Method and apparatus for generating driving conditions

Abstract

This invention discloses a method and apparatus for generating driving conditions. The method includes: acquiring measured driving condition data; classifying the measured driving condition data; acquiring driving condition segments of a predetermined type of measured driving condition data; and generating driving conditions based on the acquired driving condition segments. According to the method and apparatus for generating driving conditions of this invention, driving conditions that meet different vehicle design requirements can be generated by using different types of measured driving condition data from the measured driving condition data.

Description

Technical Field

[0001] This invention relates to the field of vehicles, and in particular to a method and apparatus for generating driving conditions. Background Technology

[0002] During the design and development phase of vehicles (including electric vehicles and new energy vehicles), it is necessary to understand the possible operating environment of the vehicle based on the driving cycle. Based on the possible operating environment of the vehicle, relevant components of the vehicle, such as drive hardware, are designed and evaluated to ensure the required performance, safety and stability of the relevant components of the vehicle.

[0003] Typically, the driving conditions used in the vehicle design phase are standard driving conditions (such as WLTC, FTP, etc.), which usually meet basic design requirements. In addition, for some specific application environments (such as airport shuttle buses, mining engineering vehicles, etc.), the measured driving conditions under that specific environment can be used as the design basis for the design of vehicle-related components.

[0004] However, standard driving conditions only cover typical, limited working environments. In actual vehicle use, the actual working environment may differ significantly from that covered by standard driving conditions. Therefore, designing vehicle components based on standard driving conditions may result in components that do not meet actual usage requirements. Furthermore, the measured driving conditions are typically only for a specific working environment (e.g., airport shuttle), thus limiting their applicability to vehicle designs operating in that specific environment and lacking versatility, leading to higher design costs.

[0005] Therefore, the existing driving conditions are insufficient to meet the different design requirements of different customers. Summary of the Invention

[0006] The purpose of this invention is to provide a method and apparatus for generating driving conditions.

[0007] According to one aspect of the present invention, a method for generating driving conditions is provided, the method comprising: acquiring measured driving condition data; classifying the measured driving condition data; acquiring driving condition segments of a predetermined type of measured driving condition data; and generating driving conditions based on the acquired driving condition segments.

[0008] Optionally, the step of classifying the measured driving condition data includes: dividing the measured driving condition data into multiple segments; determining at least one valid segment among the multiple segments; dividing each valid segment into at least one sub-segment; extracting the feature vector of the measured driving condition data of each sub-segment; and classifying the measured driving condition data of each sub-segment based on the extracted feature vector.

[0009] Optionally, the step of dividing the measured driving condition data into multiple segments includes: dividing the measured driving condition data into multiple segments by measuring the continuity of speed relative to time included in the measured driving condition data.

[0010] Optionally, the step of determining at least one valid segment among the plurality of segments includes: determining a segment among the plurality of segments that includes a number of valid speeds greater than a threshold as a valid segment, wherein the valid speed indicates a speed greater than a predetermined speed, and wherein the speed is included in the measured driving condition data.

[0011] Optionally, the step of dividing each valid segment into at least one sub-segment includes: performing a fast Fourier transform on the measured driving condition data in each valid segment to obtain the corresponding frequency domain condition data; dividing the frequency domain condition data according to a predetermined threshold range; and determining each part of the valid segment corresponding to the division result of the frequency domain condition data as the at least one sub-segment.

[0012] Optionally, the step of classifying the measured driving condition data of each sub-segment based on the extracted feature vectors includes: classifying the extracted feature vectors through supervised learning or unsupervised learning to classify the measured driving condition data of each sub-segment.

[0013] Optionally, the step of obtaining driving condition segments of a predetermined type of measured driving condition data includes: calculating the transition probability matrix of the measured driving condition data for each type in the predetermined type; obtaining driving condition segments corresponding to the corresponding type of measured driving condition data through each transition probability matrix, wherein the obtained driving condition segments are fitted as driving conditions according to a predetermined sorting.

