Risk detection method, device and system
By collecting and uploading environmental data during the risk period in the vehicle end in the autonomous driving system, the problem of insufficient coverage of hazardous scenarios in the prior art is solved, and the risk detection efficiency of the system and the safety of the autonomous driving function are improved.
Patent Information
- Application Number
- CN202411172274.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-12
AI Technical Summary
The existing autonomous driving system cannot effectively respond when facing unknown hazard scenarios. The main reason is that the number of hazard scenarios considered in system specification design and testing is insufficient, which leads to a large amount of manpower and material resources required to process massive data when achieving the expected functional safety SOTIF. .
By continuously collecting environmental data on the vehicle side and uploading data within the preset period from the time when risks occur to the cloud when risks occur, the amount of data storage in the cloud is reduced, the consumption of manpower and material resources is reduced, and efficiency is improved.
It realizes efficient detection and processing of risk data in the autonomous driving system, reduces the need for manual screening of useless data, and improves the system's risk response capabilities and efficiency.
Smart Images

Figure CN118928470A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of automobile technology and relates to a risk detection method, device and system. Background Art
[0002] Safety of the Intended Functionality (SOTIF) refers to a safety assessment of self-driving cars, connected cars, and other advanced cars with highly automated functions, which aims to ensure their safety performance. Different from traditional car safety assessments, SOTIF focuses on the "intention" of the vehicle, that is, whether the vehicle can accurately identify and understand the surrounding environment when driving automatically, and can avoid danger and accidents through correct decisions and behaviors.
[0003] However, the current autonomous driving system cannot achieve a high degree of automation. The main reason is that in the system specification design and system testing, the number of hazard scenarios considered is not enough to cover the actual driving environment, which makes the system unable to cope with unknown hazard scenarios. Therefore, collecting and mining a large number of hazard scenarios is one of the reasonable ways to achieve the expected functional safety. However, the current hazard scenario mining method is mainly based on the overall processing of massive data collected from real vehicles uploaded to the cloud, which consumes a lot of manpower and material resources.
[0004] Therefore, it is necessary to provide an improved technical solution to overcome the above technical problems existing in the prior art. Summary of the invention
[0005] The purpose of this application is to provide a risk detection method, device and system. When a risk is detected, the environmental data during the risk period is uploaded to the cloud, thereby reducing the storage volume of cloud data, reducing the consumption of manpower and material resources, and improving efficiency.
[0006] To achieve the above objectives:
[0007] In a first aspect, the present invention provides a risk detection method, which is applied to a vehicle, and comprises the following steps:
[0008] S101, in response to the vehicle turning on the automatic driving function, continuously collecting environmental data of the driving area of the vehicle, and controlling the vehicle to automatically drive according to the planned driving route;
[0009] S102, determining whether there is a risk when the vehicle is automatically driving;
[0010] S103: If yes, the environmental data collected within a preset upload period starting from when the risk occurs is defined as recorded data, and the recorded data is uploaded to a storage device in the cloud.
[0011] In one implementation, the step S101 includes:
[0012] In response to the vehicle starting the automatic driving function, continuously collecting the environmental data through the vehicle-mounted collection device of the vehicle, and preprocessing the environmental data to obtain preprocessed data;
[0013] Performing prediction processing on the preprocessed data to obtain predicted data;
[0014] Performing fusion processing on the predicted data to obtain fused data;
[0015] The driving route of the vehicle is planned according to the fused data, and the automatic driving of the vehicle is controlled according to the driving route.
[0016] In one implementation, step S102 includes:
[0017] Determine whether the hardware data of the acquisition device is at risk, determine whether the environmental data is at risk, determine whether the predicted data is at risk, determine whether the fused data is at risk, and determine whether the driving route is at risk.
[0018] In one implementation, step S103 includes:
[0019] If there is a risk in at least one of the hardware data, the environmental data, the predicted data, the fusion data, and the driving route, the environmental data collected within a preset upload time period starting from the time when the risk occurs is defined as recorded data, and the recorded data is transmitted to the storage device in the cloud.
[0020] In one embodiment, after step S103, the method further includes:
[0021] S104, in response to receiving the algorithm model update data sent from the cloud, updating the algorithm model in the vehicle for collecting environmental data, planning driving routes, and controlling automatic driving.
