Vehicle road spectrum generation method and related device

Abstract

The invention discloses a vehicle road spectrum generation method and a related device, and the method comprises the steps: obtaining an overall speed interval and a plurality of short driving samples, carrying out the interval division of the overall speed interval, obtaining a plurality of speed subintervals, and according to the speed subinterval in which the average speed of the short driving samples is located, obtaining a plurality of road spectrums; determining a short driving sample set corresponding to each speed sub-interval, and for a target speed sub-interval in the plurality of speed sub-intervals, according to the short driving sample set corresponding to the target speed sub-interval, according to the change relation of the driving speed described by the short driving samples included in the short driving sample set along with the driving duration, determining the driving speed of the target speed sub-interval in the plurality of speed sub-intervals; the method comprises the following steps: obtaining a sub-vehicle road spectrum corresponding to a target speed sub-interval, taking each speed sub-interval as the target speed sub-interval, obtaining a sub-vehicle road spectrum corresponding to each speed sub-interval, and splicing the sub-vehicle road spectrums corresponding to each speed sub-interval to obtain a vehicle road spectrum so as to obtain approximate overall speed distribution. And a representative vehicle road spectrum is provided.

Description

Technical Field

[0001] The present application relates to the field of vehicle testing, and in particular to a vehicle road spectrum generation method and related devices. Background Art

[0002] With the development of the automobile industry and the advancement of technology, research on vehicle energy consumption, pollutant emissions, etc. has become increasingly important. The vehicle road spectrum is a curve that reflects the relationship between the vehicle's driving speed and time. Researchers usually rely on the vehicle road spectrum to reflect the characteristics of the vehicle's actual operating status and conduct research on this.

[0003] In the related art, a vehicle road spectrum is generated by extracting features from driving data to obtain a vehicle road spectrum, which has poor representativeness. Summary of the invention

[0004] In view of this, the present application provides a vehicle road spectrum generation method and related devices to solve the problem that the generated vehicle road spectrum lacks representativeness.

[0005] In a first aspect, the present application provides a method for generating a vehicle road spectrum, the method comprising:

[0006] Acquire an overall speed interval and a plurality of short driving samples, wherein the short driving sample is a relationship between a driving speed and a driving duration from when the vehicle starts to move to when the vehicle stops, and the overall speed interval is a driving speed coverage range of the plurality of short driving samples;

[0007] Dividing the overall speed interval into multiple speed sub-intervals, wherein the speed sub-intervals are parts of the overall speed interval;

[0008] Determine, according to the speed subinterval in which the average speed of the short driving sample is located, a short driving sample set corresponding to each of the speed subintervals, wherein the short driving sample set includes at least one short driving sample located in the corresponding speed subinterval;

[0009] For a target speed subinterval among the plurality of speed subintervals, according to a short driving sample set corresponding to the target speed subinterval, a sub-vehicle road spectrum corresponding to the target speed subinterval is obtained through a relationship between a driving speed described by short driving samples included in the short driving sample set and a driving duration, wherein the sub-vehicle road spectrum is used to indicate a driving speed of a vehicle that changes with a driving duration in the target speed subinterval;

[0010] Taking each of the speed sub-intervals as the target speed sub-interval, and obtaining a sub-vehicle road spectrum corresponding to each of the speed sub-intervals;

[0011] The sub-vehicle road spectra corresponding to each of the speed sub-intervals are spliced ​​to obtain a vehicle road spectrum, and the vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time.

[0012] Optionally, the sub-vehicle road spectrum corresponding to the target speed sub-interval is obtained according to the short driving sample set corresponding to the target speed sub-interval by the relationship between the driving speed and the driving time described by the short driving samples included in the short driving sample set, including:

[0013] Obtaining a plurality of short driving sample categories according to the short driving sample set corresponding to the target speed subinterval, determining representative short driving samples corresponding to each of the short driving sample categories, wherein the similarity between the short driving samples in different short driving sample categories is less than a similarity threshold;

[0014] The relationship between the driving speed and the driving time described by the multiple representative short driving samples is spliced ​​together to obtain the sub-vehicle road spectrum corresponding to the target speed sub-interval.

[0015] Optionally, determining representative short driving samples corresponding to each of the short driving sample categories includes:

[0016] It is determined that each of the short driving sample categories corresponds to a representative short driving sample, and the representative short driving samples correspond to the short driving sample categories one by one.

[0017] Optionally, obtaining a plurality of short driving sample categories according to the short driving sample set corresponding to the target speed subinterval, and determining representative short driving samples corresponding to each of the short driving sample categories, respectively, comprises:

[0018] According to the short driving sample set corresponding to the target speed sub-interval, a plurality of clusters are obtained through a clustering operation, wherein the cluster includes a cluster center, and each cluster corresponds to a short driving sample category;

[0019] If the cluster center of the cluster corresponding to the short driving sample category has a corresponding representative short driving sample, the short driving sample corresponding to the cluster center is determined as the representative short driving sample corresponding to the short driving sample category;

[0020] If the cluster center of the cluster corresponding to the short driving sample category does not have a corresponding representative short driving sample, the short driving sample with the smallest feature difference from the cluster center of the cluster corresponding to the short driving sample category is determined as the representative short driving sample corresponding to the short driving sample category.

[0021] Optionally, the method further comprises:

[0022] Obtain speed threshold and driving time threshold;

[0023] The determining, according to the speed subinterval in which the average speed of the short driving samples is located, a set of short driving samples corresponding to each speed subinterval respectively comprises:

[0024] Determine a high-speed sub-interval according to the lowest speed of each speed sub-interval, wherein the high-speed sub-interval is a speed sub-interval whose lowest speed is greater than the speed threshold;

[0025] Determining initial short driving sample sets corresponding to the high-speed subintervals respectively according to the speed subintervals in which the average speed of the short driving samples is located;

[0026] From the initial short driving sample sets corresponding to the high-speed sub-intervals, short driving samples whose driving duration is greater than the driving duration threshold are deleted to obtain first short driving sample sets corresponding to the high-speed sub-intervals;

[0027] The first short driving sample sets respectively corresponding to the high-speed sub-intervals are determined as the short driving sample sets respectively corresponding to the high-speed sub-intervals.

