WO2024051661A1 - 路面不平度的分类识别方法、装置、车辆及存储介质 - Google Patents

路面不平度的分类识别方法、装置、车辆及存储介质 Download PDF

Info

Publication number
WO2024051661A1
WO2024051661A1 PCT/CN2023/116866 CN2023116866W WO2024051661A1 WO 2024051661 A1 WO2024051661 A1 WO 2024051661A1 CN 2023116866 W CN2023116866 W CN 2023116866W WO 2024051661 A1 WO2024051661 A1 WO 2024051661A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
target
road surface
time
time point
Prior art date
Application number
PCT/CN2023/116866
Other languages
English (en)
French (fr)
Inventor
洪日
张建
刘秋铮
王超
王御
谢飞
韩亚凝
闫善鑫
李扬
李雅欣
Original Assignee
中国第一汽车股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国第一汽车股份有限公司 filed Critical 中国第一汽车股份有限公司
Publication of WO2024051661A1 publication Critical patent/WO2024051661A1/zh

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/30—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring roughness or irregularity of surfaces
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06—Systems determining position data of a target
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
    • G01S13/865—Combination of radar systems with lidar systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/06—Systems determining position data of a target
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Definitions

  • This application relates to the field of image recognition technology, for example, to methods, devices, vehicles and storage media for classifying and identifying road unevenness.
  • the field of view of the vehicle's forward vision sensor is frequently occupied by other vehicles on the road, making it unable to perceive the road conditions ahead, which in turn leads to the failure of the road unevenness recognition function.
  • This function affects the vehicle's mode switching and vehicle dynamics. For other functions such as prediction and active suspension, if the road unevenness recognition function fails, it will lead to a decrease in vehicle safety, comfort and other performance.
  • Related technologies usually use visual sensors to identify the elevation and unevenness of the road ahead.
  • the related technology has the following shortcomings: the identification method of the unevenness of the road ahead does not take into account the widespread problem of obstructed vision in urban working conditions in actual applications, resulting in the inability to perceive the condition of the road ahead.
  • This application provides a method, device, vehicle and storage medium for classifying and identifying road surface unevenness to increase the accuracy and robustness of the road surface unevenness prediction model and improve the accuracy of identifying road surface unevenness ahead.
  • a method for classifying and identifying road surface unevenness includes:
  • Obtain at least one set of training data sequences perform at least one round of training on the preset neural network model based on the at least one set of training data sequences, and obtain a road surface roughness prediction model;
  • the real-time pose data is input into the road surface roughness prediction model to determine the unevenness classification and identification results that match the driving road surface of the preceding vehicle.
  • a device for classifying and identifying road surface unevenness which device includes:
  • the road surface roughness prediction model acquisition module is configured to obtain at least one set of training data sequences, perform at least one round of training on the preset neural network model based on the at least one set of training data sequences, and obtain a road surface roughness prediction model;
  • the pose data acquisition module is configured to collect the real-time pose data of the preceding vehicle after the preceding vehicle is blocked on the predicted driving trajectory of the vehicle;
  • the classification and recognition result determination module is configured to input the real-time pose data into the road surface roughness prediction model and determine the unevenness classification and recognition result that matches the driving road surface of the preceding vehicle.
  • a vehicle comprising:
  • the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can perform the above-mentioned method for classifying and identifying road surface roughness. .
  • a computer-readable storage medium stores computer instructions.
  • the computer instructions are used to cause a processor to execute the above-mentioned method for classifying and identifying road surface roughness.
  • Figure 1A is a flow chart of a method for classifying and identifying road surface unevenness provided in Embodiment 1 of the present application;
  • Figure 1B is an application scenario diagram of a method for classifying and identifying road surface unevenness provided in Embodiment 1 of the present application;
  • Figure 2 is a flow chart of another method for classifying and identifying road surface unevenness provided in Embodiment 2 of the present application;
  • Figure 3 is a schematic structural diagram of a device for classifying and identifying road surface unevenness provided in Embodiment 3 of the present application;
  • Figure 4 is a schematic structural diagram of a vehicle that implements a method for classifying and identifying road unevenness provided in Embodiment 4 of the present application.
  • Figure 1A is a flow chart of a method for classifying and identifying road surface unevenness provided in Embodiment 1 of the present application. This embodiment can be applied to situations where the vehicle in front blocks the road surface on the predicted driving trajectory of the vehicle. This method can be based on the uneven road surface.
  • the device for classifying and identifying road surface roughness can be implemented in the form of hardware and/or software, and the device for classifying and identifying road surface roughness can be configured in the main controller of the vehicle. As shown in Figure 1A, the method includes:
  • the field of view of the vehicle's own forward vision sensor is frequently occupied by other vehicles on the road, making it unable to perceive the road conditions ahead, which in turn leads to the failure of the road unevenness recognition function. Based on this, the field of view of the vehicle's own forward vision sensor can be occupied by other vehicles on the road. When occupying, the unevenness of the road surface is determined based on the position information of the vehicle in front.
  • At least one set of training data sequences may be obtained from data collected by a large number of user vehicles during actual driving.
  • Each set of training data sequences may include pose data of the preceding vehicle and matching road surface roughness label data, for example, a time point ago The vehicle's pose data and the unevenness label of the road surface driven by the vehicle ahead at the same time.
  • the default neural network model can be a recurrent neural network (RNN) suitable for time series signal calculation, such as a long short-term memory network (Long Short-Term Memory, LSTM); the default neural network model can also be a convolution Neural Networks (Convolutional Neural Networks, CNN).
  • the road surface roughness prediction model may be a model obtained by training a preset neural network model using at least one set of training data sequences.
  • At least one set of training data sequences can be obtained in advance, and at least one round of training is performed on the preset neural network model based on multiple sets of training data sequences to obtain a trained model with enhanced accuracy and robustness as a road surface. Roughness prediction model.
  • the vehicle's driving prediction trajectory can be predicted based on the steering wheel angle at that time.
  • the vehicle's forward visual transmission If the sensor's viewing angle range is blocked by the vehicle in front, it means that the forward vision sensor cannot identify the road unevenness of the vehicle in the future, and the vehicle will pass through the road under the wheels of the vehicle in front of it in the future. After that, the vehicle in front can be collected.
  • Real-time pose data identify the unevenness of the road surface under the front wheel by analyzing the pose data of the vehicle in front.
  • S130 Input the real-time pose data into the road surface roughness prediction model, and determine the unevenness classification and identification results that match the driving road surface of the preceding vehicle.
  • the unevenness classification identification results may include unevenness grade classification and the confidence level of each unevenness grade.
  • the unevenness grade classification may include, for example, slight, average and severe categories.
  • the collected pose data of the preceding vehicle can be input into a pre-trained road surface roughness prediction model to obtain the unevenness level classification results that match the driving road surface of the leading vehicle and the confidence of each unevenness level.
  • the technical solution of the embodiment of the present application is to obtain at least one set of training data sequences, and subject the preset neural network model to at least one round of training according to the at least one set of training data sequences to obtain a road surface roughness prediction model; in vehicle driving prediction After the front vehicle is blocked on the trajectory, the real-time pose data of the front vehicle is collected; the real-time pose data is input into the road surface roughness prediction model to determine the unevenness classification that matches the driving road surface of the front vehicle.
  • the recognition result is to obtain a road surface roughness prediction model by training a neural network model with a large amount of training data.
  • the road surface roughness prediction model determines the unevenness classification and identification results of the driving road surface of the front vehicle based on the position and posture signal of the front vehicle, solving problems in related technologies.
  • the vehicle's own forward vision sensor's field of view is blocked, resulting in the inability to perceive the road conditions ahead. This increases the accuracy and robustness of the road unevenness prediction model and improves the accuracy of identifying the unevenness of the road ahead.
  • Figure 2 is a flow chart of another method for classifying and identifying road surface unevenness provided in Embodiment 2 of the present application. Based on the above embodiment, this embodiment will obtain at least one set of training data sequences for explanation. As shown in Figure 2, the method includes:
  • the first time point may refer to the time point when the forward vision sensor of the target vehicle detects the vehicle in front of the target on the predicted driving trajectory.
  • the target position may refer to the actual position of the vehicle in front of the target at the first point in time.
  • At least one set of training data columns may be obtained in advance.
  • S220 Preliminarily determine based on the real-time pose data of the target vehicle in front that the target road surface where the target vehicle in front is located is seriously uneven, and determine whether the target vehicle is traveling according to the predicted driving trajectory after the first time point. to the target location.
  • the real-time pose data of the target vehicle in front can include real-time vehicle speed and real-time roll angle.
  • the real-time pose data of the vehicle in front of the target can be collected by the sensors configured on the target vehicle.
  • preliminarily determining that the target road surface where the target vehicle in front is located is severely uneven based on the real-time pose data may include: obtaining the current vehicle speed and current roll angle of the target vehicle in front; and the current roll angle, calculate the current roll variance value of the target vehicle ahead; determine the target road surface where the target vehicle ahead is located at the current moment based on the current roll variance value and the preset variance threshold severely uneven.
  • the current roll variance value of the target vehicle ahead is calculated.