[0022] In a second aspect, the present invention provides a risk detection method, which is applied to the cloud and includes the following steps:
[0023] S201, after receiving, in the cloud, recorded data sent by the vehicle and corresponding to the risk existing in the automatic driving of the vehicle, reading the recorded data;
[0024] S202, comparing the recorded data with preset risk comparison data, and determining a risk index for characterizing the severity of the risk corresponding to the recorded data according to the comparison result;
[0025] S203, if it is determined that the risk index is greater than a preset first risk index threshold, optimizing an algorithm model used in the vehicle for collecting environmental data, planning a driving route, and controlling automatic driving, and determining the optimized second risk index threshold;
[0026] S204: If it is determined that the risk index is not greater than the second risk index threshold, algorithm model update data is sent to the vehicle, and the recorded data is transmitted to a risk environment database.
[0027] In one implementation, the step S202 further includes:
[0028] The recorded data is transmitted to the risk environment database according to the comparison result.
[0029] In one implementation, the step S203 further includes:
[0030] If it is determined that the risk index is not greater than the first risk index threshold, the recorded data is transmitted to the risk environment database.
[0031] In one implementation, the step S204 further includes:
[0032] If it is determined that the risk index is greater than the second risk index threshold, step S203 is performed again.
[0033] In a third aspect, the present invention provides a vehicle-side risk detection device, comprising:
[0034] A collection and control module, configured to continuously collect environmental data of a driving area of the vehicle in response to the vehicle starting an automatic driving function, and control the vehicle to drive automatically according to a planned driving route;
[0035] A risk judgment module, used to determine whether there is a risk when the vehicle is driving automatically;
[0036] The data uploading module is used to, in response to determining that there is a risk during the automatic driving of the vehicle, define the environmental data collected within a preset uploading period starting from the time when the risk occurs as recorded data, and upload the recorded data to a storage device in the cloud.
[0037] In a fourth aspect, the present application provides a cloud risk detection device, comprising:
[0038] A data reading module, configured to read the recorded data corresponding to the risk of the vehicle during autonomous driving after receiving the recorded data sent by the vehicle in the cloud;
[0039] A risk index assessment module, used to compare the recorded data with preset risk comparison data, and determine a risk index used to characterize the severity of the risk corresponding to the recorded data according to the comparison result;
[0040] an optimization module, configured to optimize an algorithm model used in the vehicle for collecting environmental data, planning a driving route, and controlling automatic driving if it is determined that the risk index is greater than a preset first risk index threshold, and determine the optimized second risk index threshold;
[0041] An updating module is used to send algorithm model update data to the vehicle and transmit the recorded data to a risk environment database if it is determined that the risk index is not greater than the second risk index threshold.
[0042] In a fifth aspect, the present application provides a risk detection system, characterized in that it includes the vehicle-side risk detection device as described in the third aspect, and the cloud-side risk detection device as described in the fourth aspect.
[0043] The risk detection method, device and system provided by the present application include: in response to the vehicle turning on the automatic driving function, continuously collecting environmental data of the driving area of the vehicle, and controlling the vehicle to drive automatically according to the planned driving route; determining whether there is a risk when the vehicle is driving automatically; if so, defining the environmental data collected within the preset upload period starting from the time when the risk occurs as recorded data, and uploading the recorded data to a storage device in the cloud. When the present invention detects that a risk has occurred, the environmental data within the risk duration period is uploaded to the cloud, which reduces the storage volume of cloud data, reduces the consumption of manpower and material resources, and improves efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0045] Figure 1 A schematic diagram of a process flow of a risk detection method applied to a vehicle end provided by an embodiment of the present invention;
[0046] Figure 2 A schematic diagram of a process flow of a risk detection method applied to the cloud provided by an embodiment of the present invention;
[0047] Figure 3 A schematic diagram of the structure of a vehicle-side risk detection device provided by an embodiment of the present invention;
[0048] Figure 4 A schematic diagram of the structure of a cloud risk detection device provided by an embodiment of the present invention;
[0049] Figure 5 A schematic diagram of the workflow of a risk detection system provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0050] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of systems and methods consistent with some aspects of the present application as detailed in the appended claims.
[0051] It should be noted that in this article, step codes such as S1 and S2 are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the order. When implementing the step, those skilled in the art may execute S2 first and then S1, etc., but these should all be within the scope of protection of this application.
