Efficient on-site district training method for obtaining off-site road driving skills

Abstract

The invention belongs to the technical field of motor vehicle driving training, and particularly relates to a high-efficiency in-site district training method for acquiring out-of-site road driving skills, which comprises the following steps of: acquiring multi-source signals, positioning a vehicle by a site base station and a vehicle-mounted radar, acquiring a state by an OBD (On-Board Diagnostic), updating skill data of trainees in real time, and importing configurable site parameters; dynamically planning an intra-field training area by fusing subject complexity, student mastering degree and field adaptation degree based on a task pushing algorithm; and vehicle-mounted voice teaching is triggered through matching of the vehicle position and the preset skill point location, and the skill model is iteratively updated in real time based on the training data. According to the invention, efficient simulation of an off-site road scene by an in-site environment is realized, traditional training limitations are broken through from the dimensions of safety controllability, skill pertinence, site utilization rate and the like, driving training is promoted to be upgraded to a data-driven intelligent mode, and a systematic solution is provided for cost reduction and efficiency improvement of driving schools and skill improvement of students.

Description

Technical Field

[0001] This invention belongs to the field of motor vehicle driver training technology, specifically relating to an efficient on-site training method for acquiring off-site road driving skills, and is particularly applicable to on-site targeted training of off-site driving skills for driving school subject three. Background Technology

[0002] With the increase in vehicle ownership per capita, the basic qualities of drivers are receiving more and more attention, and the diversity of traffic conditions places higher demands on the comprehensiveness and rigor of drivers' skills. Among these, road driving skills in Subject 3 are an important component of essential driver abilities. However, the road driving training environment in most driving schools is currently significantly limited, making it difficult to provide comprehensive simulated scenarios, which prevents students from fully mastering the necessary off-site road driving skills.

[0003] Current driving school solutions primarily rely on experienced instructors to guide students through simulated driving scenarios required for the third stage of the driving test on public roads or designated training areas. These scenarios include traffic lights, overtaking, meeting oncoming traffic, and handling rain or fog. However, this approach has significant drawbacks: firstly, public roads have high traffic volume and complex traffic conditions, resulting in poor safety and controllability; secondly, within the prescribed training hours, limitations imposed by the training environment and the subjectivity of human instruction prevent comprehensive and targeted training to address students' skill weaknesses, leading to low training efficiency and incomplete skill coverage.

[0004] Therefore, there is an urgent need for a training method that can efficiently simulate off-road driving environments within a limited on-site area and impart skills in a targeted manner, in order to solve the problems of site limitations and insufficient training targeting in existing technologies. Summary of the Invention

[0005] In view of the limitations of existing technologies, this invention proposes an efficient on-site training method for acquiring off-site driving skills. This method involves setting up GPS base stations and deploying vehicle-mounted equipment to acquire vehicle location and status signals within a limited on-site area. Combined with a task-pushing algorithm, and based on multiple dimensions such as site length, subject complexity weights, and trainees' mastery of each subject, the method dynamically plans subject areas and generates corresponding teaching content. This fully utilizes site resources to simulate various off-site road environments required for driving test subject three in stages, achieving targeted teaching within a specified training time. This overcomes site limitations, improves training efficiency, ensures training safety, and enables trainees to efficiently master off-site driving skills.

[0006] To achieve the above objectives, this invention provides an efficient on-site training method for acquiring off-road driving skills, comprising the following steps:

[0007] (1) Signal acquisition and input:

[0008] ① The vehicle's position relative to the site is obtained through the collaboration of the on-site base station and the vehicle radar, and the vehicle status signals, including vehicle speed, braking status and steering status, are collected through the on-board OBD device.

