Bumpy pavement lane line fusion method and device, equipment and storage medium

Abstract

The invention discloses a bumpy road surface lane line fusion method, device and equipment and a storage medium, and the method comprises the steps: obtaining the historical lane line data of a current scene and the power spectrum density of the real-time vertical vibration of a current vehicle, and carrying out the recognition of a bumpy working condition according to the power spectrum density; acquiring a jolting type of the jolting working condition, and determining a corresponding jolting scene lane line fusion strategy according to the jolting type; a lane line fusion result is generated according to the bumpy scene lane line fusion strategy, the current vehicle is controlled to slow down according to the lane line fusion result and the corresponding deceleration signal, the situation that the perceptibility cannot be handled due to the fact that vibration is too large when the current vehicle runs on a bumpy road surface can be avoided, various short-time and continuous complex bumpy working conditions can be handled, and the driving experience of the current vehicle is improved. The recognition precision of the bumpy road surface is ensured, the comfort and safety of vehicle driving are improved, and the speed and efficiency of bumpy road surface lane line fusion are improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle assisted driving technology, and in particular to a method, device, equipment and storage medium for merging lane lines on a bumpy road surface. Background Art

[0002] Commercial vehicles generally experience much greater vibration than passenger cars due to their heavy weight and stiff suspension design. For commercial vehicle assisted driving, excessive vibration can cause instability in the visual recognition of lane lines, leading to abnormal lateral control input and posing a significant safety risk. The existing technical solutions usually add anti-shake strategies through perception technology to meet actual usage needs.

[0003] In CN202311441006.8, by obtaining the position and target information at the first moment, combining the vehicle driving information and the position at the first moment, the target state at the second moment is predicted, so that the virtual scene displayed in the virtual image area is consistent with the real scene environment where the vehicle is actually located. Its core goal is to enhance the experience.

[0004] CN201810929374.X uses an onboard camera to obtain multiple frames of images of the vehicle's surroundings; determines a first jitter of the vehicle based on the multiple frames of images; determines a second jitter of the vehicle using an inertial measurement unit (IMU); and fuses the first and second jitters to determine the vehicle's jitter. By determining the amount of vehicle jitter, errors caused by jitter can be compensated during monocular vision ranging, thereby achieving more accurate ranging.

[0005] Existing technologies mainly implement anti-shake processing by posture prediction at the sensing end to stabilize image quality and improve perception accuracy. However, there are the following problems: 1. When this type of technology responds to continuous jitter or bumps, the lack of stable input will result in no comparison frames.

[0006] 2. When the working conditions are extremely bumpy, excessive jitter on the sensing end may cause back-end processing failure. Summary of the Invention

[0007] The main purpose of the present invention is to provide a method, device, equipment and storage medium for fusion of lane lines on bumpy roads, aiming to solve the technical problems in the existing technology that the existing technology cannot cope with short-term, continuous bumpy conditions, the recognition accuracy of complex bumpy roads is poor, and excessive vibration is prone to occur, resulting in the inability of the perception end to cope.

[0008] In a first aspect, the present invention provides a method for merging lane lines on a bumpy road surface, the method comprising the following steps: Obtaining historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification based on the power spectrum density; Obtaining a bump type of the bumpy working condition, and determining a corresponding bumpy scene lane line fusion strategy according to the bump type; A lane line fusion result is generated according to the bumpy scene lane line fusion strategy, and the current vehicle is controlled to decelerate according to the lane line fusion result and a corresponding deceleration signal.

[0009] Optionally, acquiring historical lane line data of the current scene and a power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification according to the power spectrum density, includes: Obtain historical lane line data for the current scene, and obtain the vertical vibration acceleration data of the current vehicle using the inertial measurement unit (IMU) at a preset recognition cycle; Analyzing the vertical vibration acceleration data to obtain a power spectrum density of the real-time vertical vibration of the current vehicle; The power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0010] Optionally, comparing the power spectrum density with the power spectrum density of a previous frame and identifying the bumpy working condition according to the comparison result includes: intercepting a target power spectrum density below a preset frequency from the power spectrum density, and obtaining a power spectrum density maximum value of the target power spectrum density; The maximum value of the power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0011] Optionally, comparing the maximum value of the power spectrum density with the power spectrum density of a previous frame, and identifying the bumpy working condition according to the comparison result, includes: Comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculating a change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; When the change ratio is greater than a first preset ratio threshold, obtaining a current frame longitudinal acceleration corresponding to the current frame from the longitudinal acceleration data provided by the IMU; When the ratio of the change between the longitudinal acceleration of the current frame and the longitudinal acceleration of the previous frame exceeds a second preset ratio threshold, the absolute value of the longitudinal acceleration of the current frame is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy operating condition flag is valid; After the bumpy operating condition is confirmed, bumpy operating condition identification is performed.

[0012] Optionally, obtaining the bump type of the bumpy working condition and determining a corresponding bumpy scene lane line fusion strategy according to the bump type includes: Obtaining target longitudinal acceleration data of a preset time period of the current frame, and calculating a root mean square value of acceleration based on the target longitudinal acceleration data; Calculating a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of a previous frame; When the first root mean square value change ratio exceeds a preset root mean square value threshold, triggering a ratio calculation of the second frame and the third frame to obtain a second root mean square value change ratio and a third root mean square value change ratio; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, determining that the bump type of the bump working condition is normal bump; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the bumping duration exceeds the preset bumping duration threshold, determining that the bumping type of the bumping operating condition is continuous bumping; When one of the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, determining that the turbulence type of the turbulence operating condition is temporary turbulence, and the turbulence degree of the temporary turbulence is more severe than the turbulence degree of the normal turbulence; The corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type.

[0013] Optionally, determining a corresponding bumpy scene lane line fusion strategy according to the bumpy type includes: determining a vibration level of the current vehicle according to the power spectrum density, and evaluating a confidence level of a current lane line according to the vibration level; When the confidence level is greater than a preset confidence threshold, mathematically fitting the lane line shape of the lane where the current vehicle is located in the preset map using a cubic polynomial function; When the bump type is the common bump, calculating the difference between the cubic polynomial coefficients obtained by fitting and the fitting coefficients of the current vehicle camera detection result; When all coefficient differences are less than a preset difference threshold, the lane line data detected in real time by the current vehicle camera is used; when all coefficient differences are not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line formed by the fusion of the lane line of the previous frame and the lane line of the map is output; When the bump type is the continuous bump or the temporary bump, find data from the historical lane line data where the comparison data between the lane line of the historical frame and the lane line of the preset map is less than the preset comparison threshold, and obtain the target time point; The current state lane line of the current vehicle is obtained by calculating the current state lane line of the current vehicle based on the wheel speed of the current vehicle and the position of the lane line in the historical frame at the target time point.