[0028] Optionally, the method further comprises:

[0029] Obtaining the exclusion strategy obtained by the i-th update, wherein the exclusion strategy is used to exclude isolated short driving samples, wherein the isolated short driving samples are short driving samples that deviate from the characteristics of other short driving samples in the same short driving sample set, and i is a positive integer;

[0030] The overall speed interval is divided into a plurality of speed categories, and the speed difference between short driving samples in different speed categories is greater than a speed category threshold;

[0031] Obtaining an overall speed distribution according to a proportion of the number of times each of the speed categories appears in the multiple short driving samples to the total number of times the speed categories are generated in the multiple short driving samples;

[0032] According to the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution, updating the exclusion strategy obtained by the i-th update to obtain the exclusion strategy obtained by the i+1-th update;

[0033] The determining, according to the speed subinterval in which the average speed of the short driving samples is located, a set of short driving samples corresponding to each speed subinterval respectively comprises:

[0034] According to the speed subinterval in which the average speed of the short driving sample is located, determining the initial short driving sample sets corresponding to the speed subintervals respectively;

[0035] Based on the exclusion strategy obtained by the (i+1)th update, the isolated short driving samples corresponding to each speed subinterval are deleted from the short driving sample sets corresponding to each speed subinterval, so as to obtain the second short driving sample sets corresponding to each speed subinterval;

[0036] The second short driving sample sets corresponding to the speed sub-intervals are determined as short driving sample sets corresponding to the speed sub-intervals.

[0037] Optionally, the exclusion strategy is used to exclude the isolated short driving samples, and includes excluding a proportion of the isolated short driving samples in each of the initial short driving sample sets.

[0038] In a second aspect, the present application provides a vehicle road spectrum generation device, the device comprising: an acquisition unit, a division unit, a set determination unit, a target sub-road spectrum unit, a sub-road spectrum unit, and a road spectrum splicing unit;

[0039] The acquisition unit is used to acquire an overall speed interval and a plurality of short driving samples, wherein the short driving sample is a relationship between a driving speed and a driving duration from when the vehicle starts to move to when the vehicle stops, and the overall speed interval is a driving speed coverage range of the plurality of short driving samples;

[0040] The dividing unit is used to divide the overall speed interval into multiple speed sub-intervals, where the speed sub-intervals are parts of the overall speed interval;

[0041] The set determination unit is used to determine the short driving sample sets corresponding to the speed subintervals respectively according to the speed subintervals in which the average speed of the short driving samples is located, wherein the short driving sample sets include at least one short driving sample located in the corresponding speed subinterval;

[0042] The target sub-road spectrum unit is used to obtain, for a target speed sub-interval in the plurality of speed sub-intervals, a sub-vehicle road spectrum corresponding to the target speed sub-interval according to a short driving sample set corresponding to the target speed sub-interval and a relationship between a driving speed and a driving duration described by the short driving samples included in the short driving sample set, wherein the sub-vehicle road spectrum is used to indicate a driving speed of a vehicle that changes with a driving duration in the target speed sub-interval;

[0043] The sub-road spectrum unit is used to use each of the speed sub-intervals as the target speed sub-interval to obtain the sub-vehicle road spectrum corresponding to each of the speed sub-intervals;

[0044] The road spectrum splicing unit is used to splice the sub-vehicle road spectra corresponding to each of the sub-intervals to obtain a vehicle road spectrum, and the vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time.

[0045] In a third aspect, the present application provides a vehicle road spectrum generation device, the device comprising a memory and a processor:

[0046] The memory is used to store program codes and transmit the program codes to the processor;

[0047] The processor is used to execute the method described in any one of the first aspects according to the instructions in the program code.

[0048] In a fourth aspect, the present application provides a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store a computer program, and the computer program is used to execute any one of the methods described in the first aspect.

[0049] It can be seen from the above technical solutions that this application has the following beneficial effects:

[0050] In view of the lack of representativeness of the generated vehicle road spectrum, the overall speed interval is obtained to lay the foundation for speed interval division, and multiple short driving samples are obtained to determine the data source for generating the vehicle road spectrum. The overall speed interval is divided into intervals to obtain multiple speed sub-intervals, so that multiple short driving samples are classified according to the working conditions represented by the speed intervals. According to the speed sub-intervals in which the average speed of the short driving samples is located, the short driving sample sets corresponding to each speed sub-interval are determined, so that the data source reflecting the specific working conditions is more targeted and the accuracy of road spectrum generation is improved. For the target speed sub-interval in the multiple speed sub-intervals, according to the short driving sample set corresponding to the target speed sub-interval, the sub-vehicle road spectrum corresponding to the target speed sub-interval is obtained through the relationship between the driving speed described by the short driving samples included in the short driving sample set and the driving time. Based on the more targeted short driving sample set, the speed distribution of the sub-vehicle road spectrum is closer to the speed distribution of the working condition corresponding to the speed sub-interval, and has local representativeness. Each speed sub-interval is used as the target speed sub-interval, and the sub-vehicle road spectrum corresponding to each speed sub-interval is obtained, so that the sub-vehicle road spectrum corresponding to each speed sub-interval is locally representative. The sub-vehicle road spectra corresponding to each speed sub-interval are spliced ​​together to obtain the vehicle road spectrum, so that the vehicle road spectrum comprehensively covers the driving status of the vehicle in different speed sub-intervals, reflects the speed distribution of corresponding working conditions in different speed sub-intervals, improves the overall representativeness and authenticity of the vehicle road spectrum, and has a higher reference value for studying vehicle energy consumption and pollutant emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1A method flow chart of a vehicle road spectrum generation method provided in an embodiment of the present application;

