An autonomous vehicle health data management method and system

Abstract

The present application relates to the technical field of vehicle data management, and particularly relates to a kind of automatic driving vehicle health data management method and system, method includes: obtaining the multi-source health data of automatic driving vehicle, forms operation health data set;According to the continuous state of vehicle operation link, it is judged whether vehicle has the condition of starting;According to the state propagation relationship of health abnormal event in corresponding vehicle operation link, it is judged whether health abnormality affects actual vehicle operation influence, and determine the affected operation link;According to the residual control link of vehicle determined by the affected operation link, and determine the safety control mode of vehicle;According to the control response state in safety control process, it is judged whether safety control result meets the requirements, to update the residual control link and safety control mode, generate vehicle operation health management result.The present application improves the driving safety of automatic driving vehicle.

Description

Technical Field

[0001] This invention relates to the field of vehicle data management technology, and in particular to a method and system for managing health data of autonomous vehicles. Background Technology

[0002] In existing technologies, autonomous vehicle health management typically involves separately detecting and processing vehicle inspection data, onboard function status, and subsystem warning events. Departure restrictions or safety controls are then implemented directly based on individual detection results or warning events. This approach lacks comprehensive analysis of the operational dependencies between various onboard functions and the propagation process of abnormal states. It struggles to determine whether an anomaly actually disrupts the vehicle's operational chain and cannot dynamically determine executable safety control methods based on the vehicle's remaining control capabilities after an anomaly occurs. Furthermore, the lack of a mechanism to verify the effectiveness of safety controls based on actual control responses after execution leads to issues such as accidental control triggering due to momentary anomalies, inaccurate assessment of anomaly impacts, and mismatches between control methods and the vehicle's actual available control capabilities. These factors negatively impact the reliability and operational safety of autonomous vehicle health management.

[0003] Chinese Patent Publication No. CN122126297A discloses an autonomous driving health assessment system, comprising: a first detection module, disposed within a chassis domain controller, for generating first-type health status information based on reported signals from vehicle actuators; a second detection module, disposed within the chassis domain controller, for generating second-type health status information based on real-time information of durability parameters of vehicle actuators; a third detection module, disposed within an autonomous driving controller, for generating third-type health status information based on the actual response information of the autonomous driving system to driving control information; and an assessment module, disposed within the autonomous driving controller, communicatively connected to the first, second, and third detection modules, for generating a health status assessment result of the autonomous driving system based on at least one of the first, second, and third-type health status information.

[0004] Therefore, the aforementioned autonomous driving health assessment system suffers from the problem that, after generating health status information based on the signals reported by the actuators, durability parameters, and driving control responses for comprehensive assessment, it lacks correlation analysis of the operational dependencies between various vehicle functions and the propagation process of abnormal states along the vehicle's operating link. This makes it difficult to determine whether local anomalies actually affect vehicle operation and whether the vehicle still possesses control capabilities after an anomaly occurs, resulting in a discrepancy between the health status assessment results and the actual safe operating status of the vehicle. Summary of the Invention

[0005] To address this, the present invention provides a method and system for managing health data of autonomous vehicles, which overcomes the problem in the prior art where health status information is generated based on signals reported by actuators, durability parameters, and driving control responses, and then comprehensively evaluated. This lacks correlation analysis of the operational dependencies between various onboard functions and the propagation process of abnormal states along the vehicle's operational link, making it difficult to determine whether local anomalies actually affect vehicle operation and the vehicle's remaining control capabilities after an anomaly occurs. This results in a discrepancy between the health status assessment results and the actual safe operating status of the vehicle.

[0006] To achieve the above objectives, the present invention provides a method for managing health data of autonomous vehicles, comprising: Acquire multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset; Before the vehicle is driven, the vehicle operation link is constructed based on the operational dependencies between the vehicle functions, and the vehicle operation link is used to determine whether the vehicle is ready to depart. During vehicle operation, acquire abnormal health events and vehicle operating status. Based on the state propagation relationship of abnormal health events in the corresponding vehicle operation links, determine whether the abnormal health events affect the actual vehicle operation and identify the affected operation links. The remaining control links of the vehicle are determined based on the affected operating links, and the safety control mode of the vehicle is determined based on the remaining control links; Based on the control response status during the safety control process, determine whether the safety control result meets the requirements, update the remaining control links and safety control methods, and generate vehicle operation health management results.

