Intelligent vehicle navigation system and method

Abstract

The present invention discloses an intelligent vehicle navigation system and method, and relates to the field of intelligent navigation. The intelligent vehicle navigation system comprises: a control module, which is used to access the vehicle system of a target vehicle, obtain access rights, and edit and send control instructions of various functional modules; a route planning module, which is used to receive a destination input by a user, call a map service and real-time traffic information to calculate a path, and output a driving route; a data acquisition module, which is used to acquire real-time environmental images outside the vehicle and associated environmental images of other navigation node vehicles in the same driving route; by acquiring external environmental images and associated environmental images in real time, the system can obtain more comprehensive road information, allow environmental data sharing between vehicles, enhance perception and understanding of road conditions, predict and identify abnormal trends, and warn of potential dangers in advance, so as to help the driver take measures in advance and flexibly adjust the driving strategy.

Description

Technical Field

[0001] The present invention relates to the field of intelligent navigation technology, and in particular to an intelligent vehicle navigation system and method. Background Art

[0002] With the development of satellite communication and intelligent analysis technology, intelligent navigation systems have begun to enter users' lives. Intelligent vehicle navigation systems are an important part of intelligent transportation systems, which have promoted the intelligence and automation of traffic management and helped to achieve efficient management of urban traffic. By analyzing real-time traffic data and user needs, intelligent navigation systems can provide users with the shortest or optimal driving routes, saving time and fuel and improving travel efficiency. With the rise of autonomous driving and shared travel, the application of intelligent navigation systems will become more extensive and become one of the important infrastructures for future travel. However, existing in-vehicle navigation systems can usually only provide navigation guidance along a predetermined route, and are unable to proactively identify abnormal road conditions and make corresponding adjustments. They lack the ability to identify abnormal situations and make adaptive adjustments, and usually only focus on the situation of a single vehicle. It is difficult to utilize environmental information collected by surrounding vehicles, and lacks sharing and coordination of environmental information between vehicles, making the information acquisition surface narrow. Once the existing system determines the driving route, it is difficult to dynamically adjust and optimize the strategy. Summary of the invention

[0003] 1. Technical issues to be resolved In view of the above-mentioned shortcomings of the prior art, the present invention provides an intelligent vehicle navigation system and method, which can effectively solve the problems of the prior art.

[0004] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: The present invention discloses an intelligent vehicle navigation system, comprising: The control module is used to access the vehicle system of the target vehicle, obtain access rights, and edit and send control instructions for each functional module; The route planning module is used to receive the destination input by the user, call the map service and real-time traffic information to calculate the path, and output the driving route; The navigation guidance module is used to update the driving direction and road conditions in real time according to the driving route planned by the route planning module, and provide voice and visual instructions; A data acquisition module is used to collect real-time environmental images outside the vehicle and associated environmental images of other navigation node vehicles on the same driving route; The model building module is used to build a prediction and recognition model through a machine learning algorithm, input the collected real-time environmental images and associated environmental images, output the current abnormal trend and future abnormal trend, and generate an abnormal event report; The anomaly identification module is used to set the security anomaly threshold, classify the identified anomaly event reports, conduct evaluation based on the security anomaly threshold, and compare the evaluation results with the security anomaly threshold; The strategy adjustment module is used to determine whether the driving strategy needs to be adjusted based on the comparison results of the abnormal recognition module. When the adjustment strategy is executed, the label is extracted from the abnormal event report and matched with the preset label strategy library. According to the matching result, the vehicle driving strategy is adjusted; The adjustment database is used to provide preset editing of driving strategy adjustment plans, conduct usability evaluation on applied adjustment plans based on preset evaluation criteria, and synchronously store the evaluated adjustment plans and their corresponding abnormal data as reference items.

[0005] Furthermore, the data acquisition module is deployed with submodules at the lower level, and the submodules include: an environment acquisition module, a shared acquisition module and an integration module, and the environment acquisition module, the shared acquisition module and the integration module are interactively connected through a wireless network, wherein: The environment acquisition module, as an image acquisition component, receives acquisition instructions, acquires the image of the current vehicle's surrounding environment, and pre-processes the image; The shared acquisition module is used to upload the images collected by the environment acquisition module to the cloud database for sharing, and to open up the analysis permissions for the image data collected by other vehicles within a preset range; The integration module is used to integrate the collected data of the environment collection module and the shared data of the shared collection module associated with the environment collection module, and submit them to the model construction module.