[0052] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0053] First embodiment
[0054] A risk detection method provided in an embodiment of the present application is applied to a vehicle side, see Figure 1 , the method comprising:
[0055] Step S101, in response to the vehicle turning on the automatic driving function, continuously collecting environmental data of the vehicle's driving area, and controlling the vehicle to automatically drive according to the planned driving route.
[0056] In one embodiment, it further comprises:
[0057] In response to the vehicle turning on the automatic driving function, the environmental data is continuously collected by the vehicle-mounted collection device of the vehicle, and the environmental data is preprocessed to obtain preprocessed data; the preprocessed data is predictively processed to obtain predicted data; the predicted data is fused to obtain fused data; the vehicle's driving route is planned according to the fused data, and the vehicle's automatic driving is controlled according to the driving route.
[0058] Specifically, the vehicle-mounted acquisition device includes a vehicle-mounted camera, a vehicle-mounted laser radar, and a vehicle-mounted millimeter-wave radar. In addition, in addition to collecting environmental data, the vehicle-mounted acquisition device also collects its own hardware data, such as chip temperature, current operating voltage, etc.
[0059] It can be understood that the preprocessing of data collected by the on-board camera in the on-board acquisition device includes but is not limited to image distortion correction, image resolution adjustment, image denoising, and image enhancement; the preprocessing of data collected by the on-board laser radar in the on-board acquisition device includes but is not limited to filtering, denoising, abnormal point removal, and point cloud registration; the preprocessing of data collected by the on-board millimeter-wave radar in the on-board acquisition device includes but is not limited to filtering, denoising, and abnormal point removal.
[0060] Furthermore, predictive processing of the preprocessed data includes but is not limited to target detection, lane line segmentation, drivable area segmentation, target speed estimation, and distance estimation.
[0061] Preferably, the method of performing fusion processing on the predicted data is post-fusion.
[0062] Step S102, determining whether there is a risk when the vehicle is automatically driving.
[0063] In one embodiment, it further comprises:
[0064] Determine whether the hardware data of the vehicle-mounted acquisition device is at risk, determine whether the environmental data is at risk, determine whether the predicted data is at risk, determine whether the fused data is at risk, and determine whether the driving route is at risk.
[0065] It can be understood that for different types of hardware data in the vehicle-mounted acquisition device, this application adopts different targeted hardware level detection methods. Exemplarily, determining whether the hardware data of the vehicle-mounted camera in the vehicle-mounted acquisition device is at risk includes determining whether the current working voltage output is within the normal range, whether the current working temperature is within the normal range, whether the current working environment humidity is within the normal range, whether the parameters in the camera are normal, etc.; determining whether the hardware data of the vehicle-mounted laser radar in the vehicle-mounted acquisition device is at risk includes determining whether the current working voltage output is within the normal range, whether the current working temperature is within the normal range, whether the laser transmitter power is within the normal range, etc.; determining whether the hardware data of the vehicle-mounted laser radar in the vehicle-mounted acquisition device is at risk includes determining whether the current working voltage output is within the normal range, whether the current working temperature is within the normal range, whether the current working humidity is within the normal range, whether the transmitter power is within the normal range, etc.
[0066] Furthermore, for different types of environmental data in the vehicle-mounted acquisition device, the present application adopts different targeted physical level detection methods. Exemplarily, determining whether the environmental data of the vehicle-mounted camera in the vehicle-mounted acquisition device is at risk includes determining whether there is local exposure or global exposure in the output image, whether the camera lens is dirty or blocked, and detecting whether the camera has relative jitter based on the output image; determining whether the environmental data of the vehicle-mounted laser radar in the vehicle-mounted acquisition device is at risk includes determining whether there is excessive noise in the output point cloud, whether there are too many outliers, whether it is affected by electromagnetic interference, etc.; determining whether the environmental data of the vehicle-mounted millimeter-wave radar in the vehicle-mounted acquisition device is at risk includes determining whether there is excessive noise in the output wave data, whether there is electromagnetic interference, whether there is interference wave, etc.
[0067] Furthermore, for different types of predicted data, the present application adopts different targeted domain and target level detection methods. Exemplarily, determining whether the predicted data is at risk includes determining whether there is a domain offset. For example, the driving environment of the vehicle is different from the ODD, or the weather conditions are different from the ODD. Although the millimeter wave radar cannot directly detect the weather conditions, it can infer the weather conditions by measuring the water vapor content in the air; determining whether the predicted data is at risk also includes determining whether there are abnormal targets, uncertain targets, and target loss situations.
