionally, performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set includes:

[0151] Extracting trajectory features from the preceding trajectory data, wherein the trajectory features include trajectory continuity features, directionality features, and speed features.

[0152] In this step, key information can be extracted from the above-mentioned previous trajectory data to reflect the continuity, directionality and speed change of the trajectory. Specifically, by calculating the direction difference, distance difference and time difference between the current GPS point and the previous GPS point, the speed change of the driver during driving and the continuity of the trajectory can be understood. At the same time, based on the observation probability, direction angle and projection distance between the starting point of the route calculation and the candidate path (recall link), the distance from the projection point to the endpoint, the distance between the candidate path and the last navigation point, and other information, directional features can be extracted. These features help to evaluate the degree of match between the candidate path and the driver's current driving direction. Among them, in the path planning process, the starting point of the route calculation is the starting point for the algorithm to calculate the optimal path. Alternatively, the starting point of the route calculation also refers to the driver's corresponding last navigation end point.

[0153] Acquire first attribute information of a current point corresponding to the current position information.

[0154] In this step, the attribute information of the current point corresponding to the current location information includes direction, speed, positioning accuracy, number of recalled links, positioning type, GPS direction angle of the end point, straight-line distance to the GPS end point, etc., which reflect the current driving status of the driver.

[0155] Obtain second attribute information of a preset candidate path.

[0156] In this step, the second attribute information of the candidate path may include direction angle, road type (such as expressway, city street, etc.), length, width, speed limit level, etc. This information is helpful to evaluate the characteristics and applicability of the candidate path.

[0157] The trajectory feature, the first attribute information, and the second attribute information are combined to obtain corresponding combined data.

[0158] In this step, the combined data is composed of the trajectory feature, the first attribute information and the second attribute information.

[0159] The combined data is used as the target feature set.

[0160] In this step, the target feature set is a constructed feature set containing multiple dimensions. These features in the target feature set will be used as inputs to the ranking model, and together constitute a basic data set for subsequent ranking model analysis, which is used to score and rank candidate paths.

[0161] In an embodiment of the present invention, trajectory features are extracted from the preceding trajectory data; wherein the trajectory features include trajectory continuity features, directional features, and speed features; then the first attribute information of the current point corresponding to the current position information is obtained; and the second attribute information of the preset candidate path is obtained; then the trajectory features, the first attribute information, and the second attribute information are combined to obtain corresponding combined data; and the combined data is subsequently used as the target feature set. The present invention extracts trajectory features from the preceding trajectory data, then obtains the first attribute information of the current point corresponding to the current position information, and obtains the second attribute information of the preset candidate path, and then combines the trajectory features, the first attribute information, and the second attribute information, so that the feature extraction process of the second trajectory data can be automatically and accurately completed and the corresponding target feature set can be obtained, which effectively ensures the data accuracy of the obtained target feature set and is conducive to providing necessary data support for subsequent sorting model analysis.

[0162] Optionally, after the target navigation path is processed based on the preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application, the method further includes:

[0163] Determine whether feedback information corresponding to the target navigation result returned by the driver is received.

[0164] In this step, the target navigation result is fed back to the navigation application for the driver's reference and use, thereby providing the driver with a friendly navigation experience. The driver can also provide feedback on the target navigation result returned by the system according to actual needs.

[0165] If so, the target navigation result is updated based on the feedback information to obtain an updated navigation result.

[0166] In this step, after receiving the feedback information corresponding to the target navigation result returned by the driver, the above-mentioned target navigation result can be updated accordingly based on the feedback information, which may include adjusting the path planning algorithm, optimizing map data, improving the user interface, etc., so as to improve the accuracy and practicality of navigation.

[0167] The updated navigation result is fed back to the navigation application.

[0168] In this step, by feeding back the updated navigation results to the navigation application, the driver is intelligently provided with navigation results that meet personalized needs, thereby further improving the driver's navigation experience.

[0169] In an embodiment of the present invention, it is determined whether feedback information corresponding to the target navigation result returned by the driver is received; if so, the target navigation result is updated based on the feedback information to obtain an updated navigation result; and the updated navigation result is subsequently fed back to the navigation application. After receiving feedback information corresponding to the target navigation result returned by the driver, the present invention will automatically and intelligently update the target navigation result based on the feedback information to obtain an updated navigation result, and then feed back the updated navigation result to the navigation application, thereby automatically and accurately providing the driver with navigation results that meet personalized needs based on the driver's personal feedback, which is conducive to further improving the driver's navigation experience.

[0170] In some optional implementations, the user information obtained is subject to the user's consent and complies with relevant laws and policies.

