CN117889880A

CN117889880A - Vehicle planning track testing method, device, computer equipment and storage medium

Info

Publication number: CN117889880A
Application number: CN202410016013.1A
Authority: CN
Inventors: 倪灿斌
Original assignee: Ningbo Lutes Robotics Co ltd
Current assignee: Ningbo Lutes Robotics Co ltd
Priority date: 2024-01-05
Filing date: 2024-01-05
Publication date: 2024-04-16

Abstract

The present application relates to the field of intelligent driving test technologies, and in particular, to a method and apparatus for testing a planned trajectory of a vehicle, a computer device, and a storage medium. The method comprises the following steps: acquiring running data acquired frame by frame, wherein each frame of running data comprises a planning track corresponding to the acquisition time; the driving data acquired frame by frame are deserialized and regulated to be under the same preset coordinate system; according to the execution sequence of different detection modes, selecting a detection mode matched with each frame of running data, and according to the selected detection mode, carrying out abnormal detection on each frame of running data until each detection mode matched with each frame of running data is detected. The method can improve the test efficiency of the planned track in the real vehicle or simulation process.

Description

Vehicle planning track testing method, device, computer equipment and storage medium

Technical Field

The present application relates to the field of intelligent driving test technologies, and in particular, to a method and apparatus for testing a planned trajectory of a vehicle, a computer device, and a storage medium.

Background

The intelligent driving test verification work is an important link in the landing process of the intelligent driving technology and is also an important support for the development of the intelligent driving technology. The intelligent driving integrates the traditional vehicle technology, artificial intelligence, sensor, pattern recognition and other technologies. The intelligent driving test needs to traverse massive test working conditions and cover as many scenes as possible so as to optimize and improve the capability of the intelligent driving algorithm. Besides the traditional real vehicle test, the test means also have various simulation tests, such as WorldSim, logSim, hardware-in-the-loop test and the like.

Whether the test is a real vehicle test or a simulation test, an important test link is that the test verification of the planned track is realized. In the intelligent driving process, an intelligent driving algorithm plans a driving track in advance with a certain frequency according to the factors of the current driving state, road conditions, environment, traffic participants and the like, and guides the vehicle to drive, and each time generates a frame. A frame of planned trajectory consists of a number of trajectory points, each point including, but not limited to, information of coordinates, speed, acceleration, attitude angle, etc. expected by the vehicle. At present, whether a real vehicle test or a simulation test is carried out, the test verification work for the planned track is mainly carried out or is stopped in a manual identification stage, namely, the planned track in the vehicle running process is observed by a person frame by frame, and when an abnormality occurs, calibration and behavior judgment are carried out. This approach is straightforward and efficient, but is inefficient and has a high false positive rate.

Therefore, there is a need for a vehicle planned trajectory testing method, apparatus, computer device, and storage medium that can improve the testing efficiency of planned trajectories in real vehicles or simulation processes.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a vehicle planned trajectory test method, apparatus, computer device, computer-readable storage medium, and computer program product that can improve the test efficiency of planned trajectories in a real vehicle or simulation process.

In a first aspect, the present application provides a vehicle planned trajectory testing method, including:

Acquiring running data acquired frame by frame, wherein each frame of running data comprises a planning track corresponding to the acquisition time;

The driving data acquired frame by frame are deserialized and regulated to be under the same preset coordinate system;

according to the execution sequence of different detection modes, selecting a detection mode matched with each frame of running data, and according to the selected detection mode, carrying out abnormal detection on each frame of running data until each detection mode matched with each frame of running data is detected.

In one embodiment, the detection modes are at least classified into jitter detection, continuity detection and centering detection; aiming at the current frame running data, under the condition that the current frame running data is not the first frame running data, the detection mode adapted to the current frame running data comprises jitter detection, continuity detection and centering detection; in the case that the current frame running data is the first frame running data, the detection mode adapted to the current frame running data includes jitter detection and centering detection.

In one embodiment, the detecting the abnormality of the driving data of each frame according to the selected detection mode includes:

aiming at the current frame of running data, determining an overlapping period between a current planning period corresponding to a planning track in the current frame of running data and a previous planning period corresponding to the planning track in the previous frame of running data under the condition that the selected detection mode is continuity detection and the current frame of running data is not the first frame of running data;

According to the overlapping time period, the planned track in the current frame of running data and the planned track in the previous frame of running data are intercepted respectively, and an overlapping planned track corresponding to the current frame and an overlapping planned track corresponding to the previous frame are obtained;

Sampling the corresponding overlapped planning track of the current frame and the corresponding overlapped planning track of the previous frame respectively to obtain a current sampling point sequence corresponding to the current frame and a previous sampling point sequence corresponding to the previous frame, wherein the number of sampling points in the current sampling point sequence is the same as that of the previous sampling points in the previous sampling point sequence;

And determining the continuity of the previous frame and the current frame by adopting an algorithm capable of describing the spatial similarity based on the distance between the sampling points of the current sampling point sequence and the previous sampling point sequence.

