Intelligent driving teaching method and device based on USS and RTK data fusion

Abstract

The invention relates to the technical field of data analysis, and discloses an intelligent driving teaching method and device based on USS and RTK data fusion, and the method comprises the steps: collecting environment perception data and global positioning data corresponding to a vehicle driven by a student during the driving training of the student; the environment sensing data comprises USS data, and the global positioning data comprises RTK data; determining a dynamic local map of the environment where the vehicle is located and motion state data of the vehicle according to the environment perception data and the global positioning data; generating a dynamic scene view according to the dynamic local map and the motion state data; according to the dynamic local map and the motion state data, analyzing the driving behavior of the student to obtain a driving behavior real-time analysis result; and according to the driving behavior real-time analysis result, real-time driving guidance information for the driving behavior is fed back to the trainee. Therefore, the guidance flexibility and the guidance accuracy of driving teaching can be improved by implementing the method, so that the safety and the reliability of the driving teaching process are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, and particularly relates to an intelligent driving teaching method and device based on USS and RTK data fusion. BACKGROUND

[0002] In the traditional motor vehicle driver training, a coach usually carries out one-to-one or one-to-many teaching guidance to the trainee according to a teaching scheme, so as to assist the trainee in learning driving skills.

[0003] However, in this driving teaching mode, if the coach has limited ability to judge and intervene in actual emergencies in a complex traffic environment, it may be difficult to guide the trainee to handle the emergency situation in time, and therefore the traditional driving teaching mode has certain safety hazards. In view of the defects of the above-mentioned driving teaching mode, in the prior art, an obstacle detection and early warning mode is usually used to prompt the driver to handle the driving emergency situation, but this existing mode cannot flexibly and accurately guide the trainee to operate the vehicle in combination with the actual driving situation.

[0004] It can be seen that it is particularly important to propose a technical solution capable of improving the guidance flexibility and accuracy of driving teaching, thereby improving the safety and reliability of the driving teaching process. SUMMARY

[0005] The present application provides an intelligent driving teaching method and device based on USS and RTK data fusion, which can improve the guidance flexibility and accuracy of driving teaching, thereby improving the safety and reliability of the driving teaching process.

[0006] In order to solve the above technical problems, the present application discloses an intelligent driving teaching method based on USS and RTK data fusion, which comprises: During the driving training of the trainee, the environmental perception data corresponding to the vehicle driven by the trainee and the global positioning data corresponding to the vehicle are collected; wherein the environmental perception data comprises USS data, and the global positioning data comprises RTK data; According to the environmental perception data and the global positioning data, the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle are determined; According to the dynamic local map and the motion state data, a dynamic scene view is generated; the dynamic scene view is used to assist the trainee to complete the driving training; According to the dynamic local map and the motion state data, the driving behavior of the trainee is analyzed to obtain a real-time driving behavior analysis result; According to the real-time driving behavior analysis result, real-time driving guidance information for the driving behavior is fed back to the trainee.

[0007] As an optional implementation, in the first aspect of the present application, the determining of the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle according to the environment perception data and the global positioning data comprises: performing spatio-temporal synchronization processing on the environment perception data based on the global positioning data to obtain synchronized environment perception data; performing coordinate transformation processing on the synchronized environment perception data to convert the coordinate system corresponding to the synchronized environment perception data from a local coordinate system corresponding to the synchronized environment perception data to a global coordinate system corresponding to the global positioning data to obtain target environment perception data; performing data filtering fusion processing on the target environment perception data and the global positioning data based on a pre-set data filtering fusion algorithm to obtain the motion state data corresponding to the vehicle; constructing the dynamic local map of the environment where the vehicle is located according to the motion state data; wherein the map form of the dynamic local map is a grid map form or a feature map form, and the dynamic local map is used to record the spatial position relationship and dynamic change data of the current environment where the vehicle is located in real time.

[0008] As an optional implementation, in the first aspect of the present application, the performing of the spatio-temporal synchronization processing on the environment perception data based on the global positioning data to obtain synchronized environment perception data comprises: determining a first timestamp of the global positioning data and a second timestamp of the environment perception data; calculating a time difference value between the first timestamp and the second timestamp; determining current driving data corresponding to the vehicle; the current driving data comprises a current vehicle speed and a current driving direction; based on the determined extrapolation correction formula, correcting the environment perception data according to the current vehicle speed, the current driving direction and the time difference value to obtain synchronized environment perception data.

[0009] As an optional implementation, in the first aspect of the present application, the current driving data further comprises a current heading angle and a current acceleration; wherein the performing of the data filtering fusion processing on the target environment perception data and the global positioning data based on the pre-set data filtering fusion algorithm to obtain the motion state data corresponding to the vehicle comprises: determining position coordinate information corresponding to the vehicle according to the global positioning data; determine a raw state vector corresponding to the vehicle according to the position coordinate information, the current heading angle, the current vehicle speed and the current acceleration; perform state transition prediction on the raw state vector based on a predetermined vehicle motion model to obtain a predicted state vector; update the predicted state vector based on a predetermined observation model according to the target environment perception data and the global positioning data to obtain a target state vector; determine the target state vector as the motion state data corresponding to the vehicle.

[0010] As an optional implementation, in the first aspect of the present application, the method further comprises: loading static map data corresponding to a site where the vehicle is located before the trainee starts driving training; the static map data comprises a static global map and scene configuration data corresponding to the static global map; the scene configuration data comprises a combination of one or more of reference distance configuration data, coordinate configuration data, safety function configuration data and subject position configuration data; configuring the static global map based on the scene configuration data to obtain a configured global map; wherein, the generating a dynamic scene view according to the dynamic local map and the motion state data comprises: determining real-time training scene information corresponding to the vehicle according to the dynamic local map and the motion state data; the real-time training scene information comprises a combination of one or more of real-time scene identification data, real-time position data, virtual guide identification data, obstacle data and real-time boundary data; superimposing and displaying the real-time training scene information on the configured global map as a dynamic scene view to display the dynamic scene view on the man-machine interaction unit.

[0011] As an optional implementation, in the first aspect of the present application, the analyzing the driving behavior of the trainee according to the dynamic local map and the motion state data to obtain a driving behavior real-time analysis result comprises: obtaining standard operation information corresponding to a driving training subject; the standard operation information comprises standard operation rules, and the standard operation rules record operation safety value intervals corresponding to a plurality of driving operations; collecting operation data of the trainee in the driving training for the vehicle; determining the driving behavior of the trainee according to the operation data, the dynamic local map and the motion state data; determine whether the driving behavior meets a preset standard operation condition and / or whether the driving behavior has a driving safety risk according to the standard operation information, the dynamic local map and the motion state data, and obtain a determination result; determine a driving behavior real-time analysis result according to the determination result.

[0012] As an optional implementation form, in the first aspect of the present application, the method further comprises: generate a driving training evaluation result corresponding to the trainee according to the operation data and the driving behavior real-time analysis result; the driving training evaluation result comprises a quantitative score result corresponding to a plurality of evaluation dimensions and / or a comprehensive evaluation result; generate a training analysis result for the driving training evaluation result; the training analysis result comprises at least one of a training advantage analysis report, a training error point analysis report and a training improvement suggestion; feed back the driving training evaluation result and the training analysis result to the trainee.

[0013] The second aspect of the present application discloses an intelligent driving teaching device based on USS and RTK data fusion, which comprises: a collection module, configured to collect environmental perception data corresponding to a vehicle driven by a trainee and global positioning data corresponding to the vehicle during driving training of the trainee; wherein the environmental perception data comprises USS data, and the global positioning data comprises RTK data; a determination module, configured to determine a dynamic local map of an environment in which the vehicle is located and motion state data corresponding to the vehicle according to the environmental perception data and the global positioning data; a generation module, configured to generate a dynamic scene view according to the dynamic local map and the motion state data; the dynamic scene view is used to assist the trainee to complete the driving training; an analysis module, configured to analyze driving behavior of the trainee according to the dynamic local map and the motion state data, and obtain a driving behavior real-time analysis result; a feedback module, configured to feed back real-time driving guidance information for the driving behavior to the trainee according to the driving behavior real-time analysis result.