[0052] Figure 2 A schematic diagram of overall speed distribution and speed interval division provided in an embodiment of the present application;

[0053] Figure 3 A schematic diagram of a sub-vehicle road spectrum in a low-speed sub-interval provided in an embodiment of the present application;

[0054] Figure 4 A schematic diagram of a sub-vehicle road spectrum in a medium-speed sub-interval provided in an embodiment of the present application;

[0055] Figure 5 A schematic diagram of a sub-vehicle road spectrum of a high-speed sub-interval provided in an embodiment of the present application;

[0056] Figure 6 A schematic diagram of a vehicle road spectrum provided in an embodiment of the present application;

[0057] Figure 7 A schematic diagram of velocity distribution comparison provided in an embodiment of the present application;

[0058] Figure 8 A schematic diagram of the structure of a vehicle road spectrum generating device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0060] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0061] The overall speed distribution reflects the frequency of occurrence of different vehicle speeds in a large amount of sample data, and is an important indicator for determining whether the vehicle road spectrum is representative. If the speed distribution of the vehicle road spectrum is close to the overall speed distribution, it means that the speed frequency information of different speed ranges of the vehicle road spectrum is highly consistent with a large amount of sample data and is representative. It has a high reference value for studying vehicle energy consumption, pollutant emissions, and other aspects.

[0062] By simulating the vehicle's driving speed as it changes with driving time as indicated by the vehicle road spectrum, and applying the general speed change pattern of the vehicle model corresponding to the vehicle road spectrum in actual traffic, we can analyze and study the actual driving pattern, energy consumption, pollutant emissions, etc. of the vehicle model based on this change pattern.

[0063] In the related technology, the vehicle road spectrum is mainly generated directly by extracting features from driving data. This makes it impossible to capture the speed distribution characteristics in different speed ranges in detail and ignores the speed change relationship under different working conditions. The generated vehicle road spectrum mainly reflects the speed change relationship under the intermediate speed condition and lacks the speed change relationship under other working conditions. It is too different from the overall speed distribution and is therefore not representative. The vehicle road spectrum generated by this method has a low reference value.

[0064] In order to solve the above problems, the embodiments of the present application provide a vehicle road spectrum generation method and related devices. By dividing the overall speed interval into multiple speed sub-intervals, the speed change relationship within different speed sub-intervals is more carefully explored. The representative short driving samples in each sub-interval can more specifically reflect the speed change relationship under the corresponding working conditions of the speed interval, and the generated corresponding sub-vehicle road spectrum is also closer to the speed distribution under the working conditions. Therefore, the final vehicle road spectrum is obtained by splicing the sub-road spectra of each speed sub-interval, which can ensure that the speed distribution of the vehicle road spectrum is closer to the speed distribution of the overall sample data and more representative.

[0065] Therefore, a vehicle driving state that is closer to reality can be simulated based on the vehicle road spectrum, which has a high reference value for studying vehicle energy consumption, pollutant emissions and other aspects.

[0066] See also Figure 1 , Figure 1 A method flow chart of a vehicle road spectrum generation method provided in an embodiment of the present application, the method specifically comprises the following steps:

[0067] S101: Obtain an overall speed range and multiple short driving samples.

[0068] Among them, the short driving sample is the relationship between the change of driving speed and driving time from the start of vehicle movement to the stop of the vehicle, and the overall speed range is the driving speed coverage range of multiple short driving samples.

[0069] A short driving sample is the smallest unit of data source in generating a vehicle road spectrum. During the driving process, the vehicle will experience multiple starts and stops due to road conditions and driving purposes. The data recorded during the process from the vehicle starting to move (i.e., changing from a stationary state to a moving state), driving for more than a preset time or preset distance, to the vehicle stopping again (i.e., returning from a moving state to a stationary state) is the relationship between the driving speed and the driving time, and is usually stored in the form of comma-separated values ​​(CSV). Among them, the definition of stopping includes a driving speed of 0 kilometers per hour (km / h) or a driving speed less than a preset speed threshold.

[0070] Since short driving samples are generated frequently, the driving data of one vehicle can generate a large number of short driving samples. For example, 8612 short driving samples can be obtained through the driving data of eight vehicles. Using short driving samples as the data unit for generating vehicle road spectra helps to more carefully analyze the speed change patterns in different driving cycles, thereby better reflecting actual driving behavior.

[0071] The method of obtaining short driving samples includes directly obtaining short driving samples, obtaining them from long-distance driving data, and other methods, which are not limited in the embodiments of the present application. The process of obtaining data may have missing data or data anomalies. In this case, in the process of obtaining multiple short driving samples, the abnormal values ​​will be filtered and the missing data items will be interpolated to ensure the validity and completeness of the short driving sample data.

[0072] The overall speed range is the speed coverage of multiple short driving samples, that is, the overall speed range can be determined by the maximum and minimum values ​​of multiple short driving samples. For example, if the speed range of all short driving samples is from 0km / h to 114km / h, then the overall speed range is [0,114]km / h.

[0073] Through multiple short driving samples, the data source for generating the vehicle road spectrum is obtained. By obtaining the overall speed range and laying the foundation for dividing the speed range, it is ensured that the subsequent generation of the road spectrum is based on accurate and comprehensive data.

[0074] S102: Divide the overall speed interval into intervals to obtain a plurality of speed sub-intervals.

[0075] The speed sub-range is a part of the overall speed range, which is a smaller driving speed range obtained by dividing the overall speed range. Each speed sub-range corresponds to a working condition or driving scenario.

[0076] According to actual needs and the purpose of data analysis, determine how to divide the overall speed interval into multiple speed sub-intervals, usually including equal interval division or non-equal interval division.