[0007] Furthermore, a health dataset is generated, including: Based on the data source and vehicle identifier corresponding to the multi-source health data, the multi-source health data belonging to the same autonomous vehicle are associated; Based on the vehicle operation objects reflected by the multi-source health data, the multi-source health data is associated with the corresponding in-vehicle functions to obtain an operation health dataset.

[0008] Furthermore, the vehicle operation link is constructed, including: Determine the data transmission relationships and control dependencies of onboard functions during autonomous driving operation; Based on the sequential operation of in-vehicle functions, in-vehicle functions with continuous dependencies are connected to form a vehicle operation link.

[0009] Further, determining whether a vehicle meets the conditions for departure includes: Acquire health data corresponding to onboard functions in the vehicle operation link, and determine the connection status of operation between adjacent onboard functions; The vehicle is deemed ready to depart when the vehicle operation link on which the departure depends remains continuous. When at least one vehicle operation link on which a vehicle relies for departure is interrupted, it is determined that the vehicle does not meet the conditions for departure, and the on-board function that caused the operation link interruption is identified.

[0010] Further, determining whether the health abnormality has an actual impact on vehicle operation includes: If the status changes of adjacent vehicle functions propagate downstream in the order of transmission of the vehicle operation link and cause corresponding changes in the vehicle operation status, then the health abnormality is determined to have an actual impact on vehicle operation, and the corresponding vehicle operation link is identified as the affected operation link. If an adjacent vehicle function does not generate a state change that propagates continuously downstream along the vehicle's operating link, or does not correspondingly cause a change in the vehicle's operating state, then the health anomaly is determined to have no actual impact on vehicle operation.

[0011] Further, determine the remaining control links of the vehicle, including: The vehicle functions affected by the health anomaly are determined based on the affected operational links, and control paths that depend on the vehicle functions are excluded from the vehicle control process. Based on the control dependencies between other vehicle functions, the control paths that can still continuously transmit and execute control commands are determined to obtain the remaining control links.

[0012] Furthermore, the vehicle's safety control method is determined based on the remaining control links, including: Determine the vehicle motion control functions that the remaining control links can achieve, and obtain the current motion state of the vehicle; Based on the vehicle's current motion state and the vehicle motion control functions required for safety control, a target control link capable of completing safety control is determined from the remaining control links, and the corresponding safety control operation is executed based on the target control link.

[0013] Furthermore, the safety control results are determined based on the control response status during the safety control process to determine whether they meet the requirements, including: Acquire the control commands corresponding to safety control operations, the responses of on-board actuators, and changes in vehicle motion status, and determine the response relationship among the three; When the changes in vehicle motion state correspond to the control commands and the responses of on-board actuators, the safety control results are deemed to meet the requirements. When the change in vehicle motion state does not correspond to the aforementioned response relationship, the affected on-board functions are re-determined, and the remaining control links and safety control methods are updated.

[0014] Furthermore, vehicle operational health management results are generated, including: The vehicle departure conditions, affected operational links, remaining control links, and safety control results are associated according to vehicle identification and occurrence time. The operational health dataset corresponding to the autonomous vehicle is updated based on the associated operational status, and the updated operational health data is associated and stored to generate vehicle operational health management results.

[0015] The present invention also provides an autonomous vehicle health data management system, comprising: The data acquisition module is used to acquire multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset. The vehicle departure determination module is used to construct the vehicle operation link based on the operational dependencies between vehicle functions before the vehicle departs, and to determine whether the vehicle meets the departure conditions based on the continuous state of the vehicle operation link. The impact detection module is used to acquire abnormal health events and vehicle operating status during vehicle operation. Based on the state propagation relationship of abnormal health events in the corresponding vehicle operation links, it determines whether the abnormal health events affect the actual vehicle operation and identifies the affected operation links. A safety control module is used to determine the remaining control links of the vehicle based on the affected operating links, and to determine the safety control mode of the vehicle based on the remaining control links; The result generation module is used to determine whether the safety control result meets the requirements based on the control response status during the safety control process, so as to update the remaining control links and safety control methods and generate vehicle operation health management results.