[0009] ② Collect real-time updated data on students' driving skills, including average mastery of each subject and qualification status of sub-items;

[0010] ③ Import configurable on-site parameters, including site perimeter, subject simulation length threshold, and subject complexity weight;

[0011] (2) Algorithm decision:

[0012] The algorithm inputs location information, vehicle status, student scores, and on-site parameters into the task push algorithm. Based on multiple dimensions such as subject complexity, student mastery, and site adaptability, it calculates the priority of subject recommendations and the priority of teaching content.

[0013] (3) Regional dynamic planning:

[0014] Based on priority results, plan the simulation area for the next subject within the limited space; if the total required length of the subject exceeds the perimeter of the space, reuse the space in segments to ensure that the subject and the area are compatible.

[0015] (4) Teaching triggers and outputs:

[0016] When the vehicle enters the simulation area and its real-time location matches the preset "teachable skill location" for the subject, the on-board equipment outputs teaching content dynamically generated based on the student's mastery of the sub-item.

[0017] (5) Iterative update:

[0018] Real-time collection of vehicle operation data and student practice results, updating of mastery model, and repeating steps (2)-(4) until the required training hours are completed.

[0019] Furthermore, in step (1), the vehicle's position information relative to the site is obtained collaboratively by the site base station and the vehicle-mounted radar.

[0020] Configurable parameters include: drivable perimeter of the course, subject simulation length threshold, and subject complexity weight;

[0021] The student's driving ability performance data is updated in real time, covering the mastery of the subject level and the pass status of sub-items.

[0022] Furthermore, in step (2), the formula for calculating the priority of subject recommendations is:

[0023]

[0024] in:

[0025] c i The subject complexity weights imported in step (1);

[0026] S i The average subject-level mastery level (0-1) of students is updated in real time in step (1);

[0027] R i The formula for calculating site suitability is:

[0028]

[0029] L is the perimeter of the site, L i | represents the simulated length of subject i;

[0030] αβγ are adjustable weighting coefficients.

[0031] Furthermore, in step (2), the formula for calculating the priority of teaching content is:

[0032]

[0033] in:

[0034] M ij The mastery level (0-1) is mapped to the qualified status of the student's sub-item operation in step (1);

[0035] F j The basic importance weight (preset value) of teaching content j;

[0036] δ∈ is an adjustable weight coefficient.

[0037] Furthermore, in step (3), the regional planning rules are as follows:

[0038] Prioritize selection based on lower complexity weights (C i Small), low student mastery (S) i Small size, high site adaptability (R) i Large-scale subjects; when the total required length of a subject exceeds the perimeter of the site, the track is reused in segments according to the planned sequence to adapt to the limited site.

[0039] Furthermore, in step (4):

[0040] The timing for imparting driving skills is determined by comparing the real-time vehicle position with the preset "teachable skill location information" for the subject (the preset location is the subject teaching trigger coordinates);

[0041] The teaching content is output via voice from the in-vehicle computer, based on the student's sub-item mastery (M). ij The depth of the explanation can be dynamically adjusted.

[0042] The technical solution of the present invention brings at least the following significant advantages:

[0043] 1. This invention constructs a controllable simulated environment within a closed training area, replacing the training scenario on public roads. This completely avoids safety hazards such as interference from public vehicles and uncontrollable traffic flow, achieving a fully enclosed and monitored training process. Compared to traditional public road training, this method reduces the risk of accidents to near zero, creating "undisturbed and traceable" safe practice conditions for trainees.

[0044] 2. This invention relies on multi-source real-time data acquisition from "site base stations, vehicle-mounted radar, and OBD," combined with multi-dimensional analysis of the task push algorithm (integrating subject complexity weights, student sub-item mastery, and site suitability):

[0045] It can identify both "subject-level weaknesses" (such as the "overtaking" subject, which is generally weak among trainees) and "sub-item operation defects" (such as "turn signals not returning to their original position in time when overtaking").

[0046] By dynamically generating customized teaching content (such as enhanced guidance for incorrect turn signal operation), the limitations of human instructors in "vague experience judgment and incomplete guidance coverage" are overcome, and "targeted repair of skill gaps" is achieved.