[0068] Furthermore, for fused data, this application adopts targeted data fusion level and scene snapshot level detection methods. Exemplarily, determining whether the fused data has risks includes determining whether there are differences in the identified target categories, whether different types of prediction data match, whether there are abnormal semantic relationships, etc. Among them, the detection at the scene snapshot level is the detection of abnormal semantic relationships between various types of targets in the scene snapshot.
[0069] Furthermore, for the driving route, the present application adopts a targeted scene-level detection method. Exemplarily, determining whether the driving route is at risk includes determining whether the output driving route has abnormal trajectories and behaviors. The scene-level detection is the detection of abnormal target trajectories and behaviors in the driving scene.
[0070] Step S103: If yes, the environmental data collected within a preset upload period starting from when the risk occurs is defined as recorded data, and the recorded data is uploaded to a storage device in the cloud.
[0071] In one embodiment, it further comprises:
[0072] If there is a risk in at least one of the hardware data, the environmental data, the predicted data, the fusion data, and the driving route, the environmental data collected within a preset upload time period starting from the time when the risk occurs is defined as recorded data, and the recorded data is transmitted to the storage device in the cloud.
[0073] Furthermore, after step S103, the method further includes:
[0074] Step S104, in response to receiving the algorithm model update data sent from the cloud, updating the algorithm model in the vehicle for collecting environmental data, planning driving routes, and controlling automatic driving.
[0075] Through the above method, risks in autonomous driving scenarios can be detected, and when risks are detected, environmental data collected during a preset upload period starting from the time the risk occurs can be uploaded to the cloud, avoiding a large amount of useless data when manually screening data, reducing the consumption of manpower and material resources, and improving efficiency.
[0076] Second embodiment
[0077] Based on the first embodiment of the present application, another embodiment of the present application is provided. In this embodiment, a risk detection method is provided, which is applied to the cloud. Figure 2 , the method comprising:
[0078] Step S201, after receiving the recorded data corresponding to the risk of the vehicle during automatic driving sent by the vehicle in the cloud, read the recorded data.
[0079] It can be understood that when a risk is detected, the environmental data collected during the preset upload period starting from the time when the risk occurs is uploaded to the cloud. Then, the engineer extracts the recorded data stored in the cloud storage for screening, thereby optimizing the algorithm model on the vehicle side.
[0080] Step S202: compare the recorded data with preset risk comparison data, and determine a risk index for characterizing the severity of the risk corresponding to the recorded data according to the comparison result.
[0081] In one embodiment, it further comprises:
[0082] The recorded data is transmitted to the risk environment database according to the comparison result.
[0083] Step S203: If it is determined that the risk index is greater than a preset first risk index threshold, the algorithm model used in the vehicle to collect environmental data, plan driving routes, and control automatic driving is optimized, and the optimized second risk index threshold is determined.
[0084] In one embodiment, it further comprises:
[0085] If it is determined that the risk index is not greater than the first risk index threshold, the recorded data is transmitted to the risk environment database.
[0086] Among them, the known risk environment database contains all environmental data with identified risks, as well as simulated environment data with risks considered in function design and testing, which can be used to guide function design, generate test cases, and generate parameter distribution in test scenarios.
[0087] Preferably, the algorithm model used in the vehicle to collect environmental data, plan driving routes and control automatic driving is optimized, including optimizing the algorithm model from at least one of the collection end and the regulation and control end.
[0088] Further, optimizing the algorithm model from the acquisition end includes labeling the recorded data and training the acquisition model; optimizing the algorithm model from the regulation and control end includes changing the logic of the regulation and control model. The training can be completed based on transfer learning.
[0089] Step S204: If it is determined that the risk index is not greater than the second risk index threshold, algorithm model update data is sent to the vehicle, and the recorded data is transmitted to a risk environment database.
[0090] In one embodiment, it further comprises:
[0091] If it is determined that the risk index is greater than the second risk index threshold, step S203 is performed again.
[0092] It can be understood that when the risk index is not greater than the second risk index threshold, it means that the current intelligent driving function can cope with the analyzed risk; after optimizing and updating the algorithm model used in the vehicle to collect environmental data, plan driving routes, and control automatic driving, the risk response capability of the current intelligent driving function increases, so the second risk index threshold will also increase.