[0077] For example, if the overall speed range is [0,114] km / h, according to the actual driving scenarios of modern traffic, based on the uniformity of sample magnitude and the number of samples used to generate vehicle road spectra, the speed range is divided into equal intervals according to the ratio of 1 / 4, 1 / 4 and 1 / 2, and the speed range A [0,29) km / h, speed range B [29,57) km / h and speed range C [57,114] km / h are obtained, which correspond to the urban road scenes, rural suburban road scenes and highway scenes in real life respectively. Figure 2 , Figure 2 A schematic diagram of overall speed distribution and speed interval division is provided in an embodiment of the present application, wherein the dotted line divides the overall speed interval corresponding to the horizontal axis into three speed sub-intervals, corresponding to speed interval A, speed interval B and speed interval C.

[0078] The speed data in each speed sub-range reflects the driving behavior in a specific speed range. By dividing the overall speed range into multiple speed sub-ranges, the speed change patterns in different speed ranges can be analyzed in more detail, which helps to better understand the operating status of the vehicle under different working conditions, thereby improving the accuracy and representativeness of the vehicle road spectrum in a specific speed sub-range.

[0079] S103: Determine short driving sample sets corresponding to respective speed subintervals according to the speed subintervals in which the average speeds of the short driving samples are located.

[0080] The short driving sample set includes at least one short driving sample located in a corresponding speed sub-interval.

[0081] For a short driving sample, the corresponding speed sub-interval is determined by counting the average speed in the time period corresponding to the short driving sample. Based on this classification process, multiple short driving samples are placed in the corresponding speed sub-interval, thereby determining the short driving sample sets corresponding to each speed sub-interval.

[0082] For example, if the average speed of a short driving sample is 60 km / h, there are three speed sub-intervals: speed interval A [0, 29) km / h, speed interval B [29, 57) km / h and speed interval C [57, 114] km / h, then the short driving sample is classified as a short driving sample in the short driving sample set in speed interval C.

[0083] By determining the short driving sample sets corresponding to each speed sub-interval, the short driving sample set in each speed sub-interval includes relevant short driving samples reflecting the working conditions corresponding to the speed segment, thereby making the data source for generating the sub-vehicle road spectrum corresponding to the speed sub-interval more targeted.

[0084] S104: For a target speed subinterval among the multiple speed subintervals, according to a short driving sample set corresponding to the target speed subinterval, a sub-vehicle road spectrum corresponding to the target speed subinterval is obtained by a relationship between a driving speed and a driving time described by short driving samples included in the short driving sample set.

[0085] The sub-vehicle road spectrum is used to indicate the driving speed of the vehicle in the target speed sub-interval that varies with the driving time.

[0086] In the embodiment of the present application, the sub-vehicle road spectrum is generated, including but not limited to the following implementation methods:

[0087] Method 1: For the short driving samples in the short driving sample set corresponding to the target speed sub-interval, the average speed of multiple short driving samples at the same time point is calculated to form a curve of the average speed changing with time.

[0088] Method 2: For the short driving samples in the short driving sample set corresponding to the target speed sub-interval, multiple short driving samples are selected, and the curves of the relationship between the recorded driving speed and the driving time are spliced ​​end to end to generate the sub-vehicle road spectrum. The main purpose is to reflect its true speed distribution, that is, in this speed sub-interval, the proportion of the number of different speeds in the multiple short driving samples to the total number of different speeds generated by the multiple short driving samples. Therefore, there is no limitation on the splicing order of the short driving samples.

[0089] Since the initial speed and the final speed of the short driving sample are equal, for the target speed subinterval in the multiple speed subintervals, by splicing the multiple driving speed vs. driving time relationship curves corresponding to the multiple short driving samples end to end, a continuous driving speed curve of the vehicle varying with driving time in the target speed subinterval can be obtained.

[0090] For example, see Figure 3 , Figure 3 A schematic diagram of a sub-vehicle road spectrum of a low-speed sub-interval is provided in an embodiment of the present application, wherein speed interval A is a low-speed sub-interval obtained by division, and the sub-vehicle road spectrum corresponding to speed interval A is obtained by splicing multiple short driving samples in a short driving sample set corresponding to speed interval A.

[0091] In the process of the sub-vehicle road spectrum corresponding to the target speed sub-interval, it is generated based on the relationship between the change of driving speed and driving time described by the short driving samples included in the short driving set corresponding to the target speed sub-interval. In this process, the speed change law under the specific working condition corresponding to the target speed sub-interval is mapped to the corresponding sub-vehicle road spectrum. Due to the specific targeting of the short driving sample set, the sub-vehicle road spectrum can better reflect the actual working condition speed information of the target speed sub-interval, is close to the actual speed distribution of the target speed sub-interval, and is more representative in this speed sub-interval.

[0092] S105: Taking each speed sub-interval as a target speed sub-interval, and obtaining a sub-vehicle road spectrum corresponding to each speed sub-interval.

[0093] Each speed sub-interval is used as a target speed sub-interval, and the process of S104 is repeated to obtain a sub-vehicle road spectrum corresponding to each speed sub-interval, thereby obtaining the vehicle's driving speed that varies with driving time in each speed sub-interval.

[0094] For example, see Figure 4 and Figure 5 , Figure 4 A schematic diagram of a sub-vehicle road spectrum in a medium-speed sub-interval provided in an embodiment of the present application, Figure 5 A schematic diagram of a sub-vehicle road spectrum of a high-speed sub-interval provided in an embodiment of the present application. By dividing the speed interval, three speed sub-intervals are obtained, namely, a low-speed sub-interval (speed interval A), a medium-speed sub-interval (speed interval B) and a high-speed sub-interval (speed interval C). By taking these three speed sub-intervals as target speed sub-intervals, the sub-vehicle road spectra corresponding to each speed sub-interval are obtained, namely, the sub-vehicle road spectrum of speed interval A, the sub-vehicle road spectrum of speed interval B and the sub-vehicle road spectrum of speed interval C.