[0006] Furthermore, during the judgment stage, if the comparison result of the anomaly identification module exceeds the safety anomaly threshold, the strategy adjustment module determines that the adjustment strategy needs to be executed and jumps to the adjustment database for further operation. If the comparison result of the anomaly identification module does not exceed the safety anomaly threshold, it determines that the adjustment strategy does not need to be executed and jumps to the data acquisition module for further operation.

[0007] Furthermore, the abnormal identification module is interactively connected to a marking module via a wireless network. The marking module is used to receive a comparison result between an evaluation result of the abnormal identification module and a safety abnormality threshold. When the comparison result exceeds the threshold, it is triggered to encrypt and back up the collected image data for storage, and mark the abnormal image area.

[0008] Furthermore, the abnormality identification module is interactively connected to a feedback module via a wireless network, and the feedback module is used to synchronously feed back the comparison result of the abnormality identification module to the data acquisition module and the marking module.

[0009] Furthermore, the management and control module is interactively connected to the route planning module through a wireless network, the route planning module is interactively connected to the navigation guidance module through a wireless network, the navigation guidance module is interactively connected to the data acquisition module through a wireless network, the data acquisition module is interactively connected to the model building module through a wireless network, the anomaly identification module is interactively connected to the policy adjustment module through a wireless network, and the policy adjustment module is interactively connected to the adjustment database through a wireless network.

[0010] The intelligent vehicle navigation method comprises the following steps: Step 1: Receive the destination information input by the user, verify the validity of the destination information, plan the route based on the valid destination information, and calculate the best driving route using map data and real-time traffic information; Step 2: Confirm the planned route and pass it to the navigation application, and update the driving direction and road condition information in real time according to the planned route; Step 3: Collect the real-time environment image outside the vehicle and the associated environment images of other vehicles on the same driving route; Step 4: Build a prediction and recognition model through machine learning algorithms, input the collected real-time environmental images and associated environmental images, output current and future abnormal trends, and generate abnormal event reports; Step 5: Set the security anomaly threshold, classify the identified abnormal event reports, evaluate them based on the security anomaly threshold, and compare them with the threshold; Step 6: Based on the comparison result of step 5, determine whether the driving strategy needs to be adjusted. If it exceeds the safety abnormality threshold, it is determined that adjustment needs to be performed; Step 7: Provide preset editing of the driving strategy adjustment library, assign several labels to the edited content, extract labels from the currently submitted abnormal event report, and match them with the preset labeled strategy adjustment library to adjust the vehicle driving strategy. If the threshold is not exceeded, continue to step 4; Step 8: Conduct availability evaluation on the applied adjustment plan, and synchronously store the evaluation results and their corresponding abnormal data as reference items.

[0011] Furthermore, the operation logic of the prediction and recognition model outputting the current abnormal trend is: ; In the formula, represents the current abnormal trend at time t, n represents the number of features that affect the abnormal trend, a represents the activation function, represents the model weight, represents several characteristics that affect the abnormal trend at time t, represents the jth weight coefficient associated with historical abnormal data, represents the jth historical abnormal event data set, and m represents the number of historical abnormal data.

[0012] Furthermore, the several features that affect the abnormal trend in the prediction and recognition model include: a current image feature vector, an environment state vector, a current abnormal state vector, and an abnormal average value at several past time points.

[0013] Furthermore, in the process of outputting future abnormal trends by the prediction and recognition model, the prediction function based on the current and historical trends is recursively calculated to predict the abnormal trend at a certain moment in the future.

[0014] (III) Beneficial effects Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: By collecting external environment images and related environment images in real time, the system can obtain more comprehensive road information, allow environmental data sharing between vehicles, and conduct collaborative analysis by sharing real-time data of the surrounding environment, thereby enhancing collective intelligence and data analysis capabilities, and being able to better respond to complex traffic conditions, improve overall traffic safety, and enhance perception and understanding of road conditions.