  • the length of the judgment interval at the current moment can be calculated based on the current vehicle speed, the preset collection period and the preset unevenness.
  • the demand quantity for the historical target front vehicle pose data according to the demand quantity, obtain the corresponding number of historical target front vehicle roll angles; according to the demand quantity and the corresponding number of historical target front vehicle roll angles, calculate the target The current roll variance value of the vehicle in front.
  • the preset collection period may refer to the collection period of the pose data of the vehicle in front of the target.
  • the preset unevenness judgment interval may refer to the length within which the unevenness of the road surface is judged.
  • the unevenness judgment interval can be predetermined based on the target vehicle's body length, shock-proof performance and other information.
  • the target vehicle's body length is 5 meters, and the unevenness judgment interval can be preset to be 5 meters.
  • the required quantity of historical target leading vehicle pose data at the current moment may refer to the required quantity of target leading vehicle pose data collected within the time before the current moment.
  • the current roll variance value can be passed Calculation, where S 2 represents the target The degree of fluctuation of the vehicle's body roll in the period before the current moment, Represents the roll angle of the target vehicle in front at different times, n represents the number of sampling points, represents the length of time intercepted before the current moment, represents the number of historical target vehicle pose data requirements, and is used to combine the front vehicle pose data with the actual road surface roughness Corresponding in space, the number of sampling points should change with the vehicle speed, which can be Determine, where L c is the length of the preset unevenness judgment interval, which is a calibration quantity; t s is the preset sampling period of the target front vehicle's pose data, and u f is the speed of the target front vehicle.
  • the current roll variance value and the preset variance threshold it is determined that the target road surface where the target vehicle in front is located at the current moment is seriously uneven, and it can be determined whether the current roll variance value is greater than the preset variance threshold. ; When the current roll variance value is greater than the preset variance threshold, it is determined that the target road surface where the target front vehicle is located at the current moment is seriously uneven.
  • the roll angle data features in the pose data of the target vehicle ahead are more obvious and more precise. Therefore, the variance of the roll angle data can be used to determine the unevenness of the road surface under the wheels of the target vehicle ahead. in accordance with. If the calculated roll variance value S 2 of the target front vehicle is greater than the preset variance threshold, it can be considered that the target front vehicle is passing through a road with greater unevenness. Correspondingly, if S 2 is not greater than the preset variance threshold, the pose data of the vehicle in front of the target can be discarded, that is, it does not need to continue to be a sampling object.
  • determining whether the target vehicle drives to the target location according to the predicted driving trajectory after the first time point may include:
  • the predicted length between the target vehicle and the target preceding vehicle on the predicted driving trajectory and the current distance of the target vehicle are obtained.
  • Vehicle speed according to the predicted length and the current vehicle speed, obtain the predicted duration for the target vehicle to travel to the target location according to the predicted driving trajectory, and obtain the second predicted time for the target vehicle to travel to the target location.
  • Time point collect the first yaw angular velocity of the target vehicle at the first time point, the second yaw angular velocity of the target vehicle actually traveling to the second time point, and collect the target vehicle at the second time point.
  • the predicted length may refer to S, and the predicted length may be, for example, loaded by the target vehicle. Radar sensors and laser sensors are obtained.
  • the predicted duration may refer to the duration predicted by driving S according to the current speed of the target vehicle.
  • the second time point may refer to the time point when the predicted target vehicle travels to the target position of the target vehicle in front according to S.
  • the actual driving length of the target vehicle between the first time point and the second time point can be passed Determine, where t 1 represents the first time point, t 2 represents the second time point, and u 0 represents the actual speed of the target vehicle.
  • the formula can be the actual speed of the target vehicle between the first time point and the second time point. Accumulate over time.
  • the integrated value of the yaw angular velocity of the target vehicle between the first time point and the second time point can be obtained by Determine, where w yaw represents the second yaw angular velocity of the target vehicle, Represents the first yaw angular velocity of the target vehicle.
  • the target vehicle when the first difference between the actual travel length and the predicted length does not exceed the preset difference threshold, and the yaw angular velocity integral value is less than the preset integral threshold , then it is determined that the target vehicle will drive to the target location according to the predicted driving trajectory after the first time point.
  • the first difference between the actual travel length and the predicted length of the target vehicle can be calculated.
  • the first difference does not exceed the preset difference threshold and the yaw angular velocity integral value is less than the preset integral threshold, it is determined that the target vehicle is in After the first time point, drive to the target location according to the predicted driving trajectory.
  • the first difference exceeds the preset difference threshold, and/or, the yaw angular velocity integral value is not less than the preset integral threshold, it can be determined that the target vehicle has not traveled to the destination according to the predicted driving trajectory after the first time point. At this time, the target vehicle does not need to be used as the sampling object.
  • Dynamic response signals can include suspension vertical acceleration signals, sprung mass vertical acceleration signals, body pitch angular velocity signals, body roll angle signals, etc., which can be collected through sensors configured on the vehicle itself.
  • S240 Determine again the unevenness classification and identification result of the target road surface according to the real-time dynamic response signal of the target vehicle, and use the unevenness classification and identification result as a label data sequence matching the target road surface.
  • the unevenness classification and identification results of the target road surface that the target vehicle in front of the target traveled on can be determined again based on the fluctuation of the real-time dynamic response signal of the target vehicle, thereby classifying the unevenness into Classification recognition results are used as label data sequences matching the target road surface.
  • S250 Align the label data sequence with the real-time pose data of the target front vehicle in time series to obtain an alignment result, and use the alignment result as a set of training data sequences.
  • the advantage of setting up the S210-S250 operation is that it can improve the authenticity of the training data set, thereby increasing the robustness of the road surface roughness prediction model algorithm.
  • S260 Determine whether the number of training data sequences meets the preset conditions. If the number of training data sequences meets the preset conditions, obtain at least one set of training data sequences and perform S270. If the number of training data sequences does not meet the preset conditions, return to execution. S210 operation.
  • the preset conditions may refer to the quantity requirements of the preset training data sequence, for example, hundreds, thousands of groups, etc.
  • the number of training data sequences is multiple sets, that is, it is necessary to obtain at least one set of training data sequences based on multiple sets of target vehicles and target preceding vehicles.
  • S290 Input the real-time pose data into the road surface roughness prediction model, and determine the unevenness classification and identification results that match the driving road surface of the preceding vehicle.
  • the technical solution of the embodiment of the present application first determines the unevenness of the target road surface by collecting the position and orientation data of the target vehicle in front of the target during actual driving.
  • the target vehicle travels to the target road surface where the target vehicle in front of the target is traveling according to the predicted driving trajectory
  • the target vehicle is collected by collecting the position and orientation data of the target vehicle in front.
  • the dynamic response signal again determines the unevenness of the target road surface, obtains at least one set of training data sequences based on the pose data of the target vehicle in front and the final roughness label sequence of the target road surface, and sets the preset neural network based on at least one set of training data sequences.
  • the network model After at least one round of training, the network model obtains a road surface roughness prediction model; after the front vehicle is blocked on the predicted driving trajectory of the vehicle, the real-time pose data of the front vehicle is collected; the real-time pose data is input
  • the road surface roughness prediction model is used to determine the unevenness classification and identification results that match the driving road surface of the preceding vehicle. That is, a neural network model is trained through a large amount of training data to obtain a road surface unevenness prediction model.
  • the posture signal of the vehicle determines the unevenness classification and recognition results of the road ahead of the vehicle, which solves the problem in related technologies that the vehicle's own forward vision sensor's field of view is blocked and cannot perceive the road conditions ahead, and increases the accuracy and accuracy of the road unevenness prediction model. Robustness to improve the accuracy of identifying unevenness on the road ahead.
  • Figure 3 is a schematic structural diagram of a device for classifying and identifying road surface unevenness provided in Embodiment 3 of the present application.
  • the device includes: a road surface roughness prediction model acquisition module 310, a pose data acquisition module 320, and a classification recognition result determination module 330. in:
  • the road surface roughness prediction model acquisition module 310 is configured to obtain at least one set of training data sequences, perform at least one round of training on the preset neural network model according to the at least one set of training data sequences, and obtain a road surface roughness prediction model; pose data
  • the acquisition module 320 is configured to collect the real-time pose data of the preceding vehicle after the preceding vehicle is blocked on the predicted driving trajectory of the vehicle; the classification recognition result determination module 330 is configured to input the real-time pose data to the The road surface roughness prediction model determines the unevenness classification and recognition results that match the driving road surface of the preceding vehicle.