In one embodiment, the planned trajectory includes at least one travel data item of the vehicle; the abnormal detection of each frame of driving data according to the selected detection mode comprises the following steps:

aiming at the current frame running data, under the condition that the selected detection mode is jitter detection, performing smoothing treatment on a planned track in the current frame running data to obtain a current reference track;

Respectively sampling the current reference track and the planned track in the current frame running data to obtain a first sampling point sequence corresponding to the current reference track and a second sampling point sequence corresponding to the planned track in the current frame running data;

Aiming at each group of counterpoint sampling points between the first sampling point sequence and the second sampling point sequence, acquiring a difference curve between the first sampling point sequence and the second sampling point sequence under each driving data item according to the difference between the driving data items at different sampling points in the counterpoint sampling points;

converting a difference curve serving as a time domain signal into a frequency domain signal, and calculating the integral energy value of the frequency domain signal in different frequency intervals;

And determining whether the current frame of running data has jitter or not and corresponding jitter intensity under the condition of the jitter according to the integral energy values of the frequency domain signals in different frequency intervals.

In one embodiment, each frame of driving data further comprises two side lane lines near the current position of the vehicle corresponding to the acquisition time; the abnormal detection of each frame of driving data according to the selected detection mode comprises the following steps:

aiming at the current frame running data, determining track points on a planned track in the current frame running data under the condition that the selected detection mode is centered detection;

for each track point, acquiring the distance between the track point and lane lines on two sides respectively;

And under the condition that the distance difference between the track points and the lane lines at the two sides exceeds a preset threshold value, determining that the planned track in the current frame of driving data has an out-of-centering phenomenon.

In one embodiment, each frame of travel data further includes a vehicle control signal; the method further comprises the steps of:

Under the condition that the planned track in the current frame of running data is not centered and the vehicle control signal comprises a control signal for actively changing centered running, determining that the current frame of running data belongs to normal running behavior in planning;

And determining that the planned track in the current frame of running data is abnormal under the condition that the planned track in the current frame of running data is not centered and the vehicle control signal does not comprise a control signal for actively changing centered running.

In a second aspect, the present application also provides a vehicle planned trajectory testing device, including:

The acquisition module is used for acquiring the driving data acquired frame by frame, and each frame of driving data comprises a planning track corresponding to the acquisition time;

the processing module is used for deserializing and regulating the driving data acquired frame by frame to the same preset coordinate system;

The detection module is used for selecting a detection mode matched with the running data of each frame according to the execution sequence of different detection modes, and carrying out abnormal detection on the running data of each frame according to the selected detection mode until the detection of each detection mode matched with the running data of each frame is completed.

In a third aspect, the present application also provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

In a fifth aspect, the application also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of:

According to the vehicle planning track testing method, the vehicle planning track testing device, the computer equipment, the computer readable storage medium and the computer program product, by automatically detecting the abnormality of the frame-by-frame driving data, a great deal of manpower and time cost can be saved compared with manual frame-by-frame observation. And the adaptive detection mode is adopted to detect the running data of each frame, so that the abnormal condition can be more accurately identified and judged, and the misjudgment rate is reduced. By adjusting the preset execution sequence to the same coordinate system, the test flow can be more standardized and standardized. By using the method, various test working conditions and scenes can be more comprehensively covered, so that the accuracy and the credibility of the test are improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for those skilled in the art.

FIG. 1 is an application environment diagram of a vehicle planned trajectory test method in one embodiment;

FIG. 2 is a flow chart of a method of testing a planned trajectory of a vehicle in one embodiment;

FIG. 3 is a flow chart of a continuity check in another embodiment;

FIG. 4 is a schematic diagram of a time-domain overlapping portion of a two-frame planned trajectory in continuity testing, in one embodiment;

FIG. 5 is a schematic diagram of the Frechet distance of a two-frame planned trajectory in continuity testing, in one embodiment;

FIG. 6 is a flow chart of jitter detection in another embodiment;

FIG. 7 is a flow chart of centering detection in another embodiment;

FIG. 8 is a schematic diagram of planned trajectories and lane lines in centering detection in one embodiment;

FIG. 9 is a block diagram of a vehicle planned trajectory testing device in one embodiment;

fig. 10 is an internal structural view of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The vehicle planning track testing method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server.

The server 104 acquires the driving data acquired frame by frame, wherein each frame of driving data comprises a planning track corresponding to the acquisition time; the driving data acquired frame by frame are deserialized and regulated to be under the same preset coordinate system; according to the execution sequence of different detection modes, selecting a detection mode matched with each frame of running data, and according to the selected detection mode, carrying out abnormal detection on each frame of running data until each detection mode matched with each frame of running data is detected.

The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.