[0014] Optionally, generating a lane line fusion result according to the bumpy scene lane line fusion strategy, and controlling the current vehicle to decelerate according to the lane line fusion result and a corresponding deceleration signal includes: Generate a lane line fusion result according to the lane line fusion strategy for the bumpy scene; When the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, controlling the current vehicle to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; When the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0015] In a second aspect, to achieve the above-mentioned purpose, the present invention further provides a device for merging lane lines on a bumpy road surface, the device comprising: A bumpy working condition recognition module is used to obtain historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and perform bumpy working condition recognition based on the power spectrum density; a strategy determination module, configured to obtain a bump type of the bumpy working condition and determine a corresponding bumpy scene lane line fusion strategy according to the bump type; A deceleration control module is used to generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the deceleration of the current vehicle according to the lane line fusion result and the corresponding deceleration signal.

[0016] On the third aspect, in order to achieve the above-mentioned purpose, the present invention also proposes a bumpy road lane line fusion device, which includes: a memory, a processor, and a bumpy road lane line fusion program stored in the memory and runnable on the processor, and the bumpy road lane line fusion program is configured to implement the steps of the bumpy road lane line fusion method described above.

[0017] In a fourth aspect, in order to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which a bumpy road lane line fusion program is stored. When the bumpy road lane line fusion program is executed by a processor, the steps of the bumpy road lane line fusion method described above are implemented.

[0018] The method for fusion of lane lines on bumpy roads proposed in the present invention obtains historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and identifies the bumpy working condition according to the power spectrum density; obtains the bump type of the bumpy working condition, and determines the corresponding bumpy scene lane line fusion strategy according to the bump type; generates a lane line fusion result according to the bumpy scene lane line fusion strategy, and controls the deceleration of the current vehicle according to the lane line fusion result and the corresponding deceleration signal. This can avoid the situation where the perception cannot cope with excessive vibration when driving on a bumpy road, can cope with various short-term and continuous complex bumpy working conditions, ensures the recognition accuracy of bumpy roads, improves the comfort and safety of vehicle driving, and improves the speed and efficiency of lane line fusion on bumpy roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention; Figure 2 This is a flowchart of a first embodiment of a method for merging lane lines on a bumpy road according to the present invention; Figure 3 This is a flow chart of a second embodiment of the method for merging lane lines on a bumpy road surface according to the present invention; Figure 4 This is a flowchart of a third embodiment of the method for merging lane lines on a bumpy road according to the present invention; Figure 5 This is a flowchart of a fourth embodiment of a method for merging lane lines on a bumpy road according to the present invention; Figure 6 This is a flowchart of a fifth embodiment of a method for merging lane lines on a bumpy road according to the present invention; Figure 7 This is a functional module diagram of the first embodiment of the bumpy road lane line fusion device of the present invention.

[0020] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] The solution of the embodiment of the present invention is mainly: by obtaining the historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, the bumpy working condition is identified according to the power spectrum density; the bump type of the bumpy working condition is obtained, and the corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type; the lane line fusion result is generated according to the bumpy scene lane line fusion strategy, and the current vehicle is controlled to decelerate according to the lane line fusion result and the corresponding deceleration signal. This can avoid the situation where the perception cannot cope with excessive vibration when driving on a bumpy road surface, can cope with various short-term and continuous complex bumpy working conditions, ensure the recognition accuracy of bumpy roads, improve the comfort and safety of vehicle driving, and improve the speed and efficiency of lane line fusion on bumpy roads, solving the technical problems in the prior art that the existing technology cannot cope with short-term and continuous bumpy working conditions, the recognition accuracy of complex bumpy roads is poor, and it is easy for the perception end to be unable to cope with excessive vibration.

[0023] Reference Figure 1 , Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0024] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to implement communication between these components. The user interface 1003 may include a display and an input unit such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory, such as a disk storage. The memory 1005 may also be a storage device independent of the processor 1001.

[0025] Those skilled in the art will understand that Figure 1 The device structure shown in the figure does not constitute a limitation of the device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0026] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating device, a network communication module, a user interface module, and a bumpy road lane line fusion program.

[0027] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001 and performs the following operations: Obtaining historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification based on the power spectrum density; Obtaining a bump type of the bumpy working condition, and determining a corresponding bumpy scene lane line fusion strategy according to the bump type; A lane line fusion result is generated according to the bumpy scene lane line fusion strategy, and the current vehicle is controlled to decelerate according to the lane line fusion result and a corresponding deceleration signal.

[0028] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: Obtain historical lane line data for the current scene, and obtain the vertical vibration acceleration data of the current vehicle using the inertial measurement unit (IMU) at a preset recognition cycle; Analyzing the vertical vibration acceleration data to obtain a power spectrum density of the real-time vertical vibration of the current vehicle; The power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0029] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: intercepting a target power spectrum density below a preset frequency from the power spectrum density, and obtaining a power spectrum density maximum value of the target power spectrum density; The maximum value of the power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0030] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: Comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculating a change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; When the change ratio is greater than a first preset ratio threshold, obtaining a current frame longitudinal acceleration corresponding to the current frame from the longitudinal acceleration data provided by the IMU; When the ratio of the change between the longitudinal acceleration of the current frame and the longitudinal acceleration of the previous frame exceeds a second preset ratio threshold, the absolute value of the longitudinal acceleration of the current frame is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy operating condition flag is valid; After the bumpy operating condition is confirmed, bumpy operating condition identification is performed.