  • the technical solution of the embodiment of the present application obtains at least one set of training data sequences and performs at least one round of training on the preset neural network model based on the at least one set of training data sequences to obtain a road surface roughness prediction model; in vehicle driving prediction After the front vehicle is blocked on the trajectory, the real-time pose data of the front vehicle is collected; the real-time pose data is input into the road surface roughness prediction model to determine the unevenness classification that matches the driving road surface of the front vehicle.
  • the identification result is to obtain a road surface roughness prediction model by training a neural network model with a large amount of training data.
  • the road surface roughness prediction model determines the unevenness classification and identification results of the road ahead of the vehicle in front based on the posture signal of the vehicle in front.
  • the road surface roughness prediction model acquisition module 310 includes:
  • the real-time pose data acquisition unit is configured to record the first time point and the target position of the target preceding vehicle when the target preceding vehicle blocks the predicted driving trajectory on the predicted driving trajectory of the target vehicle, and The real-time pose data of the target vehicle in front is collected after the first time point;
  • the target vehicle driving position determination unit is configured to initially determine based on the real-time pose data of the target vehicle in front When the target road surface where the target vehicle in front is located is seriously uneven, it is determined whether the target vehicle drives to the target position according to the predicted driving trajectory after the first time point;
  • the real-time dynamic response signal acquisition unit is set In response to the target vehicle traveling to the target location according to the predicted traveling trajectory after the first time point, after the target vehicle travels to the target location according to the predicted traveling trajectory, the a real-time dynamic response signal of the target vehicle;
  • a label data sequence acquisition unit configured to re-determine the unevenness classification identification result of the target road surface based on the real-time dynamic response signal of the target vehicle, and obtain the unevenness classification identification
  • the real-time pose data of the target vehicle in front includes real-time vehicle speed and real-time roll angle; correspondingly, the target vehicle driving position judgment unit includes:
  • the vehicle speed and roll angle acquisition subunit is configured to obtain the current vehicle speed and current roll angle of the target vehicle ahead; the roll variance value calculation subunit is configured to calculate the vehicle speed and the current roll angle based on the current vehicle speed and the current roll angle.
  • the current roll variance value of the target vehicle in front; the target road surface is seriously uneven determination subunit, which is configured to determine the target road surface where the target vehicle in front is located at the current moment based on the current roll variance value and the preset variance threshold.
  • Roll variance value calculation subunit set to:
  • the demand quantity for the historical target front vehicle pose data at the current moment is calculated; according to the demand quantity, the corresponding number of historical target front vehicle roll angles are obtained ; Calculate the current roll variance value of the target front vehicle based on the required quantity and the corresponding number of historical target front vehicle roll angles.
  • the target road surface is severely uneven to determine the sub-unit, set as:
  • the target vehicle driving position judgment unit also includes:
  • the current speed acquisition subunit of the target vehicle is configured to obtain the distance between the target vehicle and the target front vehicle on the predicted driving trajectory when the target vehicle in front is blocked on the predicted driving trajectory of the target vehicle.
  • the second time point acquisition subunit is configured to obtain the target vehicle according to the predicted driving trajectory to the destination according to the predicted length and the current vehicle speed.
  • the yaw angular velocity integral value calculation subunit is configured to calculate the yaw rate according to the first yaw acceleration, the second yaw angular velocity, The first time point and the second time point, calculate the yaw angular velocity integral value of the target vehicle between the first time point and the second time point;
  • the target vehicle driving position determination subunit It is configured to determine whether the target vehicle travels to the destination according to the predicted driving trajectory after the first time point based on the actual driving length, the predicted length, the yaw angular velocity integral value and the preset integral threshold. Describe the target location.
  • the target vehicle driving position judgment subunit is set to:
  • the device for classifying and identifying road surface roughness provided in the embodiments of this application can execute the method for classifying and identifying road surface unevenness provided in any embodiment of this application, and has functional modules and effects corresponding to the execution method.
  • FIG. 4 shows a schematic structural diagram of a vehicle 400 that can be used to implement embodiments of the present application.
  • Vehicle 400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.
  • Vehicle 400 may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (eg, helmets, glasses, watches, etc.), and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions are examples only and are not intended to limit the implementation of the present application as described and/or claimed herein.
  • the vehicle 400 includes at least one processor 401, and a memory communicatively connected to the at least one processor 401, such as a read-only memory (Read-Only Memory, ROM) 402, a random access memory (Random Access Memory, RAM). ) 403, etc., in which the memory stores a computer program that can be executed by at least one processor.
  • the processor 401 can execute multiple programs according to the computer program stored in the ROM 402 or loaded from the storage unit 18 into the RAM 403. appropriate actions and handling. In RAM 403, vehicle 400 operations can also be stored. various programs and data required for operation.
  • the processor 401, the ROM 402 and the RAM 403 are connected to each other through the bus 14.
  • An input/output (I/O) interface 15 is also connected to the bus 14 .
  • the I/O interface 15 includes: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc. ; And communication unit 19, such as network card, modem, wireless communication transceiver, etc.
  • the communication unit 19 allows the vehicle 400 to exchange information/data with other devices via a computer network such as the Internet and/or various telecommunications networks.
  • Processor 401 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the processor 401 include a central processing unit (CPU), a graphics processing unit (GPU), a variety of dedicated artificial intelligence (Artificial Intelligence, AI) computing chips, and a variety of running machine learning models Algorithm processor, digital signal processor (Digital Signal Processor, DSP), and any appropriate processor, controller, microcontroller, etc.
  • the processor 401 executes a plurality of methods and processes described above, such as a method for classifying and identifying road surface roughness.
  • the method for classifying and identifying road surface roughness may be implemented as a computer program, which is tangibly included in a computer-readable storage medium, such as the storage unit 18 .
  • part or all of the computer program may be loaded and/or installed onto the vehicle 400 via the ROM 402 and/or the communication unit 19.
  • the processor 401 may be configured to perform the classification identification method of road surface unevenness in any other suitable manner (eg, by means of firmware).
  • FPGAs Field Programmable Gate Arrays
  • ASICs Application Specific Integrated Circuits
  • ASSP Application Specific Standard Products
  • SOC System on Chip
  • CPLD Complex Programmable Logic Device
  • These various embodiments may include implementation in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor
  • the processor which may be a special purpose or general purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
  • An output device may be a special purpose or general purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
  • An output device may be a special purpose or general purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
  • Computer programs for implementing the methods of the present application may be written in any combination of one or more programming languages. These computer programs may be provided for general-purpose computers, special-purpose computers, or other
  • the processor of the data processing apparatus is programmable such that the computer program, when executed by the processor, causes the functions/operations specified in the flowcharts and/or block diagrams to be performed.
  • a computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
  • a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device.
  • Computer-readable storage media may include electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any suitable combination of the foregoing.
  • the computer-readable storage medium may be a machine-readable signal medium.
  • machine-readable storage media examples include one or more wire-based electrical connections, laptop disks, hard drives, RAM, ROM, Erasable Programmable Read-Only Memory (EPROM), or flash memory ), optical fiber, portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the systems and techniques described herein may be implemented on a vehicle having a display device (e.g., a cathode ray tube (CRT)) or a liquid crystal display (e.g., a cathode ray tube (CRT)) configured to display information to a user.
  • a display device e.g., a cathode ray tube (CRT)
  • a liquid crystal display e.g., a cathode ray tube (CRT)
  • LCD Liquid Crystal Display
  • keyboard and pointing device e.g., a mouse or trackball
  • the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and may be provided in any form, including Acoustic input, voice input or tactile input) to receive input from the user.
  • feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and may be provided in any form, including Acoustic input, voice input or tactile input) to receive input from the user.
  • the systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., A user's computer having a graphical user interface or web browser through which the user can interact with implementations of the systems and technologies described herein), or including such backend components, middleware components, or any combination of front-end components in a computing system.
  • the components of the system may be interconnected by any form or medium of digital data communication (eg, a communications network). Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN), blockchain network, and the Internet.
  • Computing systems may include clients and servers.
  • Clients and servers are generally remote from each other and typically interact over a communications network.
  • the relationship of client and server is created by computer programs running on corresponding computers and having a client-server relationship with each other.
  • the server can be a cloud Server, also known as cloud computing server or cloud host, is a host product in the cloud computing service system to solve the management difficulty in traditional physical host and virtual private server (VPS) services.
  • VPN virtual private server
  • Steps can be reordered, added, or removed using various forms of the process shown above.
  • multiple steps described in this application can be executed in parallel, sequentially, or in different orders.
  • the desired results of the technical solution of this application can be achieved, there is no limitation here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Electromagnetism (AREA)
  • Multimedia (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)