[0015] By analyzing real-time environmental images and related images, generating abnormal event reports, classifying and evaluating identified abnormal events, ensuring driving safety, predicting and identifying abnormal trends, and providing early warnings of potential dangers, the system helps drivers take measures in advance. It can flexibly adjust driving strategies to deal with emergencies, and ensure driving safety and improve driving experience by dynamically adjusting driving strategies.

[0016] When an abnormal situation occurs, the system can automatically encrypt and back up image data to ensure data security and user privacy. It provides a preset driving strategy adjustment plan editing function and performs usability evaluation to make driving strategies more flexible and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of the framework of the intelligent vehicle navigation system of the present invention; Figure 2 It is a schematic diagram of the framework of the data acquisition module in the present invention.

[0019] The numbers in the figure represent: 1. Management and control module; 2. Route planning module; 3. Navigation guidance module; 4. Data collection module; 41. Environment collection module; 42. Shared collection module; 43. Integration module; 5. Model building module; 6. Anomaly identification module; 7. Strategy adjustment module; 8. Adjustment database; 9. Marking module; 10. Feedback module. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] The present invention will be further described below in conjunction with the embodiments.

[0022] Example 1 The intelligent vehicle navigation system of this embodiment is as follows: Figure 1 As shown, including: Control module 1 is used to access the vehicle system of the target vehicle, obtain access rights, and edit and send control instructions for each functional module; Route planning module 2, used to receive the destination input by the user, call the map service and real-time traffic information to calculate the path, and output the driving route; The navigation guidance module 3 is used to update the driving direction and road conditions in real time according to the driving route planned by the route planning module 2, and provide voice and visual instructions; The data acquisition module 4 is used to collect real-time environment images outside the vehicle and the associated environment images of other navigation node vehicles in the same driving route; the data acquisition module 4 is deployed with submodules, including: environment acquisition module 41, shared acquisition module 42 and integration module 43, the environment acquisition module 41, shared acquisition module 42 and integration module 43 are interactively connected through a wireless network, wherein: The environment acquisition module 41, as an image acquisition component, receives acquisition instructions, acquires the current vehicle surrounding environment image, and pre-processes the image; The shared acquisition module 42 is used to upload the images collected by the environment acquisition module 41 to the cloud database for sharing, and to open the analysis authority of the image data collected by other vehicles within a preset range; The integration module 43 is used to integrate the collected data of the environment collection module 41 and the shared data of the shared collection module 42 associated with the environment collection module 41, and submit them to the model building module 5; thereby realizing the sharing and analysis of the environment image and enhancing the perception ability of the overall road condition; The model building module 5 is used to build a prediction and recognition model through a machine learning algorithm, input the collected real-time environmental image and the associated environmental image, output the current abnormal trend and the future abnormal trend, and generate an abnormal event report; The abnormality identification module 6 is used to set the safety abnormality threshold, classify the identified abnormal event reports, evaluate based on the safety abnormality threshold, and compare the evaluation result with the safety abnormality threshold; the abnormality identification module 6 is interactively connected to the marking module 9 through a wireless network, and the marking module 9 is used to receive the comparison result of the evaluation result of the abnormality identification module 6 and the safety abnormality threshold, and is triggered when the comparison result exceeds the threshold, and the collected image data is encrypted and backed up for storage, and the abnormal image area is marked; the abnormality identification module 6 is interactively connected to the feedback module 10 through a wireless network, and the feedback module 10 is used to synchronously feed back the comparison result of the abnormality identification module 6 to the data acquisition module 4 and the marking module 9; The strategy adjustment module 7 is used to determine whether the driving strategy needs to be adjusted according to the comparison result of the abnormal identification module 6. When executing the adjustment strategy, the label is extracted from the abnormal event report and matched with the preset label strategy library. According to the matching result, the vehicle driving strategy is adjusted. In the judgment stage, if the comparison result of the abnormal identification module 6 exceeds the safety abnormality threshold, the strategy adjustment module 7 determines that the adjustment strategy needs to be executed and jumps to the adjustment database 8 for further operation. If the comparison result of the abnormal identification module 6 does not exceed the safety abnormality threshold, it is determined that the adjustment strategy does not need to be executed, and jumps to the data acquisition module 4 for further operation. By matching with the preset label strategy library, the strategy adjustment module 7 can automatically select the optimal driving strategy according to the actual situation, reduce manual intervention, and improve the intelligent level of decision-making; The adjustment database 8 is used to provide preset editing of driving strategy adjustment schemes, conduct usability evaluation on applied adjustment schemes based on preset evaluation criteria, and synchronously store the evaluated adjustment schemes and their corresponding abnormal data as reference items; allowing users to adjust driving strategies according to real-time environment and driving conditions, ensuring that vehicles can flexibly respond to various emergencies in a dynamic traffic environment, and by conducting usability evaluation on applied adjustment schemes, it can be ensured that the adopted strategies are effective, thereby improving driving safety and efficiency. By synchronously storing abnormal data as reference items, historical data can be used to make more accurate decisions, reducing the possibility of human errors; like Figure 1As shown, the management and control module 1 is interactively connected to the route planning module 2 through a wireless network, the route planning module 2 is interactively connected to the navigation guidance module 3 through a wireless network, the navigation guidance module 3 is interactively connected to the data acquisition module 4 through a wireless network, the data acquisition module 4 is interactively connected to the model building module 5 through a wireless network, the anomaly identification module 6 is interactively connected to the policy adjustment module 7 through a wireless network, and the policy adjustment module 7 is interactively connected to the adjustment database 8 through a wireless network.