[0016] Compared with existing technologies, the beneficial effects of the present invention are as follows: The method of the present invention establishes a correlation between multi-source health data and vehicle functions, constructs vehicle operation links based on the operational dependencies between various vehicle functions, determines whether the vehicle meets the departure conditions based on the continuous state of the operation links, and improves the matching between departure control and the actual operating capability of the vehicle; it determines whether the anomaly has an actual impact on vehicle operation based on the state propagation relationship of the abnormal health event in the corresponding operation link, reducing misjudgments caused by instantaneous anomalies and warnings without actual impact; it determines the remaining control links of the vehicle based on the affected operation links, and determines the safety control mode based on the remaining control capability, avoiding the dependence of safety control on functions already affected by anomalies; and it updates the remaining control links and safety control modes based on the control response state during the safety control process, realizing closed-loop management of vehicle health anomaly identification, safety control, and feedback correction, thereby improving the driving safety of autonomous vehicles.

[0017] Furthermore, the method of the present invention determines whether a vehicle is ready to depart based on the continuous state of the vehicle operation link. Since autonomous driving operation relies on multiple functional links such as perception, positioning, decision-making, steering, and braking, it is difficult to determine whether an abnormality truly affects vehicle operation based solely on whether each vehicle inspection item is qualified. By establishing a correspondence between health data and functional nodes, and judging whether necessary functional nodes are continuously effective along the autonomous driving operation process, abnormalities that cause the operation link to be interrupted can be identified, and departure permissions can be controlled accordingly. This avoids treating non-critical abnormalities as critical functional abnormalities, further improving the driving safety of autonomous vehicles.

[0018] Furthermore, the method of the present invention determines whether a health anomaly affects the actual operation of the vehicle based on the state propagation relationship of the health anomaly event in the corresponding vehicle operation link. Since communication jitter, instantaneous noise, and short-term data loss can all generate anomalies, relying solely on rules such as duration and frequency of occurrence can easily lead to misjudgment. By tracking the propagation relationship of the anomaly event between adjacent functional nodes and the vehicle's motion state, it can be determined whether the anomaly forms a continuous state change along the corresponding operation link. If no continuous propagation is formed, it can be excluded or continued to be observed. If continuous propagation is formed, it is determined that the corresponding operation link is actually affected, which further improves the driving safety of autonomous vehicles.

[0019] Furthermore, the method of the present invention determines whether the safety control result meets the requirements based on the control response state during the safety control process. Since different anomalies have different effects on control capabilities such as steering, braking, and driving, the fixed parking strategy may rely on control links that have failed. By excluding affected functional nodes and determining the remaining control links, corresponding safety control operations can be generated based on the vehicle's current control capabilities. After execution, the control effect is verified by combining the actuator response and the vehicle's motion state. When the response does not meet expectations, the remaining control links are re-determined and the safety control operations are adjusted, further improving the driving safety of autonomous vehicles. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of the autonomous vehicle health data management method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the process by which the autonomous vehicle health data management method of this invention determines whether a vehicle meets the conditions for departure. Figure 3 This is a flowchart illustrating the process by which the autonomous vehicle health data management method of this invention determines whether a health abnormality affects the actual operation of the vehicle. Figure 4 This is a block diagram of the overall structure of the autonomous vehicle health data management system according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] Please see Figure 1 The diagram shown is an overall flowchart of the autonomous vehicle health data management method according to an embodiment of the present invention. The method described in this embodiment includes: Step S1: Obtain multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset; Step S2: Before the vehicle departs, a vehicle operation link is constructed based on the operational dependencies between vehicle functions, and the vehicle is deemed ready to depart based on the continuous state of the vehicle operation link. Step S3: During vehicle operation, acquire health abnormality events and vehicle operating status. Based on the state propagation relationship of health abnormality events in the corresponding vehicle operating links, determine whether the health abnormality affects the actual vehicle operation and identify the affected operating links. Step S4: Determine the remaining control links of the vehicle based on the affected operating links, and determine the vehicle's safety control mode based on the remaining control links; Step S5: Determine whether the safety control result meets the requirements based on the control response status during the safety control process, so as to update the remaining control links and safety control methods and generate vehicle operation health management results.

[0024] Specifically, multi-source health data includes vehicle inspection data, tire pressure monitoring data, and early warning data from autonomous driving subsystems.

[0025] Specifically, abnormal health events include perceptual abnormal events, localized abnormal events, and execution abnormal events.