Abstract

申请路面不平度的分类识别方法、装置、车辆及存储介质。该路面不平度的分类识别方法包括获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型(S110)；在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据(S120)；将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果(S130)。

Description

路面不平度的分类识别方法、装置、车辆及存储介质
本申请要求在2022年09月07日提交中国专利局、申请号为202211088233.2的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像识别技术领域，例如涉及路面不平度的分类识别方法、装置、车辆及存储介质。
背景技术
城市道路工况下，车辆自身前向视觉传感器视野频繁遭路面上的其他车辆占据，无法感知前方路面状况，进而导致路面不平度识别功能失效，此功能影响车辆整车模式切换、车辆动力学状态预测、主动悬架等其余功能，若路面不平度识别功能失效将导致车辆安全性、舒适性等性能的下降。相关技术通常使用视觉传感器对前方路面高程、路面不平度进行识别的技术。
相关技术存在如下缺陷：对前方路面不平度的识别方法并没有考虑到实际应用过程中广泛存在的城市工况下视野受遮挡的问题，导致无法感知前方路面状况。
发明内容
本申请提供了路面不平度的分类识别方法、装置、车辆及存储介质，以增加路面不平度预测模型的准确性与鲁棒性，提高对前方路面不平度识别的准确性。
根据本申请的一方面，提供了一种路面不平度的分类识别方法，该方法包括：
获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；
在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；
将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
根据本申请的另一方面，提供了一种路面不平度的分类识别装置，该装置包括：
路面不平度预测模型获取模块，设置为获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；
位姿数据采集模块，设置为在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；
分类识别结果确定模块，设置为将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
根据本申请的另一方面，提供了一种车辆，所述车辆包括：
至少一个处理器；以及
与所述至少一个处理器通信连接的存储器；其中，
所述存储器存储有可被所述至少一个处理器执行的计算机程序，所述计算机程序被所述至少一个处理器执行，以使所述至少一个处理器能够执行上述的路面不平度的分类识别方法。
根据本申请的另一方面，提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机指令，所述计算机指令用于使处理器执行上述的路面不平度的分类识别方法。
附图说明
图1A为本申请实施例一提供的一种路面不平度的分类识别方法的流程图；
图1B为本申请实施例一提供的一种路面不平度的分类识别方法的应用场景图；
图2为本申请实施例二提供的另一种路面不平度的分类识别方法的流程图；
图3为本申请实施例三提供的一种路面不平度的分类识别装置的结构示意图；
图4为本申请实施例四提供的一种实现路面不平度的分类识别方法的车辆的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述，所描述的实施例仅仅是本申请一部分的实施例。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这 样使用的数据在适当情况下可以互换，以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
实施例一
图1A为本申请实施例一提供的一种路面不平度的分类识别方法的流程图，本实施例可适用于在车辆的行驶预测轨迹上出现前车遮挡路面的情况，该方法可以由路面不平度的分类识别装置来执行，该路面不平度的分类识别装置可以采用硬件和/或软件的形式实现，该路面不平度的分类识别装置可配置于车辆的主控制器中。如图1A所示，该方法包括：
S110、获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型。
城市道路工况下，车辆自身前向视觉传感器视野频繁遭路面上的其他车辆占据，无法感知前方路面状况，进而导致路面不平度识别功能失效，基于此，可以在车辆自身前向视觉传感器视野被占据时，通过前车的位姿信息判别路面不平度。
至少一组训练数据序列可以是从大量用户车辆在实际行驶中采集的数据中获得，每组训练数据序列可以包括前车的位姿数据以及匹配的路面不平度标签数据，例如，一时间点前车的位姿数据以及同时间前车行驶路面的不平度标签。预设神经网络模型可以是适用于时间序列信号计算的循环神经网络(Recurrent Neural Network，RNN)，例如长短期记忆网络(Long Short-Term Memory，LSTM)；预设神经网络模型还可以是卷积神经网络(Convolutional Neural Networks，CNN)。路面不平度预测模型可以是使用至少一组训练数据序列训练预设神经网络模型得到的模型。
在本实施例中，可以预先获取至少一组的训练数据序列，根据多组训练数据序列对预设神经网络模型进行至少一轮训练，得到准确性与鲁棒性增强的训练后的模型作为路面不平度预测模型。
S120、在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据。
本实施例中，参考图1B，可以当车辆在路面行驶时根据此时的方向盘转角预测得到车辆的行驶预测轨迹，在该行驶预测轨迹上若车辆的前向视觉传 感器视角范围受前车遮挡，则表示前向视觉传感器无法识别车辆在未来时刻的路面不平度，且车辆在未来将经过前车此刻轮下的路面，那么在此之后，可以采集前车的实时位姿数据，通过分析前车的位姿数据对前车轮下路面进行不平度识别。
S130、将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
不平度分类识别结果可以包括不平度等级分类以及每个不平度等级的置信度。其中，不平度等级分类例如可以包括轻微、一般和严重等类别。
在本实施例中，可以将采集的前车的位姿数据输入至预先训练的路面不平度预测模型，得到与前车的行驶路面匹配的不平度等级分类结果以及每个不平度等级的置信度。
在确定前车的行驶路面的不平度分类结果后，可以根据自身车辆状况为车辆的其他控制系统提供输入信号，以提升车辆驾驶的舒适性和安全性。
本申请实施例的技术方案，通过获取至少一组训练数据序列，根据所述至少一组训练数据序列将预设神经网络模型经过至少一轮训练，得到路面不平度预测模型；在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果，即通过大量训练数据训练神经网络模型得到路面不平度预测模型，通过路面不平度预测模型根据前车的位姿信号确定前车的行驶路面的不平度分类识别结果，解决了相关技术中车辆自身前向视觉传感器视野受遮挡导致无法感知前方路面状况的问题，增加路面不平度预测模型的准确性与鲁棒性，提高对前方路面不平度识别的准确性。
实施例二
图2为本申请实施例二提供的另一种路面不平度的分类识别方法的流程图，本实施例在上述实施例的基础上，将获取至少一组训练数据序列进行说明。如图2所示，该方法包括：
S210、在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡所述行驶预测轨迹时，记录第一时间点和所述目标前车的目标位置，并在所述第一时间点后采集所述目标前车的实时位姿数据。
第一时间点可以指目标车辆的前向视觉传感器在行驶预测轨迹上检测到目标前车的时间点。目标位置可以指第一时间点时目标前车的实际位置。
在本实施例中，为训练预设神经网络模型，可以预先获取至少一组的训练数据列。以获取一组训练数据序列为例：在目标车辆的行驶预测轨迹上出现目标前车遮挡行驶预测轨迹时，记录第一时间点和目标前车的目标位置，并在第一时间点后采集目标前车的实时位姿数据。
S220、根据所述目标前车的实时位姿数据初步确定所述目标前车所处的目标路面严重不平时，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置。
目标前车的实时位姿数据可以包括实时车速和实时侧倾角。目标前车的实时位姿数据可以由目标车辆上配置的传感器采集。
由于车辆与车辆之间的车身尺寸、悬架系统性能等存在差异，不同车辆在同一不平度等级路面上的车身姿态响应情况也存在不同，因此无法直接根据前车的位姿数据判断路面不平度，还需要本车驶过前方路面后，以自身动力学响应信号判断路面不平度等级。
在本实施例中，可以在根据目标前车的实时车速和实时侧倾角初步确定目标前车所处目标路面严重不平时，还需要判断目标车辆在第一时间点后是否按照行驶预测轨迹行驶至目标位置。
在一个实施方式中，根据所述实时位姿数据初步确定所述目标前车所处的目标路面严重不平，可以包括：获取所述目标前车的当前车速和当前侧倾角；根据所述当前车速和所述当前侧倾角，计算所述目标前车的当前侧倾方差值；根据所述当前侧倾方差值和预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平。
根据所述当前车速和所述当前侧倾角，计算所述目标前车的当前侧倾方差值，可以根据所述当前车速、预设采集周期和预设不平度判断区间长度，计算在当前时刻对历史目标前车位姿数据的需求数量；根据所述需求数量，获取对应数量的历史目标前车侧倾角；根据所述需求数量和所述对应数量的历史目标前车侧倾角，计算所述目标前车的当前侧倾方差值。
预设采集周期可以指对目标前车位姿数据的采集周期。预设不平度判断区间可以指判断多少长度内路面的不平度。例如，可以根据目标车辆的车身长度、防震性能等信息预先确定不平度判断区间，例如目标车辆车身长5米，可以预设不平度判断区间为5米。当前时刻对历史目标前车位姿数据的需求数量可以指在当前时刻之前时间内采集的目标前车位姿数据的需求数量。