[0014] Optionally, the step of classifying the measured driving condition data of each sub-segment by classifying the extracted feature vectors through supervised learning includes: classifying the extracted feature vectors using a predetermined classifier, wherein the predetermined classifier is a classifier trained using feature vectors extracted from standard driving condition data and corresponding type labels as a training set; and taking the type of the feature vectors as the type of the measured driving condition data of the corresponding sub-segment.

[0015] Optionally, the step of classifying the measured driving condition data of each sub-segment by classifying the extracted feature vectors through unsupervised learning includes: classifying the extracted feature vectors using an unsupervised classifier; and using the classification type of the feature vectors as the classification type of the measured driving condition data of the corresponding sub-segment.

[0016] Optionally, the measured driving condition data includes driving condition data measured under different driving environments, wherein the driving environment corresponds to the driving area, driving mode and / or weather conditions.

[0017] According to another aspect of the present invention, an apparatus for generating driving conditions is provided, the apparatus comprising: a data acquisition unit configured to acquire measured driving condition data; a classification unit configured to classify the measured driving condition data; a segment acquisition unit configured to acquire driving condition segments of a predetermined type of measured driving condition data; and a driving condition generation unit configured to generate driving conditions based on the acquired driving condition segments.

[0018] According to another aspect of the present invention, a computer-readable recording medium storing a computer program is provided, wherein the computer program is configured to implement, when executed by a processor, a method for generating driving conditions according to the present invention.

[0019] According to another aspect of the present invention, a system for generating driving conditions is provided, the system comprising: a processor; and a memory storing a computer program, wherein when the computer program is executed by the processor, the method for generating driving conditions according to the present invention is implemented.

[0020] The method and apparatus for generating driving conditions according to the present invention can generate driving conditions that meet different vehicle design requirements by using different types of measured driving condition data from the measured driving condition data, according to different design requirements of the vehicle. Attached Figure Description

[0021] The foregoing and other aspects of the invention will be more fully understood from the following detailed description taken in conjunction with the accompanying drawings, including:

[0022] Figure 1 A flowchart of a method for generating driving conditions according to an exemplary embodiment of the present invention is shown.

[0023] Figure 2 A flowchart illustrating the step of classifying measured driving condition data in a method for generating driving conditions according to an exemplary embodiment of the present invention is shown.

[0024] Figure 3An example of segmentation according to an exemplary embodiment of the present invention is shown.

[0025] Figure 4 An example of dividing sub-segments according to an exemplary embodiment of the present invention is shown.

[0026] Figure 5 A block diagram of an apparatus for generating driving conditions according to an exemplary embodiment of the present invention is shown. Detailed Implementation

[0027] Below, some exemplary embodiments of the present invention will be described in more detail with reference to the accompanying drawings in order to better understand the basic ideas and advantages of the present invention.

[0028] Figure 1 A flowchart of a method for generating driving conditions according to an exemplary embodiment of the present invention is shown.

[0029] Reference Figure 1 In step S1, the measured driving condition data is obtained.

[0030] As an example, the measured driving condition data may include driving condition data measured under different driving environments, where the driving environment corresponds to the driving area, driving mode, and / or weather conditions. For example, driving condition data under different driving environments can be obtained by driving an existing vehicle for an extended period of time (e.g., driving for multiple days) in various driving areas, using various driving modes, and under various weather conditions. For example, the driving area may include urban areas, suburban areas, rural areas, highway areas, etc., where urban areas may also correspond to congested environments, smooth traffic environments, etc. Driving modes may include low-speed cruising, medium-speed cruising, high-speed cruising, etc. Weather conditions include extreme weather conditions (e.g., weather conditions with strong winds, rain, snow, fog, etc.) and normal weather conditions (sunny or weather conditions with light winds, rain, snow, fog, etc.).

[0031] It should be understood that the above driving areas, driving modes and weather conditions are only examples, and other forms may be included as needed.

[0032] Furthermore, according to this disclosure, the measured driving condition data may include speed, acceleration, and time data. Speed ​​may correspond to real-time speed, average speed, maximum speed, minimum speed, etc., and acceleration may correspond to real-time acceleration (deceleration), average acceleration (deceleration), maximum acceleration (deceleration), minimum acceleration (deceleration), etc.