1. And (3) data source acquisition: the running data are collected frame by frame in the process of real vehicles or simulation tests through various data collection tools in software/hardware forms. The driving data collected by each frame need to include: time stamps, planned trajectory data, lane line data near the current position of the vehicle, vehicle control signals, etc. Each piece of planning trajectory data is composed of a series of points, each of which contains, but is not limited to, information of coordinates, velocity, acceleration, attitude angle, and the like. Each lane line data is composed of a series of points, each of which contains, but is not limited to, information such as coordinates, curvature, and the like. Currently, it is common practice to collect each frame of data by recording ROS packets.

2. Data preprocessing: after the data source is collected, specific information such as planning track data, lane line data and the like is extracted and deserialized through automatic software/hardware. Different types of data sources may provide points in different coordinate systems, such as WGS-84, ENU, UTM, bicycle, etc. For such a small-scale observation of the planned trajectory, a vehicle coordinate system, an ENU coordinate system, or the like is generally used according to the actual situation. After data acquisition is completed, the information under different coordinate systems is firstly regulated to a specific coordinate system, so that subsequent analysis is facilitated. The data frames after preprocessing are arranged in the order of time stamps.

3. Detecting service traversal: each frame of running data corresponds to different detection services, the detection services are to perform abnormality detection on each frame of running data in a detection mode, each detection mode serves a specific detection service, and specific detection content, detection weight (priority), abnormality processing strategies and the like can be configured.

A) And (5) sequencing weights: and sequencing the execution sequence of the detection service according to the configured weight value, wherein the high weight item is preferentially executed.

B) Determining detection service: and determining the service to be detected of the driving data of each frame in sequence from high to low according to the weight sequence.

C) Executing the service to be detected: the detection service is executed according to the internal logic.

D) And recording a detection result.

E) And (3) judging a detection result: detecting that the result is abnormal, and jumping to f); and (5) the detection result is normal, and the process jumps to g).

F) Abnormality processing policy determination: if the abnormality severity is higher, the subsequent service to be detected is not required to be detected, and the user directly exits; if the severity of the abnormality is within an acceptable range, only record is needed, jump to g).

G) Detecting mode residual judgment: if not, exiting; if the service to be detected still exists, the step b) is skipped to continue traversing.

4. And (3) recording a detection result: recording the detection result for the test and developer to review and analyze.

Specifically, in an exemplary embodiment, as shown in fig. 2, a vehicle planned trajectory test method is provided, and an example of application of the method to the server in fig. 1 is described, including the following steps S202 to S206. Wherein:

step S202, acquiring driving data acquired frame by frame, wherein each frame of driving data comprises a planning track corresponding to the acquisition time.

Specifically, in the real vehicle or simulation process, data are collected frame by frame, and the data of each frame comprise planning track information of the vehicle at the time of collection, lane lines on two sides of a lane where the vehicle is currently located and vehicle control signals. The planned trajectory information is a prediction of the vehicle driving state over a future period of time. The origin of the planned trajectory is the current position of the vehicle. Each track point in the planned track sequentially comprises data such as the position, the speed, the acceleration, the attitude angle and the like of the vehicle at a certain future moment, and a time sequence is formed. The data for each frame will include the position of the vehicle at that time and its trajectory according to the plan. The running data collected frame by frame can cover the states of the vehicle at different moments, and the track path of the vehicle in the whole running process can be restored through the data. The acquisition mode can provide detailed vehicle motion information, and is beneficial to subsequent analysis and processing. The lane lines on both sides of the lane where the vehicle is currently located are road markings indicating the running direction of the road vehicle. On most roads, lane lines are drawn as solid or dashed lines for separating lanes of different directions of travel. The solid line indicates the lane change prohibited, and the broken line indicates the lane change permitted. During the central driving of the vehicle, the vehicle control signal information is used for controlling the driving of the vehicle and keeping the vehicle at the correct position.

The collected travel data may be used for a variety of purposes, such as:

and (3) analyzing the running track: by means of the frame-by-frame data, the accurate track path of the vehicle can be restored and used for analyzing the running mode, the path planning effect and the like of the vehicle.

Data driven algorithm: the acquired driving data can be used for training and improving a driving control algorithm, improving the automatic driving performance of the vehicle or optimizing a path planning algorithm.

Step S204, the driving data collected frame by frame are deserialized and normalized to the same preset coordinate system.

In particular, during real vehicles or simulation, different sensors and devices may use different coordinate systems to record vehicle position and motion information. For data analysis and processing, it is necessary to uniformly transform these data into the same coordinate system to ensure consistency and comparability of the data.

The step of normalizing the travel data to the same preset coordinate system generally comprises the following aspects:

Coordinate system definition: a predetermined coordinate system, such as the earth longitude and latitude coordinate system or the vehicle coordinate system, is determined.