[0047] The algorithm-driven automated planning in this invention: The task push algorithm automatically calculates the subject priority and recommends subjects with "high complexity, low mastery, and good site suitability", replacing manual scheduling and compressing subject planning time from "minutes" to "seconds".

[0048] Breakthrough in time and space utilization of venues: By using a segmented and cyclical venue utilization mechanism (such as training in segments according to the planned sequence when the total length of a subject exceeds the perimeter of the venue), the "time utilization rate" of limited venues is improved. Traditional single-subject training requires exclusive use of the venue, but this method can carry out multi-subject training in the same venue in cycles, ensuring that more skill items are covered within the prescribed study time.

[0049] In summary, this invention, through technological innovation, systematically solves the core pain points of traditional driver training, namely "high safety risks, insufficient targeting, low efficiency, and limited resources." While improving the efficiency of students' skill mastery, it provides a technical path for driving schools to reduce costs and increase efficiency, and has significant economic value and potential for industry transformation. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of the efficient on-site training method of the present invention;

[0052] Figure 2 This is a flowchart illustrating the specific implementation method of the efficient on-site training method of the present invention. Detailed Implementation

[0053] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0054] Example 1

[0055] like Figure 1 As shown, this embodiment provides an efficient on-site training method for acquiring off-site road driving skills, including the following steps:

[0056] (1) Signal acquisition and input:

[0057] ① The vehicle's position relative to the site is obtained through the collaboration of the on-site base station and the vehicle radar, and the vehicle status signals, including vehicle speed, braking status and steering status, are collected through the on-board OBD device.

[0058] ② Collect real-time updated data on students' driving skills, including average mastery of each subject and qualification status of sub-items;

[0059] ③ Import configurable on-site parameters, including site perimeter, subject simulation length threshold, and subject complexity weight;

[0060] (2) Algorithm decision:

[0061] The algorithm inputs location information, vehicle status, student scores, and on-site parameters into the task push algorithm. Based on multiple dimensions such as subject complexity, student mastery, and site adaptability, it calculates the priority of subject recommendations and the priority of teaching content.

[0062] (3) Regional dynamic planning:

[0063] Based on priority results, plan the simulation area for the next subject within the limited space; if the total required length of the subject exceeds the perimeter of the space, reuse the space in segments to ensure that the subject and the area are compatible.

[0064] (4) Teaching triggers and outputs:

[0065] When the vehicle enters the simulation area and its real-time location matches the preset "teachable skill location" for the subject, the on-board equipment outputs teaching content dynamically generated based on the student's mastery of the sub-item.

[0066] (5) Iterative update:

[0067] Real-time collection of vehicle operation data and student practice results, updating of mastery model, and repeating steps (2)-(4) until the required training hours are completed.

[0068] As one implementation method, in step (1) of this embodiment, the vehicle's position information relative to the site is obtained collaboratively by the site base station and the vehicle-mounted radar.

[0069] Configurable parameters include: drivable perimeter of the course, subject simulation length threshold, and subject complexity weight;

[0070] The student's driving ability performance data is updated in real time, covering the mastery of the subject level and the pass status of sub-items.

[0071] Furthermore, in step (2), the formula for calculating the priority of subject recommendations is:

[0072]

[0073] in:

[0074] c i The subject complexity weights imported in step (1);

[0075] S i The average subject-level mastery level (0-1) of students is updated in real time in step (1);

[0076] R i The formula for calculating site suitability is:

[0077]

[0078] L is the perimeter of the site, L i | represents the simulated length of subject i;

[0079] αβγ are adjustable weighting coefficients.

[0080] As one implementation method, in step (2) of this embodiment, the formula for calculating the priority of teaching content is:

[0081]

[0082] in:

[0083] M ij The mastery level (0-1) is mapped to the qualified status of the student's sub-item operation in step (1);

[0084] F j The basic importance weight (preset value) of teaching content j;

[0085] δ∈ is an adjustable weight coefficient.