[0093] In this embodiment, by utilizing the collected and mined environmental data with risks, the algorithm models used in the vehicle to collect environmental data, plan driving routes, and control autonomous driving are optimized, thereby achieving a continuously iterative data closed loop, reducing costs while effectively improving the system's risk response capabilities, improving the autonomous driving function, and enhancing the user's vehicle experience.
[0094] Third embodiment
[0095] Based on the first embodiment of the present application, in this embodiment, a vehicle-side risk detection device is provided, referring to Figure 3 , the device comprises:
[0096] The collection and control module 31 is used to continuously collect environmental data of the vehicle's driving area in response to the vehicle turning on the automatic driving function, and control the vehicle to automatically drive according to the planned driving route.
[0097] The risk judgment module 32 is used to determine whether there is a risk when the vehicle is automatically driving.
[0098] The data uploading module 33 is used to define the environmental data collected within a preset uploading period starting from the time when the risk occurs as recorded data in response to determining that there is a risk during the automatic driving of the vehicle, and upload the recorded data to a storage device in the cloud.
[0099] Fourth embodiment
[0100] Based on the second embodiment of the present application, in this embodiment, a cloud risk detection device is provided, referring to Figure 4 , the device comprises:
[0101] The data reading module 41 is used to read the recorded data after receiving the recorded data corresponding to the risk of the vehicle during automatic driving sent by the vehicle in the cloud.
[0102] The risk index evaluation module 42 is used to compare the recorded data with preset risk comparison data, and determine a risk index used to characterize the severity of the risk corresponding to the recorded data according to the comparison result.
[0103] The optimization module 43 is used to optimize the algorithm model used in the vehicle for collecting environmental data, planning driving routes and controlling automatic driving if it is determined that the risk index is greater than a preset first risk index threshold, and determine the optimized second risk index threshold.
[0104] The updating module 44 is configured to send algorithm model update data to the vehicle and transmit the recorded data to a risk environment database if it is determined that the risk index is not greater than the second risk index threshold.
[0105] Fifth embodiment
[0106] Based on the same inventive concept as the above embodiments, the present application provides a risk detection system. Figure 5 , is a schematic diagram of the workflow of a risk detection system provided in an embodiment of the present application, and the risk detection system can be implemented in software and / or hardware, such as electronic devices such as vehicle-mounted terminals. The system includes the vehicle-side risk detection device and the cloud-side risk detection device provided in the aforementioned embodiment.
[0107] like Figure 5 As shown, step S501, in response to the vehicle turning on the automatic driving function, continuously collects environmental data of the driving area of the vehicle, and controls the vehicle to automatically drive according to the planned driving route, and then continues to execute step S502; step S502, determines whether there is a risk when the vehicle is automatically driving, if so, continues to execute step S503, if not, executes step S502 again; step S503, defines the environmental data collected within a preset upload period starting from the time when the risk occurs as recorded data, uploads the recorded data to a cloud storage device, and then continues to execute step S504; step S504, reads the recorded data uploaded to the cloud storage device, and then continues to execute step S505; step S505, determines whether the recorded data matches the preset risk comparison data, if so, executes step S512, if not, continues to execute step S506; step S506 determines the data used to characterize the comparison with the recorded data Whether the risk index of the severity of the corresponding risk is greater than the preset first risk index threshold, if so, continue to step S507, if not, continue to step S512; step S507, determine whether to optimize the algorithm model from the acquisition end, if so, continue to step S508, if otherwise, execute step S509; step S508, optimize the algorithm model from the acquisition end, and determine the optimized second risk index threshold, and then execute step S510; step S509, optimize the algorithm model from the regulation and control end, and determine the optimized second risk index threshold, and then continue to step S510; step S510, determine whether the risk index is greater than the second risk index threshold, if so, execute step S507 again, if otherwise, continue to step S511; step S511, send algorithm model update data to the vehicle, and then continue to step S512; step S512, transmit the recorded data to the risk environment database.
[0108] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other. Figure 2 In step S202, Figure 5 The specific implementation is through step S505 and step S506; Figure 2 In step S203, Figure 5 The specific implementation is through step S507, step S508, step S509 and step S510; reference Figure 2 In step S204, Figure 5 It is specifically implemented through step S511 and step S512.