Calibration conversion: and converting the coordinate data acquired by each sensor from the original coordinate system to a preset coordinate system by using a calibration method and related calibration data. This may include translation, rotation, scaling, etc. conversion processes.

Alignment of data: and the time mark of each frame of acquired driving data is aligned with the corresponding planning track data, so that the time consistency is ensured.

By deserializing and regularizing all the driving data to the same preset coordinate system, the data can have a uniform reference frame, and subsequent data processing, analysis and application are facilitated. Therefore, integration and comparison of the data across the sensors can be realized, and the effectiveness and usability of the data are further improved.

Step S206, selecting a detection mode matched with each frame of running data according to the execution sequence of different detection modes, and performing abnormal detection on each frame of running data according to the selected detection mode until each detection mode matched with each frame of running data is detected.

Specifically, characteristics and requirements of each frame of driving data are selected to perform abnormality detection in a most appropriate detection mode. Different detection modes may be suitable for different types of anomaly detection, such as track offset, speed anomalies, acceleration anomalies, and so on. And carrying out anomaly detection on each frame of driving data according to the execution sequence of the selected detection mode. Each detection mode analyzes and judges the data according to the algorithm or rule thereof, and identifies possible abnormal conditions. And for each frame of running data, sequentially applying the selected detection modes to detect until all the detection modes detect each frame of running data, and recording whether each detection mode detects abnormal conditions. By selecting the detection mode which is adapted to the running data of each frame according to the execution sequence of the different detection modes and detecting the abnormality of the detection mode, the capability of discovering the potential abnormality in the running data can be improved. Thus, the safety and reliability of the driving data can be enhanced, and a trust basis is provided for subsequent data analysis and processing.

According to the vehicle planning track testing method, the abnormality detection is carried out on the frame-by-frame driving data through automation, and compared with manual frame-by-frame observation, a large amount of manpower and time cost can be saved. And the adaptive detection mode is adopted to detect the running data of each frame, so that the abnormal condition can be more accurately identified and judged, and the misjudgment rate is reduced. By adjusting the preset execution sequence to the same coordinate system, the test flow can be more standardized and standardized. By using the method, various test working conditions and scenes can be more comprehensively covered, so that the accuracy and the credibility of the test are improved.

In an exemplary embodiment, the detection modes are at least classified into jitter detection, continuity detection, and centering detection; aiming at the current frame running data, under the condition that the current frame running data is not the first frame running data, the detection mode adapted to the current frame running data comprises jitter detection, continuity detection and centering detection; in the case that the current frame running data is the first frame running data, the detection mode adapted to the current frame running data includes jitter detection and centering detection.

Specifically, when the current frame driving data is not the first frame driving data, the detection mode adapted to the current frame driving data includes jitter detection, continuity detection and centering detection. When the current frame driving data is the first frame driving data, the detection mode adapted to the current frame driving data comprises jitter detection and centering detection. Wherein:

And (3) jitter detection: for detecting the presence of a shaking or vibration condition in the driving data. Continuous fluctuations or instabilities in the detection data may be involved. If the single frame planning track has the shaking phenomenon, the vehicle can be guided to swing left and right in the subsequent running, and the vehicle is in dangerous driving behavior.

Note that, the object of jitter detection may be:

jitter on coordinates: refers to a sharp change or continuous fluctuation of position in the travel track. When the running track has frequent position change or intense shake, the unstable condition of the vehicle during running is indicated.

Jitter in speed: the rapid change of the vehicle speed is meant to indicate that the abnormal acceleration or deceleration phenomenon, such as sudden rapid acceleration or rapid braking, occurs in the vehicle during running. This abrupt change in speed is also known as a dithering behavior.

Dithering of attitude angle: refers to a drastic change in the direction of travel, typically as a result of the steering behavior of the vehicle. Abnormal steering behavior such as frequent lane change or sharp turning is regarded as a shake phenomenon of the attitude angle.

And (3) continuity detection: for detecting continuity or persistence in the travel data. It may involve detecting whether the continuity state or behavior in the data is consistent with expectations. The track of the front frame and the track of the rear frame should have higher continuity, namely the rear frame should be the natural extension of the front frame, the planned track of the front frame and the rear frame is too different, the vehicle can not adapt to the track of the abrupt change of the next frame in the current running state, the vehicle can not run according to the track, and meanwhile, the track which is frequently changed also causes panic and trouble to passengers. It should be noted that, in general, continuity detection is performed by determining continuity of two adjacent frames of data using a coordinate position of a track point.

Centering detection: for detecting a center position or a focus condition in the traveling data. It may involve detecting whether the location or value in the data is off-center or in an aggregate area. When the actions such as lane changing, avoidance and the like are not triggered, the planned track should be kept in the middle of the lane all the time.

In this embodiment, according to different conditions of the driving data, different detection modes are adapted, and shake detection, continuity detection and centering detection are performed respectively. Through the combination of the detection modes, the abnormal detection and analysis can be more comprehensively carried out on the running data so as to improve the quality and reliability of the running data.