[0023] In the specific implementation of this embodiment, the control module 1 controls the global function module, the route planning module 2 performs route planning by inputting the destination, and the navigation guidance module 3 continues to perform navigation after confirming the route. The data acquisition module 4 collects the environment image outside the vehicle and the abnormal data of other navigation node vehicles in the same route within a preset range, and the abnormal prediction model is constructed by the model construction module 5. Based on the environment image collected by the current vehicle and the abnormal data of other vehicles, the model outputs the abnormal change trend of the destination roadside; The abnormal identification module 6 determines whether the adjustment should be triggered, the feedback module 10 provides feedback on each judgment information, the marking module 9 encrypts and stores the abnormal image and marks it, and the strategy adjustment module 7 extracts labels based on the current abnormal data and the corresponding abnormal change trend. The extracted labels are matched in the abnormal identification module 6 of the labeled storage adjustment strategy, and the associated strategy is applied.

[0024] Example 2 In other aspects, this embodiment also provides an intelligent vehicle navigation method, comprising the following steps: Step 1: Receive the destination information input by the user, verify the validity of the destination information, plan the route based on the valid destination information, and calculate the best driving route using map data and real-time traffic information; Step 2: Confirm the planned route and pass it to the navigation application, and update the driving direction and road condition information in real time according to the planned route; Step 3: Collect the real-time environment image outside the vehicle and the associated environment images of other vehicles on the same driving route; Step 4: Build a prediction and recognition model through machine learning algorithms, input the collected real-time environmental images and associated environmental images, output current and future abnormal trends, and generate abnormal event reports; Step 5: Set the security anomaly threshold, classify the identified abnormal event reports, evaluate them based on the security anomaly threshold, and compare them with the threshold; Step 6: Based on the comparison result of step 5, determine whether the driving strategy needs to be adjusted. If it exceeds the safety abnormality threshold, it is determined that adjustment needs to be performed; Step 7: Provide preset editing of the driving strategy adjustment library, assign several labels to the edited content, extract labels from the currently submitted abnormal event report, and match them with the preset labeled strategy adjustment library to adjust the vehicle driving strategy. If the threshold is not exceeded, continue to step 4; Step 8: Conduct availability evaluation on the applied adjustment plan, and synchronously store the evaluation results and their corresponding abnormal data as reference items.

[0025] Compared with existing technologies, it can provide more accurate and dynamic route planning and navigation by combining map data, real-time traffic information and images of the vehicle's external environment. By analyzing real-time environmental images through machine learning algorithms, it can identify potential safety abnormal trends in advance and enhance driving safety. Based on the comparison of abnormal event reports with safety thresholds, it can intelligently determine whether the driving strategy needs to be adjusted, thereby improving the vehicle's ability to respond to emergencies. Through the preset labeled strategy adjustment library, the driving strategy can be adaptively adjusted to respond to different driving environments and conditions more flexibly. The adjustment plan can be evaluated and the results can be stored to form a closed-loop feedback mechanism, which helps to continuously optimize navigation and driving strategies, set safety anomaly thresholds, and help to respond to and deal with safety hazards in a timely manner, thereby reducing the risk of traffic accidents.