[0033] Furthermore, after obtaining the measured driving condition data, it can be determined whether the measured driving condition data needs to be preprocessed, so as to obtain more accurate driving conditions using the preprocessed measured driving condition data.

[0034] For example, it can be determined whether there are missing or erroneous data in the measured driving condition data. When such data is found to exist, it can be preprocessed through methods such as prediction, interpolation, and deletion.

[0035] In step S2, the measured driving condition data is classified.

[0036] As an example, the classification of the measured driving condition data can be determined based on any one or a combination of the driving area, driving mode, and weather conditions corresponding to the different driving environments mentioned above. Furthermore, the classification of the measured driving condition data can be determined based on other standards according to actual circumstances.

[0037] In step S3, the driving condition segment for measuring driving condition data of a predetermined type is obtained.

[0038] For example, since measured driving condition data includes driving condition data measured under various driving conditions, it can be classified into many different types. However, in vehicle design, it may not be necessary to use all types of measured driving condition data. Therefore, the measured driving condition data can first be categorized as a whole according to the vehicle's design requirements. Then, some types of measured driving condition data related to the design requirements of the vehicle or related components can be obtained, thus obtaining the driving condition corresponding to each of these related types (i.e., predetermined types) as a driving condition segment.

[0039] For example, when the vehicle to be designed is a city bus, measurement driving condition data related to urban areas, low-to-medium speed cruising, and all weather conditions can be acquired to obtain driving condition segments for these types of measurement driving condition data. It should be understood that this is only an example, and the specific classification method and the relevant types (predetermined types) to be used may vary depending on the actual situation.

[0040] In step S4, driving conditions are generated based on the acquired driving condition segments.

[0041] As an example, after obtaining the driving condition segments corresponding to each type, the obtained driving condition segments can be fitted into the final driving condition according to a predetermined sorting.

[0042] Using the above method, driving condition data measured under different driving environments can be classified in different ways according to the actual needs of the vehicle or related components, and driving conditions that meet different needs can be generated based on the predetermined type of driving condition data required in the classified driving condition data.

[0043] Figure 2A flowchart of step S2, which involves classifying measured driving condition data, is shown in a method for generating driving conditions according to an exemplary embodiment of the present invention.

[0044] Reference Figure 2 In step S21, the measured driving condition data is divided into multiple segments.

[0045] As an example, the measured driving condition data can be divided into multiple segments by measuring the continuity of speed relative to time. For instance, speed and time data can be extracted from the measured driving condition data to obtain a speed-time curve, and then multiple segments can be divided based on the continuity of speed relative to time, such as based on idling time. For example, segments with continuous idling time greater than or equal to an idling threshold (e.g., idling segments) and segments with continuous idling time less than the idling threshold (e.g., non-idling segments) can be obtained.

[0046] In step S22, at least one valid segment among the plurality of segments is determined.

[0047] As an example, a valid segment may be defined as a segment that includes a number of valid speeds greater than a threshold among the plurality of segments, wherein the valid speed indicates a speed greater than a predetermined speed, and wherein the speed is included in the measured driving condition data.

[0048] For example, the predetermined speed can be zero, or it can be any other value greater than zero. For example, the number of effective speeds can be determined based on the number of sampling points used to sample the speed data, or it can be determined based on the duration of the effective speed.

[0049] Here, through step S22, segments with less speed data in the idling and non-idling segments can be filtered out, thereby identifying segments in the non-idling segment that include more speed data as valid segments.

[0050] Figure 3 An example of segmentation according to an exemplary embodiment of the present invention is shown.

[0051] refer to Figure 3 , Figure 3 It can be a portion of the speed-time curve obtained from the measured driving condition data, for example, a speed-time curve from 0 seconds to 50,000 seconds. Figure 3 The segments 1, 2, and 3, which contain more data about speed, are considered valid segments. The idling segment between segments 1 and 2, the idling segment between segments 2 and 3, and the segment between segments 2 and 3 that contains less data about speed are also considered valid segments. Figure 3 Sections marked with circles are filtered out.

[0052] Return to reference Figure 2 In step S23, each valid segment is divided into at least one sub-segment.