[0014] As an optional implementation form, in the second aspect of the present application, the specific manner in which the determination module determines the dynamic local map of the environment in which the vehicle is located and the motion state data corresponding to the vehicle according to the environmental perception data and the global positioning data comprises: perform space-time synchronization processing on the environmental perception data based on the global positioning data, and obtain synchronized environmental perception data; perform coordinate transformation processing on the synchronous environment perception data to convert a coordinate system corresponding to the synchronous environment perception data from a local coordinate system corresponding to the synchronous environment perception data to a global coordinate system corresponding to the global positioning data, to obtain target environment perception data; perform data filtering and fusion processing on the target environment perception data and the global positioning data based on a pre-set data filtering and fusion algorithm, to obtain motion state data corresponding to the vehicle; construct a dynamic local map of an environment in which the vehicle is located according to the motion state data; The dynamic local map is in the form of a grid map or a feature map, and is used to record spatial position relationships and dynamic change data of an environment in which the vehicle is currently located in real time.

[0015] As an optional implementation, in the second aspect of the present application, the specific manner in which the determination module performs spatio-temporal synchronization processing on the environment perception data based on the global positioning data to obtain synchronous environment perception data includes: determining a first timestamp of the global positioning data and a second timestamp of the environment perception data; calculating a time difference value between the first timestamp and the second timestamp; determining current driving data corresponding to the vehicle; the current driving data includes a current vehicle speed and a current driving direction; based on the determined extrapolation correction formula, correcting the environment perception data according to the current vehicle speed, the current driving direction and the time difference value to obtain synchronous environment perception data.

[0016] As an optional implementation, in the second aspect of the present application, the current driving data further includes a current heading angle and a current acceleration; The specific manner in which the determination module performs data filtering and fusion processing on the target environment perception data and the global positioning data based on a pre-set data filtering and fusion algorithm to obtain motion state data corresponding to the vehicle includes: determining position coordinate information corresponding to the vehicle according to the global positioning data; determining an original state vector corresponding to the vehicle according to the position coordinate information, the current heading angle, the current vehicle speed and the current acceleration; performing state transition prediction on the original state vector based on a pre-determined vehicle motion model to obtain a predicted state vector; updating the predicted state vector according to the target environment perception data and the global positioning data based on a predetermined observation model, to obtain a target state vector; determining the target state vector as the motion state data corresponding to the vehicle.

[0017] As an optional implementation, in the second aspect of the present application, the device further comprises: a loading module configured to load static map data corresponding to a site where the vehicle is located before the trainee starts the driving training; the static map data comprises a static global map and scene configuration data corresponding to the static global map; the scene configuration data comprises a combination of one or more of reference distance configuration data, coordinate configuration data, safety function configuration data, and subject position configuration data; a configuration module configured to configure the static global map based on the scene configuration data to obtain a configured global map; The specific manner in which the generation module generates the dynamic scene view according to the dynamic local map and the motion state data comprises: determining real-time training scene information corresponding to the vehicle according to the dynamic local map and the motion state data; the real-time training scene information comprises a combination of one or more of real-time scene identification data, real-time position data, virtual guide identification data, obstacle data, and real-time boundary data; superimposing and displaying the real-time training scene information on the configured global map as a dynamic scene view to display the dynamic scene view on the human-computer interaction unit.

[0018] As an optional implementation, in the second aspect of the present application, the analysis module analyzes the driving behavior of the trainee according to the dynamic local map and the motion state data to obtain a driving behavior real-time analysis result in the following specific manner: obtaining standard operation information corresponding to a driving training subject; the standard operation information comprises standard operation rules, and the standard operation rules record operation safety value intervals corresponding to a plurality of driving operations; collecting operation data of the trainee on the vehicle in the driving training; determining the driving behavior of the trainee according to the operation data, the dynamic local map, and the motion state data; judging whether the driving behavior meets a pre-set standard operation condition and / or whether the driving behavior has a driving safety risk according to the standard operation information, the dynamic local map, and the motion state data to obtain a judgment result; determining a driving behavior real-time analysis result according to the judgment result.

[0019] As an optional implementation, in the second aspect of the present application, the generating module is further configured to generate the driving training evaluation result corresponding to the trainee according to the operation data and the real-time analysis result of the driving behavior; the driving training evaluation result comprises a quantitative score result corresponding to a plurality of evaluation dimensions and / or a comprehensive evaluation result; The generating module is further configured to generate a training analysis result for the driving training evaluation result; the training analysis result comprises at least one of a training advantage analysis report, a training error point analysis report and a training improvement suggestion; The feedback module is further configured to feed back the driving training evaluation result and the training analysis result to the trainee.

[0020] The third aspect of the present application discloses another intelligent driving teaching device based on USS and RTK data fusion, which comprises: a memory storing executable program codes; a processor coupled with the memory; The processor invokes the executable program codes stored in the memory to execute part or all of the steps of the intelligent driving teaching method based on USS and RTK data fusion disclosed in the first aspect of the present application.

[0021] The fourth aspect of the present application discloses a computer storage medium storing computer instructions, which, when invoked, is configured to execute part or all of the steps of the intelligent driving teaching method based on USS and RTK data fusion disclosed in the first aspect of the present application.

[0022] Compared with the prior art, the present application has the following beneficial effects: In the present application, during the driving training of the trainee, the environmental perception data corresponding to the vehicle driven by the trainee and the global positioning data corresponding to the vehicle are collected; the environmental perception data includes USS data, and the global positioning data includes RTK data; the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle are determined according to the environmental perception data and the global positioning data; the dynamic scene view is generated according to the dynamic local map and the motion state data; the dynamic scene view is used to assist the trainee to complete the driving training; the driving behavior of the trainee is analyzed according to the dynamic local map and the motion state data, and the real-time analysis result of the driving behavior is obtained; the real-time driving guidance information for the driving behavior is fed back to the trainee according to the real-time analysis result of the driving behavior. It can be seen that, by implementing the present application, the environmental perception data including the USS data corresponding to the vehicle driven by the trainee and the global positioning data including the RTK data can be collected during the driving training of the trainee, and then the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle are determined in combination with the environmental perception data and the global positioning data. Then, the dynamic scene view is generated according to the dynamic local map and the motion state data, and the driving behavior of the trainee is analyzed to obtain the real-time analysis result of the driving behavior. According to the real-time analysis result of the driving behavior, the corresponding real-time driving guidance information is fed back to the trainee. By fusing the depth data of the USS and the RTK, the accurate perception and high-precision positioning of the vehicle itself state and the complex environment in the near distance are realized, a powerful safety monitoring system for driving teaching is constructed, thereby being beneficial to providing timely, flexible and accurate driving guidance for the vehicle motion state and the surrounding environment, and further being beneficial to improving the guidance flexibility and accuracy of the driving teaching, and being beneficial to improving the safety and reliability of the driving teaching process, and further being beneficial to improving the driving training efficiency and the driving training effect of the trainee. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating any inventive labor.

[0024] Figure 1 is a flowchart of an intelligent driving teaching method based on fusion of USS and RTK data disclosed by the embodiments of the present application; Figure 2 is a flowchart of another intelligent driving teaching method based on fusion of USS and RTK data disclosed by the embodiments of the present application; Figure 3 is a structural schematic diagram of an intelligent driving teaching device based on fusion of USS and RTK data disclosed by the embodiments of the present application; Figure 4 is another structure schematic view of the intelligent driving teaching device based on USS and RTK data fusion disclosed by the embodiment of the present application; Figure 5 is another structure schematic view of the intelligent driving teaching device based on USS and RTK data fusion disclosed by the embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor fall within the scope of protection of the present application.

[0026] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, and are not used to describe a particular order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or end including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or end.

[0027] In this document, the term "embodiment" means that the specific features, structures, or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor is it independent or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] The application discloses an intelligent driving teaching method and device based on USS and RTK data fusion, which can collect environment perception data including USS data and global positioning data including RTK data corresponding to a vehicle driven by a trainee during driving training of the trainee, combine the environment perception data and the global positioning data, determine a dynamic local map of an environment where the vehicle is located and motion state data corresponding to the vehicle, generate a dynamic scene view according to the dynamic local map and the motion state data, analyze driving behavior of the trainee, and obtain real-time analysis results of the driving behavior, so as to feed back real-time driving guidance information corresponding to the real-time analysis results of the driving behavior to the trainee. The intelligent driving teaching method and device based on USS and RTK data fusion can realize accurate perception and high-precision positioning of a state of the vehicle and a complex environment in a short distance through deep data fusion of the USS and the RTK, construct a powerful safety monitoring system for driving teaching, and thus is favorable for providing timely, flexible and accurate driving guidance for the motion state of the vehicle and the surrounding environment, and further favorable for improving guidance flexibility and guidance accuracy of the driving teaching, and improving safety and reliability of the driving teaching process, and further improving driving training efficiency and driving training effect of the trainee. The following will be described in detail.