[0026] Example 3 In this embodiment, a calculation process for predicting the current and future abnormal trends of the recognition model output is provided, which is specifically: ; In the formula, represents the current abnormal trend at time t, n represents the number of features that affect the abnormal trend, and a represents the activation function, which is used to convert the model output into a probability value, indicating the possibility of abnormality. Represents the model weight, which is used to measure the importance of each feature for abnormal trend prediction. represents several characteristics that affect the abnormal trend at time t, represents the jth weight coefficient associated with historical abnormal data. These weights are used to adjust the impact of historical abnormal events on the current abnormal trend prediction. represents the jth historical abnormal event data set, m represents the number of historical abnormal data, and the total number of historical abnormal events considered when calculating the impact on the current abnormal trend; Several features that affect the abnormal trend include: current image feature vector, environment state vector, current abnormal state vector and the average abnormality value at several past time points; In the process of predicting and identifying the future abnormal trend output by the prediction model, the prediction function based on the current and historical trends is recursively calculated to predict the abnormal trend at a certain moment in the future; In this way, the present embodiment can integrate current environmental information and historical abnormal data to accurately predict current and future abnormal trends so as to adjust the driving strategy in a timely manner.

[0027] In summary, the present invention enhances the vehicle-mounted system's ability to identify and respond to abnormal situations through real-time environmental perception, abnormal trend prediction and automatic adjustment of driving strategies, focuses on sharing and coordinating environmental information between vehicles, and realizes sharing and analysis of environmental images through information collection and interaction between vehicles, thereby enhancing the ability to perceive the overall road conditions. Through a preset strategy library, abnormal data storage and availability evaluation, continuous optimization of driving strategies is achieved, thereby improving overall safety and efficiency.

[0028] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will 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. Intelligent vehicle navigation system, characterized in that, include: A control module (1) is used to access the vehicle computer system of the target vehicle, obtain access rights, and edit and send control instructions for each functional module; The route planning module (2) is used to receive the destination input by the user, call the map service and real-time traffic information to calculate the path, and output the driving route; A navigation guidance module (3) is used to update the driving direction and road conditions in real time according to the driving route planned by the route planning module (2), and to provide voice and visual instructions; A data acquisition module (4) is used to acquire real-time environmental images outside the vehicle and associated environmental images of other navigation node vehicles within the same driving route; A model building module (5) is used to build a prediction and recognition model through a machine learning algorithm, input the collected real-time environmental image and the associated environmental image, output the current abnormal trend and the future abnormal trend, and generate an abnormal event report; An anomaly identification module (6) is used to set a security anomaly threshold, classify the identified anomaly event report, perform an evaluation based on the security anomaly threshold, and compare the evaluation result with the security anomaly threshold; A strategy adjustment module (7) is used to determine whether the driving strategy needs to be adjusted based on the comparison result of the abnormality identification module (6); when the strategy is adjusted, a label is extracted from the abnormal event report and matched with a preset label strategy library; and the vehicle driving strategy is adjusted based on the matching result; The adjustment database (8) is used to provide preset editing of driving strategy adjustment schemes, conduct usability evaluation on applied adjustment schemes based on preset evaluation criteria, and synchronously store the evaluated adjustment schemes and their corresponding abnormal data as reference items.

2. The intelligent vehicle navigation system according to claim 1, characterized in that: The data acquisition module (4) has submodules deployed at the lower level, and the submodules include: an environment acquisition module (41), a shared acquisition module (42) and an integration module (43). The environment acquisition module (41), the shared acquisition module (42) and the integration module (43) are interactively connected via a wireless network, wherein: The environment acquisition module (41) serves as an image acquisition component, receives an acquisition instruction, acquires an image of the current vehicle surrounding environment, and pre-processes the image; A shared acquisition module (42) is used to upload the images collected by the environment acquisition module (41) to a cloud database for sharing, and to open up analysis permissions for image data collected by other vehicles within a preset range; The integration module (43) is used to integrate the collected data of the environment collection module (41) and the shared data of the shared collection module (42) associated with the environment collection module (41), and submit them to the model construction module (5).