In an exemplary embodiment, as shown in fig. 3, the step of detecting abnormality of the driving data of each frame according to the selected detection manner includes steps S302 to S308. Wherein:

Step S302, aiming at the current frame of running data, determining an overlapping period between a current planning period corresponding to a planning track in the current frame of running data and a previous planning period corresponding to the planning track in the previous frame of running data under the condition that the selected detection mode is continuity detection and the current frame of running data is not first frame of running data;

Step S304, according to the overlapping time period, the planned track in the current frame of running data and the planned track in the previous frame of running data are intercepted respectively, and the corresponding overlapping planned track of the current frame and the corresponding overlapping planned track of the previous frame are obtained;

step S306, sampling the corresponding overlapped planned track of the current frame and the corresponding overlapped planned track of the previous frame respectively to obtain a corresponding current sampling point sequence of the current frame and a corresponding previous sampling point sequence of the previous frame, wherein the number of sampling points in the current sampling point sequence is the same as that of the previous sampling point sequence;

Step S308, for the sampling points between the current sampling point sequence and the previous sampling point sequence, determining the continuity of the previous frame and the current frame by adopting an algorithm capable of describing the spatial similarity based on the distance between the sampling points.

Specifically, first, on the premise that the selected detection mode is continuity detection, for the current frame of running data, an overlapping period corresponding to the previous frame of running data needs to be determined. This overlap period represents a period in which the planned trajectory of the current frame and the planned trajectory of the previous frame overlap each other. And then, according to the overlapping time period, intercepting the planned track of the current frame and the planned track of the previous frame to obtain the overlapped planned track of the current frame and the overlapped planned track of the previous frame. And then, sampling the overlapped planning track of the current frame and the overlapped planning track of the previous frame to obtain a corresponding current sampling point sequence and a corresponding previous sampling point sequence. Importantly, the number of sampling points of the two sequences should be the same, ensuring consistency of subsequent alignment comparisons.

In the continuity test, a Freche distance (Fre CHET DISTANCE) algorithm can be used, which is used to compare the similarity between two sequences and which calculates the degree of overall difference between the sequences.

Specifically, when comparing the sample point sequence of the previous frame with the sample point sequence of the current frame, the fraiche distance algorithm calculates a distance (e.g., euclidean distance or manhattan distance) between each point in the sample point sequence of the previous frame and each point in the sample point sequence of the current frame, and generates a distance matrix. This distance matrix describes the distance between each point in the sequence of sampling points of the previous frame and each point in the sequence of sampling points of the current frame. Then, by calculating the infinitesimal bounds of the maximum values of all elements in the distance matrix, a comprehensive fraiche distance is obtained for representing the spatial similarity between the sampling point sequence of the previous frame and the sampling point sequence of the current frame. The smaller the value of the friendship distance, the higher the similarity in space between the sampling point sequence of the previous frame and the sampling point sequence of the current frame, namely the better the continuity. Thus, whether the planned track has continuity or continuity in time sequence can be judged.

Illustratively, the current frame travel data is as follows:

Current frame period: [ t2, t5]

Current frame planning trajectory: [ Point A, point B, point C, point D ]

The previous frame of travel data is as follows:

The previous frame time: [ t0, t3]

Planning a track in the previous frame: [ Point X ', point Y ', point A ', point B ]

The overlap period may be determined to be [ t2, t3], i.e., there is overlap in time of the previous frame travel data and the current frame travel data. In the overlapping period [ t2, t3], the overlapping planned trajectories are respectively: the overlapping planning trajectory of [ point a, point B ] and [ point a ', point B' ] is shown in fig. 4.

For the overlapping planned track of the current frame and the overlapping planned track of the previous frame, it is assumed that the two tracks are sampled at the same interval to obtain a sampling point sequence, and the number of the sampling points is the same.

Sample point sequence of current frame: sample point 1, sample point 2, sample point 3, sample point 4;

sample point sequence of previous frame: sample point 5, sample point 6, sample point 7, sample point 8);

And comparing the similarity according to the sampling results of the two tracks. In the present application, as shown in fig. 5, the fraiche distance algorithm is used to evaluate the similarity of two tracks. A smaller distance indicates that the tracks are spatially closer, indicating a higher continuity between the previous frame and the current frame.

In this embodiment, the planned trajectory in the current frame of travel data is compared with the planned trajectory in the previous frame of travel data, and the continuity thereof is determined according to the distance between the sampling points. This makes it possible to evaluate the continuity of the driving data and to identify possible anomalies.