当前侧倾方差值可以通过计算，其中，S2表示目 标前车在当前时刻之前的一段时间内车身侧倾的波动程度，表示目标前车不同时刻的侧倾角，n表示采样点个数，表示在当前时刻之前截取的时间长度，表示对历史目标前车位姿数据的需求数量，为将前车位姿数据与实际路面不平度在空间上对应，采样点个数应随车速改变，可以通过确定，其中Lc为预设不平度判断区间长度，为一标定量；ts为目标前车位姿数据的预设采样周期，uf为目标前车的车速。
根据所述当前侧倾方差值和预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平，可以判断所述当前侧倾方差值是否大于所述预设方差阈值；在所述当前侧倾方差值大于所述预设方差阈值时，确定所述目标前车在当前时刻所处的目标路面严重不平。
受目标车辆的传感器安装高度所限，目标前车的位姿数据中的侧倾角数据特征更为明显、精度更高，因此可以使用侧倾角数据的方差作为目标前车车轮下路面不平度的判断依据。若计算的目标前车侧倾方差值S2大于预设方差阈值，可以认为目标前车正在经过不平度较大的路面。相应的，若S2未大于预设方差阈值，可以舍弃该目标前车的位姿数据，即可以不将其继续作为采样对象。
在另一个实施方式中，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置，可以包括：
在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡时，获取所述目标车辆与所述目标前车之间在所述行驶预测轨迹上的预测长度，以及所述目标车辆的当前车速；根据所述预测长度和所述当前车速，获取所述目标车辆按照所述行驶预测轨迹行驶至所述目标位置的预测时长，并获取所述目标车辆预测行驶至所述目标位置的第二时间点；采集所述目标车辆在所述第一时间点的第一横摆角速度，所述目标车辆实际行驶至第二时间点的第二横摆角速度，并采集所述目标车辆在所述第一时间点和所述第二时间点之间的实时车速；根据所述实时车速、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的实际行驶长度；根据所述第一横摆加速度、所述第二横摆角速度、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的横摆角速度积分值；根据所述实际行驶长度、所述预测长度、所述横摆角速度积分值和预设积分阈值，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置。
参考图1B，预测长度可以指S，预测长度例如可以通过目标车辆装载的 雷达传感器和激光传感器等获得。预测时长可以指按照目标车辆的当前车速行驶S所预计的时长。第二时间点可以指预测的目标车辆按照S行驶至目标前车所在目标位置的时间点。
目标车辆在第一时间点与第二时间点之间的实际行驶长度可以通过确定，其中，t1表示第一时间点，t2表示第二时间点，u0表示目标车辆的实际车速，该公式可以为将目标车辆的实际车速在第一时间点和第二时间点之间进行累加。
目标车辆在所述第一时间点和所述第二时间点之间的横摆角速度积分值可以通过确定，其中，wyaw表示目标车辆的第二横摆角速度，表示目标车辆的第一横摆角速度。
在上述实施方式的基础上，在所述实际行驶长度与所述预测长度之间的第一差值不超过预设差值阈值，且所述横摆角速度积分值小于所述预设积分阈值时，则确定所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置。
可以计算目标车辆的实际行驶长度与预测长度之间的第一差值，在该第一差值不超过预设差值阈值，且横摆角速度积分值小于预设积分阈值时，确定目标车辆在第一时间点后按照行驶预测轨迹行驶至所述目标位置。相应的，若第一差值超过预设差值阈值，和/或，横摆角速度积分值不小于预设积分阈值时，可以确定目标车辆在第一时间点后未按照行驶预测轨迹行驶至所述目标位置，此时可以不将目标车辆继续作为采样对象。
S230、响应于所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置，在所述目标车辆按照所述行驶预测轨迹行驶至所述目标位置后，采集所述目标车辆的实时动力学响应信号。
动力学响应信号可以包括悬架垂向加速度信号、簧载质量垂向加速度信号、车身俯仰角速度信号和车身侧倾角信号等，可以通过车辆自身配置的传感器采集。
S240、根据所述目标车辆的实时动力学响应信号再次确定所述目标路面的不平度分类识别结果，并将所述不平度分类识别结果，作为与所述目标路面匹配的标签数据序列。
本实施例中，可以根据目标车辆的实时动力学响应信号的波动情况再一次确定目标前车之前行驶的目标路面的不平度分类识别结果，从而将不平度 分类识别结果，作为与目标路面匹配的标签数据序列。
S250、将所述标签数据序列与所述目标前车的实时位姿数据在时序上对齐，得到对齐结果，并将所述对齐结果作为一组训练数据序列。
本实施例中，由于目标车辆与目标前车之间存在距离差异，因此根据目标车辆的动力学响应信号得到的标签数据序列与目标前车的实时位姿数据在时序上存在差异，因此，可以将标签数据序列与目标前车的实时位姿数据在时序上进行对齐，从而将对齐结果作为一组训练数据序列。
S210-S250操作这样设置的好处在于，可以提升训练数据集的真实性，进而增加路面不平度预测模型算法的鲁棒性。
S260、判断训练数据序列的数量是否满足预设条件，若训练数据序列的数量满足预设条件，获取至少一组训练数据序列执行S270操作，若训练数据序列的数量不满足预设条件，返回执行S210操作。
预设条件可以指预设训练数据序列的数量需求值，例如，几百、几千组等。
在本实施例中，训练数据序列的数量为多组，即需要根据多组目标车辆和目标前车获取至少一组训练数据序列。在获取一组训练数据序列后，可以判断训练数据序列的数量是否满足预设条件，若训练数据序列的数量满足预设条件满足预设条件，则获取至少一组训练数据序列，若训练数据序列的数量满足预设条件不满足预设条件，则返回重新执行S210-S250的操作，直至训练数据序列满足预设条件。
S270、根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型。
S280、在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据。
S290、将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
本申请实施例的技术方案，通过采集实际驾驶中目标前车的位姿数据初次判断目标路面不平度，在目标车辆按照行驶预测轨迹行驶至目标前车行驶的目标路面时，通过采集目标车辆的动力学响应信号再次判断目标路面的不平度，根据目标前车的位姿数据以及最终得到的目标路面的不平度标签序列获取至少一组训练数据序列，根据至少一组训练数据序列将预设神经网络模型经过至少一轮训练，得到路面不平度预测模型；在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；将所述实时位姿数据输入 至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果，即通过大量训练数据训练神经网络模型得到路面不平度预测模型，通过路面不平度预测模型根据前车的位姿信号确定前车行驶路面的不平度分类识别结果，解决了相关技术中车辆自身前向视觉传感器视野受遮挡导致无法感知前方路面状况的问题，增加路面不平度预测模型的准确性与鲁棒性，提高对前方路面不平度识别的准确性。
实施例三
图3为本申请实施例三提供的一种路面不平度的分类识别装置的结构示意图。如图3所示，该装置包括：路面不平度预测模型获取模块310、位姿数据采集模块320和分类识别结果确定模块330。其中：
路面不平度预测模型获取模块310，设置为获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；位姿数据采集模块320，设置为在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；分类识别结果确定模块330，设置为将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
本申请实施例的技术方案，通过获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；在车辆的行驶预测轨迹上出现前车遮挡后，采集所述前车的实时位姿数据；将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果，即通过大量训练数据训练神经网络模型得到路面不平度预测模型，通过路面不平度预测模型根据前车的位姿信号确定前车行驶路面的不平度分类识别结果，解决了相关技术中车辆自身前向视觉传感器视野受遮挡导致无法感知前方路面状况的问题，增加路面不平度预测模型的准确性与鲁棒性，提高对前方路面不平度识别的准确性。
路面不平度预测模型获取模块310，包括：
针对一组目标车辆和目标前车执行下述操作：
实时位姿数据采集单元，设置为在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡所述行驶预测轨迹时，记录第一时间点和所述目标前车的目标位置，并在所述第一时间点后采集所述目标前车的实时位姿数据；目标车辆行驶位置判断单元，设置为根据所述目标前车的实时位姿数据初步确定 所述目标前车所处的目标路面严重不平时，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置；实时动力学响应信号采集单元，设置为响应于所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置，在所述目标车辆按照所述行驶预测轨迹行驶至所述目标位置后，采集所述目标车辆的实时动力学响应信号；标签数据序列获取单元，设置为根据所述目标车辆的实时动力学响应信号再次确定所述目标路面的不平度分类识别结果，并将所述不平度分类识别结果，作为与所述目标路面匹配的标签数据序列；训练数据序列获取单元，设置为将所述标签数据序列与所述目标前车的实时位姿数据在时序上对齐，得到对齐结果，并将所述对齐结果作为一组训练数据序列。