[0171] In addition, the present invention proposes an innovative method, which uses the previous trajectory and navigation behavior cached by the front end (making full use of the previous trajectory data to cache when the vehicle or equipment moves, and using the cached previous trajectory to extract historical continuity features to provide a more reliable reference basis for subsequent positioning), and applies the sorting model to predict the path and direction of the current location (optimizing multiple candidate l paths through the sorting model, combining the current point status and previous trajectory, selecting the optimal positioning result, and improving the overall performance of the model; selecting the lambdaMART model for training and reasoning, the effect and performance can meet the requirements). This method significantly enhances the robustness and accuracy of positioning, and is of great significance for solving the starting point positioning problem in complex path planning (the determination of location information not only depends on single-point data, but also combines dynamic trajectory characteristics; so that accurate starting point positioning can be achieved in complex scenes such as bridges, main and auxiliary roads, intersections or obstacles). Specific effects include:

[0172] Improve robustness: effectively deal with signal interference in complex environments and enhance the system's adaptability under changing conditions.

[0173] Improved accuracy: By integrating historical data, current positioning is optimized to ensure more accurate path identification.

[0174] Facilitate complex path planning: Provide reliable starting point positioning for complex path planning to ensure the effectiveness and accuracy of subsequent navigation.

[0175] Further references Figure 3 , as a response to the above Figure 2 In order to realize the method shown in the figure, the present application provides an embodiment of a navigation processing device 300 based on artificial intelligence, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0176] An embodiment of the present invention provides an artificial intelligence-based navigation processing device 300, and the artificial intelligence-based navigation processing device 300 includes:

[0177] The first processing module 310 is used to obtain the current location information of the driver after the driver starts the navigation application, and collect the previous trajectory data of the driver based on the preset number of trajectory points and the current location information;

[0178] The first acquisition module 320 is used to acquire the target trajectory data corresponding to the driver at the end of the last navigation;

[0179] An integration module 330 is used to integrate the preceding trajectory data and the target trajectory data to obtain corresponding first trajectory data;

[0180] A preprocessing module 340, configured to preprocess the first trajectory data to obtain corresponding second trajectory data;

[0181] An extraction module 350 is used to perform feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set;

[0182] The second processing module 360 ​​is used to perform positioning processing on the target feature set based on a preset sorting model and output a corresponding target navigation path;

[0183] The third processing module 370 is used to process the target navigation path based on a preset route calculation service to obtain a corresponding target navigation result, and feed the target navigation result back to the navigation application.

[0184] Optionally, the artificial intelligence-based navigation processing device 300 further includes:

[0185] The second acquisition module is used to acquire the driver's driving log data collected in advance;

[0186] The labeling module is used to label the driver's driving log data based on a preset offline trajectory flow algorithm to obtain the corresponding target data;

[0187] A fourth processing module is used to perform feature engineering on the target data to obtain corresponding sample data;

[0188] A calling module is used to call the preset initial model;

[0189] A construction module, used to train and optimize the initial model based on the sample data to obtain a specified model that meets the preset construction requirements;

[0190] A determination module is used to use the specified model as the sorting model.

[0191] Optionally, the building block includes:

[0192] A division submodule, used for dividing the sample data into a training set and a validation set based on a preset division ratio;

[0193] A training submodule, used to train the initial model using the training set to obtain a trained first model;

[0194] An evaluation submodule, configured to perform a model evaluation on the first model using the validation set based on a preset gain value evaluation index to obtain a corresponding performance evaluation result;

[0195] An optimization submodule, configured to optimize the first model using a preset optimization strategy based on the performance evaluation result to obtain an optimized second model;

[0196] The first determining submodule is used to use the second model as the designated model.

[0197] Optionally, the first acquisition module 320 includes:

[0198] A first acquisition submodule is used to acquire first historical track data of the GPS point corresponding to the driver at the end of the last navigation;

[0199] The second acquisition submodule is used to acquire the second historical track data of the GPS point after route matching at the end of the last navigation corresponding to the driver;

[0200] An integration submodule, used for integrating the first historical trajectory data and the second historical trajectory data to obtain corresponding designated trajectory data;

[0201] The second determining submodule is configured to use the designated trajectory data as the target trajectory data.

[0202] Optionally, the preprocessing module 340 includes:

[0203] A first processing submodule, configured to perform noise removal processing on the first trajectory data to obtain corresponding first processed data;

[0204] A second processing submodule, configured to perform duplicate point removal processing on the first processed data to obtain corresponding second processed data;

[0205] A third processing submodule, configured to perform outlier filtering processing on the second processed data to obtain corresponding third processed data;

[0206] The third determining submodule is configured to use the third processed data as the second trajectory data.

[0207] Optionally, the extraction module 350 includes:

[0208] An extraction submodule, used to extract trajectory features from the preceding trajectory data; wherein the trajectory features include trajectory continuity features, directionality features and speed features;

[0209] A third acquisition submodule is used to acquire first attribute information of a current point corresponding to the current position information;

[0210] A fourth acquisition submodule is used to acquire second attribute information of a preset candidate path;

[0211] A combining submodule, used for combining the trajectory feature, the first attribute information and the second attribute information to obtain corresponding combined data;

[0212] The fourth determination submodule is used to use the combined data as the target feature set.