[0026] In this embodiment, the method establishes a correlation between multi-source health data and vehicle functions, constructs vehicle operation links based on the operational dependencies between various vehicle functions, determines whether the vehicle meets the departure conditions based on the continuous state of the operation links, and improves the matching between departure control and the actual operating capability of the vehicle; it determines whether the anomaly has an actual impact on vehicle operation based on the state propagation relationship of the abnormal health event in the corresponding operation link, reducing misjudgments caused by instantaneous anomalies and warnings without actual impact; it determines the remaining control links of the vehicle based on the affected operation links, and determines the safety control mode based on the remaining control capability, avoiding the dependence of safety control on functions already affected by the anomaly; and it updates the remaining control links and safety control mode based on the control response state during the safety control process, realizing closed-loop management of vehicle health anomaly identification, safety control, and feedback correction, thereby improving the driving safety of autonomous vehicles.

[0027] Specifically, a healthy operational dataset is generated, including: Based on the data source and vehicle identifier corresponding to the multi-source health data, the multi-source health data belonging to the same autonomous vehicle are associated; Based on the vehicle operation objects reflected by the multi-source health data, the multi-source health data is associated with the corresponding in-vehicle functions to obtain an operation health dataset.

[0028] Specifically, constructing the vehicle operation link includes: Determine the data transmission relationships and control dependencies of onboard functions during autonomous driving operation; Based on the sequential operation of in-vehicle functions, in-vehicle functions with continuous dependencies are connected to form a vehicle operation link.

[0029] Specifically, in-vehicle functions include environmental perception, vehicle positioning, and motion control.

[0030] Specifically, runtime dependencies include data transfer relationships, function call relationships, and control execution relationships.

[0031] Specifically, the vehicle operation link includes the perception and decision-making link, the positioning and decision-making link, and the control and execution link.

[0032] Please see Figure 2 The diagram shown is a logical flowchart illustrating the process of determining whether a vehicle meets the departure conditions according to an embodiment of the present invention. In this embodiment, the determination of whether a vehicle meets the departure conditions is based on the continuous state of the vehicle's operating link. Acquire health data corresponding to onboard functions in the vehicle operation link, and determine the connection status of operation between adjacent onboard functions; The vehicle is deemed ready to depart when the vehicle operation link on which the departure depends remains continuous. When at least one vehicle operation link on which a vehicle relies for departure is interrupted, it is determined that the vehicle does not meet the conditions for departure, and the on-board function that caused the operation link interruption is identified.

[0033] Understandably, the normal operation of autonomous vehicles relies on continuous data transmission and control coordination between onboard functions such as perception, localization, decision-making, control, and execution. Any abnormality in any necessary function may cause the corresponding operational link to be interrupted, preventing subsequent functions from obtaining effective input or completing control execution. Whether the vehicle's operational link remains continuous can reflect whether the vehicle has the basic operational capabilities required to complete autonomous driving tasks. Determining the departure conditions based on the continuous status of the operational link can avoid making isolated judgments based solely on whether a single health data point is normal or not, and improve the consistency between the determination of departure conditions and the actual operational capabilities of the vehicle.

[0034] In this embodiment, the method determines whether a vehicle is ready to depart based on the continuous state of the vehicle's operational link. Since autonomous driving relies on multiple functional links such as perception, positioning, decision-making, steering, and braking, it is difficult to determine whether an anomaly truly affects vehicle operation simply by checking whether each vehicle inspection item is qualified. By establishing a correspondence between health data and functional nodes, and judging whether necessary functional nodes are continuously effective along the autonomous driving operation process, anomalies that cause the operational link to be interrupted can be identified, and departure permissions can be controlled accordingly. This avoids treating non-critical anomalies as critical functional anomalies, further improving the driving safety of autonomous vehicles.

[0035] Please see Figure 3 The diagram shown is a logical flowchart illustrating the process of determining whether a health anomaly affects actual vehicle operation according to an embodiment of the present invention. In this embodiment, the impact of a health anomaly on actual vehicle operation is determined based on the state propagation relationship of the health anomaly event in the corresponding vehicle operation link. If the status changes of adjacent vehicle functions propagate downstream in the order of transmission of the vehicle operation link and cause corresponding changes in the vehicle operation status, then the health abnormality is determined to have an actual impact on vehicle operation, and the corresponding vehicle operation link is identified as the affected operation link. If an adjacent vehicle function does not generate a state change that propagates continuously downstream along the vehicle's operating link, or does not correspondingly cause a change in the vehicle's operating state, then the health anomaly is determined to have no actual impact on vehicle operation.