[0086] As one implementation method, in step (3) of this embodiment, the regional planning rule is as follows:

[0087] Prioritize selection based on lower complexity weights (C i Small), low student mastery (S) i Small size, high site adaptability (R) i Large-scale subjects; when the total required length of a subject exceeds the perimeter of the site, the track is reused in segments according to the planned sequence to adapt to the limited site.

[0088] As one implementation method, in step (4) of this embodiment:

[0089] The timing for imparting driving skills is determined by comparing the real-time vehicle position with the preset "teachable skill location information" for the subject (the preset location is the subject teaching trigger coordinates);

[0090] The teaching content is output via voice from the in-vehicle computer, based on the student's sub-item mastery (M). ij The depth of the explanation can be dynamically adjusted.

[0091] This embodiment utilizes a complete process of "signal acquisition and input → algorithm decision → regional dynamic planning → teaching triggering and output → iterative update," relying on multi-source data acquisition (site base station, vehicle radar, OBD, and student scores), multi-dimensional priority calculation of task push algorithms (subject complexity, student mastery, and site adaptability), combined with a site segmented cyclical reuse mechanism and targeted teaching content output. This achieves efficient simulation of the off-site road driving environment required for Subject 3 within a limited site area. It can accurately match students' skill deficiencies and maximize the use of site resources, ensuring that comprehensive and targeted driving skills training is completed within the prescribed training hours.

[0092] Example 2

[0093] The current driving school solution involves human instructors guiding students through simulated driving scenarios required for the third stage of the driving test, based on their experience and by issuing commands to simulate various road conditions, such as traffic lights, overtaking maneuvers, meeting oncoming traffic, and rain / fog. However, the off-site driving roads used for the third stage of the driving test at driving schools are currently congested with other vehicles, resulting in complex traffic conditions. Within the existing prescribed training hours, it is impossible to provide comprehensive and targeted skills instruction in such an environment.

[0094] This embodiment obtains GPS information by setting up a GPS base station on the training ground and arranging equipment connected to the base station inside the vehicle. An onboard computer is installed in the vehicle and connected to the onboard OBD to obtain vehicle status signals. Through the above, the vehicle's position and status signals in the training ground are obtained as a whole. More importantly, an algorithm is designed to fully utilize the training ground and roads within the prescribed training hours, combining the signals obtained above, to reasonably simulate all the required training environment of Subject 3 in stages, and to impart the corresponding road driving operation skills appropriately according to the students' practice.

[0095] Based on the students' mastery of the skills required for different subjects, combined with the road environment and the varying complexity of the subjects themselves, the system dynamically generates corresponding subject simulations and provides appropriate teaching recommendations.

[0096] like Figure 2 The diagram illustrates the overall process by which the onboard computer in a driving school vehicle outputs driving skill voice content to the student at an appropriate time. The execution flow of this method is implemented through modular signal transfer. The correspondence and logic between each module and the steps in the embodiment are as follows:

[0097] 1. Signal Acquisition Module

[0098] Site base station + vehicle radar: The site base station constructs the spatial coordinate system within the site, and the vehicle radar senses the displacement data of the vehicle relative to the base station in real time. The two work together to generate "vehicle relative to site position information" (corresponding to the position acquisition logic of step (1) ① in the embodiment, to achieve centimeter-level accurate positioning of the vehicle within the site).

[0099] Onboard OBD signal: Directly collects vehicle speed, braking status, steering status and other operating data, and outputs "vehicle status information" (corresponding to the status collection logic of step (1) ① in the embodiment, providing real-time basis for judging the standardization of driving operation).