[0029] Embodiment one Please refer to Figure 1 , Figure 1 is a flowchart of an intelligent driving teaching method based on USS and RTK data fusion disclosed by the embodiment of the application. Wherein, Figure 1 The intelligent driving teaching method based on USS and RTK data fusion described above can be applied to an intelligent driving teaching device based on USS and RTK data fusion. The intelligent driving teaching device can include one of an intelligent device, an intelligent terminal, an intelligent system and a server. The server can include a local server or a cloud server, and the embodiment of the application does not limit the server. Further, the device can be applied to a vehicle used for driving training, or can be applied to a control system corresponding to the vehicle used for driving training, and the embodiment of the application does not limit the device. As shown in Figure 1 The intelligent driving teaching method based on USS and RTK data fusion can include the following operations: 101、In the process of driving training of the trainee, the environment perception data corresponding to the vehicle driven by the trainee and the global positioning data corresponding to the vehicle are collected.

[0030] In the embodiment of the present application, the environment perception data can include USS (Ultrasonic Sensor, ultrasonic sensor) data, and optionally, the environment perception data can also include other types of environment perception data, which are not limited in the embodiment of the present application; and the global positioning data can include RTK (Real-Time Kinematic Positioning, real-time dynamic differential positioning) data, and optionally, the global positioning data can also include other types of high-precision positioning data, which are not limited in the embodiment of the present application.

[0031] In the embodiment of the present application, optionally, the USS data corresponding to the vehicle driven by the trainee can be collected based on an ultrasonic sensor array; further optionally, the ultrasonic sensor array is arranged around the vehicle body and is used to detect obstacle information within a preset distance range around the vehicle; wherein, for example, the preset distance range can be in the range of 0.1 meters to 5 meters, which is not limited in the embodiment of the present application; further optionally, the environment perception data can include the above-mentioned obstacle information, and the obstacle information can include one or more of the distance between the obstacle and the vehicle, the direction, and the obstacle contour, which are not limited in the embodiment of the present application.

[0032] In the embodiment of the present application, optionally, the global positioning data corresponding to the vehicle can be obtained based on an RTK-GNSS (Real-Time Kinematic - Global Navigation Satellite System, real-time dynamic differential - global navigation satellite system) receiver; further optionally, the global positioning data can be centimeter-level positioning data, and the global positioning data can include one or more combinations of three-dimensional position coordinates, speed, heading angle, and vehicle attitude information, which are not limited in the embodiment of the present application.

[0033] 102. Determine the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle according to the environment perception data and the global positioning data.

[0034] In the embodiment of the present application, the dynamic local map can be a dynamic high-precision local map, and further, the dynamic local map can be a dynamic high-precision map corresponding to a local area centered on the current position of the vehicle.

[0035] 103. Generate a dynamic scene view according to the dynamic local map and the motion state data.

[0036] In the embodiment of the present application, the dynamic scene view is used to assist the trainee to complete the driving training; optionally, the dynamic scene view can be displayed on the display screen of the human-computer interaction unit corresponding to the vehicle, which is not limited in the embodiment of the present application.

[0037] 104. Analyze the driving behavior of the trainee according to the dynamic local map and the motion state data, to obtain a driving behavior real-time analysis result.

[0038] 105. According to the driving behavior real-time analysis result, feedback real-time driving guidance information for driving behavior to the trainee.

[0039] In the embodiment of the present application, optionally, the real-time driving guidance information can include one or a combination of operation guidance instructions, prompt information, warning information and error correction information, and the embodiment of the present application is not limited.

[0040] In the embodiment of the present application, optionally, step 105 can include the following operations: According to the driving behavior real-time analysis result, determine the real-time driving guidance information for driving behavior; According to the information type corresponding to the preset feedback mode contained in the real-time driving guidance information, feedback the real-time driving guidance information to the trainee through the human-computer interaction unit corresponding to the vehicle; Optionally, the human-computer interaction unit corresponding to the vehicle can include a display screen and a voice interaction module; wherein the display screen can be a vehicle-mounted display screen, or a display screen of other equipment, and the embodiment of the present application is not limited.

[0041] Optionally, the preset feedback mode can include one or a combination of voice broadcast mode, visual display mode, vibration prompt mode and auxiliary control mode, and the embodiment of the present application is not limited; wherein, for example, the visual display mode can be to highlight the real-time driving guidance information on the display screen of the human-computer interaction unit, and the embodiment of the present application is not limited.

[0042] It can be seen that by implementing the method described in the embodiment of the present application, the environmental perception data including USS data and the global positioning data including RTK data corresponding to the vehicle driven by the trainee can be collected during the driving training of the trainee, and then the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle are determined in combination with the environmental perception data and the global positioning data. Then, according to the dynamic local map and the motion state data, a dynamic scene view is generated and the driving behavior of the trainee is analyzed to obtain a driving behavior real-time analysis result. According to the driving behavior real-time analysis result, corresponding real-time driving guidance information is fed back to the trainee. Through the deep data fusion of USS and RTK, accurate perception and high-precision positioning of the state of the vehicle itself and the complex environment in the near distance are realized, a powerful safety monitoring system for driving teaching is constructed, thereby facilitating timely, flexible and accurate driving guidance for the motion state of the vehicle and the surrounding environment, and further improving the guidance flexibility and accuracy of driving teaching, and improving the safety and reliability of the driving teaching process, and further improving the driving training efficiency and driving training effect of the trainee.

[0043] In an optional embodiment, determining the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle according to the environment perception data and the global positioning data can include the following operations: performing spatio-temporal synchronization processing on the environment perception data based on the global positioning data to obtain synchronized environment perception data; performing coordinate transformation processing on the synchronized environment perception data to convert the coordinate system corresponding to the synchronized environment perception data from a local coordinate system corresponding to the synchronized environment perception data to a global coordinate system corresponding to the global positioning data to obtain target environment perception data; performing data filtering fusion processing on the target environment perception data and the global positioning data based on a pre-set data filtering fusion algorithm to obtain motion state data corresponding to the vehicle; constructing a dynamic local map of the environment where the vehicle is located according to the motion state data.

[0044] The dynamic local map is in the form of a grid map or a feature map, and is used to record the spatial position relationship and dynamic change data of the current environment where the vehicle is located in real time.

[0045] Optionally, when the environment perception data includes USS observation data collected by a plurality of ultrasonic sensors, each ultrasonic sensor is configured with a corresponding local coordinate system, and therefore, in the coordinate transformation processing, the USS observation data in different local coordinate systems needs to be converted to the global coordinate system.

[0046] Optionally, the global coordinate system corresponding to the global positioning data can be a global coordinate system or a local coordinate system, which is not limited by the embodiments of the present application.

[0047] Optionally, the data filtering fusion algorithm can be an extended Kalman filter (EKF) algorithm or an unscented Kalman filter (UKF) algorithm, or other filtering fusion algorithms, which are not limited by the embodiments of the present application.

[0048] Wherein, optionally, the dynamic local map can record one or more combinations of the following: close-range obstacle profile and distance information, vehicle self-state information (position, heading, speed, etc.), and relative position relationship between the vehicle and the obstacle; further optionally, the dynamic local map can adopt a sliding window mechanism or a vehicle center following strategy, so that the dynamic local map is updated in real time as the vehicle travels; further optionally, the map coverage range of the dynamic local map is set according to the needs, with the vehicle as the center and a radius of, for example, 5 meters, 10 meters, or other numerical values. Wherein, for example, in the "reverse parking" training subject, the dynamic local map continuously records the obstacle profiles (such as warehouse corner markers and stake barrels) within a range of 0.5-3 meters behind and on both sides of the vehicle, and updates synchronously with the slow movement of the vehicle, for the purpose of determining whether there is a risk of line compression, deviation, or collision.