[0053] As an example, each valid segment can be divided into at least one sub-segment as follows: the measured driving condition data in each valid segment is subjected to a Fast Fourier Transform (FFT) to obtain the corresponding frequency domain condition data; the frequency domain condition data is divided according to a predetermined threshold range; and each part of the valid segment corresponding to the division result of the frequency domain condition data is determined as the at least one sub-segment.

[0054] For example, the velocity-time curve of the effective segment obtained in step S22 can first be subjected to a fast Fourier transform to obtain the curve in the frequency domain. Here, a low-pass filter (LPF) can be optionally applied to the obtained frequency domain curve according to the actual situation. Then, the frequency domain curve can be divided into at least one segment according to the amplitude range of the frequency domain curve. Finally, each part of the effective segment corresponding to each segment of the frequency domain curve is determined as the at least one sub-segment.

[0055] Figure 4 An example of dividing sub-segments according to an exemplary embodiment of the present invention is shown.

[0056] Reference Figure 4 , Figure 4 The diagram below shows the velocity-time curves corresponding to a valid segment. The horizontal axis indicates time, and the vertical axis indicates velocity. It should be understood that this segment may differ from... Figure 3 Any of the sections shown in the image.

[0057] It is possible Figure 4 The diagram below illustrates the windowing method indicated by double-headed arrows, where a Fast Fourier Transform (FFT) is performed on the effective segment. By performing an FFT on the curve within each window, the following can be obtained: Figure 4 The graph above shows a curve in the frequency domain. Optionally, this curve can be low-pass filtered. Figure 4 The horizontal axis in the diagram above indicates the window number, and the vertical axis indicates the amplitude.

[0058] Reference Figure 4 The frequency domain curve in the above figure is divided into three parts according to the relevant threshold range. Figure 4 The curve below is also divided into three parts, namely, three sub-segments, according to the division results of the frequency domain curve.

[0059] It should be understood that Figure 4 The sub-segments shown are for illustrative purposes only, and the effective segment can be divided into different numbers of sub-segments according to actual usage needs.

[0060] Return to reference Figure 2 In step S24, feature vectors of the measured driving condition data for each sub-segment are extracted.

[0061] Here, feature vectors can be extracted from the measured driving condition data of each sub-segment in any way. For example, the extracted feature vectors may include data about speed, acceleration, time, etc. It should be understood that the data included in the feature vectors is not limited to this and may include different data depending on actual needs and the different measured driving condition data.

[0062] In step S25, the measured driving condition data of each sub-segment are classified based on the extracted feature vectors.

[0063] As an example, the extracted feature vectors can be classified using supervised or unsupervised learning methods to classify the measured driving condition data for each sub-segment.

[0064] According to the first embodiment, when classifying the extracted feature vectors through supervised learning, a predetermined classifier can be used to classify the extracted feature vectors, wherein the predetermined classifier is a classifier trained using feature vectors extracted from standard driving condition data and corresponding type labels as a training set; then, the type in which the feature vectors are classified can be used as the type in which the measured driving condition data of the corresponding sub-segment are classified.

[0065] In this context, as an example, feature vectors can be extracted from standard driving conditions, and different type labels can be matched to the feature vectors according to design requirements. Then, the feature vectors and the corresponding type labels can be used as a training set to train a classifier, thereby using the trained classifier to classify the feature vectors extracted from the measured driving condition data.

[0066] As another example, standard driving condition data can be classified first according to design requirements using predetermined standards, i.e., type labels can be added. Then, feature vectors of standard driving condition data for each type can be extracted, and the feature vectors and corresponding type labels can be used as training sets to train a classifier. The trained classifier can then be used to classify the feature vectors extracted from the measured driving condition data.

[0067] According to the second embodiment, when classifying the extracted feature vectors through unsupervised learning, an unsupervised classifier can be used to classify the extracted feature vectors; then, the classification type of the feature vectors can be used as the classification type of the measurement driving condition data of the corresponding sub-segment. For example, the unsupervised classifier may include any unsupervised classifier such as K-means or support vector machine classifiers.