[0100] 2. Dynamic Data Module

[0101] The student's driving ability performance data for each subject: record the student's average mastery of the subject (such as the overall score of the "overtaking" subject), the qualification status of sub-item operation (such as whether the "turn signal operation when overtaking" is qualified), and continuously update it as the training process (corresponding to the real-time characteristics of step (1) ② in the embodiment, supporting the algorithm to dynamically identify the student's skill shortcomings).

[0102] Configurable inputs (circumference of the track, length required for each subject, relative weight of subject complexity, etc.): provide preset parameters for the algorithm.

[0103] Site perimeter, subject simulation length → site adaptability in support step (2)

[0104]

[0105] The calculation (determining the matching degree between the subject and the venue);

[0106] Subject complexity weight → Subject recommendation priority in supporting step (2)

[0107]

[0108] Calculation of c i To preset the complexity weight, the parameter import logic corresponds to step (1) ③ of the embodiment).

[0109] 3. Algorithm Decision Module (Task Push Algorithm + Generation of Next Subject)

[0110] Task push algorithm: Receives a continuously updated signal source consisting of "vehicle location, vehicle status, student performance, and configurable parameters," and calculates priority based on three-dimensional conditions.

[0111] Subject complexity (preset weight c) i ), student mastery level (real-time data S) i ), Site suitability (calculated using the formula R) i (The priority calculation logic corresponding to step (2) in the embodiment);

[0112] Output subject recommendation priority (which subject to practice first) and teaching content priority (which sub-item to teach first within a subject day).

[0113] Generating the next subject: Based on the priority results, the algorithm dynamically plans the next training subject and its corresponding simulation area within a limited space.

[0114] If the total required length of the subject area is less than or equal to the perimeter of the site: directly allocate a continuous area;

[0115] If the total required length of the subject is greater than the perimeter of the site: trigger the segmented cyclic reuse mechanism (e.g., split the "overtaking" subject into 3 segments, and train 3 times in a 1km site, corresponding to the regional planning logic of step (3) in the implementation example).

[0116] 4. Teaching Trigger Module (Real-time Updated Stage Teaching + Diamond-shaped Decision Box)

[0117] Real-time updated stage-based teaching: Subject-specific teaching content that carries the algorithm's output, including preset values ​​for "location information of teachable skills" (e.g., "in the overtaking subject, 'turn signal operation' needs to be explained 200 meters from the starting point"), and updated according to the student's mastery level M. ij Dynamically adjust the teaching depth (corresponding to the content generation logic of step (4) in the embodiment).

[0118] Whether the timing for driving skill transfer is met: via "real-time vehicle location (such as GPS positioning)". Subject preset skill position (such as 200-meter marker)" comparison and judgment (corresponding to the trigger condition logic of step (4) in the embodiment):

[0119] If the following conditions are met: the onboard computer is triggered to output the skill operation content (such as "Please turn on the left turn signal 3 seconds in advance when overtaking"), and a single teaching session is completed;

[0120] If not satisfied: loop waiting for position matching (e.g., if the vehicle has not reached the 200-meter marker, continue to monitor the position), supporting the iterative update logic of step (5) of the implementation example (ensuring that the teaching timing and scenario are accurately synchronized).

[0121] Through the above-mentioned full-process mapping of "hardware signal → dynamic data → algorithm decision → teaching trigger", the attached figure fully presents the technical architecture of "data-driven, closed-loop iteration" of this method, which corresponds one by one with the steps of the embodiment, and clearly explains the implementation path of "efficiently simulating the off-site driving environment in a limited field area".

[0122] The technical solution of this invention is mainly embodied in the "task push algorithm". Task push mainly consists of dynamically generating a subject database location and dynamically pushing a specific teaching content into that database location.