[0036] Specifically, vehicle operating status includes vehicle speed status, vehicle attitude status, and vehicle trajectory status.

[0037] Specifically, state propagation relationships include the propagation of abnormal states to adjacent functions, the propagation of abnormal states to downstream functions, and the propagation of abnormal states to vehicle motion states.

[0038] Specifically, the affected operational links include those affected by perception and decision-making, those affected by positioning and decision-making, and those affected by control and execution.

[0039] Understandably, genuine health anomalies typically propagate downstream from the point of occurrence along the data transmission or control dependencies between vehicle functions, causing corresponding changes in the status of subsequent functions and the vehicle's motion. In contrast, short-term anomalies such as communication jitter and transient noise may only manifest at a single functional node and will not propagate continuously. By determining whether adjacent vehicle functions and the vehicle's operating status change continuously according to the operational link after a health anomaly occurs, we can distinguish between transient anomalies that have not yet had a real impact and effective anomalies that have already affected vehicle operation, thereby improving the accuracy of health anomaly identification.

[0040] Specifically, determining the remaining control links of the vehicle includes: The vehicle functions affected by the health anomaly are determined based on the affected operational links, and control paths that depend on the vehicle functions are excluded from the vehicle control process. Based on the control dependencies between other vehicle functions, the control paths that can still continuously transmit and execute control commands are determined to obtain the remaining control links.

[0041] Specifically, the remaining control links include the steering control link, the braking control link, and the drive control link.

[0042] In this embodiment, the method determines whether a health anomaly affects the actual operation of the vehicle based on the state propagation relationship of the health anomaly event in the corresponding vehicle operation link. Since communication jitter, instantaneous noise, and short-term data loss can all generate abnormal events, relying solely on rules such as duration and frequency of occurrence can easily lead to misjudgment. By tracking the propagation relationship of the abnormal event between adjacent functional nodes and the vehicle's motion state, it can be determined whether the anomaly forms a continuous state change along the corresponding operation link. If no continuous propagation is formed, it can be excluded or continued to be observed. If continuous propagation is formed, it is determined that the corresponding operation link is actually affected, further improving the driving safety of autonomous vehicles.

[0043] Specifically, determining the vehicle's safety control method based on the remaining control links includes: Determine the vehicle motion control functions that the remaining control links can achieve, and obtain the current motion state of the vehicle; Based on the vehicle's current motion state and the vehicle motion control functions required for safety control, a target control link capable of completing safety control is determined from the remaining control links, and the corresponding safety control operation is executed based on the target control link.

[0044] In this embodiment, determining the vehicle's safety control method based on the remaining control links includes: The operating status of each vehicle function in the remaining control links is obtained, the current steering, braking, driving and vehicle state adjustment capabilities of the vehicle are determined, and the vehicle motion control operations that each remaining control link can support are determined accordingly. The system acquires the vehicle's current speed, direction of travel, road environment, and abnormal impact status. Based on the vehicle's current operating status, it determines the motion control capabilities required for safety control and correlates them with the vehicle motion control operations that each remaining control link can support. When the remaining control links can support both steering and deceleration control simultaneously, control the vehicle to decelerate and move to a safe stopping position; when steering control is unavailable but longitudinal control remains effective, control the vehicle to decelerate and stop in the currently feasible direction; when partial braking control is limited, utilize the still effective drive or braking control capabilities to reduce the vehicle speed. The remaining control links that can complete the current safety control operation are identified as target control links, and corresponding control commands are generated based on the target control links to execute vehicle safety control.

[0045] In this embodiment, the safety control result is determined to meet the requirements based on the control response status during the safety control process. Acquire the control commands corresponding to safety control operations, the responses of on-board actuators, and changes in vehicle motion status, and determine the response relationship among the three; When the changes in vehicle motion state correspond to the control commands and the responses of on-board actuators, the safety control results are deemed to meet the requirements. When the change in vehicle motion state does not correspond to the aforementioned response relationship, the affected on-board functions are re-determined, and the remaining control links and safety control methods are updated.