[0049] It can be seen that the optional embodiment can determine the motion state of the vehicle and construct the corresponding dynamic local map of the vehicle by performing spatio-temporal synchronization, coordinate transformation, and filtering fusion on the global positioning data and the environment perception data, which can improve the processing efficiency and accuracy of the global positioning data and the environment perception data, thereby facilitating the improvement of the data fusion accuracy of the global positioning data and the environment perception data, and further facilitating the improvement of the determination accuracy of the vehicle motion state and the construction accuracy and real-time performance of the dynamic local map, and further facilitating the provision of timely, flexible, and accurate driving guidance for the vehicle motion state and the surrounding environment.

[0050] In the optional embodiment, optionally, based on the global positioning data, the spatio-temporal synchronization processing of the environment perception data to obtain synchronized environment perception data can include the following operations: determining a first timestamp of the global positioning data and a second timestamp of the environment perception data; calculating a time difference value between the first timestamp and the second timestamp; determining the current driving data corresponding to the vehicle; the current driving data includes the current vehicle speed and the current driving direction; based on the determined extrapolation correction formula, correcting the environment perception data according to the current vehicle speed, the current driving direction, and the time difference value to obtain the synchronized environment perception data.

[0051] Wherein, optionally, the current driving direction can be represented by a driving direction unit vector, which is not limited by the embodiments of the present application.

[0052] Wherein, optionally, the first timestamp of the RTK data is , the second timestamp of the USS data is , and the time difference value can be . and assuming that the current vehicle speed is v and the unit vector of the travel direction is The extrapolation correction formula can be as follows:

[0053] wherein, is the original ranging vector corresponding to the USS data, is the synchronized ranging vector; the above-mentioned synchronized environmental perception data can include a plurality of synchronized ranging vectors.

[0054] It can be seen that the optional embodiment can also perform time correction on the environmental perception data according to the time difference between the first timestamp of the global positioning data and the second timestamp of the environmental perception data and the current travel data, to obtain synchronized environmental perception data, so as to realize the space-time synchronization between the global positioning data and the environmental perception data, improve the efficiency and accuracy of the space-time synchronization of the data, and thus be conducive to improving the data fusion accuracy of the subsequent global positioning data and environmental perception data.

[0055] In the optional embodiment, optionally, the coordinate transformation processing is performed on the synchronized environmental perception data to convert the coordinate system corresponding to the synchronized environmental perception data from the local coordinate system corresponding to the synchronized environmental perception data to the global coordinate system corresponding to the global positioning data, to obtain target environmental perception data, which can include the following operations: First, based on the calibration, the external parameter transformation matrix of the USS sensor relative to the vehicle coordinate system is obtained:

[0056] wherein, is a rotation matrix, is a translation vector; and the vehicle attitude information output by the RTK-GNSS system in the global coordinate system is represented as:

[0057] Finally, the global coordinate position of any detection point in the USS data can be obtained through the following transformation formula:

[0058] wherein, is the global coordinate position of the detection point .

[0059] It can be seen that the optional embodiment can also convert the coordinate system of the environment perception data to the global coordinate system through a plurality of predetermined transformation matrices and transformation formulas, thereby improving the coordinate transformation efficiency and transformation accuracy of the environment perception data, and thus facilitating more efficient and accurate analysis of the environment perception data and the global positioning data, and further facilitating improvement of the data fusion accuracy of the subsequent global positioning data and environment perception data.

[0060] In this optional embodiment, optionally, the current driving data can further include a current heading angle and a current acceleration.

[0061] In this optional embodiment, optionally, based on a pre-set data filtering fusion algorithm, the target environment perception data and the global positioning data are subjected to data filtering fusion processing to obtain the motion state data corresponding to the vehicle, which can include the following operations: determining the position coordinate information corresponding to the vehicle according to the global positioning data; determining the original state vector corresponding to the vehicle according to the position coordinate information, the current heading angle, the current vehicle speed and the current acceleration; performing state transition prediction on the original state vector based on a pre-determined vehicle motion model to obtain a predicted state vector; updating the predicted state vector based on a pre-determined observation model according to the target environment perception data and the global positioning data to obtain a target state vector; determining the target state vector as the motion state data corresponding to the vehicle.

[0062] Optionally, the motion state data can be calculated by the following formula: First, the motion state vector x is defined as:

[0063] wherein, represents the position coordinate information of the vehicle in the global coordinate system, θ represents the current heading angle (i.e., the angle between the vehicle heading and the positive east, and the unit is radian), v is the vehicle speed, and a is the acceleration.

[0064] Then, in the filtering process, it is divided into two stages of a prediction stage and an update stage: (1) Prediction stage: according to the vehicle motion model, the state vector is subjected to state transition prediction:

[0065] wherein, is the vehicle motion model, is the control input vector, is the process noise; Further, the control input vector may be as follows:

[0066] wherein, is the steering wheel angle (unit: radian), and a is the longitudinal acceleration input from the accelerator / brake.

[0067] Further optionally, the state transition equation in discrete form (i.e., the above vehicle motion model) can be as follows: wherein, the equation is in time step Prediction, the estimated value of the vehicle motion state at time k+1 can be calculated by the following formula:

[0068]

[0069]

[0070]

[0071] wherein, L is the wheelbase of the vehicle (unit: meters), and is the vehicle position coordinate at time k+1, is the heading angle at time k+1, is the vehicle speed at time k+1; and is the vehicle position coordinate at time k, is the heading angle at time k, is the vehicle speed at time k, is the steering wheel angle at time k.

[0072] In this way, by jointly optimizing with the USS ranging constraint (the actual distance between the vehicle and the obstacle), the state estimation can be further corrected, and the accuracy in a small scene can be improved.

[0073] (2) Update phase: combine target environment perception data and global positioning data, i.e., RTK data and USS observation data, to update the state:

[0074] wherein, is the observation model, is the observation noise.

[0075] After filtering and fusion processing, the fused and optimized vehicle accurate state estimation value can be obtained as the target state vector; and the relative spatial relationship model between the vehicle and the obstacle can be obtained, which can be used as the basis for generating a dynamic local map.

[0076] It can be seen that the optional embodiment can also determine the position coordinate information corresponding to the vehicle according to the global positioning data, and then determine the original state vector corresponding to the vehicle according to the position coordinate information and the current driving data of the vehicle, and then sequentially perform state transition prediction and update on the state vector based on the vehicle motion model and the observation model to obtain the target state vector, that is, the motion state data, which can improve the efficiency and accuracy of data filtering and data fusion of the global positioning data and the global positioning data, thereby facilitating to improve the analysis comprehensiveness and analysis accuracy of the vehicle position and driving condition, and further improving the determination accuracy of the vehicle motion state, and further facilitating to improve the analysis accuracy of the subsequent driving behavior.

[0077] Embodiment two Please refer to Figure 2 , Figure 2 is a flowchart of an intelligent driving teaching method based on USS and RTK data fusion disclosed by the embodiment of the application. Among them, Figure 2 The intelligent driving teaching method based on USS and RTK data fusion described can be applied to an intelligent driving teaching device based on USS and RTK data fusion. The intelligent driving teaching device can include one of an intelligent device, an intelligent terminal, an intelligent system and a server, wherein the server can include a local server or a cloud server, and the embodiments of the application are not limited; further, the device can be applied to a vehicle used for driving training, and can also be applied to a control system corresponding to the vehicle used for driving training, and the embodiments of the application are not limited. As shown in the figure, Figure 2 The intelligent driving teaching method based on USS and RTK data fusion can include the following operations: 201、In the process of driving training of the trainee, the environmental perception data corresponding to the vehicle driven by the trainee and the global positioning data corresponding to the vehicle are collected.

[0078] In the embodiment of the application, the environmental perception data includes USS data, and the global positioning data includes RTK data.

[0079] 202、According to the environmental perception data and the global positioning data, the dynamic local map of the environment where the vehicle is located and the motion state data corresponding to the vehicle are determined.

[0080] 203、According to the dynamic local map and the motion state data, a dynamic scene view is generated.

[0081] In the embodiment of the application, the dynamic scene view is used to assist the trainee to complete the driving training.

[0082] 204、Obtain the standard operation information corresponding to the driving training subject.

[0083] In the embodiment of the present application, the standard operation information includes standard operation rules, and the standard operation rules record operation safety value intervals corresponding to a plurality of driving operations, each operation safety value interval can include at least one operation safety value threshold; optionally, the standard operation information can also include safe driving rules, which are not limited in the embodiment of the present application.