In one exemplary embodiment, the planned trajectory includes at least one travel data item of the vehicle; according to the selected detection mode, the anomaly detection is performed on each frame of running data, as shown in fig. 6, including:

Step S602, for the current frame running data, performing smoothing processing on a planned track in the current frame running data to obtain a current reference track under the condition that the selected detection mode is jitter detection;

step S606, sampling the current reference track and the planned track in the current frame running data respectively to obtain a first sampling point sequence corresponding to the current reference track and a second sampling point sequence corresponding to the planned track in the current frame running data;

Step S606, aiming at each group of counterpoint sampling points between the first sampling point sequence and the second sampling point sequence, acquiring a difference curve between the first sampling point sequence and the second sampling point sequence under each driving data item according to the difference between the driving data items at different sampling points in the counterpoint sampling points;

Step S608, converting the difference curve as a time domain signal into a frequency domain signal, and calculating the integral energy value of the frequency domain signal in different frequency intervals;

In step S610, it is determined whether the current frame of driving data has jitter and the corresponding jitter intensity in the presence of jitter according to the overall energy values of the frequency domain signal in different frequency intervals.

The smoothing processing refers to performing smoothing processing on a planned track in the current frame of running data to obtain a current reference track. The purpose of the smoothing process is to reduce noise and jitter in the track, making it smoother and more continuous. And a polynomial fitting algorithm based on a least square method can be adopted to perform track fitting, so that a required reference track is obtained. The data to be fitted may be selected as desired, including but not limited to: coordinates (independent variable: track point time, dependent variable: track point coordinate position), speed (independent variable: track point time, dependent variable: track point speed), acceleration (independent variable: track point time, dependent variable: track point acceleration), attitude angle (independent variable: track point time, dependent variable: track point attitude angle), and the like.

Sampling refers to sampling the current reference track and the planned track in the current frame driving data to obtain a corresponding first sampling point sequence and a corresponding second sampling point sequence. The purpose of the sampling is to discretize the trajectory data for subsequent variance analysis. It is necessary to ensure that the sampling points of the two planned trajectories are identical.

The difference curve calculation refers to calculating a difference curve according to differences between track point running data items at different sampling points for each group of pairs of sampling points between the first sampling point sequence and the second sampling point sequence. The difference curve describes the difference between the two sequences, which is used to characterize the extent of change of the track.

The frequency domain conversion refers to performing frequency domain conversion by using the difference curve as a time domain signal. By converting the time domain signal into a frequency domain signal, the signal energy distribution over different frequency bins can be analyzed. In this embodiment, FFT may be used to perform frequency domain conversion, which includes the following steps:

a) Time domain windowing: in order to avoid the problems of the fence effect, the spectrum leakage and the like, the difference curve needs to be windowed. Windowing is the multiplication of window functions on the curve in the time domain, and common window functions include hamming windows, hanning windows, blackman windows and the like. The function of the window is to reduce spectral leakage, reduce spectral amplitude spectral fluctuations and the effects of spectral boundaries.

B) And (3) FFT sampling point filling: to avoid frequency domain deviations, sample point alignment of the difference curve is required. The purpose of the filling is to fill the number of sampling points to the power of 2 n. Common padding methods include head padding and tail padding, and the number of sampling points is ensured to be the power of 2 n, which is helpful for optimizing the calculation efficiency.

C) FFT processing using butterfly: the FFT algorithm is an efficient Fourier transform algorithm, and the Fourier transform is calculated rapidly by means of divide-and-conquer and iteration. The butterfly method is an implementation method of the FFT algorithm, and the fast Fourier transform process is decomposed into a plurality of butterfly operations in a divide-and-congregate mode. The butterfly operation is a basic complex operation, and a representation of the difference curve in the frequency domain, i.e. a frequency domain signal, can be obtained by repeatedly performing the butterfly operation.

Energy calculation refers to calculating the overall energy value of the frequency domain signal in different frequency bins. By calculating the energy values, the degree of signal variation over different frequency bins can be quantified.

Jitter evaluation refers to determining whether or not there is jitter in the current frame of running data based on the overall energy value of the frequency domain signal in different frequency intervals, and the intensity of the jitter in the presence of jitter. A higher energy value indicates that there is greater jitter.

In this embodiment, the jitter condition and the intensity of the jitter in the current frame of running data can be evaluated by performing smoothing, sampling, differential analysis and frequency domain conversion on the planned trajectory. Such an approach may help analyze and identify anomalies and instabilities in the vehicle travel data.

In an exemplary embodiment, each frame of travel data further includes two side lane lines near the current position of the vehicle at the corresponding time of acquisition; according to the selected detection mode, the anomaly detection is performed on each frame of running data, as shown in fig. 7, including:

step S702, determining track points on a planned track in the current frame running data under the condition that the selected detection mode is centered detection according to the current frame running data;

Step S704, for each track point, obtaining the distance between the track point and the lane lines on two sides respectively;

Step S706, determining that the planned track in the current frame of driving data is not centered when the difference value of the distances between the track points and the lane lines at the two sides exceeds a preset threshold value.