[0046] It is understandable that safety control operations require the generation of control commands, the response of actuators, and ultimately their effect on the vehicle's motion state. If the control links on which safety control depends are functioning normally, the response of the actuators and changes in vehicle motion should correspond to the control commands. However, when there are still unidentified abnormal functions in the control links, there may be situations where control commands have been output but the vehicle has not produced a corresponding response. The control response status during the safety control process can be used to verify whether the remaining control links are actually effective, and when the control response does not meet the requirements, the affected functions and safety control methods can be re-determined to improve the reliability of safety control.

[0047] In this embodiment, the method determines whether the safety control result meets the requirements based on the control response status during the safety control process. Since different anomalies have different effects on control capabilities such as steering, braking, and driving, the fixed parking strategy may rely on control links that have failed. By excluding affected functional nodes and determining the remaining control links, corresponding safety control operations can be generated based on the vehicle's current control capabilities. After execution, the control effect is verified by combining the actuator response and the vehicle's motion state. When the response does not meet expectations, the remaining control links are re-determined and the safety control operations are adjusted, further improving the driving safety of autonomous vehicles.

[0048] Specifically, generating vehicle operational health management results includes: The vehicle departure conditions, affected operational links, remaining control links, and safety control results are associated according to vehicle identification and occurrence time. The operational health dataset corresponding to the autonomous vehicle is updated based on the associated operational status, and the updated operational health data is associated and stored to generate vehicle operational health management results.

[0049] In this embodiment, the generation of vehicle operational health management results includes: The vehicle departure condition determination results, affected operation links, remaining control links, and safety control execution results are associated according to vehicle identification and occurrence time to form the corresponding vehicle's operation status record; Based on the chronological order of the various operational status records, update the vehicle's current departure status, affected onboard function status, remaining control capabilities, and safety control execution status to obtain the vehicle's current operational health status. The updated operational health status is associated with the corresponding abnormal health events, operational link changes and control processing results, and stored in the vehicle operational health database; The associated storage results are used to generate vehicle operation health management records for subsequent vehicle status queries, anomaly tracing, and operation and maintenance.

[0050] Please see Figure 4 The diagram shown is an overall structural block diagram of the autonomous vehicle health data management system according to an embodiment of the present invention. The system described in this embodiment includes: The data acquisition module is used to acquire multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset. The departure determination module is connected to the data acquisition module. Before the vehicle departs, it constructs the vehicle operation link based on the operational dependencies between vehicle functions and determines whether the vehicle meets the departure conditions based on the continuous state of the vehicle operation link. The impact detection module, which is connected to the departure determination module, is used to acquire health abnormality events and vehicle operating status during vehicle operation. Based on the state propagation relationship of the health abnormality event in the corresponding vehicle operation link, it determines whether the health abnormality affects the actual vehicle operation and identifies the affected operation link. A safety control module, connected to the impact detection module, is used to determine the remaining control links of the vehicle based on the affected operating links, and to determine the safety control mode of the vehicle based on the remaining control links. The result generation module, which is connected to the safety control module, is used to determine whether the safety control result meets the requirements based on the control response status during the safety control process, so as to update the remaining control links and safety control methods and generate vehicle operation health management results.

[0051] Specifically, safety control methods include deceleration control, sidewalk parking control, and parking control.

[0052] Specifically, vehicle operation health management results include vehicle departure status, affected operation link status, and safety control execution status.

[0053] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for managing health data of autonomous vehicles, characterized in that, include: Acquire multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset; Before the vehicle is driven, the vehicle operation link is constructed based on the operational dependencies between the vehicle functions, and the vehicle is determined to be ready to depart based on the continuous status of the vehicle operation link. During vehicle operation, acquire abnormal health events and vehicle operating status. Based on the state propagation relationship of abnormal health events in the corresponding vehicle operation links, determine whether the abnormal health events affect the actual vehicle operation and identify the affected operation links. The remaining control links of the vehicle are determined based on the affected operating links, and the safety control mode of the vehicle is determined based on the remaining control links. Based on the control response status during the safety control process, determine whether the safety control result meets the requirements, update the remaining control links and safety control methods, and generate vehicle operation health management results.

2. The method for managing health data of autonomous vehicles according to claim 1, characterized in that, Create a runtime health dataset, including: Based on the data source and vehicle identifier corresponding to the multi-source health data, the multi-source health data belonging to the same autonomous vehicle are associated; Based on the vehicle operation objects reflected by the multi-source health data, the multi-source health data is associated with the corresponding in-vehicle functions to obtain an operation health dataset.