[0031] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: Obtaining target longitudinal acceleration data of a preset time period of the current frame, and calculating a root mean square value of acceleration based on the target longitudinal acceleration data; Calculating a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of a previous frame; When the first root mean square value change ratio exceeds a preset root mean square value threshold, triggering a ratio calculation of the second frame and the third frame to obtain a second root mean square value change ratio and a third root mean square value change ratio; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, determining that the bump type of the bump working condition is normal bump; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the bumping duration exceeds the preset bumping duration threshold, determining that the bumping type of the bumping operating condition is continuous bumping; When one of the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, determining that the turbulence type of the turbulence operating condition is temporary turbulence, and the turbulence degree of the temporary turbulence is more severe than the turbulence degree of the normal turbulence; The corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type.

[0032] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: determining a vibration level of the current vehicle according to the power spectrum density, and evaluating a confidence level of a current lane line according to the vibration level; When the confidence level is greater than a preset confidence threshold, mathematically fitting the lane line shape of the lane where the current vehicle is located in the preset map using a cubic polynomial function; When the bump type is the common bump, calculating the difference between the cubic polynomial coefficients obtained by fitting and the fitting coefficients of the current vehicle camera detection result; When all coefficient differences are less than a preset difference threshold, the lane line data detected in real time by the current vehicle camera is used; when all coefficient differences are not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line formed by the fusion of the lane line of the previous frame and the lane line of the map is output; When the bump type is the continuous bump or the temporary bump, find data from the historical lane line data where the comparison data between the lane line of the historical frame and the lane line of the preset map is less than the preset comparison threshold, and obtain the target time point; The current state lane line of the current vehicle is obtained by calculating the current state lane line of the current vehicle based on the wheel speed of the current vehicle and the position of the lane line in the historical frame at the target time point.

[0033] The device of the present invention calls the bumpy road lane line fusion program stored in the memory 1005 through the processor 1001, and further performs the following operations: Generate a lane line fusion result according to the lane line fusion strategy for the bumpy scene; When the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, controlling the current vehicle to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; When the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0034] This embodiment adopts the above scheme to obtain historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and identify the bumpy working condition according to the power spectrum density; obtain the bump type of the bumpy working condition, and determine the corresponding bumpy scene lane line fusion strategy according to the bump type; generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the deceleration of the current vehicle according to the lane line fusion result and the corresponding deceleration signal. This can avoid the situation where the perception cannot cope with excessive vibration when driving on bumpy roads, can cope with various short-term and continuous complex bumpy working conditions, ensure the recognition accuracy of bumpy roads, improve the comfort and safety of vehicle driving, and improve the speed and efficiency of lane line fusion on bumpy roads.

[0035] Based on the above hardware structure, an embodiment of the method for fusion of lane lines on bumpy roads of the present invention is proposed.

[0036] Reference Figure 2 , Figure 2 Schematic diagram of the flow of the first embodiment of the method for merging lane lines on a bumpy road according to the present invention.

[0037] In a first embodiment, the method for merging lane lines on a bumpy road surface includes the following steps: Step S10: Acquire historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and perform bumpy working condition recognition based on the power spectrum density.

[0038] It should be noted that the historical lane line data of the current scene and the power spectral density (PSD) of the real-time vertical vibration of the current vehicle are obtained, and then the bumpy working condition can be identified based on the power spectral density to identify which bumpy working condition the current vehicle is in.

[0039] Step S20: Obtain the bump type of the bumpy working condition, and determine the corresponding bumpy scene lane line fusion strategy according to the bump type.

[0040] It should be understood that after obtaining the bump type of the bumpy working condition, the corresponding bumpy scene lane line fusion strategy can be determined according to the bump type, and the bumpy scene lane line fusion strategy is a lane line fusion strategy corresponding to different bumpy scenes.

[0041] Step S30: Generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the current vehicle to decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0042] It can be understood that different lane line fusion strategies for bumpy scenarios correspond to different lane line fusion results and different deceleration signals. According to the lane line fusion strategy for bumpy scenarios, a lane line fusion result can be generated, and then the current vehicle can be controlled to decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0043] This embodiment adopts the above scheme to obtain historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and identify the bumpy working condition according to the power spectrum density; obtain the bump type of the bumpy working condition, and determine the corresponding bumpy scene lane line fusion strategy according to the bump type; generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the deceleration of the current vehicle according to the lane line fusion result and the corresponding deceleration signal. This can avoid the situation where the perception cannot cope with excessive vibration when driving on bumpy roads, can cope with various short-term and continuous complex bumpy working conditions, ensure the recognition accuracy of bumpy roads, improve the comfort and safety of vehicle driving, and improve the speed and efficiency of lane line fusion on bumpy roads.

[0044] Further, Figure 3 This is a flow chart of the second embodiment of the method for merging lane lines on a bumpy road according to the present invention. Figure 3As shown, a second embodiment of the method for merging lane lines on a bumpy road of the present invention is proposed based on the first embodiment. In this embodiment, step S10 specifically includes the following steps: Step S11: Acquire historical lane line data of the current scene, and acquire vertical vibration acceleration data of the current vehicle through an inertial measurement unit (IMU) with a preset recognition period.

[0045] It should be noted that historical lane line data of the current scene can be obtained from a database, storage device or storage medium, and then the vertical vibration data of the current vehicle can be obtained through an inertial measurement unit (IMU) with a preset recognition period.

[0046] Step S12: Analyze the vertical vibration acceleration data to obtain the power spectrum density of the real-time vertical vibration of the current vehicle.

[0047] It is understandable that, by analyzing the vertical vibration acceleration data, the power spectrum density of the real-time vertical vibration of the current vehicle can be obtained.

[0048] In a specific implementation, the power spectrum density of the vertical vibration can be analyzed in real time using 1s of data.

[0049] Step S13: Compare the power spectrum density with the power spectrum density of the previous frame, and identify the bumpy working condition based on the comparison result.

[0050] It should be understood that after comparing the power spectrum density with the power spectrum density of the previous frame, a comparison result is generated, and the bumpy working condition can be identified based on the comparison result.

[0051] In the specific implementation, the power spectrum density is compared with the power spectrum density of the previous frame, and combined with the vertical and longitudinal pulse excitation, it can be determined whether the current working condition is bumpy.