[0068] For example, in Figure 3 The numbers 1 and 2 marked on the velocity-time curves corresponding to segments 1, 2, and 3 indicate the classification type of each sub-segment (i.e., type 1, type 2, etc.). It should be understood that... Figure 3 The examples shown are merely illustrative; depending on actual needs, the segments within the valid segment can be classified into many more types.

[0069] After classifying the measured driving condition data of each sub-section through step S25, as an example, it can be performed as follows: Figure 1 Step S3, the step of obtaining the driving condition segment of the predetermined type of measured driving condition data: Calculate the transition probability matrix of the measured driving condition data for each type in the predetermined type; obtain the driving condition segment corresponding to the corresponding type of measured driving condition data through each transition probability matrix. Then, in Figure 1 In step S4, the acquired driving condition segments can be fitted into the final driving condition according to a predetermined order through a global state offset process. That is, in steps S3 and S4, the final driving condition can be obtained through a Markov chain method.

[0070] The method for generating driving conditions according to the present invention can generate driving conditions that meet different vehicle design requirements by using different types of measured driving condition data from the measured driving condition data, based on different vehicle design requirements.

[0071] Figure 5 A block diagram of an apparatus for generating driving conditions according to an exemplary embodiment of the present invention is shown.

[0072] Reference Figure 5 An exemplary embodiment of the apparatus for generating driving conditions according to the present invention includes: a data acquisition unit 1, a classification unit 2, a segment acquisition unit 3, and a driving condition generation unit 4.

[0073] The data acquisition unit 1 is configured to acquire measured driving condition data.

[0074] Classification unit 2 is configured to classify the measured driving condition data.

[0075] The section acquisition unit 3 is configured to acquire a predetermined type of measurement driving condition data for a driving condition section.

[0076] The driving condition generation unit 4 is configured to generate driving conditions based on the acquired driving condition segments.

[0077] Here, we have already referred to the above. Figures 1 to 4The measurement of driving condition data, the classification of driving condition data, the acquisition of driving condition sections, and the acquisition of final driving condition data are described in detail, and will not be repeated here.

[0078] The apparatus for generating driving conditions according to the present invention can generate driving conditions that meet different vehicle design requirements by using different types of measured driving condition data from the measured driving condition data, based on different vehicle design requirements.

[0079] An exemplary embodiment of the present invention also provides a system for generating driving conditions. The system for generating driving conditions includes a processor and a memory. The memory is configured to store a computer program. The computer program can be executed by the processor to implement the method for generating driving conditions according to the present invention.

[0080] An exemplary embodiment of the present invention also provides a computer-readable recording medium storing a computer program configured to implement, when executed by a processor, a method for generating driving conditions according to the present invention. The computer-readable recording medium is any data storage device capable of storing data readable by a computer system. Examples of computer-readable recording media include: read-only memory, random access memory, read-only optical disc, magnetic tape, floppy disk, optical data storage device, and carrier waves (such as data transmission via the Internet through wired or wireless transmission paths). The computer-readable recording medium can also be distributed across a networked computer system, thereby allowing computer-readable code to be stored and executed in a distributed manner. Furthermore, the functional programs, code, and code segments that perform the present invention can be readily interpreted by a programmer of ordinary skill in the art related to the present invention within the scope of the invention.

[0081] Furthermore, the various units in the above-described apparatus and device according to exemplary embodiments of the present invention can be implemented as hardware components or software modules. Moreover, those skilled in the art can implement each unit, for example, using a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or a processor, depending on the processing performed by each defined unit.

[0082] Although the invention has been described and illustrated herein with reference to specific embodiments, the invention is not limited to the details shown. Rather, various modifications to these details can be made within the scope of the invention.

[0083] List of reference numerals

[0084] S1 acquires the measured driving condition data.

[0085] S2 classifies the measured driving condition data.

[0086] S3 Obtains driving condition data of a predetermined type for the driving condition section.

[0087] S4 generates driving conditions based on the acquired driving condition segments.

[0088] S21 divides the measured driving condition data into multiple sections.

[0089] S22 Determine at least one valid segment among the plurality of segments.

[0090] S23 divides each valid segment into at least one sub-segment.

[0091] S24 Extracts the feature vector of the measured driving condition data for each sub-segment.