[0213] Optionally, the artificial intelligence-based navigation processing device 300 further includes:

[0214] A judgment module, used for judging whether feedback information corresponding to the target navigation result returned by the driver is received;

[0215] An updating module, configured to update the target navigation result based on the feedback information to obtain an updated navigation result;

[0216] A feedback module is used to feed back the updated navigation result to the navigation application.

[0217] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0218] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0219] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0220] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as the program code of the navigation processing method based on artificial intelligence, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0221] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the program code stored in the memory 41 or process data, such as running the program code of the navigation processing method based on artificial intelligence.

[0222] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0223] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores the application crash processing program, and the application crash processing program can be executed by at least one processor to enable the at least one processor to perform the steps of the artificial intelligence-based navigation processing method as described above.

[0224] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware online platform, and of course, by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0225] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0226] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A navigation processing method based on artificial intelligence, characterized in that: include: After the driver starts the navigation application, the current location information of the driver is obtained, and the previous trajectory data of the driver is collected based on a preset number of trajectory points and the current location information; Acquire target trajectory data corresponding to the driver at the end of the last navigation; Integrating the preceding trajectory data with the target trajectory data to obtain corresponding first trajectory data; Preprocessing the first trajectory data to obtain corresponding second trajectory data; Performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set; Performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path; The target navigation path is processed based on a preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application.

2. The method according to claim 1, characterized in that Before performing positioning processing on the target feature set based on the preset sorting model and outputting the corresponding target navigation path, the method further includes: Obtaining pre-collected driver driving log data; The driver's driving log data is labeled and processed based on the preset offline trajectory flow algorithm to obtain the corresponding target data; Performing feature engineering on the target data to obtain corresponding sample data; Call the preset initial model; Training and optimizing the initial model based on the sample data to obtain a specified model that meets the preset construction requirements; The specified model is used as the sorting model.

3. The method according to claim 2, characterized in that The training and optimization of the initial model based on the sample data to obtain a specified model that meets the preset construction requirements includes: Dividing the sample data into a training set and a validation set based on a preset division ratio; Using the training set to train the initial model to obtain a trained first model; Based on a preset gain value evaluation index, using the validation set to perform a model evaluation on the first model to obtain a corresponding performance evaluation result; Based on the performance evaluation result, the first model is optimized using a preset optimization strategy to obtain an optimized second model; The second model is used as the designated model.

4. The method according to claim 1, characterized in that The step of obtaining the target trajectory data corresponding to the driver at the end of the last navigation comprises: Acquire the first historical track data of the GPS point corresponding to the driver at the end of the last navigation; Acquire the second historical track data of the GPS point after the route matching at the end of the last navigation corresponding to the driver; Integrate the first historical trajectory data and the second historical trajectory data to obtain corresponding designated trajectory data; The designated trajectory data is used as the target trajectory data.

5. The method according to claim 1, characterized in that The preprocessing of the first trajectory data to obtain corresponding second trajectory data includes: Performing noise removal processing on the first trajectory data to obtain corresponding first processed data; Performing duplicate point removal processing on the first processed data to obtain corresponding second processed data; Performing outlier filtering processing on the second processed data to obtain corresponding third processed data; The third processed data is used as the second trajectory data.

6. The method according to claim 1, characterized in that The performing feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set includes: Extracting trajectory features from the preceding trajectory data; wherein the trajectory features include trajectory continuity features, directionality features and speed features; Acquire first attribute information of a current point corresponding to the current position information; Obtaining second attribute information of a preset candidate path; Combining the trajectory feature, the first attribute information, and the second attribute information to obtain corresponding combined data; The combined data is used as the target feature set.

7. The method according to claim 1, characterized in that After the target navigation path is processed based on the preset route calculation service to obtain a corresponding target navigation result, and the target navigation result is fed back to the navigation application, the method further includes: Determining whether feedback information corresponding to the target navigation result returned by the driver is received; If yes, updating the target navigation result based on the feedback information to obtain an updated navigation result; The updated navigation result is fed back to the navigation application.

8. A navigation processing device based on artificial intelligence, characterized in that: include: A first processing module, configured to obtain the current location information of the driver after the driver starts the navigation application, and collect the previous trajectory data of the driver based on a preset number of trajectory points and the current location information; A first acquisition module is used to acquire target trajectory data corresponding to the driver at the end of the last navigation; An integration module, used for integrating the preceding trajectory data with the target trajectory data to obtain corresponding first trajectory data; A preprocessing module, used for preprocessing the first trajectory data to obtain corresponding second trajectory data; An extraction module, used to perform feature extraction processing on the second trajectory data based on a preset feature extraction strategy to obtain a corresponding target feature set; A second processing module, used for performing positioning processing on the target feature set based on a preset sorting model, and outputting a corresponding target navigation path; The third processing module is used to process the target navigation path based on a preset route calculation service to obtain a corresponding target navigation result, and feed the target navigation result back to the navigation application.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the artificial intelligence-based navigation processing method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the artificial intelligence-based navigation processing method according to any one of claims 1 to 7 are implemented.

Images