[0052] Through the above scheme, this embodiment obtains historical lane line data of the current scene, obtains the vertical vibration acceleration data of the current vehicle through the inertial measurement unit (IMU) with a preset recognition period; analyzes the vertical vibration acceleration data to obtain the power spectrum density of the real-time vertical vibration of the current vehicle; compares the power spectrum density with the power spectrum density of the previous frame, and identifies bumpy conditions based on the comparison results. This can quickly identify bumpy conditions and improve the speed and efficiency of lane line fusion on bumpy roads.

[0053] Further, Figure 4 This is a flow chart of the third embodiment of the method for merging lane lines on a bumpy road according to the present invention. Figure 4As shown, a third embodiment of the method for merging lane lines on a bumpy road surface of the present invention is proposed based on the second embodiment. In this embodiment, step S13 specifically includes the following steps: Step S131 : intercepting a target power spectrum density below a preset frequency from the power spectrum density, and obtaining a maximum power spectrum density of the target power spectrum density.

[0054] It should be noted that the preset frequency is a pre-set power spectrum screening frequency, and the preset frequency can be used to screen out the power spectrum density below the preset frequency as the target power spectrum density, thereby obtaining the maximum power spectrum density of the target power spectrum density.

[0055] Step S132: Compare the maximum value of the power spectrum density with the power spectrum density of the previous frame, and identify the bumpy working condition based on the comparison result.

[0056] It is understandable that after comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, the bumpy working condition can be identified based on the comparison result.

[0057] In specific implementation, the vertical vibration acceleration data provided by the IMU can be used to analyze and calculate the power spectrum density based on the data 1 second before the current frame, intercept the data below 15 Hz, and obtain the maximum power spectrum density within 15 Hz (in actual applications, the natural frequency corresponding to the commercial vehicle suspension is generally below 5 Hz). The ratio with the previous frame is calculated. When the ratio is greater than the set threshold a1 (1.5), the bumpy working condition is confirmed.

[0058] Furthermore, the step S132 specifically includes the following steps: Comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculating a change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; When the change ratio is greater than a first preset ratio threshold, obtaining a current frame longitudinal acceleration corresponding to the current frame from the longitudinal acceleration data provided by the IMU; When the ratio of the change between the longitudinal acceleration of the current frame and the longitudinal acceleration of the previous frame exceeds a second preset ratio threshold, the absolute value of the longitudinal acceleration of the current frame is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy operating condition flag is valid; After the bumpy operating condition is confirmed, bumpy operating condition identification is performed.

[0059] It should be understood that by calculating the change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame, and comparing the change ratio with the corresponding threshold, the current frame longitudinal acceleration corresponding to the current frame can be obtained from the longitudinal acceleration data provided by the IMU, and then when the change ratio between the current frame longitudinal acceleration and the previous frame longitudinal acceleration exceeds the second preset ratio threshold, the absolute value of the current frame longitudinal acceleration is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy condition flag position and the bumpy condition is valid. After the bumpy condition is confirmed, the bumpy condition is identified.

[0060] In the specific implementation, the current frame data is collected through the longitudinal acceleration data provided by the IMU. If the change ratio of the longitudinal acceleration of the current frame relative to the previous frame exceeds the set threshold a2 (1.5), and its longitudinal acceleration absolute value is greater than the set threshold b1 (0.8g), and the vertical acceleration absolute value is greater than the set threshold b2 (0.8g), then it is considered that the current bumpy working condition flag is valid.

[0061] Through the above scheme, this embodiment intercepts the target power spectrum density below the preset frequency from the power spectrum density and obtains the power spectrum density maximum value of the target power spectrum density; compares the power spectrum density maximum value with the power spectrum density of the previous frame, and identifies bumpy working conditions based on the comparison results. This can quickly identify bumpy working conditions and improve the speed and efficiency of lane line fusion on bumpy roads.

[0062] Further, Figure 5 This is a flow chart of the fourth embodiment of the method for merging lane lines on a bumpy road according to the present invention. Figure 5 As shown, a fourth embodiment of the method for merging lane lines on a bumpy road surface of the present invention is proposed based on the first embodiment. In this embodiment, step S20 specifically includes the following steps: Step S21 : obtaining target longitudinal acceleration data of a preset time period of a current frame, and calculating a root mean square value of acceleration based on the target longitudinal acceleration data.

[0063] It should be noted that after the target longitudinal acceleration data of the preset time period of the current frame is obtained, the acceleration root mean square value can be calculated based on the target longitudinal acceleration data.

[0064] Step S22: Calculate a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of the previous frame.

[0065] It is understandable that after the acceleration root mean square value is obtained, a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of the previous frame may be calculated.

[0066] Step S23: When the first RMS value change ratio exceeds a preset RMS value threshold, triggering the calculation of the ratio of the second frame to the third frame to obtain a second RMS value change ratio and a third RMS value change ratio.

[0067] It should be understood that when the first RMS value change ratio exceeds a preset RMS value threshold, the second frame and third frame ratio calculation can be triggered to obtain the second RMS value change ratio and the third RMS value change ratio.

[0068] In the specific implementation, the longitudinal acceleration data of 1s before the current frame is used to calculate the RMS value of acceleration during this period of time and save several frames. 1avg The root mean square value of acceleration relative to the previous frame a 0avg Change ratio 1 (ratio = a 1avg / a 0avg ) exceeds the set threshold c1, triggering the calculation of the ratio of the second and third frames.

[0069] The calculation formula is as follows: The second frame changes the ratio 2 = a 2avg / a 0avg The third frame changes the ratio 3 = a 3avg / a 0avg where a 2avg , a 3avg is the RMS value of acceleration of the second and third frames.

[0070] Step S24: When the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, determine that the bump type of the bump condition is normal bump.

[0071] It should be noted that when the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, it can be determined that the bump type of the bump condition is ordinary bump.

[0072] Step S25: When the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the turbulence duration exceeds the preset turbulence duration threshold, it is determined that the turbulence type of the turbulence working condition is continuous turbulence.

[0073] It can be understood that when the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the turbulence duration exceeds the preset turbulence duration threshold, it can be determined that the turbulence type of the turbulence working condition is continuous turbulence.

[0074] Step S26: When one of the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, it is determined that the turbulence type of the turbulence working condition is temporary turbulence, and the turbulence degree of the temporary turbulence is heavier than the turbulence degree of the ordinary turbulence.