3. The intelligent vehicle navigation system according to claim 1, characterized in that: During the judgment phase, if the comparison result of the anomaly identification module (6) exceeds the safety anomaly threshold, the strategy adjustment module (7) determines that the adjustment strategy needs to be executed and jumps to the adjustment database (8) for further operation; if the comparison result of the anomaly identification module (6) does not exceed the safety anomaly threshold, the strategy adjustment module (7) determines that the adjustment strategy does not need to be executed and jumps to the data acquisition module (4) for further operation.

4. The intelligent vehicle navigation system according to claim 1, characterized in that: The abnormality identification module (6) is interactively connected to a marking module (9) via a wireless network. The marking module (9) is used to receive a comparison result between an evaluation result of the abnormality identification module (6) and a safety abnormality threshold value, and is triggered when the comparison result exceeds the threshold value, to encrypt and back up the collected image data for storage, and to mark the abnormal image area.

5. The intelligent vehicle navigation system according to claim 1, characterized in that: The abnormality identification module (6) is interactively connected to a feedback module (10) via a wireless network, and the feedback module (10) is used to synchronously feed back the comparison result of the abnormality identification module (6) to the data acquisition module (4) and the marking module (9).

6. The intelligent vehicle navigation system according to claim 1, characterized in that: The control module (1) is interactively connected to the route planning module (2) via a wireless network, the route planning module (2) is interactively connected to the navigation guidance module (3) via a wireless network, the navigation guidance module (3) is interactively connected to the data acquisition module (4) via a wireless network, the data acquisition module (4) is interactively connected to the model building module (5) via a wireless network, the anomaly identification module (6) is interactively connected to the strategy adjustment module (7) via a wireless network, and the strategy adjustment module (7) is interactively connected to the adjustment database (8) via a wireless network.

7. An intelligent vehicle navigation method, the method being an implementation method of an intelligent vehicle navigation system according to any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: Receive the destination information input by the user, verify the validity of the destination information, plan the route based on the valid destination information, and calculate the best driving route using map data and real-time traffic information; Step 2: Confirm the planned route and pass it to the navigation application, and update the driving direction and road condition information in real time according to the planned route; Step 3: Collect the real-time environment image outside the vehicle and the associated environment images of other vehicles on the same driving route; Step 4: Build a prediction and recognition model through machine learning algorithms, input the collected real-time environmental images and associated environmental images, output current and future abnormal trends, and generate abnormal event reports; Step 5: Set the security anomaly threshold, classify the identified abnormal event reports, evaluate them based on the security anomaly threshold, and compare them with the threshold; Step 6: Based on the comparison result of step 5, determine whether the driving strategy needs to be adjusted. If it exceeds the safety abnormality threshold, it is determined that adjustment needs to be performed; Step 7: Provide preset editing of the driving strategy adjustment library, assign several labels to the edited content, extract labels from the currently submitted abnormal event report, and match them with the preset labeled strategy adjustment library to adjust the vehicle driving strategy. If the threshold is not exceeded, continue to step 4; Step 8: Conduct availability evaluation on the applied adjustment plan, and synchronously store the evaluation results and their corresponding abnormal data as reference items.

8. The intelligent vehicle navigation method according to claim 7, characterized in that: The operation logic of the prediction and recognition model outputting the current abnormal trend is: ； In the formula, represents the current abnormal trend at time t, n represents the number of features that affect the abnormal trend, and a represents the activation function, which is used to convert the model output into a probability value, indicating the possibility of abnormality. Represents the model weight, which is used to measure the importance of each feature for abnormal trend prediction. represents several characteristics that affect the abnormal trend at time t, represents the jth weight coefficient associated with historical abnormal data, represents the jth historical abnormal event data set, m represents the number of historical abnormal data, and the total number of historical abnormal events considered when calculating the impact on the current abnormal trend.

9. The intelligent vehicle navigation method according to claim 8, characterized in that: The several features that affect the abnormal trend in the prediction and recognition model include: a current image feature vector, an environment state vector, a current abnormal state vector, and an abnormal average value at several past time points.

10. The intelligent vehicle navigation method according to claim 7, characterized in that: In the process of outputting future abnormal trends by the prediction recognition model, the prediction function based on the current and historical trends is recursively calculated to predict the abnormal trend at a certain moment in the future.

Images