[0095] In the overall speed range, the sub-vehicle road spectrum corresponding to each speed sub-range determines the more representative speed change law under the corresponding working condition of each speed sub-range. Therefore, different speed sub-ranges in the overall speed range have sub-vehicle road spectra that can reflect the general speed distribution of the corresponding working condition, and each sub-vehicle road spectrum is locally representative.

[0096] S106: Connect the sub-vehicle road spectra corresponding to each speed sub-interval to obtain a vehicle road spectrum.

[0097] Among them, the vehicle road spectrum is used to indicate the vehicle's driving speed that changes with driving time.

[0098] In this step, the process of splicing the sub-vehicle road spectra corresponding to each speed sub-interval is consistent with the process of splicing multiple short driving samples in S104, and will not be repeated here.

[0099] For example, see Figure 6 , Figure 6 A schematic diagram of a vehicle road spectrum provided in an embodiment of the present application, if the sub-vehicle road spectra corresponding to the three speed sub-intervals are obtained, the sub-vehicle road spectra corresponding to the three speed sub-intervals are spliced ​​to obtain the following Figure 6 The vehicle road spectrum (final road spectrum) shown.

[0100] The vehicle road spectrum is obtained by splicing the sub-vehicle road spectra corresponding to each speed sub-interval, and the local representativeness of the sub-vehicle road spectra is converted into the comprehensive representativeness of the vehicle road spectrum. That is, the vehicle road spectrum in each speed sub-interval of the overall speed range is closer to the speed distribution corresponding to the speed sub-interval, and is representative in each speed sub-interval, thereby obtaining a road spectrum with higher reference value and more accurate reflection of the general speed distribution of the overall speed range.

[0101] In view of the lack of representativeness of the generated vehicle road spectrum, the overall speed interval is obtained to lay the foundation for speed interval division, and multiple short driving samples are obtained to determine the data source for generating the vehicle road spectrum. The overall speed interval is divided into intervals to obtain multiple speed sub-intervals, so that multiple short driving samples are classified according to the working conditions represented by the speed intervals. According to the speed sub-intervals in which the average speed of the short driving samples is located, the short driving sample sets corresponding to each speed sub-interval are determined, so that the data source reflecting the specific working conditions is more targeted and the accuracy of road spectrum generation is improved. For the target speed sub-interval in the multiple speed sub-intervals, according to the short driving sample set corresponding to the target speed sub-interval, the sub-vehicle road spectrum corresponding to the target speed sub-interval is obtained through the relationship between the driving speed described by the short driving samples included in the short driving sample set and the driving time. Based on the more targeted short driving sample set, the speed distribution of the sub-vehicle road spectrum is closer to the speed distribution of the working condition corresponding to the speed sub-interval, and has local representativeness. Each speed sub-interval is used as the target speed sub-interval, and the sub-vehicle road spectrum corresponding to each speed sub-interval is obtained, so that the sub-vehicle road spectrum corresponding to each speed sub-interval is locally representative. The sub-vehicle road spectra corresponding to each speed sub-interval are spliced ​​together to obtain the vehicle road spectrum, so that the vehicle road spectrum comprehensively covers the driving status of the vehicle in different speed sub-intervals, reflects the speed distribution of corresponding working conditions in different speed sub-intervals, improves the overall representativeness and authenticity of the vehicle road spectrum, and has a higher reference value for studying vehicle energy consumption and pollutant emissions.

[0102] This embodiment of the application provides a specific implementation method for generating a sub-vehicle road spectrum, see A1-A2:

[0103] A1: A plurality of short driving sample categories are obtained according to a short driving sample set corresponding to a target speed sub-interval, and representative short driving samples corresponding to each short driving sample category are determined.

[0104] Among them, the similarity between the short driving samples in different short driving sample categories is less than the similarity threshold.

[0105] The short driving sample categories are obtained by further subdividing the target speed sub-intervals according to similarity. The short driving samples in the same short driving sample category have similarities.

[0106] The similarity threshold is the standard for judging whether two short driving samples belong to the same category. If the similarity between two short driving samples is less than the similarity threshold, the two short driving samples are considered to be dissimilar and should be classified into different short driving sample categories.

[0107] One or more representative short driving samples are determined for each short driving sample category, and the representative short driving samples are used to reflect the characteristics of the short driving sample category to which they belong.

[0108] A2: Multiple short driving samples representing the relationship between driving speed and driving time are concatenated to obtain the sub-vehicle road spectrum corresponding to the target speed sub-interval.

[0109] For example, assuming the target speed sub-interval is [60,80] km / h, based on the similarity threshold, the short driving sample set corresponding to the target speed sub-interval is divided into three main categories. A representative sample is selected from each category. The categories of these three representative short driving samples are:

[0110] Representative short driving sample 1: Smoothly accelerate to 75km / h and then decelerate to stop.

[0111] Representative short driving sample 2: Rapidly accelerate to 80km / h and maintain for a while before decelerating.

[0112] Representative short driving sample 3: Slowly accelerate to 65km / h and then maintain a constant speed.

[0113] Next, the time-varying relationship curves of the three representative samples are spliced ​​together to create the sub-vehicle road spectrum of the speed sub-interval.

[0114] By determining the representative short driving samples in the short driving sample category, the different behavior patterns of the target speed sub-intervals can be better reflected, so that the sub-vehicle road spectra generated in this way can reflect the behavior patterns corresponding to each short driving category and be representative in each short driving category.

[0115] In a possible implementation, it is determined that each short driving sample category corresponds to a representative short driving sample, and the representative short driving samples correspond to the short driving sample categories one by one.