[0075] It should be understood that when one of the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, it can be determined that the turbulence type of the turbulence working condition is temporary turbulence, and the turbulence degree of the temporary turbulence is heavier than the turbulence degree of the ordinary turbulence.

[0076] In a specific implementation, if the RMS value change ratios 1, 2, and 3 all exceed the set thresholds, the bumpy condition flag is considered valid and the bumpy condition flag is invalid. Bumpy scenarios can be divided into the following three categories: Scenario 1: Bumpy Scenario (Normal Bumping) When the flag m (bumpy condition) appears in a single frame, it is determined that the scene has entered a bumpy scene. Scenario 2: Continuous Bumping Scenario (Continuous Bumping) When the previous frame flag is m and the current frame flag is M, it is determined that the scene has entered a continuous bumpy scene. Scenario 3: Short-term turbulence scenario (temporary turbulence) When the flag M (bumping condition) appears in a single frame, it is judged that a short-term bumping scene has been entered. Step S27: Determine a corresponding bumpy scene lane line fusion strategy according to the bumpy type.

[0077] It is understandable that different bump types correspond to different lane line fusion strategies for bumpy scenarios.

[0078] Furthermore, the step S27 specifically includes the following steps: determining a vibration level of the current vehicle according to the power spectrum density, and evaluating a confidence level of a current lane line according to the vibration level; When the confidence level is greater than a preset confidence threshold, mathematically fitting the lane line shape of the lane where the current vehicle is located in the preset map using a cubic polynomial function; When the bump type is the common bump, calculating the difference between the cubic polynomial coefficients obtained by fitting and the fitting coefficients of the current vehicle camera detection result; When all coefficient differences are less than a preset difference threshold, the lane line data detected in real time by the current vehicle camera is used; when all coefficient differences are not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line formed by the fusion of the lane line of the previous frame and the lane line of the map is output; When the bump type is the continuous bump or the temporary bump, find data from the historical lane line data where the comparison data between the lane line of the historical frame and the lane line of the preset map is less than the preset comparison threshold, and obtain the target time point; The current state lane line of the current vehicle is obtained by calculating the current state lane line of the current vehicle based on the wheel speed of the current vehicle and the position of the lane line in the historical frame at the target time point.

[0079] It should be noted that the vibration level of the current vehicle is determined based on the power spectral density, the confidence of the current lane line is evaluated based on the vibration level, the confidence of the current lane line recognition is evaluated through the vibration level, and the lane line is integrated with the perceived lane line through high-precision map output data to achieve stable lane line recognition.

[0080] It can be understood that when the confidence level is too low, the lane lines are fused with the map lane lines based on historical lane line information; when the confidence level meets the requirements, the visual lane lines are compared with the map lane lines. If there is a large difference, the fusion strategy is switched. If the difference is small, the visual lane lines captured by the vehicle camera are used.

[0081] It should be understood that when the confidence level is greater than a preset confidence threshold, the lane line topology points of the lane where the current vehicle is located in the preset map are mathematically fitted with a cubic polynomial function to obtain a lane line shape, and then different lane line fusion strategies are adopted according to different bump types.

[0082] In specific implementation, three fusion strategy examples are as follows: 1. Fusion Strategy 1 (Normal Bumping, i.e., Severe Bumping) When the system determines that the scene is bumpy, it first determines the lane line confidence. A. If the lane line confidence is higher than the set threshold, the lane line in the high-precision map is fitted with a cubic polynomial based on its topological points. If the difference between the corresponding C0\C1\C2\C3 is less than the set threshold, the camera lane line is used; B. If the lane line confidence is higher than the set threshold, the lane line in the high-precision map is fitted with a cubic polynomial based on its topological points. When the corresponding C0\C1\C2\C3 difference is higher than the set threshold, the moment when the comparison data between the lane line in the previous frame and the lane line in the high-precision map is less than the set threshold is found. The lane line after the previous frame and the high-precision map are fused (lateral positioning is achieved through the lane line C0 value, and longitudinal positioning is achieved through line matching of the remaining values. The weight of the lane line in the high-precision map is also provided. This mature method is not yet expanded). A level 1 deceleration signal is issued (level 1 deceleration is the coasting state) to reduce vehicle vibration through deceleration.

[0083] 2. Fusion Strategy 2 (Continuous Bumping, i.e. Continuous Bumping Scenario) The continuous bumpy scene is a transition from a relatively bumpy scene. When in a relatively bumpy scene, the fusion strategy 1 is executed. When jumping from a relatively bumpy scene to a bumpy scene, the processing logic is as follows: Because the lane lines in the previous frame are basically unusable in bumpy scenarios, it is necessary to compare the lane lines in the historical frame with the lane lines in the historical high-precision map. The data with a value less than the set threshold is input, and the moment at that time is obtained. The vehicle's position in the historical fused lane lines and the lane lines in the current state are calculated based on the wheel speed, and a level 2 deceleration signal is issued (level 2 deceleration requires the use of the service brake) to reduce vehicle vibration through level 2 deceleration.

[0084] During the bumpy scene, the lane line confidence needs to be continuously monitored. If the confidence is too low for a set period of time, a downgrade request needs to be sent to the assisted driving system.

[0085] 3. Fusion Strategy 3 (Temporary Bumps, i.e. Short-term Bumps) When a bumpy scene is identified, the same processing as fusion strategy 2 is performed.

[0086] It should be noted that the subsequent fusion process with the high-precision map will only be initiated when the confidence level of the lane lines detected by the camera (i.e. the reliability probability of the detection results) is higher than the preset threshold.

[0087] Based on the lane line topology points of the lane in the high-precision map (i.e., the key coordinate points of the lane line recorded in the map), a cubic polynomial function is used to mathematically fit the lane line shape.

[0088] Calculate the difference between the cubic polynomial coefficients (C0, C1, C2, C3) obtained by high-precision map fitting and the fitting coefficients of the camera detection results; If all coefficient differences are less than the set threshold, it means that the lane line shape of the map and the camera are highly matched. In this case, the lane line data detected in real time by the camera is used (probably because the camera data is closer to the real-time environment).

[0089] y = C3·x³ + C2·x² + C1·x + C0 Among them, C0 is a constant term that determines the intercept of the lane line on the y-axis (i.e., the y-coordinate when x=0); if C0 increases, the lane line as a whole shifts in the y-axis direction.