所述目标前车的实时位姿数据包括实时车速和实时侧倾角；相应的，目标车辆行驶位置判断单元，包括：
车速和侧倾角获取子单元，设置为获取所述目标前车的当前车速和当前侧倾角；侧倾方差值计算子单元，设置为根据所述当前车速和所述当前侧倾角，计算所述目标前车的当前侧倾方差值；目标路面严重不平确定子单元，设置为根据所述当前侧倾方差值和预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平。
侧倾方差值计算子单元，设置为：
根据所述当前车速、预设采集周期和预设不平度判断区间长度，计算在当前时刻对历史目标前车位姿数据的需求数量；根据所述需求数量，获取对应数量的历史目标前车侧倾角；根据所述需求数量和所述对应数量的历史目标前车侧倾角，计算所述目标前车的当前侧倾方差值。
目标路面严重不平确定子单元，设置为：
判断所述当前侧倾方差值是否大于所述预设方差阈值；在所述当前侧倾方差值大于所述预设方差阈值时，确定所述目标前车在当前时刻所处的目标路面严重不平。
目标车辆行驶位置判断单元，还包括：
目标车辆的当前车速获取子单元，设置为在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡时，获取所述目标车辆与所述目标前车之间在所述行驶预测轨迹上的预测长度，以及所述目标车辆的当前车速；第二时间点获取子单元，设置为根据所述预测长度和所述当前车速，获取所述目标车辆按照所述的行驶预测轨迹行驶至所述目标位置的预测时长，并获取所述目标车辆预测行驶至所述目标位置的第二时间点；目标车辆的实时车速采集子单 元，设置为采集所述目标车辆在所述第一时间点的第一横摆角速度，所述目标车辆实际行驶至第二时间点的第二横摆角速度，并采集所述目标车辆在所述第一时间点和所述第二时间点之间的实时车速；实际行驶长度计算子单元，设置为根据所述实时车速、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的实际行驶长度；横摆角速度积分值计算子单元，设置为根据所述第一横摆加速度、所述第二横摆角速度、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的横摆角速度积分值；目标车辆行驶位置判断子单元，设置为根据所述实际行驶长度、所述预测长度、所述横摆角速度积分值和预设积分阈值，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置。
目标车辆行驶位置判断子单元，设置为：
在所述实际行驶长度与所述预测长度之间的第一差值不超过预设差值阈值，且所述横摆角速度积分值小于所述预设积分阈值时，则确定所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置。
本申请实施例所提供的路面不平度的分类识别装置可执行本申请任意实施例所提供的路面不平度的分类识别方法，具备执行方法相应的功能模块和效果。
实施例四
图4示出了可以用来实施本申请的实施例的车辆400的结构示意图。车辆400旨在表示多种形式的数字计算机，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。车辆400还可以表示多种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备(如头盔、眼镜、手表等)和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本申请的实现。
如图4所示，车辆400包括至少一个处理器401，以及与至少一个处理器401通信连接的存储器，如只读存储器(Read-Only Memory，ROM)402、随机访问存储器(Random Access Memory，RAM)403等，其中，存储器存储有可被至少一个处理器执行的计算机程序，处理器401可以根据存储在ROM 402中的计算机程序或者从存储单元18加载到RAM 403中的计算机程序，来执行多种适当的动作和处理。在RAM 403中，还可存储车辆400操 作所需的多种程序和数据。处理器401、ROM 402以及RAM 403通过总线14彼此相连。输入/输出(Input/Output，I/O)接口15也连接至总线14。
车辆400中的多个部件连接至I/O接口15，包括：输入单元16，例如键盘、鼠标等；输出单元17，例如多种类型的显示器、扬声器等；存储单元18，例如磁盘、光盘等；以及通信单元19，例如网卡、调制解调器、无线通信收发机等。通信单元19允许车辆400通过诸如因特网的计算机网络和/或多种电信网络与其他设备交换信息/数据。
处理器401可以是多种具有处理和计算能力的通用和/或专用处理组件。处理器401的一些示例包括中央处理单元(Central Processing Unit，CPU)、图形处理单元(Graphics Processing Unit，GPU)、多种专用的人工智能(Artificial Intelligence，AI)计算芯片、多种运行机器学习模型算法的处理器、数字信号处理器(Digital Signal Processor，DSP)、以及任何适当的处理器、控制器、微控制器等。处理器401执行上文所描述的多个方法和处理，例如路面不平度的分类识别方法。
在一些实施例中，路面不平度的分类识别方法可被实现为计算机程序，其被有形地包含于计算机可读存储介质，例如存储单元18。在一些实施例中，计算机程序的部分或者全部可以经由ROM 402和/或通信单元19而被载入和/或安装到车辆400上。当计算机程序加载到RAM 403并由处理器401执行时，可以执行上文描述的路面不平度的分类识别方法的一个或多个步骤。备选地，在其他实施例中，处理器401可以通过其他任何适当的方式(例如，借助于固件)而被配置为执行路面不平度的分类识别方法。
本文中以上描述的系统和技术的多种实施方式可以在数字电子电路系统、集成电路系统、场可编程门阵列(Field Programmable Gate Array，FPGA)、专用集成电路(Application Specific Integrated Circuit，ASIC)、专用标准产品(ASSP)、芯片上的系统(System on Chip，SOC)、复杂可编程逻辑设备(Complex Programmable Logic Device，CPLD)、计算机硬件、固件、软件、和/或它们的组合中实现。这些多种实施方式可以包括：实施在一个或者多个计算机程序中，该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释，该可编程处理器可以是专用或者通用可编程处理器，可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令，并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。
用于实施本申请的方法的计算机程序可以采用一个或多个编程语言的任何组合来编写。这些计算机程序可以提供给通用计算机、专用计算机或其他 可编程数据处理装置的处理器，使得计算机程序当由处理器执行时使流程图和/或框图中所规定的功能/操作被实施。计算机程序可以完全在机器上执行、部分地在机器上执行，作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。
在本申请的上下文中，计算机可读存储介质可以是有形的介质，其可以包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的计算机程序。计算机可读存储介质可以包括电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备，或者上述内容的任何合适组合。备选地，计算机可读存储介质可以是机器可读信号介质。机器可读存储介质的示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、RAM、ROM、可擦除可编程只读存储器(Erasable Programmable Read-Only Memory，EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。
为了提供与用户的交互，可以在车辆上实施此处描述的系统和技术，该车辆具有：设置为向用户显示信息的显示装置(例如，阴极射线管(Cathode Ray Tube，CRT)或者液晶显示器(Liquid Crystal Display，LCD)监视器)；以及键盘和指向装置(例如，鼠标或者轨迹球)，用户可以通过该键盘和该指向装置来将输入提供给车辆。其它种类的装置还可以设置为提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈(例如，视觉反馈、听觉反馈、或者触觉反馈)；并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如，作为数据服务器)、或者包括中间件部件的计算系统(例如，应用服务器)、或者包括前端部件的计算系统(例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如，通信网络)来将系统的部件相互连接。通信网络的示例包括：局域网(Local Area Network，LAN)、广域网(Wide Area Network，WAN)、区块链网络和互联网。
计算系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。服务器可以是云 服务器，又称为云计算服务器或云主机，是云计算服务体系中的一项主机产品，以解决了传统物理主机与虚拟专用服务器(Virtual Private Server，VPS)服务中，存在的管理难度大，业务扩展性弱的缺陷。
可以使用上面所示的多种形式的流程，重新排序、增加或删除步骤。例如，本申请中记载的多个步骤可以并行地执行也可以顺序地执行也可以不同的次序执行，只要能够实现本申请的技术方案所期望的结果，本文在此不进行限制。