[0084] In the embodiment of the present application, optionally, the driving training subject can be determined in the following manner: Before the student starts the driving training, the recommended training subject selected by the student from all recommended training subjects is obtained as the driving training subject; or, the driving training level of the student is obtained, and the driving training subject currently required by the student to be trained is determined according to the driving training level, which is not limited in the embodiment of the present application.

[0085] 205, collecting operation data of the student on the vehicle in the driving training.

[0086] In the embodiment of the present application, optionally, the operation data of the student on the vehicle in the driving training can be collected through the vehicle control interface unit; further optionally, the vehicle control interface unit can be connected with the CAN (Controller Area Network) bus of the vehicle, which is not limited in the embodiment of the present application. Optionally, the operation data of the vehicle can include one or a combination of a plurality of combinations of steering wheel angle, accelerator / brake pedal opening, gear, wheel speed, and can also include other data that can be obtained by the CAN bus, which is not limited in the embodiment of the present application.

[0087] 206, determining the driving behavior of the student according to the operation data, the dynamic local map and the motion state data.

[0088] 207, judging whether the driving behavior meets the pre-set standard operation condition and / or whether the driving behavior has a driving safety risk according to the standard operation information, the dynamic local map and the motion state data, to obtain a judgment result.

[0089] In the embodiment of the present application, the driving safety risk can include real-time safety risk and / or potential safety risk, which is not limited in the embodiment of the present application.

[0090] 208, determining the real-time analysis result of the driving behavior according to the judgment result.

[0091] 209, feeding back real-time driving guidance information for the driving behavior to the student according to the real-time analysis result of the driving behavior.

[0092] For further detailed descriptions of steps 201-203 and 209 in this embodiment of the invention, please refer to the detailed descriptions of steps 101-103 and 105 in Embodiment 1. These descriptions will not be repeated in this embodiment of the invention.

[0093] As can be seen, the method described in the embodiments of the present invention can collect environmental perception data, including USS data, and global positioning data, including RTK data, corresponding to the vehicle driven by the student during driving training. Combining the environmental perception data and global positioning data, a dynamic local map of the vehicle's environment and the vehicle's corresponding motion state data are determined. Then, based on the dynamic local map and motion state data, a dynamic scene view is generated, and the student's driving behavior is analyzed to obtain real-time driving behavior analysis results. Based on these results, corresponding real-time driving guidance information is provided to the student. Through deep data fusion of USS and RTK, accurate perception and high-precision positioning of the vehicle's own state and the complex environment at close range are achieved, constructing a powerful safety monitoring system for driving instruction. This facilitates the provision of timely, flexible, and accurate driving guidance based on the vehicle's motion state and surrounding environment, thereby improving the flexibility and accuracy of driving instruction, enhancing the safety and reliability of the driving instruction process, and ultimately improving the efficiency and effectiveness of driving training. Furthermore, it can combine the student's operational data, dynamic local maps, and motion status data during driving training to determine the student's driving behavior. Then, by combining the standard operating information, dynamic local maps, and motion status data corresponding to the driving training subject, it can determine whether the driving behavior complies with the operating specifications and / or whether there are any safety risks. Based on the judgment results, it can determine the real-time analysis results of driving behavior. It can achieve a comprehensive analysis of the student's operation, as well as the vehicle's driving environment and driving conditions, thereby improving the comprehensiveness, accuracy, and timeliness of the analysis of the student's driving behavior. This, in turn, improves the accuracy and timeliness of the assessment of the student's driving behavior, and ultimately helps to improve the flexibility and accuracy of guidance on driving behavior.

[0094] In an optional embodiment, the method may further include the following operations: Before the student begins driving training, load the static map data corresponding to the vehicle's location; the static map data includes a static global map and the scene configuration data corresponding to the static global map; the scene configuration data includes one or more of the following combinations: reference distance configuration data, coordinate configuration data, safety function configuration data, and subject location configuration data; Based on the scenario configuration data, configure a static global map to obtain the configured global map.

[0095] Optionally, the static map data can be matched with the driving training subject, and the static map data can be a high-precision map, and the embodiments of the present application are not limited thereto.

[0096] Optionally, the reference distance configuration data can include one or more combinations of USS safety distance configuration, buffer brake start distance, brake force and stroke distance, the coordinate configuration data can include base station coordinates and / or coordinates of other objects related to intelligent driving teaching, the safety function configuration data can include a safety function opening switch, and the subject position configuration data can include one or more combinations of warehouse positions for reversing into the warehouse, side parking positions, S-bend positions and other driving training subject corresponding site positions on the training site, and the embodiments of the present application are not limited thereto.

[0097] It should be noted that the dynamic local map is not a direct subset of the static global map, and the differences between the static global map and the dynamic local map in terms of data sources, generation methods, update frequencies and application purposes can be shown in Table 1 as follows:

[0098] Table 1 Therefore, the dynamic local map is an environment model constructed online based on real-time USS data and RTK data, which is used to describe the geometric structure and dynamic information of the current environment of the vehicle in real time, and is not a clipping or partitioning of the static global map.

[0099] It can be seen that the optional embodiment can load the static global map of the site where the vehicle is located before driving training and configure the corresponding scene configuration data to form a configured global map, which can improve the loading efficiency and configuration efficiency of the global map, thereby facilitating efficient construction of a static model of the training environment and efficient provision of an environmental reference benchmark for vehicle movement in the training site.

[0100] In the optional embodiment, optionally, generating a dynamic scene view according to the dynamic local map and the motion state data can include the following operations: According to the dynamic local map and the motion state data, determining real-time training scene information corresponding to the vehicle; the real-time training scene information includes one or more combinations of real-time scene identification data, real-time position data, virtual guide identification data, obstacle data and real-time boundary data; Superimposing and displaying the real-time training scene information on the configured global map as a dynamic scene view, so as to display the dynamic scene view on the man-machine interaction unit.

[0101] Optionally, the real-time scene identification data may include traffic signs, the real-time location data may include the real-time location coordinates and / or real-time driving direction of the vehicle, the obstacle data may include one or more of the obstacle location, obstacle boundary and obstacle outline, the virtual guide sign data may include virtual guide lines, and the real-time boundary data may include road boundaries. This embodiment of the invention does not impose any limitations.

[0102] As can be seen, this optional embodiment can combine dynamic local maps and vehicle motion state data to determine the corresponding real-time training scene information that needs to be displayed, thereby overlaying and displaying the real-time training scene information on the global map, so as to display a dynamic scene view to the trainee on the human-computer interaction unit. This can help trainees efficiently browse the environmental information of the current site and the guidance information for assisted driving, thereby guiding trainees to complete driving subject training in a more flexible and accurate manner.

[0103] In an optional embodiment, based on standard operating information, a dynamic local map, and motion state data, it is determined whether the driving behavior meets pre-set standard operating conditions, and / or whether the driving behavior poses a driving safety risk, and the determination result can include the following operations: Determine whether the numerical value of each driving operation in the driving behavior is within the safe operating value range corresponding to that driving operation in the standard operating rules; When it is determined that the operation value of each driving operation in the driving behavior is within the safe operation value range corresponding to that driving operation in the standard operation rules, it is determined that the driving behavior meets the pre-set standard operation conditions. When it is determined that the value of at least one driving operation in the driving behavior is not within the safe value range of the corresponding driving operation in the standard operating rules, it is determined that the driving behavior does not meet the pre-set standard operating conditions, and the type of wrong behavior corresponding to the driving behavior is determined. And / or, Based on safe driving rules, dynamic local maps, and motion status data, determine whether the distance between the vehicle and any obstacle is less than or equal to the preset safe distance, or whether the vehicle has crossed the line, or whether the vehicle is at risk of collision. When it is determined that the distance between the vehicle and any obstacle is less than or equal to the preset safe distance, or the vehicle is crossing the line, or the vehicle is at risk of collision, the driving behavior is determined to have a driving safety risk. When it is determined that the distance between the vehicle and any obstacle is less than or equal to the preset safe distance, or the vehicle is crossing the line, or there is a risk of collision, it is determined that there is no driving safety risk in the driving behavior.

[0104] Optionally, the erroneous behavior type may include one or more of the following: vehicle deviation from trajectory type, abnormal vehicle speed type, and steering error type. It may also include other behavior types that do not conform to driving standards. This embodiment of the invention does not limit the types of erroneous behavior.