[0090] C1 is a linear coefficient that affects the linear slope of the lane line (the degree of inclination of the approximately straight line portion); the larger C1 is, the greater the slope of the lane line (such as a steeper left or right deviation).

[0091] C2 is the quadratic coefficient, which determines the quadratic curvature of the lane line (such as the opening direction and curvature of the curve); when C2>0, the lane line is a parabola opening upward, and when C2<0, it opens downward.

[0092] C3 is the cubic coefficient, which affects the degree of cubic curve distortion of the lane line (such as the high-order changes in S-shaped curves). When C3 is non-zero, the lane line may appear S-shaped or more complex.

[0093] This embodiment adopts the above scheme, by obtaining the target longitudinal acceleration data of the preset time period of the current frame, calculating the acceleration root mean square value according to the target longitudinal acceleration data; calculating the first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of the previous frame; when the first root mean square value change ratio exceeds the preset root mean square value threshold, triggering the second frame and the third frame ratio calculation, obtaining the second root mean square value change ratio and the third root mean square value change ratio; when the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, determining the bump type of the bump condition as ordinary bump; when the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold value, and the bumpy duration exceeds the preset bumpy duration threshold, the bumpy type of the bumpy working condition is determined to be continuous bumpy; when one of the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the bumpy duration does not exceed the preset bumpy duration threshold, the bumpy type of the bumpy working condition is determined to be temporary bumpy, and the bumpy degree of the temporary bump is heavier than the bumpy degree of the ordinary bump; the corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type, which can avoid the situation where the perception cannot cope with the excessive vibration when driving on a bumpy road, and can cope with various short-term and continuous complex bumpy working conditions, ensure the recognition accuracy of the bumpy road, improve the success rate of the lane line fusion of the bumpy road, and improve the speed and efficiency of the lane line fusion of the bumpy road.

[0094] Further, Figure 6This is a flow chart of the fifth embodiment of the method for merging lane lines on a bumpy road according to the present invention. Figure 6 As shown, a fifth embodiment of the method for merging lane lines on a bumpy road surface of the present invention is proposed based on the first embodiment. In this embodiment, step S30 specifically includes the following steps: Step S31: Generate a lane line fusion result according to the bumpy scene lane line fusion strategy.

[0095] It should be noted that different lane line fusion strategies for bumpy scenarios correspond to different lane line fusion results, and the lane line fusion results are generated according to the lane line fusion strategies for bumpy scenarios.

[0096] Step S32: When the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, the current vehicle is controlled to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0097] It can be understood that when the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, the current vehicle can be controlled to perform level 1 deceleration, that is, coasting deceleration, based on the lane line fusion result and the corresponding deceleration signal.

[0098] Step S33: When the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0099] It should be understood that when the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle can be controlled to perform level 2 deceleration, i.e., braking deceleration, based on the lane line fusion result and the corresponding deceleration signal.

[0100] This embodiment uses the above scheme to generate a lane line fusion result through the bumpy scene lane line fusion strategy; when the bumpy scene lane line fusion strategy is the ordinary bumpy fusion strategy, the current vehicle is controlled to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; when the bumpy scene lane line fusion strategy is the continuous bumpy fusion strategy or the temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal; it can avoid the situation where the perception cannot cope with excessive vibration when driving on a bumpy road, can cope with various short-term and continuous complex bumpy working conditions, ensure the recognition accuracy of bumpy roads, improve the comfort and safety of vehicle driving, and improve the speed and efficiency of lane line fusion on bumpy roads.

[0101] Accordingly, the present invention further provides a device for merging lane lines on bumpy roads.

[0102] Reference Figure 7 , Figure 7 This is a functional module diagram of the first embodiment of the bumpy road lane line fusion device of the present invention.

[0103] In a first embodiment of the device for merging lane lines on a bumpy road surface of the present invention, the device for merging lane lines on a bumpy road surface comprises: The bumpy working condition recognition module 10 is used to obtain historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and perform bumpy working condition recognition based on the power spectrum density.

[0104] The strategy determination module 20 is used to obtain the bump type of the bumpy working condition and determine the corresponding bumpy scene lane line fusion strategy according to the bump type.

[0105] The deceleration control module 30 is configured to generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the current vehicle to decelerate according to the lane line fusion result and a corresponding deceleration signal.

[0106] The bumpy condition recognition module 10 is also used to obtain historical lane line data of the current scene, obtain vertical vibration acceleration data of the current vehicle through the inertial measurement unit IMU with a preset recognition period; analyze the vertical vibration acceleration data to obtain the power spectrum density of the real-time vertical vibration of the current vehicle; compare the power spectrum density with the power spectrum density of the previous frame, and perform bumpy condition recognition based on the comparison result.

[0107] The bumpy working condition identification module 10 is also used to intercept the target power spectrum density below the preset frequency from the power spectrum density, and obtain the power spectrum density maximum value of the target power spectrum density; compare the power spectrum density maximum value with the power spectrum density of the previous frame, and identify the bumpy working condition based on the comparison result.

[0108] The bumpy working condition identification module 10 is further used to compare the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculate the change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; when the change ratio is greater than a first preset ratio threshold, the current frame longitudinal acceleration corresponding to the current frame is obtained from the longitudinal acceleration data provided by the IMU; when the change ratio between the current frame longitudinal acceleration and the previous frame longitudinal acceleration exceeds a second preset ratio threshold, the absolute value of the current frame longitudinal acceleration is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy working condition flag position is valid; after the bumpy working condition is confirmed, the bumpy working condition identification is performed.

[0109] The strategy determination module 20 is further configured to obtain target longitudinal acceleration data for a preset time period of a current frame, calculate a root mean square (RMS) value of acceleration based on the target longitudinal acceleration data, calculate a first RMS value change ratio of the RMS value of acceleration relative to the RMS value of acceleration of a previous frame, trigger a ratio calculation for the second and third frames when the first RMS value change ratio exceeds a preset RMS value threshold, and obtain a second RMS value change ratio and a third RMS value change ratio; and determine that the bump type of the bump condition is normal bump when the first RMS value change ratio, the second RMS value change ratio, and the third RMS value change ratio all exceed the preset RMS value threshold. When the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the bump duration exceeds the preset bump duration threshold, the bump type of the bumpy working condition is determined to be continuous bump; when one of the first root mean square value change ratio, the second root mean square value change ratio and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the bump duration does not exceed the preset bump duration threshold, the bump type of the bumpy working condition is determined to be temporary bump, and the bump degree of the temporary bump is heavier than that of the ordinary bump; the corresponding bump scene lane line fusion strategy is determined according to the bump type.