Claims (10)

  1. 一种路面不平度的分类识别方法，包括：
    获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；
    在车辆的行驶预测轨迹上出现前车遮挡的情况下，采集所述前车的实时位姿数据；
    将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
  2. 根据权利要求1所述的方法，其中，所述获取至少一组训练数据序列，包括：
    针对一组目标车辆和目标前车执行下述操作：
    在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡所述行驶预测轨迹时，记录第一时间点和所述目标前车的目标位置，并在所述第一时间点后采集所述目标前车的实时位姿数据；
    在根据所述目标前车的实时位姿数据初步确定所述目标前车所处的目标路面严重不平的情况下，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置；
    响应于所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置，在所述目标车辆按照所述行驶预测轨迹行驶至所述目标位置后，采集所述目标车辆的实时动力学响应信号；
    根据所述目标车辆的实时动力学响应信号再次确定所述目标路面的不平度分类识别结果，并将所述不平度分类识别结果，作为与所述目标路面匹配的标签数据序列；
    将所述标签数据序列与所述目标前车的实时位姿数据在时序上对齐，得到对齐结果，并将所述对齐结果作为一组训练数据序列。
  3. 根据权利要求2所述的方法，其中，所述目标前车的实时位姿数据包括实时车速和实时侧倾角；
    所述根据所述目标前车的实时位姿数据确定所述目标前车所处的目标路面严重不平，包括：
    获取所述目标前车的当前车速和当前侧倾角；
    根据所述当前车速和所述当前侧倾角，计算所述目标前车的当前侧倾方差值；
    根据所述当前侧倾方差值和预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平。
  4. 根据权利要求3所述的方法，其中，所述根据所述当前车速和所述当前侧倾角，计算所述目标前车的当前侧倾方差值，包括：
    根据所述当前车速、预设采集周期和预设不平度判断区间长度，计算在当前时刻对历史目标前车位姿数据的需求数量；
    根据所述需求数量，获取对应数量的历史目标前车侧倾角；
    根据所述需求数量和所述对应数量的历史目标前车侧倾角，计算所述目标前车的当前侧倾方差值。
  5. 根据权利要求3所述的方法，其中，所述根据所述当前侧倾方差值和预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平，包括：
    判断所述当前侧倾方差值是否大于所述预设方差阈值；
    响应于所述当前侧倾方差值大于所述预设方差阈值，确定所述目标前车在当前时刻所处的目标路面严重不平。
  6. 根据权利要求2所述的方法，其中，所述判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置，包括：
    在所述目标车辆的行驶预测轨迹上出现所述目标前车遮挡时，获取所述目标车辆与所述目标前车之间在所述行驶预测轨迹上的预测长度，以及所述目标车辆的当前车速；
    根据所述预测长度和所述当前车速，获取所述目标车辆按照所述行驶预测轨迹行驶至所述目标位置的预测时长，并获取所述目标车辆预测行驶至所述目标位置的第二时间点；
    采集所述目标车辆在所述第一时间点的第一横摆角速度，所述目标车辆实际行驶至所述第二时间点的第二横摆角速度，并采集所述目标车辆在所述第一时间点和所述第二时间点之间的实时车速；
    根据所述实时车速、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的实际行驶长度；
    根据所述第一横摆加速度、所述第二横摆角速度、所述第一时间点和所述第二时间点，计算所述目标车辆在所述第一时间点和所述第二时间点之间的横摆角速度积分值；
    根据所述实际行驶长度、所述预测长度、所述横摆角速度积分值和预设积 分阈值，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置。
  7. 根据权利要求6所述的方法，其中，所述根据所述实际行驶长度、所述预测长度、所述横摆角速度积分值和预设积分阈值，判断所述目标车辆在所述第一时间点后是否按照所述行驶预测轨迹行驶至所述目标位置，包括：
    在所述实际行驶长度与所述预测长度之间的第一差值不超过预设差值阈值，且所述横摆角速度积分值小于所述预设积分阈值的情况下，确定所述目标车辆在所述第一时间点后按照所述行驶预测轨迹行驶至所述目标位置。
  8. 一种路面不平度的分类识别装置，包括：
    路面不平度预测模型获取模块，设置为获取至少一组训练数据序列，根据所述至少一组训练数据序列对预设神经网络模型进行至少一轮训练，得到路面不平度预测模型；
    位姿数据采集模块，设置为在车辆的行驶预测轨迹上出现前车遮挡的情况下，采集所述前车的实时位姿数据；
    分类识别结果确定模块，设置为将所述实时位姿数据输入至所述路面不平度预测模型，确定与所述前车的行驶路面匹配的不平度分类识别结果。
  9. 一种车辆，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的计算机程序，所述计算机程序被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1-7中任一项所述的路面不平度的分类识别方法。
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机指令，所述计算机指令用于使处理器执行时实现权利要求1-7中任一项所述的路面不平度的分类识别方法。
PCT/CN2023/116866 2022-09-07 2023-09-05 路面不平度的分类识别方法、装置、车辆及存储介质 WO2024051661A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211088233.2 2022-09-07
CN202211088233.2A CN115451901A (zh) 2022-09-07 2022-09-07 一种路面不平度的分类识别方法、装置、车辆及存储介质