[0105] As can be seen, this optional embodiment can determine whether driving behavior meets the preset standard operating conditions by judging whether the operation value of each driving operation in the driving behavior is within the operation safety value range corresponding to that driving operation in the standard operating rules, thereby improving the accuracy of judging whether driving behavior conforms to the standard; and it can also judge whether there is a driving safety risk by judging whether the distance between the vehicle and any obstacle is less than or equal to the preset safe distance, whether the vehicle crosses the line, and whether the vehicle has a collision risk, etc., which can improve the flexibility and accuracy of judging whether there is a risk in driving behavior, thereby helping to improve the accuracy of analysis and evaluation of the student's driving behavior.

[0106] In this optional embodiment, determining the real-time analysis result of driving behavior based on the judgment result may include the following operations: When the judgment result indicates that the driving behavior meets the pre-set standard operating conditions and there is no driving safety risk, the real-time analysis result of the driving behavior is determined to be that the vehicle is in a standard and safe driving state. When the judgment result indicates that the driving behavior does not meet the pre-set standard operating conditions, the real-time analysis result of the driving behavior is determined to be that the vehicle is in the non-standard driving state corresponding to the above-mentioned error behavior type. When the judgment result indicates that there is a driving safety risk in the driving behavior, the real-time analysis result of the driving behavior is determined to be that the vehicle is in a risky driving state.

[0107] As can be seen, this optional embodiment can also determine the corresponding real-time analysis results of driving behavior based on different judgment results regarding whether the student's driving behavior complies with regulations and whether there are safety risks. This can improve the accuracy and flexibility of driving behavior analysis, thereby improving the accuracy of determining the real-time analysis results of driving behavior. In turn, it can facilitate more targeted driving instruction guidance for students, thereby improving the accuracy of driving instruction guidance.

[0108] In this optional embodiment, the method may also include the following operations: When it is determined that a vehicle is at risk of collision, the risk level of the collision risk is determined. The test assesses whether trainees take appropriate emergency response actions for collision risks within the response time frame corresponding to the risk level. When the risk level is the highest risk level or no emergency response action is detected from the trainee within the response time range corresponding to the risk level, a warning signal corresponding to the highest risk level is fed back to the trainee, and auxiliary safety measures are triggered.

[0109] In this embodiment, the auxiliary safety measure is assisted braking, but this is not limited to that.

[0110] As can be seen, this optional embodiment can provide timely warnings of potential risks and provide auxiliary intervention when the highest collision risk is detected and the trainee fails to respond in a timely manner, thereby effectively avoiding training accidents and improving the safety and reliability of the driving training process.

[0111] In an optional embodiment, the method may further include the following operations: Based on operational data and real-time analysis of driving behavior, a corresponding driving training assessment result is generated for the trainee; the driving training assessment result includes quantitative scoring results and / or comprehensive assessment results corresponding to multiple assessment dimensions. Generate training analysis results based on the driving training evaluation results; the training analysis results include at least one of the following: training advantages analysis report, training error analysis report, and training improvement suggestions; Provide feedback to trainees on driving training assessment results and training analysis results.

[0112] Optionally, the evaluation dimensions may include one or more of the following: driving trajectory error, operational smoothness, steering angle change, safe distance maintained from obstacles, rule compliance, and reaction time. This embodiment of the invention does not limit these dimensions.

[0113] Optionally, during or after training, the human-computer interaction unit can provide the trainee with detailed driving training evaluation results and training analysis results, but this embodiment of the invention does not limit this.

[0114] As can be seen, this optional embodiment can generate corresponding driving training assessment results for trainees based on real-time analysis of operational data and driving behavior, and generate training analysis results based on the driving training assessment results. This allows for feedback of driving training assessment results and training analysis results to trainees, providing trainees with unified, objective, and quantifiable driving assessment standards. This ensures the stability and fairness of driving training quality, thereby facilitating more accurate assessment of trainees' driving behavior and improving the accuracy of driving training analysis. Consequently, it enables more targeted feedback of assessment results and improvement of analysis results to trainees, leading to more targeted guidance for driving training and improving the efficiency and effectiveness of driving training.

[0115] In an optional embodiment, the method may further include the following operations: Obtain the trainees' historical training data; Based on historical training data, the above-mentioned driving training evaluation results, and the above-mentioned training analysis results, the corresponding subject training information for trainees will be adjusted; among which, subject training information includes one or more combinations of subject training difficulty, subject training focus, and training content; In this optional embodiment, the method may also include the following operations: Based on all training data and evaluation results during the driving training process, a corresponding learning progress report is generated for each student.

[0116] As can be seen, this optional embodiment can dynamically adjust the driving training situation according to the actual driving level and learning progress of the trainees, realize individualized teaching, improve the matching degree between the driving subject training content and the trainees' learning situation, and thus provide more targeted guidance for trainees' driving training, thereby improving the efficiency and effectiveness of trainees' driving training.

[0117] In an optional embodiment, the method may further include the following operations: Optimize the driving instruction model based on all training data and evaluation results during the driving training process.

[0118] As can be seen, this optional embodiment can accumulate student training data, and through big data analysis, it can continuously optimize the design of teaching subjects, evaluation criteria and feedback strategies, thereby achieving continuous improvement in teaching effectiveness.

[0119] In an embodiment of the present invention, for example, the intelligent driving teaching method based on USS and RTK data fusion can be applied to an intelligent driving teaching system based on USS and RTK data fusion. This system may include an onboard perception unit, a central processing unit, a human-machine interaction unit, a vehicle control interface unit, and a data storage and analysis unit, wherein: The vehicle-mounted perception unit includes at least one ultrasonic sensor (USS) array, which is deployed around the vehicle body to detect obstacles near the vehicle in real time (e.g., within a range of 0.1 meters to 5 meters), including the distance and orientation of the obstacles; and a high-precision real-time dynamic differential positioning (RTK-GNSS) receiver to acquire the vehicle's centimeter-level three-dimensional position, velocity, heading, and attitude information in the global coordinate system. The central processing unit is electrically connected to the vehicle-mounted sensing unit and is configured to receive and process near-range obstacle data from the USS array and high-precision positioning and attitude data from the RTK-GNSS receiver; it runs built-in core algorithm modules, including USS and RTK data fusion algorithms, intelligent training scenario generation and management algorithms, real-time driving behavior monitoring and intelligent guidance algorithms, and comprehensive driving skill evaluation algorithms. The human-computer interaction unit is electrically connected to the central processing unit and includes an in-vehicle display screen and a voice interaction module. The in-vehicle display screen is used to visually display training instructions, vehicle status, virtual guidance information, obstacle warnings, real-time operation feedback and evaluation results to the trainee. The voice interaction module is used to broadcast voice instructions, prompts and evaluation feedback to the trainee, and can also receive voice instructions from the trainee. The vehicle control interface unit is electrically connected to the central processing unit and the vehicle's control system (such as the CAN bus) to read the vehicle's underlying status information (such as steering wheel angle, accelerator / brake pedal opening, gear position, wheel speed, etc.) and send instructions to the vehicle's auxiliary control system under specific conditions (such as safety warnings or auxiliary intervention teaching). The data storage and analysis unit is connected to the central processing unit and is used to store high-precision map data, a preset training subject library, trainees' personal files, detailed training process data and evaluation reports, and supports statistical analysis of training data to optimize teaching strategies.

[0120] Example 3 Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of an intelligent driving training device based on USS and RTK data fusion, as disclosed in an embodiment of the present invention. Figure 3 The described intelligent driving training device based on USS and RTK data fusion may include one of the following: intelligent device, intelligent terminal, intelligent system, and server. The server may include a local server or a cloud server; this embodiment of the invention is not limited thereto. Furthermore, the device can be applied to vehicles used for driving training, or to the control system corresponding to the aforementioned vehicles used for driving training; this embodiment of the invention is not limited thereto. Figure 3 As shown, the intelligent driving training device based on USS and RTK data fusion may include: The data acquisition module 301 is used to collect environmental perception data and global positioning data of the vehicle driven by the student during driving training. The environmental perception data includes USS data and the global positioning data includes RTK data. The determination module 302 is used to determine the dynamic local map of the environment in which the vehicle is located and the corresponding motion state data of the vehicle based on the environmental perception data and global positioning data. The generation module 303 is used to generate a dynamic scene view based on the dynamic local map and motion state data; the dynamic scene view is used to assist trainees in completing driving training. Analysis module 304 is used to analyze the driving behavior of trainees based on dynamic local maps and motion state data, and obtain real-time analysis results of driving behavior. Feedback module 305 is used to provide real-time driving guidance information to trainees based on the real-time analysis results of driving behavior.