[0116] Because the sub-vehicle road spectrum is obtained by splicing representative short driving samples, the number of representative short driving samples directly determines the driving time range of the sub-vehicle road spectrum. Too many representative short driving samples will cause the driving time corresponding to the sub-vehicle road spectrum to be too long, which is not conducive to research and analysis of the vehicle road spectrum.

[0117] A representative short driving sample is determined for each short driving sample category to reflect its characteristics, making the generated sub-vehicle road spectrum more intuitive and easy to understand. Researchers can directly analyze these representative samples to understand the speed change pattern of the vehicle under specific working conditions, which facilitates subsequent data interpretation and research work.

[0118] This embodiment of the application provides a specific implementation method for determining a representative short driving sample, see B1-B3:

[0119] B1: Based on the short driving sample set corresponding to the target speed sub-interval, multiple clusters are obtained through clustering operation.

[0120] The cluster includes a cluster center, and each cluster corresponds to a short driving sample category.

[0121] A clustering operation is performed on the short driving sample set corresponding to the target speed sub-interval. Clustering is an unsupervised learning method that can automatically divide the short driving samples in the short driving sample set into different clusters according to their characteristics. Generally speaking, clustering operations include K-means clustering (K-mean), DBSCAN (Density-Based Spatial Clustering of Applications with Noise, DBSCAN), etc.

[0122] The cluster center is a representative point of all the short driving samples in the cluster, usually the characteristic average of the short driving samples in the cluster or the data point that best reflects the characteristics of the samples in the cluster.

[0123] B2: If the cluster center of the cluster corresponding to the short driving sample category has a corresponding representative short driving sample, the short driving sample corresponding to the cluster center is determined as the representative short driving sample corresponding to the short driving sample category.

[0124] B3: If the cluster center of the cluster corresponding to the short driving sample category does not have a corresponding representative short driving sample, the short driving sample with the smallest feature difference with the cluster center of the cluster corresponding to the short driving sample category is determined as the representative short driving sample corresponding to the short driving sample category.

[0125] When the cluster center is not any specific short driving sample, it is necessary to find a short driving sample with the closest characteristics to the cluster center as the representative short driving sample. This is usually done by calculating the distance from each short driving sample to the cluster center. The distance is the difference in characteristics between the cluster center and the cluster corresponding to the short driving sample category. The short driving sample with the smallest distance is selected as the representative short driving sample of the category. In this case, although the cluster center is not used directly, the main characteristics of the cluster can still be well represented by the representative short driving sample.

[0126] The cluster center is the most representative point calculated based on the characteristics of all short driving samples in the cluster. Taking the short driving sample corresponding to the cluster center or the short driving sample closest to it as the representative short driving sample can more accurately reflect the main characteristics and behavior pattern of the cluster, so that the sub-vehicle road spectrum obtained by splicing the representative short driving samples can more accurately reflect the speed change law of the corresponding speed range, be closer to the actual general speed distribution, and be more representative.

[0127] This embodiment of the application provides a specific implementation method for determining a short driving sample set, see C1-C5:

[0128] C1: Obtain speed threshold and driving time threshold.

[0129] The speed threshold is a preset speed limit used to distinguish a high-speed sub-interval from multiple speed sub-intervals, and the driving time threshold is a preset driving time limit used to distinguish short driving samples with excessively long driving time from the high-speed sub-interval.

[0130] C2: Determine a high-speed sub-interval according to the minimum speed of each speed sub-interval, where the high-speed sub-interval is a speed sub-interval whose minimum speed is greater than the speed threshold.

[0131] According to the lowest speed of each speed sub-interval and based on the speed threshold, a speed sub-interval whose lowest speed is greater than the speed threshold is determined as a high-speed sub-interval.

[0132] For example, the speed threshold is 80 km / h, and the minimum speeds of speed sub-interval A [90, 100) km / h and speed sub-interval B [100, 120) km / h are both greater than the speed threshold, so speed sub-interval A and speed sub-interval B are determined to be high-speed sub-intervals.

[0133] C3: According to the speed sub-interval in which the average speed of the short driving samples is located, determine the initial short driving sample sets corresponding to the high-speed sub-intervals respectively.

[0134] C4: From the initial short driving sample sets corresponding to the high-speed sub-intervals, delete the short driving samples whose driving duration is greater than the driving duration threshold, and obtain the first short driving sample sets corresponding to the high-speed sub-intervals.

[0135] The initial short driving sample set corresponding to the high-speed sub-interval often comes from road scenarios such as highways and urban expressways that require long driving time. In this road scenario, there are short driving samples with long driving time, which affects the determination of representative short driving samples and the driving time range of the sub-vehicle road spectrum, thereby affecting the accuracy of the sub-vehicle road spectrum.

[0136] C5: Determine the first short driving sample sets corresponding to the high-speed sub-intervals as the short driving sample sets corresponding to the high-speed sub-intervals.

[0137] Based on the speed threshold and the driving time threshold, unrepresentative short driving samples with too long driving time in the initial short driving sample set corresponding to the high-speed sub-interval are deleted to obtain a first short driving sample set, and the first short driving sample set is determined as the short driving sample set, thereby improving the sample data quality of the short driving sample set, so that the driving time of the sub-vehicle road spectrum corresponding to the high-speed sub-interval is controlled within a reasonable range.

[0138] This embodiment of the application provides a specific implementation method for excluding isolated short driving samples, see D1-D7:

[0139] D1: Get the exclusion strategy obtained by the ith update. The exclusion strategy is used to exclude isolated short driving samples.

[0140] Among them, the isolated short driving sample is a short driving sample that deviates from the characteristics of other short driving samples in the same short driving sample set, and i is a positive integer.

[0141] The exclusion strategy is a set of rules for identifying and removing isolated short driving samples, which are short driving samples that have large feature deviations from other short driving samples in the same short driving sample set. The exclusion strategy will be updated multiple times, and each update will more accurately identify and remove isolated short driving samples.