[0123] Given the following conditions: 1. The length L of a lap around the driving school track; 2. There are 16 subjects in total for the third stage of the driving test, each with a different required length, and the total length of the 16 subjects is much greater than L. However, the track length can be reused. The next subject is dynamically recommended based on a comprehensive analysis of multiple factors; 3. Different subjects have different complexity weights, with lower complexity weights being prioritized; 4. Students' mastery of different subjects and their scores are assessed, with lower scores receiving higher weights; 5. Within each subject, operational items and scores are assessed, with failures triggering notifications. By fully utilizing the track's principles, the subject areas are dynamically planned. Upon completion of a subject, the next subject area is dynamically identified, and the corresponding skills are taught.

[0124] {

[0125] A multi-dimensional dynamic recommendation algorithm. The following are the algorithm formulas and implementation ideas:

[0126] 1. Subject recommendation priority formula: P_i=(α*(1-C_i)+β*(1-S_i)+γ*R_i) / (α+β+γ)

[0127] in:

[0128] P_i: Recommendation priority score for subject i

[0129] C_i: Complexity weight of subject i (0-1 standardized)

[0130] S_i: Students' average mastery of subject i (0-1)

[0131] R_i: Driving school simulation test track length adaptation degree = 1 - |L - L_i| / max(L,L_i)

[0132] α, β, γ: Weighting coefficients for each dimension (adjustable)

[0133] 2. Formula for prioritizing teaching content: T_ij=(δ*(1-M_ij)+ε*F_j) / (δ+ε)

[0134] in:

[0135] T_ij: Teaching priority of teaching content j in subject i

[0136] M_ij: Student's mastery of content j in subject i (0-1)

[0137] F_j: The fundamental importance weight of content j

[0138] δ, ε: Weighting coefficients

[0139] Algorithm implementation steps:

[0140] 1. Calculate the P_i value for all available subjects.

[0141] 2. Select the subject with the highest P_i as the recommended subject.

[0142] 3. Within the selected subjects, calculate the T_ij value for all teaching content.

[0143] 4. Arrange the teaching content in descending order of T_ij.

[0144] 5. Select the combination of teaching content based on the remaining length of the track.

[0145] {

[0146] The following code:

[0147] import numpyasnp

[0148] classProjectRecommender:

[0149] def__init__(self,projects,track_length):

[0150] self.projects = projects#project list

[0151] self.L = track_length # track length

[0152] self.alpha = 0.4 # Complexity weight

[0153] self.beta = 0.4 #Mastery weight

[0154] self.gamma = 0.2 # Length adaptation weight

[0155] defcalculate_priority(self,project):

[0156] #Calculate the priority of a single project

[0157] complexity=project['complexity']

[0158] mastery=np.mean(project['student_scores'])

[0159] length_fit=1-abs(self.L-project['length']) / max(self.L,project['length'])

[0160] priority=(self.alpha*(1-complexity)+

[0161] self.beta*(1-mastery)+

[0162] self.gamma*length_fit)

[0163] returnpriority

[0164] defrecommend_project(self):

[0165] #Recommended Best Projects

[0166] priorities=[self.calculate_priority(p)forpinself.projects]

[0167] best_idx=np.argmax(priorities)

[0168] return self.projects[best_idx]

[0169] defrecommend_content(self,project,delta=0.7,epsilon=0.3):

[0170] #Recommended Teaching Content

[0171] contents=project['contents']

[0172] priorities=[]

[0173] forcontentincontents:

[0174] mastery=content['student_score']

[0175] importance=content['importance']

[0176] priority=(delta*(1-mastery)+epsilon*importance)

[0177] priorities.append((priority,content))

[0178] # Sort by priority in descending order

[0179] priorities.sort(reverse=True,key=lambdax:x[0])

[0180] return[item[1]foriteminpriorities]

[0181] }

[0182] This algorithm implements a dynamic recommendation system based on multiple dimensions, taking into account factors such as project complexity, student mastery, and track suitability. When using it, you need to provide subject data and track length, and it will return a ranking of the most suitable projects and teaching content.

