Method for iteratively solving false triggering of automatic emergency braking function

Abstract

The invention discloses a method for iteratively solving false triggering of an automatic emergency braking function, and belongs to the technical field of automobiles. The method comprises the steps that 1, data are collected, and a cloud scene library is established; an automatic emergency braking function optimization scheme is designed according to the collected data characteristics and scenes in the false triggering scene library; according to automatic emergency braking function performance confirmation in the automatic emergency braking function optimization scheme, whether automatic emergency braking function optimization or data iteration needs to be carried out again is judged; and according to the data iteration information, determining whether to perform automatic emergency braking function deployment or re-perform cloud scene library establishment. False triggering, positive triggering and non-triggering are distinguished through data acquisition and establishment of a cloud scene library, and the automatic emergency braking function is continuously optimized through performance confirmation of the automatic emergency braking function in different scenes, so that the risk of false triggering of the automatic emergency braking function is solved.

Description

Technical Field

[0001] The invention relates to a method for iteratively solving false triggering of an automatic emergency braking function, belonging to the technical field of automobiles. Background Art

[0002] The automatic emergency braking function is one of the most important functions of advanced driver assistance systems and unmanned driving systems. It improves the safety of vehicle occupants and other road users by detecting potential collisions and automatically intervening in braking. According to a Chinese invention patent (application number: 202310168262.8), a method, device, system and storage medium for developing an automatic emergency braking function are disclosed. The development method includes: obtaining real vehicle data of different AEB user vehicles within a preset time range when the AEB function is triggered, wherein the real vehicle data includes vehicle condition data and trigger scene image data; judging whether the driver of the AEB user vehicle has emergency avoidance behavior when the AEB function is triggered according to the vehicle condition data; if the driver of the AEB user vehicle does not have emergency avoidance behavior when the AEB function is triggered, the trigger scene image data is defined as trigger error scene data; using a number of trigger error scene data and using a vision-based deep learning algorithm, the image-based neural network in the AEB system is trained.

[0003] False triggering is one of the key factors in evaluating the application value of automatic emergency braking function. False triggering will lead to unnecessary braking action, increasing the risk of rear-end collision or traffic jam. False triggering will reduce users' trust in automatic emergency braking function and affect its application in actual driving.

[0004] In addition, an excessively high false triggering rate cannot meet the requirements of the OEM, resulting in failure of review and acceptance. Therefore, only by solving the false triggering problem can the large-scale mass production application of the automatic emergency braking function be promoted. Summary of the invention

[0005] The main purpose of the present invention is to reduce the adverse effects caused by false triggering and realize the large-scale mass production application of the automatic emergency braking function, and a method for iteratively solving the false triggering of the automatic emergency braking function is proposed.

[0006] A method for iteratively resolving false triggering of an automatic emergency braking function, comprising:

[0007] Step 1: Collect data and build a cloud scene library;

[0008] Step 2: Design an optimization plan for the automatic emergency braking function based on the characteristics of the collected data and the scenarios in the false trigger scenario library;

[0009] Step 3: according to the automatic emergency braking function performance confirmation in the automatic emergency braking function optimization plan, determine whether it is necessary to re-optimize the automatic emergency braking function or iterate the data;

[0010] Step 4: Determine whether to deploy the automatic emergency braking function or re-establish the cloud scenario library based on the data iteration information.

[0011] Preferably, data collection includes but is not limited to: image data, perception data and chassis data.

[0012] Preferably, data collection includes two methods: offline and online. Offline data collection first stores the collected data locally, and then uploads the data to the cloud by sending a hard disk or transmitting it through Ethernet.

[0013] Online data collection directly uploads the collected data to the cloud through wireless networks.

[0014] Preferably, the image data is used to identify false triggering, and the rationality of the automatic emergency braking triggering is confirmed based on the video stream;

[0015] Perception data includes but is not limited to the position, speed, acceleration and heading angle of vehicles, pedestrians and specific targets, lane lines, road edges and fences, and other scattered points or polynomials based on scattered point fitting;

[0016] The chassis data includes but is not limited to vehicle speed, steering wheel angle, yaw rate, accelerator pedal, brake pedal, and trigger signal of the automatic emergency braking function.

[0017] Preferably, a cloud-based scene library is established, and a false triggering scene library, a true triggering scene library, and a non-triggering scene library are established.