[0121] As can be seen, the device described in the embodiments of the present invention can collect environmental perception data, including USS data, and global positioning data, including RTK data, corresponding to the vehicle driven by the student during driving training. Combining the environmental perception data and global positioning data, it determines the dynamic local map of the vehicle's environment and the vehicle's corresponding motion state data. Then, based on the dynamic local map and motion state data, it generates a dynamic scene view and analyzes the student's driving behavior to obtain real-time driving behavior analysis results. Based on these results, it provides corresponding real-time driving guidance information to the student. Through deep data fusion of USS and RTK, it achieves accurate perception and high-precision positioning of the vehicle's own state and the complex environment at close range, constructing a powerful safety monitoring system for driving instruction. This facilitates timely, flexible, and accurate driving guidance based on the vehicle's motion state and surrounding environment, thereby improving the flexibility and accuracy of driving instruction, enhancing the safety and reliability of the driving instruction process, and ultimately improving the efficiency and effectiveness of driving training.

[0122] In an optional embodiment, the specific method by which the determining module 302 determines the dynamic local map of the vehicle's environment and the vehicle's corresponding motion state data based on environmental perception data and global positioning data may include: Based on global positioning data, spatiotemporal synchronization processing is performed on environmental perception data to obtain synchronized environmental perception data. The synchronous environmental perception data is processed by coordinate transformation to convert the coordinate system corresponding to the synchronous environmental perception data from the local coordinate system corresponding to the synchronous environmental perception data to the global coordinate system corresponding to the global positioning data, so as to obtain the target environmental perception data. Based on a pre-defined data filtering and fusion algorithm, target environment perception data and global positioning data are processed by data filtering and fusion to obtain the vehicle's motion state data. Based on motion state data, a dynamic local map of the vehicle's environment is constructed. The dynamic local map is in the form of a grid map or a feature map, and is used to record the spatial location relationship and dynamic changes of the vehicle's current environment in real time.

[0123] As can be seen, the apparatus described in this optional embodiment can determine the vehicle's motion state and construct a corresponding dynamic local map by performing spatiotemporal synchronization, coordinate transformation, and filtering fusion on global positioning data and environmental perception data. This improves the processing efficiency and accuracy of global positioning data and environmental perception data, thereby enhancing the accuracy of data fusion between global positioning data and environmental perception data. Consequently, it improves the accuracy of determining the vehicle's motion state and the accuracy and real-time performance of constructing the dynamic local map. This, in turn, facilitates the provision of timely, flexible, and accurate driving guidance based on the vehicle's motion state and surrounding environment.

[0124] In this optional embodiment, optionally, the determining module 302 performs spatiotemporal synchronization processing on the environmental perception data based on global positioning data to obtain a specific method for synchronizing the environmental perception data, which may include: Determine the first timestamp of the global positioning data and the second timestamp of the environmental perception data; Calculate the time difference between the first timestamp and the second timestamp; Determine the current driving data for the vehicle; the current driving data includes the current vehicle speed and current driving direction. Based on the determined extrapolation correction formula, the environmental perception data is corrected according to the current vehicle speed, current driving direction and time difference to obtain synchronous environmental perception data.

[0125] As can be seen, the apparatus described in this optional embodiment can also perform time correction on the environmental perception data based on the time difference between the first timestamp of the global positioning data and the second timestamp of the environmental perception data, as well as the current driving data, to obtain synchronized environmental perception data, so as to achieve spatiotemporal synchronization between global positioning data and environmental perception data, improve the efficiency and accuracy of data spatiotemporal synchronization, and thus help improve the accuracy of subsequent data fusion of global positioning data and environmental perception data.

[0126] In this optional embodiment, the current driving data may optionally include the current heading angle and the current acceleration; The specific method by which the determining module 302 performs data filtering and fusion processing on the target environment perception data and global positioning data based on a pre-set data filtering and fusion algorithm to obtain the vehicle's corresponding motion state data may include: Based on global positioning data, determine the vehicle's location coordinates. Based on the position coordinates, current heading angle, current vehicle speed, and current acceleration, determine the original state vector corresponding to the vehicle; Based on a predetermined vehicle motion model, state transition prediction is performed on the original state vector to obtain the predicted state vector. Based on a predetermined observation model, the predicted state vector is updated according to the target environment perception data and global positioning data to obtain the target state vector; The target state vector is defined as the motion state data corresponding to the vehicle.

[0127] As can be seen, the apparatus described in this optional embodiment can also determine the vehicle's position coordinates based on global positioning data, and then determine the vehicle's original state vector based on the position coordinates and the vehicle's current driving data. Then, based on the vehicle motion model and the observation model, the state vector is predicted and updated to obtain the target state vector, i.e., the motion state data. This can improve the efficiency and accuracy of data filtering and data fusion for global positioning data, thereby improving the comprehensiveness and accuracy of the analysis of vehicle position and driving conditions, and further improving the accuracy of determining the vehicle's motion state, which in turn improves the accuracy of subsequent analysis of driving behavior.

[0128] In an optional embodiment, such as Figure 4 As shown, the device may further include: The loading module 306 is used to load the static map data corresponding to the vehicle's location before the student begins driving training; the static map data includes a static global map and scene configuration data corresponding to the static global map; the scene configuration data includes one or more combinations of reference distance configuration data, coordinate configuration data, safety function configuration data, and subject location configuration data; Configuration module 307 is used to configure a static global map based on scene configuration data, and obtain a configured global map. The specific method by which the generation module 303 generates a dynamic scene view based on the dynamic local map and motion state data may include: Based on dynamic local maps and motion status data, determine the real-time training scenario information corresponding to the vehicle; the real-time training scenario information includes one or more combinations of real-time scenario identification data, real-time location data, virtual guidance sign data, obstacle data, and real-time boundary data; Real-time training scene information is overlaid on the global map as a dynamic scene view, which is then displayed on the human-computer interaction unit.

[0129] As can be seen, the device described in this optional embodiment can load a static global map of the vehicle's location and configure corresponding scene configuration data before driving training to form a configured global map. This improves the loading and configuration efficiency of the global map, thereby facilitating the efficient construction of a static model of the training environment and providing an environmental reference benchmark for the vehicle's movement in the training area. Furthermore, it can combine dynamic local maps and vehicle motion state data to determine the corresponding real-time training scene information to be displayed, thereby overlaying the real-time training scene information onto the configured global map. This allows for the display of a dynamic scene view to the trainee on the human-computer interaction unit, assisting the trainee in efficiently browsing the current environment information and driving assistance guidance information. This enables more flexible and accurate guidance for the trainee to complete driving training in a standardized manner.

[0130] In an optional embodiment, the specific method by which the analysis module 304 analyzes the student's driving behavior based on the dynamic local map and motion state data to obtain real-time driving behavior analysis results may include: Obtain standard operating information corresponding to driving training subjects; standard operating information includes standard operating rules, which record the operating safety value ranges for several driving operations. Collect driver operation data related to the vehicle during driving training; Based on operational data, dynamic local maps, and motion status data, determine the student's driving behavior; Based on standard operating information, dynamic local maps, and motion status data, determine whether the driving behavior meets the pre-set standard operating conditions, and / or determine whether the driving behavior poses a driving safety risk, and obtain the judgment result; Based on the judgment results, the real-time analysis results of driving behavior are determined.

[0131] As can be seen, the device described in this optional embodiment can combine the student's operation data, dynamic local map, and motion state data during driving training to determine the student's driving behavior. It then combines this data with the standard operation information, dynamic local map, and motion state data corresponding to the driving training subject to determine whether the driving behavior conforms to operating procedures and / or whether there are safety risks. Based on the judgment results, it determines the real-time analysis results of the driving behavior. This enables a comprehensive analysis of the student's operation, vehicle driving environment, and vehicle driving conditions, thereby improving the comprehensiveness, accuracy, and timeliness of the analysis of the student's driving behavior. This, in turn, improves the accuracy and timeliness of the assessment of the student's driving behavior, ultimately enhancing the flexibility and accuracy of guidance for driving behavior.