Publications (1)

Publication Number Publication Date
WO2024051661A1 true WO2024051661A1 (zh) 2024-03-14

Family

ID=84302722

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/116866 WO2024051661A1 (zh) 2022-09-07 2023-09-05 路面不平度的分类识别方法、装置、车辆及存储介质

Country Status (2)

Country Link
CN (1) CN115451901A (zh)
WO (1) WO2024051661A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115451901A (zh) * 2022-09-07 2022-12-09 中国第一汽车股份有限公司 一种路面不平度的分类识别方法、装置、车辆及存储介质
CN115937046B (zh) * 2023-01-09 2023-05-26 禾多科技（北京）有限公司 道路地面信息生成方法、装置、设备和计算机可读介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107977641A (zh) * 2017-12-14 2018-05-01 东软集团股份有限公司 一种智能识别地形的方法、装置、车载终端及车辆
CN109050535A (zh) * 2018-07-25 2018-12-21 北京理工大学 一种基于车辆姿态的快速地形工况辨识方法
US20190344783A1 (en) * 2018-05-14 2019-11-14 GM Global Technology Operations LLC Autonomous ride dynamics comfort controller
CN111290386A (zh) * 2020-02-20 2020-06-16 北京小马慧行科技有限公司 路径规划方法及装置、运载工具
CN112896188A (zh) * 2021-02-22 2021-06-04 浙江大学 一种考虑前车遭遇的自动驾驶决策控制的系统
CN115451901A (zh) * 2022-09-07 2022-12-09 中国第一汽车股份有限公司 一种路面不平度的分类识别方法、装置、车辆及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107977641A (zh) * 2017-12-14 2018-05-01 东软集团股份有限公司 一种智能识别地形的方法、装置、车载终端及车辆
US20190344783A1 (en) * 2018-05-14 2019-11-14 GM Global Technology Operations LLC Autonomous ride dynamics comfort controller
CN109050535A (zh) * 2018-07-25 2018-12-21 北京理工大学 一种基于车辆姿态的快速地形工况辨识方法
CN111290386A (zh) * 2020-02-20 2020-06-16 北京小马慧行科技有限公司 路径规划方法及装置、运载工具
CN112896188A (zh) * 2021-02-22 2021-06-04 浙江大学 一种考虑前车遭遇的自动驾驶决策控制的系统
CN115451901A (zh) * 2022-09-07 2022-12-09 中国第一汽车股份有限公司 一种路面不平度的分类识别方法、装置、车辆及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SHENGFENG GU: "Research on Neural Network Method for Recognizing Road Surface Roughness Based on Vehicle Response", ENGINEERING SCIENCE AND TECHNOLOGY II, CHINA MASTER'S THESES FULL-TEXT DATABASE, 1 June 2018 (2018-06-01), XP093147480, [retrieved on 20240403] *

Also Published As

Publication number Publication date
CN115451901A (zh) 2022-12-09

Similar Documents

Publication Publication Date Title
WO2024051661A1 (zh) 路面不平度的分类识别方法、装置、车辆及存储介质
CN110134126B (zh) 轨迹匹配方法、装置、设备和介质
CN105444770B (zh) 基于智能手机的车道级别地图生成和定位方法
EP3957955A2 (en) Vehicle locating method and apparatus, electronic device, storage medium and computer program product
WO2022247203A1 (zh) 自动驾驶车辆的控制方法、装置、设备以及存储介质
KR20210038852A (ko) 조기 경보 방법, 장치, 전자 기기, 컴퓨터 판독 가능 저장 매체 및 컴퓨터 프로그램
WO2022222401A1 (zh) 自主泊车的方法、装置、设备及自动驾驶车辆
WO2023221963A1 (zh) 跟随误差确定方法、装置、设备及存储介质
WO2024087887A1 (zh) 鲁莽驾驶行为标记方法、车辆、云端服务器和存储介质
EP4170285A1 (en) Method and apparatus for constructing three-dimensional map in high-definition map, device and storage medium
CN115903831A (zh) 一种车辆驾驶控制方法、装置、车辆及存储介质
EP4120206A1 (en) Information processing method and apparatus, electronic device, storage medium, and product
WO2024001758A1 (zh) 一种代客泊车车速的确定方法、装置、设备及介质
CN114620013A (zh) 一种车辆前方行人的保护方法、装置、设备及介质
WO2023241556A1 (zh) 泊车控制方法、装置、设备和存储介质
WO2023237081A1 (zh) 一种泊车避障方法、装置、电子设备及存储介质
US20230126172A1 (en) Method of outputting prompt information, device, medium, and vehicle
CN116358584A (zh) 一种自动驾驶车辆路径规划方法、装置、设备及介质
WO2022257488A1 (zh) 基于自动驾驶的乘车方法、装置、设备和存储介质
CN115871683A (zh) 一种车辆横摆角速度确定方法、装置、车辆及介质
CN115583258A (zh) 自动驾驶车辆会车控制方法、装置、车辆控制设备和介质
CN115959154A (zh) 变道轨迹的生成方法、装置和存储介质
CN114282776A (zh) 车路协同评估自动驾驶安全性的方法、装置、设备和介质
WO2024007569A1 (zh) 航位预测方法、装置、设备及介质
TWI830415B (zh) 碰撞預估方法、裝置及電腦可讀存儲介質

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23862355

Country of ref document: EP

Kind code of ref document: A1