[0018] Preferably, the false trigger and positive trigger scenario library is established by using the trigger signal of the automatic emergency braking function to perform timed long capture of the image data, perception data and chassis data before and after the trigger, and judging the trigger type by observing the data before and after the trigger. If the automatic emergency braking function should be triggered in this scenario, it is placed in the positive trigger scenario library, otherwise it is placed in the false trigger scenario library.

[0019] Preferably, the non-triggering scene library is established by intercepting the remaining image data, perception data and chassis data after the false triggering and positive triggering scene libraries are formed into the non-triggering scene library.

[0020] Preferably, the automatic emergency braking function is optimized, and according to the characteristics of the perception data and chassis data, an optimization solution is designed to solve all false triggering scenarios for the scenarios in the false triggering scenario library.

[0021] Preferably, the method for confirming the performance of the automatic emergency braking function is as follows: a. for the false triggering and positive triggering scenario libraries, it is necessary to ensure that the optimized automatic emergency braking function is still triggered normally; b. for the non-triggering scenario library, it is necessary to ensure that the optimized automatic emergency braking function is still not triggered;

[0022] If a and b are satisfied, jump to data iteration; otherwise, jump to re-optimization of the automatic emergency braking function.

[0023] The purpose of the present invention can be achieved by adopting the following technical solutions: according to the method of iteratively solving the false triggering of the automatic emergency braking function of the present invention, false triggering, positive triggering and non-triggering are distinguished through data collection and establishment of a cloud scenario library, and the automatic emergency braking function is continuously optimized through automatic emergency braking function performance confirmation in different scenarios to solve the risk of false triggering of the automatic emergency braking function, reduce unnecessary braking actions and the risk of the vehicle being rear-ended or traffic jams; it also enhances users' trust in the automatic emergency braking function and enhances its application in actual driving; the reduction in the false triggering rate can meet the requirements of the main engine manufacturer, and can realize and promote the large-scale mass production application of the automatic emergency braking function. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Flow chart of a method for iteratively resolving false triggering of an automatic emergency braking function according to the present invention. DETAILED DESCRIPTION

[0025] The following describes the implementation methods of the present invention through specific specific embodiments, and those skilled in the art can fully understand other advantages and technical effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through different specific implementation methods, and the various details in this specification can also be applied based on different viewpoints, and various modifications or changes can be made without deviating from the overall design concept of the invention. It should be noted that the following embodiments and the features in the embodiments can be combined with each other in the absence of conflict. The following exemplary embodiments of the present invention can be implemented in a variety of different forms and should not be interpreted as being limited to the specific embodiments described herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical aspects of these exemplary specific embodiments to those skilled in the art.

[0026] Embodiment 1:

[0027] like Figure 1 As shown, the present invention provides a method for iteratively solving the false triggering of the automatic emergency braking function, comprising:

[0028] Step 1: Collect data and establish a cloud scene library;

[0029] Data collection includes: image data, perception data and chassis data. Image data is used to identify false triggers and confirm the rationality of automatic emergency braking triggering based on the video stream. Perception data includes the position, speed, acceleration and heading angle of vehicles, pedestrians and specific targets, lane lines, road edges and fences and other scattered points or polynomials based on scattered point fitting. Chassis data includes vehicle speed, steering wheel angle, yaw rate, accelerator pedal, brake pedal and trigger signal of automatic emergency braking function.

[0030] Data collection includes two methods: offline and online. Offline data collection first stores the collected data locally, and then uploads the data to the cloud by sending a hard disk or transmitting it through Ethernet. Online data collection directly uploads the collected data to the cloud through a wireless network.

[0031] Step 2: Design an optimization plan for the automatic emergency braking function based on the characteristics of the collected data and the scenarios in the false trigger scenario library;

[0032] The false trigger and positive trigger scenario library is established. Through the trigger signal of the automatic emergency braking function, the image data, perception data and chassis data before and after the trigger are intercepted for a long time. By observing the image video stream, combined with the perception data and chassis data, the trigger type is determined. If the automatic emergency braking function should be triggered in this scenario, it is placed in the positive trigger scenario library, otherwise it is placed in the false trigger scenario library.

[0033] The non-triggering scene library is established. The remaining image data, perception data and chassis data after intercepting the false triggering and positive triggering scene libraries form the non-triggering scene library.

[0034] Automatic emergency braking function optimization: Based on the characteristics of perception data and chassis data, we design optimization solutions to solve all false triggering scenarios in the false triggering scenario library.