[0132] In this optional embodiment, the generation module 303 is optionally further configured to generate driving training evaluation results corresponding to the trainee based on the operation data and real-time analysis results of driving behavior; the driving training evaluation results include quantitative scoring results and / or comprehensive evaluation results corresponding to multiple evaluation dimensions; The generation module 303 is also used to generate training analysis results based on the driving training evaluation results; the training analysis results include at least one of a training advantages analysis report, a training error point analysis report, and training improvement suggestions; Feedback module 305 is also used to provide feedback to trainees on driving training assessment results and training analysis results.

[0133] As can be seen, the device described in this optional embodiment can also combine dynamic local maps and vehicle motion state data to determine the corresponding real-time training scene information that needs to be displayed, thereby overlaying the real-time training scene information on the global map and displaying a dynamic scene view to the trainee on the human-computer interaction unit. This can help the trainee efficiently browse the environmental information of the current site and the guidance information for assisted driving, thereby guiding the trainee to complete the driving subject training in a more flexible and accurate manner.

[0134] Example 4 Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another intelligent driving teaching device based on USS and RTK data fusion disclosed in an embodiment of the present invention. Figure 5 As shown, the intelligent driving training device based on USS and RTK data fusion may include: Memory 401 storing executable program code; Processor 402 coupled to memory 401; The processor 402 calls the executable program code stored in the memory 401 to execute some or all of the steps in the intelligent driving teaching method based on USS and RTK data fusion described in Embodiment 1 or Embodiment 2 of the present invention.

[0135] Example 5 This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute some or all of the steps in the intelligent driving teaching method based on USS and RTK data fusion described in Embodiment 1 or Embodiment 2 of this invention.

[0136] Example 6 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform some or all of the steps in the intelligent driving teaching method based on USS and RTK data fusion described in Embodiment 1 or Embodiment 2.

[0137] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0138] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0139] Finally, it should be noted that the intelligent driving teaching method and device based on USS and RTK data fusion disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for teaching intelligent driving based on the fusion of USS and RTK data, characterized in that, The method includes: During the driving training process, environmental perception data and global positioning data corresponding to the vehicle driven by the student are collected; wherein, the environmental perception data includes USS data and the global positioning data includes RTK data. Based on the environmental perception data and the global positioning data, a dynamic local map of the environment in which the vehicle is located and the corresponding motion state data of the vehicle are determined. A dynamic scene view is generated based on the dynamic local map and the motion state data; the dynamic scene view is used to assist the trainee in completing the driving training. Based on the dynamic local map and the motion state data, the driving behavior of the trainee is analyzed to obtain real-time analysis results of driving behavior; Based on the real-time analysis results of the driving behavior, real-time driving guidance information is provided to the trainee regarding the driving behavior.

2. The intelligent driving teaching method based on USS and RTK data fusion according to claim 1, characterized in that, The step of determining the dynamic local map of the vehicle's environment and the vehicle's corresponding motion state data based on the environmental perception data and the global positioning data includes: Based on the global positioning data, the environmental perception data is spatiotemporally synchronized to obtain synchronized environmental perception data. The synchronous environment perception data is subjected to coordinate transformation processing to convert the coordinate system corresponding to the synchronous environment perception data from the local coordinate system corresponding to the synchronous environment perception data to the global coordinate system corresponding to the global positioning data, so as to obtain the target environment perception data. Based on a pre-defined data filtering and fusion algorithm, the target environment perception data and the global positioning data are subjected to data filtering and fusion processing to obtain the motion state data corresponding to the vehicle. Based on the motion state data, a dynamic local map of the vehicle's environment is constructed; The dynamic local map is in the form of a grid map or a feature map, and is used to record the spatial location relationship and dynamic changes of the vehicle's current environment in real time.

3. The intelligent driving teaching method based on USS and RTK data fusion according to claim 2, characterized in that, The step of performing spatiotemporal synchronization processing on the environmental perception data based on the global positioning data to obtain synchronized environmental perception data includes: Determine the first timestamp of the global positioning data and the second timestamp of the environmental perception data; Calculate the time difference between the first timestamp and the second timestamp; Determine the current driving data corresponding to the vehicle; the current driving data includes the current vehicle speed and current driving direction; Based on the determined extrapolation correction formula, the environmental perception data is corrected according to the current vehicle speed, the current driving direction, and the time difference to obtain synchronized environmental perception data.

4. The intelligent driving teaching method based on USS and RTK data fusion according to claim 3, characterized in that, The current driving data also includes the current heading angle and current acceleration; The step of performing data filtering and fusion processing on the target environment perception data and the global positioning data based on a pre-defined data filtering and fusion algorithm to obtain the motion state data corresponding to the vehicle includes: Based on the global positioning data, determine the location coordinates of the vehicle. Based on the position coordinates, the current heading angle, the current vehicle speed, and the current acceleration, determine the original state vector corresponding to the vehicle; Based on a predetermined vehicle motion model, state transition prediction is performed on the original state vector to obtain a predicted state vector. Based on a predetermined observation model, the predicted state vector is updated according to the target environment perception data and the global positioning data to obtain the target state vector; The target state vector is determined as the motion state data corresponding to the vehicle.

5. The intelligent driving teaching method based on USS and RTK data fusion according to any one of claims 1-4, characterized in that, The method further includes: Before the student begins driving training, static map data corresponding to the site where the vehicle is located is loaded; the static map data includes a static global map and scene configuration data corresponding to the static global map; the scene configuration data includes one or more of the following combinations: baseline distance configuration data, coordinate configuration data, safety function configuration data, and subject location configuration data. Based on the scenario configuration data, configure the static global map to obtain the configured global map; The step of generating a dynamic scene view based on the dynamic local map and the motion state data includes: Based on the dynamic local map and the motion state data, the real-time training scene information corresponding to the vehicle is determined; the real-time training scene information includes one or more combinations of real-time scene identification data, real-time location data, virtual guidance identification data, obstacle data, and real-time boundary data. The real-time training scene information is overlaid on the global map as a dynamic scene view, which is then displayed on the human-computer interaction unit.

6. The intelligent driving teaching method based on USS and RTK data fusion according to any one of claims 1-4, characterized in that, The step of analyzing the student's driving behavior based on the dynamic local map and the motion state data to obtain real-time driving behavior analysis results includes: Obtain standard operating information corresponding to driving training subjects; the standard operating information includes standard operating rules, and the standard operating rules record the operating safety value ranges corresponding to several driving operations. Collect the driving data of the trainee regarding the vehicle during the driving training; The driving behavior of the student is determined based on the operation data, the dynamic local map, and the motion state data. Based on the standard operating information, the dynamic local map, and the motion state data, determine whether the driving behavior meets the preset standard operating conditions, and / or determine whether the driving behavior poses a driving safety risk, and obtain the judgment result; Based on the judgment results, the real-time analysis results of driving behavior are determined.

7. The intelligent driving teaching method based on USS and RTK data fusion according to claim 6, characterized in that, The method further includes: Based on the operational data and the real-time analysis results of the driving behavior, a driving training assessment result corresponding to the trainee is generated; the driving training assessment result includes quantitative scoring results and / or comprehensive assessment results corresponding to multiple assessment dimensions. Generate training analysis results based on the driving training evaluation results; the training analysis results include at least one of a training advantages analysis report, a training error point analysis report, and training improvement suggestions; The driving training assessment results and training analysis results are fed back to the trainee.

8. An intelligent driving training device based on USS and RTK data fusion, characterized in that, The device includes: The data acquisition module is used to collect environmental perception data and global positioning data of the vehicle driven by the student during driving training; wherein, the environmental perception data includes USS data and the global positioning data includes RTK data. The determination module is used to determine the dynamic local map of the environment in which the vehicle is located and the motion state data corresponding to the vehicle based on the environmental perception data and the global positioning data. The generation module is used to generate a dynamic scene view based on the dynamic local map and the motion state data; the dynamic scene view is used to assist the trainee in completing the driving training. The analysis module is used to analyze the driving behavior of the trainee based on the dynamic local map and the motion state data, and obtain real-time analysis results of the driving behavior. The feedback module is used to provide real-time driving guidance information to the trainee based on the real-time analysis results of the driving behavior.

9. An intelligent driving training device based on USS and RTK data fusion, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the intelligent driving teaching method based on USS and RTK data fusion as described in any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the intelligent driving teaching method based on USS and RTK data fusion as described in any one of claims 1-7.

Images