[0110] The strategy determination module 20 is further configured to determine the vibration level of the current vehicle based on the power spectrum density, and to evaluate the confidence level of the current lane line based on the vibration level; when the confidence level is greater than a preset confidence threshold, mathematically fit the lane line shape of the lane line topology point of the lane where the current vehicle is located in the preset map using a cubic polynomial function; when the bump type is the ordinary bump, calculate the difference between the cubic polynomial coefficient obtained by fitting and the fitting coefficient of the current vehicle camera detection result; when all coefficient differences are less than a preset difference threshold, use the lane line data detected in real time by the current vehicle camera; in the system When the difference between the numbers is not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line after the previous frame lane line and the map lane line are fused is output; when the bump type is the continuous bump or the temporary bump, the data when the comparison data between the historical frame lane line and the preset map lane line is less than the preset comparison threshold is found from the historical lane line data to obtain the target time point; the current state lane line of the current vehicle is obtained by inferring the wheel speed of the current vehicle in combination with the position of the historical frame lane line at the target time point.

[0111] The deceleration control module 30 is further used to generate a lane line fusion result according to the bumpy scene lane line fusion strategy; when the bumpy scene lane line fusion strategy is the ordinary bumpy fusion strategy, the current vehicle is controlled to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; when the bumpy scene lane line fusion strategy is the continuous bumpy fusion strategy or the temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0112] Among them, the steps implemented by each functional module of the bumpy road lane line fusion device can refer to the various embodiments of the bumpy road lane line fusion method of the present invention, and will not be repeated here.

[0113] In addition, an embodiment of the present invention further provides a storage medium storing a bumpy road lane line fusion program. When the bumpy road lane line fusion program is executed by a processor, the following operations are performed: Obtaining historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification based on the power spectrum density; Obtaining a bump type of the bumpy working condition, and determining a corresponding bumpy scene lane line fusion strategy according to the bump type; A lane line fusion result is generated according to the bumpy scene lane line fusion strategy, and the current vehicle is controlled to decelerate according to the lane line fusion result and a corresponding deceleration signal.

[0114] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: Obtain historical lane line data for the current scene, and obtain the vertical vibration acceleration data of the current vehicle using the inertial measurement unit (IMU) at a preset recognition cycle; Analyzing the vertical vibration acceleration data to obtain a power spectrum density of the real-time vertical vibration of the current vehicle; The power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0115] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: intercepting a target power spectrum density below a preset frequency from the power spectrum density, and obtaining a power spectrum density maximum value of the target power spectrum density; The maximum value of the power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

[0116] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: Comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculating a change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; When the change ratio is greater than a first preset ratio threshold, obtaining a current frame longitudinal acceleration corresponding to the current frame from the longitudinal acceleration data provided by the IMU; When the ratio of the change between the longitudinal acceleration of the current frame and the longitudinal acceleration of the previous frame exceeds a second preset ratio threshold, the absolute value of the longitudinal acceleration of the current frame is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy operating condition flag is valid; After the bumpy operating condition is confirmed, bumpy operating condition identification is performed.

[0117] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: Obtaining target longitudinal acceleration data of a preset time period of the current frame, and calculating a root mean square value of acceleration based on the target longitudinal acceleration data; Calculating a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of a previous frame; When the first root mean square value change ratio exceeds a preset root mean square value threshold, triggering a ratio calculation of the second frame and the third frame to obtain a second root mean square value change ratio and a third root mean square value change ratio; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, determining that the bump type of the bump working condition is normal bump; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the bumping duration exceeds the preset bumping duration threshold, determining that the bumping type of the bumping operating condition is continuous bumping; When one of the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, determining that the turbulence type of the turbulence operating condition is temporary turbulence, and the turbulence degree of the temporary turbulence is more severe than the turbulence degree of the normal turbulence; The corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type.

[0118] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: determining a vibration level of the current vehicle according to the power spectrum density, and evaluating a confidence level of a current lane line according to the vibration level; When the confidence level is greater than a preset confidence threshold, mathematically fitting the lane line shape of the lane where the current vehicle is located in the preset map using a cubic polynomial function; When the bump type is the common bump, calculating the difference between the cubic polynomial coefficients obtained by fitting and the fitting coefficients of the current vehicle camera detection result; When all coefficient differences are less than a preset difference threshold, the lane line data detected in real time by the current vehicle camera is used; when all coefficient differences are not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line formed by the fusion of the lane line of the previous frame and the lane line of the map is output; When the bump type is the continuous bump or the temporary bump, find data from the historical lane line data where the comparison data between the lane line of the historical frame and the lane line of the preset map is less than the preset comparison threshold, and obtain the target time point; The current state lane line of the current vehicle is obtained by calculating the current state lane line of the current vehicle based on the wheel speed of the current vehicle and the position of the lane line in the historical frame at the target time point.

[0119] Furthermore, when the bumpy road lane fusion program is executed by the processor, the following operations are also implemented: Generate a lane line fusion result according to the lane line fusion strategy for the bumpy scene; When the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, controlling the current vehicle to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; When the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

[0120] Those skilled in the art will understand that all or part of the steps in the above-mentioned implementation methods can be implemented by instructing related hardware through a program. The program is stored in a storage medium and includes a number of instructions for enabling a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application; and the aforementioned storage medium is a computer-readable storage medium, including: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0121] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0122] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0123] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for merging lane lines on a bumpy road, characterized in that: The bumpy road lane line fusion method includes: Obtaining historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification based on the power spectrum density; Obtaining a bump type of the bumpy working condition, and determining a corresponding bumpy scene lane line fusion strategy according to the bump type; A lane line fusion result is generated according to the bumpy scene lane line fusion strategy, and the current vehicle is controlled to decelerate according to the lane line fusion result and a corresponding deceleration signal.