[0109] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0110] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A risk detection method, applied to a vehicle, characterized in that: The following steps are involved: S101, in response to the vehicle turning on the automatic driving function, continuously collecting environmental data of the driving area of the vehicle, and controlling the vehicle to automatically drive according to the planned driving route; S102, determining whether there is a risk when the vehicle is automatically driving; S103: If yes, the environmental data collected within a preset upload period starting from when the risk occurs is defined as recorded data, and the recorded data is uploaded to a storage device in the cloud.
2. The method according to claim 1, characterized in that The step S101 includes: In response to the vehicle starting the automatic driving function, continuously collecting the environmental data through the vehicle-mounted collection device of the vehicle, and preprocessing the environmental data to obtain preprocessed data; Performing prediction processing on the preprocessed data to obtain predicted data; Performing fusion processing on the predicted data to obtain fused data; The driving route of the vehicle is planned according to the fused data, and the automatic driving of the vehicle is controlled according to the driving route.
3. The method according to claim 2, characterized in that The step S102 includes: Determine whether the hardware data of the acquisition device is at risk, determine whether the environmental data is at risk, determine whether the predicted data is at risk, determine whether the fused data is at risk, and determine whether the driving route is at risk.
4. The method according to claim 3, characterized in that The step S103 includes: If there is a risk in at least one of the hardware data, the environmental data, the predicted data, the fusion data, and the driving route, the environmental data collected within a preset upload time period starting from the time when the risk occurs is defined as recorded data, and the recorded data is transmitted to the storage device in the cloud.
5. The method according to claim 1, characterized in that After step S103, the method further includes: S104, in response to receiving the algorithm model update data sent from the cloud, updating the algorithm model in the vehicle for collecting environmental data, planning driving routes, and controlling automatic driving.
6. A risk detection method, applied in the cloud, characterized in that: The following steps are involved: S201, after receiving, in the cloud, recorded data sent by the vehicle and corresponding to the risk existing in the automatic driving of the vehicle, reading the recorded data; S202, comparing the recorded data with preset risk comparison data, and determining a risk index for characterizing the severity of the risk corresponding to the recorded data according to the comparison result; S203, if it is determined that the risk index is greater than a preset first risk index threshold, optimizing an algorithm model used in the vehicle for collecting environmental data, planning a driving route, and controlling automatic driving, and determining the optimized second risk index threshold; S204: If it is determined that the risk index is not greater than the second risk index threshold, algorithm model update data is sent to the vehicle, and the recorded data is transmitted to a risk environment database.
7. The method according to claim 6, characterized in that The step S202 further includes: The recorded data is transmitted to the risk environment database according to the comparison result.
8. The method according to claim 6, characterized in that The step S203 further includes: If it is determined that the risk index is not greater than the first risk index threshold, the recorded data is transmitted to the risk environment database.
9. The method according to claim 6, characterized in that The step S204 further includes: If it is determined that the risk index is greater than the second risk index threshold, step S203 is performed again.
10. A vehicle-side risk detection device, characterized in that: include: A collection and control module, configured to continuously collect environmental data of a driving area of the vehicle in response to the vehicle starting an automatic driving function, and control the vehicle to drive automatically according to a planned driving route; A risk judgment module, used to determine whether there is a risk when the vehicle is driving automatically; The data uploading module is used to, in response to determining that there is a risk during the automatic driving of the vehicle, define the environmental data collected within a preset uploading period starting from the time when the risk occurs as recorded data, and upload the recorded data to a storage device in the cloud.
11. A cloud risk detection device, characterized in that: include: A data reading module, configured to read the recorded data corresponding to the risk of the vehicle during autonomous driving after receiving the recorded data sent by the vehicle in the cloud; A risk index assessment module, used to compare the recorded data with preset risk comparison data, and determine a risk index used to characterize the severity of the risk corresponding to the recorded data according to the comparison result; an optimization module, configured to optimize an algorithm model used in the vehicle for collecting environmental data, planning a driving route, and controlling automatic driving if it is determined that the risk index is greater than a preset first risk index threshold, and determine the optimized second risk index threshold; An updating module is used to send algorithm model update data to the vehicle and transmit the recorded data to a risk environment database if it is determined that the risk index is not greater than the second risk index threshold.
12. A risk detection system, characterized in that: It includes the vehicle-side risk detection device as described in claim 10 and the cloud-side risk detection device as described in claim 11.
Citation Information
Cited By
Vehicle control method, system and equipment and storage medium
CN121697589A