Specifically, lane lines on two sides of a current lane of the vehicle are obtained, the lane lines should cover the range of the planned track, specifically, the lane lines on two sides of the current lane of the vehicle are intercepted according to the length of the planned track, and the intercepting range needs to ensure the range capable of covering the planned track. And calculating the distance difference between each track point and the lane lines on two sides according to each track point on the planned track of the current frame of driving data. As shown in fig. 8, each track point of the planned track is traversed, a straight line is made perpendicular to the tangential direction of the track point with the track point as the center, the straight line intersects with the lane lines on both sides, and the distance from the intersection point to the track point is the distance from the track point to the left lane line and the right lane line. And comparing the distance difference value from each track point to the lane lines on the two sides with a preset threshold value. If the distance difference between the track point and the lane lines at the two sides exceeds a preset threshold value, the fact that the planned track of the current frame of driving data is not centered, namely that the vehicle deviates from the center position of the road, can be determined.

In this embodiment, by calculating the distance difference between the track point and the lane lines on both sides and comparing with a preset threshold value, it is possible to determine whether the vehicle deviates from the center position of the road. This process can be used to monitor the driving condition of the vehicle and to determine if there is an misalignment.

In one exemplary embodiment, each frame of travel data further includes a vehicle control signal; the method further comprises the steps of:

Specifically, during the running process of the vehicle, the control signal of the vehicle can be actively adjusted to enable the vehicle to return to the planned central track position, so that normal running is realized. On the other hand, if there is an un-centering phenomenon of the planned trajectory in the current frame of travel data, but the vehicle control signal does not include a control signal to actively change the centered travel, it may be determined that the planned trajectory in the current frame of travel data is abnormal. This means that the vehicle is not taking action to correct the off-centre trajectory during its travel, and there may be problems or anomalies that require further analysis and processing.

In this embodiment, whether the planned trajectory in the current frame of travel data is normal travel or abnormal behavior may be determined according to whether the vehicle control signal includes a control signal that actively changes the center travel. Actively adjusting the control signal of the vehicle may correct the trajectory deviation, and the lack of such a signal may mean that there is a problem or abnormality. Such a determination may help identify the driving status of the vehicle and perform relevant processing and adjustments.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a vehicle planning track testing device for realizing the vehicle planning track testing method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the vehicle planned trajectory testing device provided below may be referred to the limitation of the vehicle planned trajectory testing method hereinabove, and will not be repeated herein.

In one exemplary embodiment, as shown in fig. 9, there is provided a vehicle planned trajectory test device including:

The acquiring module 902 is configured to acquire driving data acquired frame by frame, where each frame of driving data includes a planned track corresponding to an acquisition time;

The processing module 904 is configured to deserialize and normalize the running data acquired frame by frame to the same preset coordinate system;

The detection module 906 is configured to select a detection mode adapted to each frame of running data according to an execution sequence of different detection modes, and perform anomaly detection on each frame of running data according to the selected detection mode until each detection mode adapted to each frame of running data is detected.

In an exemplary embodiment, the processing module 904 is further configured to determine, for the current frame of running data, an overlapping period between a current planning period corresponding to a planned track in the current frame of running data and a previous planning period corresponding to a planned track in the previous frame of running data, where the selected detection mode is continuity detection and the current frame of running data is not first frame of running data;

The detection module 906 is further configured to intercept, according to the overlapping period, the planned track in the current frame of running data and the planned track in the previous frame of running data, respectively, to obtain an overlapping planned track corresponding to the current frame and an overlapping planned track corresponding to the previous frame;

The detection module 906 is further configured to sample the overlapping planned track corresponding to the current frame and the overlapping planned track corresponding to the previous frame, respectively, to obtain a current sampling point sequence corresponding to the current frame and a previous sampling point sequence corresponding to the previous frame, where the number of sampling points in the current sampling point sequence is the same as the number of sampling points in the previous sampling point sequence;

the detection module 906 is further configured to determine, for a sampling point between the current sampling point sequence and the previous sampling point sequence, continuity of the previous frame and the current frame by using an algorithm capable of describing spatial similarity based on a distance between the sampling points.

In an exemplary embodiment, the detection module 906 is further configured to, for the current frame running data, perform smoothing on a planned track in the current frame running data to obtain a current reference track if the selected detection mode is jitter detection;

The detection module 906 is further configured to sample the current reference track and the planned track in the current frame running data, respectively, to obtain a first sampling point sequence corresponding to the current reference track and a second sampling point sequence corresponding to the planned track in the current frame running data;

The detection module 906 is further configured to obtain, for each set of alignment sampling points between the first sampling point sequence and the second sampling point sequence, a difference curve between the first sampling point sequence and the second sampling point sequence under each driving data item according to differences between driving data items at different sampling points in the alignment sampling points;

The detection module 906 is further configured to convert the difference curve, which is a time domain signal, into a frequency domain signal, and calculate an overall energy value of the frequency domain signal in different frequency intervals;

The detection module 906 is further configured to determine whether jitter exists in the current frame of running data and a corresponding jitter intensity degree when the jitter exists according to the overall energy value of the frequency domain signal in different frequency intervals.