2. The method for merging lane lines on a bumpy road according to claim 1, wherein: The acquiring of historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and performing bumpy working condition identification according to the power spectrum density, includes: Obtain historical lane line data for the current scene, and obtain the vertical vibration acceleration data of the current vehicle using the inertial measurement unit (IMU) at a preset recognition cycle; Analyzing the vertical vibration acceleration data to obtain a power spectrum density of the real-time vertical vibration of the current vehicle; The power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

3. The method for merging lane lines on a bumpy road as claimed in claim 2, wherein: The comparing the power spectrum density with the power spectrum density of the previous frame and identifying the bumpy working condition according to the comparison result includes: intercepting a target power spectrum density below a preset frequency from the power spectrum density, and obtaining a power spectrum density maximum value of the target power spectrum density; The maximum value of the power spectrum density is compared with the power spectrum density of the previous frame, and the bumpy working condition is identified based on the comparison result.

4. The method for merging lane lines on a bumpy road as claimed in claim 3, wherein: The comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame and identifying the bumpy working condition according to the comparison result includes: Comparing the maximum value of the power spectrum density with the power spectrum density of the previous frame, and calculating a change ratio between the maximum value of the power spectrum density and the power spectrum density of the previous frame; When the change ratio is greater than a first preset ratio threshold, obtaining a current frame longitudinal acceleration corresponding to the current frame from the longitudinal acceleration data provided by the IMU; When the ratio of the change between the longitudinal acceleration of the current frame and the longitudinal acceleration of the previous frame exceeds a second preset ratio threshold, the absolute value of the longitudinal acceleration of the current frame is greater than the first preset acceleration threshold, and the absolute value of the vertical acceleration is greater than the second preset acceleration threshold, it is determined that the current vehicle is in a bumpy operating condition flag is valid; After the bumpy operating condition is confirmed, bumpy operating condition identification is performed.

5. The method for merging lane lines on a bumpy road as claimed in claim 1, wherein: The obtaining of the bump type of the bumpy working condition and determining a corresponding bumpy scene lane line fusion strategy according to the bump type include: Obtaining target longitudinal acceleration data of a preset time period of the current frame, and calculating a root mean square value of acceleration based on the target longitudinal acceleration data; Calculating a first root mean square value change ratio of the acceleration root mean square value relative to the acceleration root mean square value of a previous frame; When the first root mean square value change ratio exceeds a preset root mean square value threshold, triggering a ratio calculation of the second frame and the third frame to obtain a second root mean square value change ratio and a third root mean square value change ratio; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, determining that the bump type of the bump working condition is normal bump; When the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio all exceed the preset root mean square value threshold, and the bumping duration exceeds the preset bumping duration threshold, determining that the bumping type of the bumping operating condition is continuous bumping; When one of the first root mean square value change ratio, the second root mean square value change ratio, and the third root mean square value change ratio exceeds the preset root mean square value threshold, and the turbulence duration does not exceed the preset turbulence duration threshold, determining that the turbulence type of the turbulence operating condition is temporary turbulence, and the turbulence degree of the temporary turbulence is more severe than the turbulence degree of the normal turbulence; The corresponding bumpy scene lane line fusion strategy is determined according to the bumpy type.

6. The method for merging lane lines on a bumpy road as claimed in claim 5, characterized in that: The determining of a corresponding bumpy scene lane line fusion strategy according to the bumpy type includes: determining a vibration level of the current vehicle according to the power spectrum density, and evaluating a confidence level of a current lane line according to the vibration level; When the confidence level is greater than a preset confidence threshold, mathematically fitting the lane line shape of the lane where the current vehicle is located in the preset map using a cubic polynomial function; When the bump type is the common bump, calculating the difference between the cubic polynomial coefficients obtained by fitting and the fitting coefficients of the current vehicle camera detection result; When all coefficient differences are less than a preset difference threshold, the lane line data detected in real time by the current vehicle camera is used; when all coefficient differences are not less than the preset difference threshold, the moment when the comparison data between the lane line of the previous frame and the map lane line of the preset map is less than the preset comparison threshold is found from the historical lane line data, and the lane line formed by the fusion of the lane line of the previous frame and the lane line of the map is output; When the bump type is the continuous bump or the temporary bump, find data from the historical lane line data where the comparison data between the lane line of the historical frame and the lane line of the preset map is less than the preset comparison threshold, and obtain the target time point; The current state lane line of the current vehicle is obtained by calculating the current state lane line of the current vehicle based on the wheel speed of the current vehicle and the position of the lane line in the historical frame at the target time point.

7. The method for merging lane lines on a bumpy road as claimed in claim 1, wherein: Generating a lane line fusion result according to the bumpy scene lane line fusion strategy, and controlling the current vehicle to decelerate according to the lane line fusion result and a corresponding deceleration signal, includes: Generate a lane line fusion result according to the lane line fusion strategy for the bumpy scene; When the bumpy scene lane line fusion strategy is a normal bumpy fusion strategy, controlling the current vehicle to coast and decelerate according to the lane line fusion result and the corresponding deceleration signal; When the bumpy scene lane line fusion strategy is a continuous bumpy fusion strategy or a temporary bumpy fusion strategy, the current vehicle is controlled to brake and decelerate according to the lane line fusion result and the corresponding deceleration signal.

8. A device for merging lane lines on a bumpy road, characterized in that: The bumpy road lane line fusion device includes: A bumpy working condition recognition module is used to obtain historical lane line data of the current scene and the power spectrum density of the real-time vertical vibration of the current vehicle, and perform bumpy working condition recognition based on the power spectrum density; a strategy determination module, configured to obtain a bump type of the bumpy working condition and determine a corresponding bumpy scene lane line fusion strategy according to the bump type; A deceleration control module is used to generate a lane line fusion result according to the bumpy scene lane line fusion strategy, and control the deceleration of the current vehicle according to the lane line fusion result and the corresponding deceleration signal.

9. A device for merging lane lines on bumpy roads, characterized in that: The bumpy road lane line fusion device includes: a memory, a processor, and a bumpy road lane line fusion program stored in the memory and executable on the processor. The bumpy road lane line fusion program is configured to implement the steps of the bumpy road lane line fusion method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a bumpy road lane line fusion program, which, when executed by a processor, implements the steps of the bumpy road lane line fusion method according to any one of claims 1 to 7.

Images