3. The method for managing health data of autonomous vehicles according to claim 2, characterized in that, Constructing the vehicle operation link includes: Determine the data transmission relationships and control dependencies of onboard functions during autonomous driving operation; Based on the sequential operation of in-vehicle functions, in-vehicle functions with continuous dependencies are connected to form a vehicle operation link.

4. The method for managing health data of autonomous vehicles according to claim 3, characterized in that, Determining whether a vehicle meets the conditions for departure includes: Acquire health data corresponding to onboard functions in the vehicle operation link, and determine the connection status of operation between adjacent onboard functions; The vehicle is deemed ready to depart when the vehicle operation link on which the departure depends remains continuous. When at least one vehicle operation link on which a vehicle relies for departure is interrupted, it is determined that the vehicle does not meet the conditions for departure, and the on-board function that caused the operation link interruption is identified.

5. The method for managing health data of autonomous vehicles according to claim 4, characterized in that, Determining whether a health abnormality has an actual impact on vehicle operation includes: If the status changes of adjacent vehicle functions propagate downstream in the order of transmission of the vehicle operation link and cause corresponding changes in the vehicle operation status, then the health abnormality is determined to have an actual impact on vehicle operation, and the corresponding vehicle operation link is identified as the affected operation link. If an adjacent vehicle function does not generate a state change that propagates continuously downstream along the vehicle's operating link, or does not correspondingly cause a change in the vehicle's operating state, then the health anomaly is determined to have no actual impact on vehicle operation.

6. The method for managing health data of autonomous vehicles according to claim 5, characterized in that, Determine the remaining control links of the vehicle, including: The affected vehicle functions are determined based on the affected operational links, and control paths that depend on the vehicle functions are excluded from the vehicle control process. Based on the control dependencies between other vehicle functions, the control paths that can still continuously transmit and execute control commands are determined to obtain the remaining control links.

7. The method for managing health data of autonomous vehicles according to claim 6, characterized in that, The vehicle's safety control method is determined based on the remaining control links, including: Determine the vehicle motion control functions that the remaining control links can achieve, and obtain the current motion state of the vehicle; Based on the vehicle's current motion state and the vehicle motion control functions required for safety control, a target control link capable of completing safety control is determined from the remaining control links, and the corresponding safety control operation is executed based on the target control link.

8. The method for managing health data of autonomous vehicles according to claim 7, characterized in that, Determine whether the safety control results meet the requirements based on the control response status during the safety control process, including: Acquire the control commands corresponding to safety control operations, the responses of on-board actuators, and changes in vehicle motion status, and determine the response relationship among the three; When the changes in vehicle motion state correspond to the control commands and the responses of on-board actuators, the safety control results are deemed to meet the requirements. When the change in vehicle motion state does not correspond to the aforementioned response relationship, the affected on-board functions are re-determined, and the remaining control links and safety control methods are updated.

9. The method for managing health data of autonomous vehicles according to claim 8, characterized in that, Generate vehicle operational health management results, including: The vehicle departure conditions, affected operational links, remaining control links, and safety control results are associated according to vehicle identification and occurrence time. The operational health dataset corresponding to the autonomous vehicle is updated based on the associated operational status, and the updated operational health data is associated and stored to generate vehicle operational health management results.

10. A management system using the autonomous vehicle health data management method according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire multi-source health data of autonomous vehicles, establish the correlation between the multi-source health data and vehicle functions, and form an operational health dataset. The vehicle departure determination module is used to construct the vehicle operation link based on the operational dependencies between vehicle functions before the vehicle departs, and to determine whether the vehicle meets the departure conditions based on the continuous state of the vehicle operation link. The impact detection module is used to acquire abnormal health events and vehicle operating status during vehicle operation. Based on the state propagation relationship of abnormal health events in the corresponding vehicle operation links, it determines whether the abnormal health events affect the actual vehicle operation and identifies the affected operation links. A safety control module is used to determine the remaining control links of the vehicle based on the affected operating links, and to determine the safety control mode of the vehicle based on the remaining control links; The result generation module is used to determine whether the safety control result meets the requirements based on the control response status during the safety control process, so as to update the remaining control links and safety control methods and generate vehicle operation health management results.

Images