In an exemplary embodiment, each frame of travel data further includes two side lane lines near the current position of the vehicle at the corresponding time of acquisition; the detection module 906 is further configured to determine, for the current frame running data, a track point on a planned track in the current frame running data, where the selected detection mode is center detection;

The detection module 906 is further configured to obtain, for each track point, a distance between the track point and the lane lines on both sides;

The detection module 906 is further configured to determine that the planned track in the current frame of driving data has an out-of-centering phenomenon when a difference between a distance between the existing track point and the lane lines on both sides exceeds a preset threshold.

In one exemplary embodiment, each frame of travel data further includes a vehicle control signal; the detection module 906 is further configured to determine that the current frame of running data belongs to a normal running behavior in the plan, where the planned track in the current frame of running data has an out-of-centering phenomenon and the vehicle control signal includes a control signal that actively alters the centered running;

The detection module 906 is further configured to determine that the planned trajectory in the current frame of running data is abnormal if the planned trajectory in the current frame of running data has an out-of-centering phenomenon and the vehicle control signal does not include a control signal for actively changing the in-center running.

The modules in the vehicle planning track testing device can be realized in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one exemplary embodiment, a computer device is provided, which may be a server, and the internal structure thereof may be as shown in fig. 10. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the driving data collected frame by frame. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a vehicle planned trajectory testing method.

It will be appreciated by those skilled in the art that the structure shown in FIG. 10 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one exemplary embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the processor when executing the computer program further performs the steps of:

The detection mode is at least divided into jitter detection, continuity detection and centering detection; aiming at the current frame running data, under the condition that the current frame running data is not the first frame running data, the detection mode adapted to the current frame running data comprises jitter detection, continuity detection and centering detection; in the case that the current frame running data is the first frame running data, the detection mode adapted to the current frame running data includes jitter detection and centering detection.

According to the selected detection mode, carrying out anomaly detection on each frame of running data, wherein the anomaly detection comprises the following steps:

According to the overlapping time period, the planned track in the current frame of running data and the planned track in the previous frame of running data are intercepted respectively, and the corresponding overlapping planned track of the current frame and the corresponding overlapping planned track of the previous frame are obtained;

Sampling the corresponding overlapped planning track of the current frame and the corresponding overlapped planning track of the previous frame respectively to obtain a current sampling point sequence corresponding to the current frame and a previous sampling point sequence corresponding to the previous frame, wherein the number of sampling points in the current sampling point sequence is the same as that of the previous sampling point sequence;

The planned trajectory comprises at least one travel data item of the vehicle; according to the selected detection mode, carrying out anomaly detection on each frame of running data, wherein the anomaly detection comprises the following steps:

Converting the difference curve serving as a time domain signal into a frequency domain signal, and calculating the integral energy value of the frequency domain signal in different frequency intervals;

And determining whether the current frame of driving data has jitter or not and the corresponding jitter intensity under the condition of the jitter according to the integral energy value of the frequency domain signal in different frequency intervals.

Each frame of driving data also comprises two side lane lines near the current position of the vehicle corresponding to the acquisition time; according to the selected detection mode, carrying out anomaly detection on each frame of running data, wherein the anomaly detection comprises the following steps:

For each track point, acquiring the distance between the track point and the lane lines on two sides respectively;

each frame of driving data also comprises a vehicle control signal; the method further comprises the steps of:

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are both information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to meet the related regulations.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high density embedded nonvolatile memory, resistive random access memory (ReRAM), magneto-resistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (PHASE CHANGE memory, PCM), graphene memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims

1. A method for testing a planned trajectory of a vehicle, the method comprising:

2. The method according to claim 1, wherein the detection means is at least divided into jitter detection, continuity detection and centering detection; aiming at the current frame running data, under the condition that the current frame running data is not the first frame running data, the detection mode adapted to the current frame running data comprises jitter detection, continuity detection and centering detection; in the case that the current frame running data is the first frame running data, the detection mode adapted to the current frame running data includes jitter detection and centering detection.

3. The method according to claim 2, wherein the anomaly detection for each frame of driving data according to the selected detection mode includes:

4. The method of claim 2, wherein the planned trajectory comprises at least one travel data item of the vehicle; the abnormal detection of each frame of driving data according to the selected detection mode comprises the following steps:

5. The method of claim 2, wherein each frame of travel data further comprises two side lane lines near the current location of the vehicle at the respective time of acquisition; the abnormal detection of each frame of driving data according to the selected detection mode comprises the following steps:

6. The method of claim 5, wherein each frame of travel data further comprises a vehicle control signal; the method further comprises the steps of:

7. A vehicle planned trajectory testing device, the device comprising:

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.