[0142] D2: The overall speed range is divided into multiple speed categories, and the speed difference between short driving samples in different speed categories is greater than the speed category threshold.

[0143] The speed category threshold is used to define the speed range of the speed category. For example, the overall speed interval [0,100] km / h is divided into multiple speed categories [0,1) km / h, [1,2) km / h, [2,3) km / h, and so on.

[0144] The process of obtaining the speed category is the process of integerizing the speed relationship. The lower limit value of each speed category represents its speed category, which is convenient for counting the proportion of the number of speed category occurrences in the total number of speed categories generated by multiple short driving samples.

[0145] For example, there are speed category A[0,1)km / h, speed category B[1,2)km / h and speed category C[2,3)km / h, and 1km / h, 2km / h and 3km / h are respectively used as representatives to perform statistics on the overall speed distribution.

[0146] D3: The overall speed distribution is obtained according to the proportion of the number of times each speed category appears in multiple short driving samples to the total number of speed categories generated by multiple short driving samples.

[0147] In the process of obtaining the overall speed distribution, data points with a speed of 0 km / h in short driving samples are usually removed.

[0148] See also Figure 2 , Figure 2 A schematic diagram of overall speed distribution and speed interval division is provided for an embodiment of the present application, wherein the corresponding curves are the speed proportions corresponding to each speed category.

[0149] D4: According to the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution, the exclusion strategy obtained by the i-th update is updated to obtain the exclusion strategy obtained by the i+1-th update.

[0150] The overall speed distribution reflects the frequency of occurrence of different vehicle speeds in a large amount of sample data, and is an important indicator for determining whether the vehicle road spectrum is representative. If the speed distribution of the vehicle road spectrum is close to the overall speed distribution, it means that the speed frequency information of different speed ranges of the vehicle road spectrum is highly consistent with a large amount of sample data and is representative. It has a high reference value for studying vehicle energy consumption, pollutant emissions, and other aspects.

[0151] D5: According to the speed sub-interval in which the average speed of the short driving sample is located, determine the initial short driving sample sets corresponding to each speed sub-interval.

[0152] D6: Based on the exclusion strategy obtained by the (i+1)th update, the isolated short driving samples corresponding to each speed subinterval are deleted from the short driving sample sets corresponding to each speed subinterval, so as to obtain the second short driving sample sets corresponding to each speed subinterval.

[0153] Before the exclusion strategy is updated, the first exclusion strategy is obtained. Based on the first exclusion strategy, isolated short driving samples corresponding to each speed sub-interval are deleted from the short driving sample sets corresponding to each speed sub-interval, so as to obtain the second short driving sample sets corresponding to each speed sub-interval.

[0154] See also Figure 7 , Figure 7 A speed distribution comparison diagram is provided for an embodiment of the present application. The dotted line curve is the speed distribution corresponding to the vehicle road spectrum, and the solid line curve is the overall speed distribution. It can be clearly seen from the figure that there is a difference between the two.

[0155] The exclusion strategy obtained by the i+1th update is obtained by updating the previous exclusion strategy based on the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution. Therefore, the vehicle road spectrum corresponding to the short driving sample set determined by the exclusion strategy obtained by the i+1th update has a speed distribution that is closer to the overall speed distribution and is more representative and of reference value.

[0156] D7: Determine the second short driving sample sets corresponding to the speed sub-intervals as the short driving sample sets corresponding to the speed sub-intervals.

[0157] The exclusion strategy corresponds to the way of determining the short driving sample set, which affects the generation process of the sub-vehicle road spectrum and the vehicle road spectrum. Therefore, according to the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution, the exclusion strategy obtained by the i-th update is updated to obtain the exclusion strategy obtained by the i+1-th update, so that the short driving sample set determined by the exclusion strategy obtained by the i+1-th update has fewer isolated short driving samples, thereby making the short driving sample set more accurate.

[0158] Specifically, the exclusion strategy includes excluding a proportion of isolated short driving samples in each initial short driving sample set.

[0159] The multiple initial short driving sample sets determined by D5, each of which has a corresponding proportion of excluding isolated short driving samples.

[0160] For example, for three speed sub-intervals: speed interval A, speed interval B and speed interval C, there are three corresponding initial short driving sample sets, set A, set B and set C, the exclusion strategy corresponding to set A includes RA (0.1), the exclusion strategy corresponding to set B includes RB (0.37), and the exclusion strategy corresponding to set C includes RC (0.37). RA indicates that 10% of the isolated short driving samples in set A are excluded, RB indicates that 37% of the isolated short driving samples in set B are excluded, and RC indicates that 37% of the isolated short driving samples in set C are excluded.

[0161] As an implementation method, the exclusion strategy is implemented through the Isolation Forest (iForest) algorithm. The Isolation Forest algorithm includes hyperparameters corresponding to the initial short driving set. The hyperparameters are used to indicate the proportion of isolated short driving samples to be excluded. The difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution is determined based on the Euclidean distance. The hyperparameters are updated by searching the step size until the optimal hyperparameters corresponding to the minimum difference are obtained. The vehicle road spectrum corresponding to the optimal hyperparameters is determined as the final vehicle road spectrum.

[0162] By updating the exclusion strategy to exclude the proportion of isolated short driving samples in each initial short driving sample set, the exclusion method of isolated short driving samples is made more accurate, and the vehicle road spectrum is iteratively updated in a more detailed manner, so that the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution gradually becomes smaller, and the data processing process is more transparent and easy to understand.

[0163] See also Figure 8 , Figure 8A schematic diagram of the structure of a vehicle road spectrum generating device provided in an embodiment of the present application, the device 200 includes: an acquisition unit 201, a division unit 202, a set determination unit 203, a target sub-road spectrum unit 204, a sub-road spectrum unit 205 and a road spectrum splicing unit 206.