[0092] S25 Classifies the measured driving condition data of each sub-segment based on the extracted feature vectors.

[0093] 1 Data Acquisition Unit

[0094] 2. Classification Units

[0095] 3-segment acquisition unit

[0096] 4 Driving Condition Generation Unit

Claims

1. A method for generating driving conditions, the method comprising: Acquire the measured driving condition data; The measured driving condition data are classified as a whole according to the vehicle's design requirements; The driving condition segment for which predetermined types of measurement driving condition data are obtained, which are relevant to the design requirements of the vehicle or related components; as well as Driving conditions are generated based on the acquired driving condition segments.

2. The method according to claim 1, wherein, The steps for classifying measured driving condition data include: The measured driving condition data is divided into multiple segments; Determine at least one valid segment among the plurality of segments; Divide each valid segment into at least one sub-segment; Extract the feature vector of the measured driving condition data for each sub-segment; and The measured driving condition data of each sub-segment are classified based on the extracted feature vectors.

3. The method according to claim 2, wherein, The steps for dividing the measured driving condition data into multiple segments include: By measuring the continuity of speed relative to time in the driving condition data, the driving condition data is divided into multiple segments.

4. The method according to claim 2, wherein, The steps for determining at least one valid segment among the plurality of segments include: The effective segments are defined as those that include a number of effective speeds greater than a threshold among the plurality of segments, wherein the effective speed indicates a speed greater than a predetermined speed, and wherein the speed is included in the measured driving condition data.

5. The method according to claim 2, wherein, The steps of dividing each valid segment into at least one sub-segment include: The measured driving condition data in each effective segment are subjected to fast Fourier transform to obtain the corresponding frequency domain condition data. The frequency domain operating condition data is divided according to a predetermined threshold range; and Each part of the effective segment corresponding to the division result of the frequency domain operating condition data is determined as the at least one sub-segment.

6. The method according to claim 2, wherein, The steps for classifying the measured driving condition data of each sub-segment based on the extracted feature vectors include: The extracted feature vectors are classified using supervised or unsupervised learning methods to classify the measured driving condition data for each sub-segment.

7. The method according to any one of claims 1 to 6, wherein, The steps for obtaining driving condition data of a predetermined type for a specific driving condition segment include: Calculate the transition probability matrix of the measured driving condition data for each of the predetermined types; and By using the various transition probability matrices, the driving condition segments corresponding to the respective types of measured driving condition data can be obtained. In this process, the acquired driving condition segments are fitted into driving conditions according to a predetermined sorting.

8. The method according to claim 6, wherein, The steps for classifying the measured driving condition data of each sub-segment by classifying the extracted feature vectors using supervised learning include: The extracted feature vectors are classified using a predetermined classifier, wherein the predetermined classifier is a classifier trained using feature vectors extracted from standard driving condition data and corresponding type labels as a training set; and The type of feature vector classification is used as the type of measurement driving condition data for the corresponding sub-segment.

9. The method according to claim 6, wherein, The steps for classifying the measured driving condition data of each sub-segment by classifying the extracted feature vectors using unsupervised learning include: The extracted feature vectors are classified using an unsupervised classifier; and The type of feature vector classification is used as the type of measurement driving condition data for the corresponding sub-segment.

10. The method according to any one of claims 1 to 6, wherein, The driving condition data includes driving condition data measured under different driving environments, wherein the driving environment corresponds to the driving area, driving mode and / or weather conditions.

11. An apparatus for generating driving conditions, the apparatus comprising: The data acquisition unit is configured to acquire measured driving condition data. The classification unit is configured to classify the measured driving condition data as a whole according to the vehicle's design requirements. A segment acquisition unit is configured to acquire a segment of driving condition data of a predetermined type that is related to the design requirements of the vehicle or related components. as well as The driving condition generation unit is configured to generate driving conditions based on the acquired driving condition segments.

12. A computer-readable recording medium storing a computer program, wherein, The computer program is configured to perform the method described in any one of claims 1 to 10 when executed by a processor.

13. A system for generating driving conditions, the system comprising: processor; as well as A memory storing a computer program that, when executed by a processor, performs the method described in any one of claims 1 to 10.

Images