[0035] Step 3: according to the automatic emergency braking function performance confirmation in the automatic emergency braking function optimization plan, determine whether it is necessary to re-optimize the automatic emergency braking function or iterate the data;

[0036] aFor the positive triggering scenario library, it is necessary to ensure that the optimized automatic emergency braking function is still triggered normally.

[0037] bFor the non-triggering scenario library, it is necessary to ensure that the optimized automatic emergency braking function is still not triggered.

[0038] If a and b are satisfied, jump to data iteration; otherwise, jump to re-optimization of the automatic emergency braking function.

[0039] Step 4: Determine whether to deploy the automatic emergency braking function or re-establish the cloud scenario library based on the data iteration information;

[0040] When adding new collected data, you need to return to re-establish the cloud scene library, otherwise jump to the automatic emergency braking function deployment.

[0041] To sum up, in this embodiment, according to the method for iteratively solving the false triggering of the automatic emergency braking function of this embodiment, the method for iteratively solving the false triggering of the automatic emergency braking function provided by this embodiment solves the risk of false triggering of the automatic emergency braking function by continuously optimizing the automatic emergency braking function through automatic emergency braking function performance confirmation in different scenarios, reduces unnecessary braking actions and the risk of the vehicle being rear-ended or stuck in traffic; it also enhances the user's trust in the automatic emergency braking function and enhances its application in actual driving; the reduction in the false triggering rate can meet the requirements of the OEM, and can realize and promote the large-scale mass production application of the automatic emergency braking function.

[0042] The above description is only a further embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and concepts of the present invention within the scope disclosed by the present invention, which belong to the protection scope of the present invention.

Claims

1. A method for iteratively resolving false triggering of an automatic emergency braking function, characterized in that: include: Step 1: Collect data and establish a cloud scene library; Step 2: Design an optimization plan for the automatic emergency braking function based on the characteristics of the collected data and the scenarios in the false trigger scenario library; Step 3: according to the automatic emergency braking function performance confirmation in the automatic emergency braking function optimization plan, determine whether it is necessary to re-optimize the automatic emergency braking function or iterate the data; Step 4: Determine whether to deploy the automatic emergency braking function or re-establish the cloud scenario library based on the data iteration information.

2. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 1, characterized in that: Data collection includes but is not limited to: image data, perception data, and chassis data.

3. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 1, characterized in that: Data collection includes two methods: offline and online. Offline data collection first stores the collected data locally, and then uploads the data to the cloud by sending a hard disk or transmitting it via Ethernet. Online data collection directly uploads the collected data to the cloud through wireless networks.

4. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 2, characterized in that: Image data is used to identify false triggers and confirm the rationality of automatic emergency braking triggering based on the video stream; Perception data includes but is not limited to the position, speed, acceleration and heading angle of vehicles, pedestrians and specific targets, lane lines, road edges and fences, and other scattered points or polynomials based on scattered point fitting; The chassis data includes but is not limited to vehicle speed, steering wheel angle, yaw rate, accelerator pedal, brake pedal, and trigger signal of the automatic emergency braking function.

5. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 1, characterized in that: The cloud scene library is established, and the false triggering, positive triggering and non-triggering scene libraries are established.

6. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 5, characterized in that: The false trigger and positive trigger scenario library is established by using the trigger signal of the automatic emergency braking function to perform timed long capture of the image data, perception data and chassis data before and after the trigger. By observing the data before and after the trigger, the trigger type is judged. If the automatic emergency braking function should be triggered in this scenario, it is placed in the positive trigger scenario library, otherwise it is placed in the false trigger scenario library.

7. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 5, characterized in that: The non-triggering scene library is established by intercepting the remaining image data, perception data and chassis data after the false triggering and positive triggering scene libraries are formed into a non-triggering scene library.

8. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 1, characterized in that: The automatic emergency braking function is optimized. Based on the characteristics of perception data and chassis data, optimization solutions are designed to solve all false triggering scenarios in the false triggering scenario library.

9. The method for iteratively resolving false triggering of automatic emergency braking function according to claim 1, characterized in that: The method for confirming the performance of the automatic emergency braking function is as follows: a. For the false triggering and positive triggering scenario libraries, it is necessary to ensure that the optimized automatic emergency braking function is still triggered normally; b. For the non-triggering scenario library, it is necessary to ensure that the optimized automatic emergency braking function is still not triggered; If a and b are satisfied, jump to data iteration; otherwise, jump to re-optimization of the automatic emergency braking function.

Images