[0183] }

[0184] Within the prescribed driving school hours and with limited space, nearly twenty subjects are categorized according to their complexity and weighted accordingly. Since each subject requires a different teaching area, a comprehensive analysis is conducted based on complexity weights and the student's success rate with previously practiced subjects to plan the next subject to be practiced, and the required area for each subject is planned in real-time. The system recognizes the subject when the vehicle enters this area.

[0185] In this embodiment, the algorithm dynamically generates the required subject area in a small training area and recommends it to trainees for instruction. Training in a small training area achieves the same effect as training driving skills on public roads, while saving trainees' practice time. At different times, based on the training environment, the trainee's ability in each subject, and the subject's characteristic weights, the algorithm dynamically generates the currently required practice subject and corresponding simulated parking spaces, and provides appropriate guidance. This fully utilizes the small training area to simulate public roads.

[0186] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A highly efficient on-site training method for acquiring off-road driving skills, characterized in that, Includes the following steps: (1) Signal acquisition and input: ① The vehicle's position relative to the site is obtained through the collaboration of the on-site base station and the vehicle radar, and the vehicle status signals, including vehicle speed, braking status and steering status, are collected through the on-board OBD device. ② Collect real-time updated data on students' driving skills, including average mastery of each subject and qualification status of sub-items; ③ Import configurable on-site parameters, including site perimeter, subject simulation length threshold, and subject complexity weight; (2) Algorithm decision: The algorithm inputs location information, vehicle status, student scores, and on-site parameters into the task push algorithm. Based on multiple dimensions such as subject complexity, student mastery, and site adaptability, it calculates the priority of subject recommendations and the priority of teaching content. (3) Regional dynamic planning: Based on priority results, plan the simulation area for the next subject within the limited space; if the total required length of the subject exceeds the perimeter of the space, reuse the space in segments to ensure that the subject and the area are compatible. (4) Teaching triggers and outputs: When the vehicle enters the simulation area and its real-time location matches the preset "teachable skill location" for the subject, the on-board equipment outputs teaching content that is dynamically generated based on the student's mastery of the sub-item. (5) Iterative update: Real-time collection of vehicle operation data and student practice results, updating of mastery model, and repeating steps (2)-(4) until the required training hours are completed.

2. The method according to claim 1, characterized in that: In step (1), the vehicle's position information relative to the site is obtained collaboratively by the site base station and the vehicle-mounted radar. Configurable parameters include: drivable perimeter of the course, subject simulation length threshold, and subject complexity weight; The student's driving ability performance data is updated in real time, covering the mastery of the subject level and the pass status of sub-items.

3. The method according to claim 1, characterized in that, In step (2), the formula for calculating the priority of subject recommendations is: in: c i The subject complexity weights imported in step (1); S i The average subject-level mastery level (0-1) of students is updated in real time in step (1); R i The formula for calculating site suitability is: L is the perimeter of the site, L i | represents the simulated length of subject i; αβγ are adjustable weighting coefficients.

4. The method according to claim 1, characterized in that, In step (2), the formula for calculating the priority of teaching content is: in: M ij The mastery level (0-1) is mapped to the qualified status of the student's sub-item operation in step (1); F j The basic importance weight (preset value) of teaching content j; δ∈ is an adjustable weight coefficient.

5. The method according to claim 1, characterized in that, In step (3), the regional planning rules are as follows: Prioritize selection based on lower complexity weights (C i Small), low student mastery (S) i Small size, high site adaptability (R) i Large-scale subjects; when the total required length of a subject exceeds the perimeter of the site, the track is reused in segments according to the planned sequence to adapt to the limited site.

6. The method according to claim 1, characterized in that, In step (4): The timing for imparting driving skills is determined by comparing the real-time vehicle position with the preset "teachable skill location information" for the subject (the preset location is the subject teaching trigger coordinates); The teaching content is output via voice from the in-vehicle computer, based on the student's sub-item mastery (M). ij The depth of the explanation can be dynamically adjusted.

Images