[0164] The acquisition unit 201 is used to acquire the overall speed interval and multiple short driving samples. The short driving sample is the relationship between the driving speed and the driving time from the start of the vehicle to the stop of the vehicle. The overall speed interval is the driving speed coverage range of the multiple short driving samples.

[0165] The dividing unit 202 is used to divide the overall speed interval into intervals to obtain a plurality of speed sub-intervals, where the speed sub-interval is a part of the overall speed interval.

[0166] The set determination unit 203 is used to determine the short driving sample sets corresponding to the speed subintervals according to the speed subintervals in which the average speed of the short driving samples is located, wherein the short driving sample sets include at least one short driving sample located in the corresponding speed subinterval.

[0167] The target sub-road spectrum unit 204 is used to obtain a sub-vehicle road spectrum corresponding to the target speed sub-interval among the multiple speed sub-intervals, based on a short driving sample set corresponding to the target speed sub-interval and a relationship between the driving speed and the driving time described by the short driving samples included in the short driving sample set, wherein the sub-vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time in the target speed sub-interval.

[0168] The sub-road spectrum unit 205 is used to take each speed sub-interval as a target speed sub-interval, and obtain a sub-vehicle road spectrum corresponding to each speed sub-interval.

[0169] The road spectrum splicing unit 206 is used to splice the sub-vehicle road spectra corresponding to each sub-interval to obtain a vehicle road spectrum. The vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time.

[0170] In view of the lack of representativeness of the generated vehicle road spectrum, the acquisition unit 201 acquires the overall speed interval to lay the foundation for speed interval division. The acquisition unit 201 acquires multiple short driving samples to determine the data source for generating the vehicle road spectrum. The division unit 202 divides the overall speed interval into intervals to obtain multiple speed sub-intervals, thereby classifying multiple short driving samples according to the working conditions represented by the speed intervals. The set determination unit 203 determines the short driving sample sets corresponding to each speed sub-interval according to the speed sub-interval in which the average speed of the short driving sample is located, so that the data source reflecting the specific working condition is more targeted and the accuracy of road spectrum generation is improved. The target sub-road spectrum unit 204 obtains the sub-vehicle road spectrum corresponding to the target speed sub-interval according to the short driving sample set corresponding to the target speed sub-interval among the multiple speed sub-intervals, through the relationship between the driving speed described by the short driving samples included in the short driving sample set and the driving time. Based on the more targeted short driving sample set, the speed distribution of the sub-vehicle road spectrum is closer to the speed distribution of the working condition corresponding to the speed sub-interval, and has local representativeness. The sub-road spectrum unit 205 takes each speed sub-interval as the target speed sub-interval, and obtains the sub-vehicle road spectrum corresponding to each speed sub-interval, so that the sub-vehicle road spectrum corresponding to each speed sub-interval is locally representative. The road spectrum splicing unit 206 splices the sub-vehicle road spectra corresponding to each speed sub-interval to obtain the vehicle road spectrum, so that the vehicle road spectrum fully covers the driving state of the vehicle in different speed sub-intervals, reflects the speed distribution of the corresponding working conditions in different speed sub-intervals, improves the overall representativeness and authenticity of the vehicle road spectrum, and has a higher reference value for studying vehicle energy consumption and pollutant emissions.

[0171] As a possible implementation manner, the target sub-path spectrum unit 204 is specifically used for:

[0172] A plurality of short driving sample categories are obtained according to a short driving sample set corresponding to a target speed subinterval, and representative short driving samples corresponding to each short driving sample category are determined, and the similarity between short driving samples in different short driving sample categories is less than a similarity threshold.

[0173] The relationship between the driving speed and the driving time described by multiple representative short driving samples is spliced ​​together to obtain the sub-vehicle road spectrum corresponding to the target speed sub-interval.

[0174] As a possible implementation manner, the target sub-path spectrum unit 204 is specifically used for:

[0175] It is determined that each short driving sample category corresponds to a representative short driving sample, and the representative short driving samples correspond to the short driving sample categories one by one.

[0176] As a possible implementation, the target sub-path spectrum unit 204 is specifically used for:

[0177] According to the short driving sample set corresponding to the target speed sub-interval, a plurality of clusters are obtained through clustering operation, wherein the cluster includes a cluster center, and each cluster corresponds to a short driving sample category.

[0178] If the cluster center of the cluster corresponding to the short driving sample category has a corresponding representative short driving sample, the short driving sample corresponding to the cluster center is determined as the representative short driving sample corresponding to the short driving sample category.

[0179] If the cluster center of the cluster corresponding to the short driving sample category does not have a corresponding representative short driving sample, the short driving sample with the smallest feature difference with the cluster center of the cluster corresponding to the short driving sample category is determined as the representative short driving sample corresponding to the short driving sample category.

[0180] As a possible implementation manner, the apparatus 200 further includes a threshold value obtaining unit, which is used to:

[0181] Get the speed threshold and driving time threshold.

[0182] The set determination unit 203 is specifically used for:

[0183] A high-speed sub-interval is determined according to the lowest speed of each speed sub-interval, and the high-speed sub-interval is a speed sub-interval in which the lowest speed is greater than a speed threshold.

[0184] According to the speed sub-interval in which the average speed of the short driving sample is located, the initial short driving sample sets corresponding to the high-speed sub-intervals are determined.

[0185] From the initial short driving sample sets corresponding to the high-speed sub-intervals, short driving samples whose driving duration is greater than the driving duration threshold are deleted to obtain the first short driving sample sets corresponding to the high-speed sub-intervals.

[0186] The first short driving sample sets respectively corresponding to the high-speed sub-intervals are determined as the short driving sample sets respectively corresponding to the high-speed sub-intervals.

[0187] As a possible implementation, the apparatus 200 further includes an updating exclusion strategy unit, which is configured to:

[0188] Get the exclusion strategy obtained by the i-th update. The exclusion strategy is used to exclude isolated short driving samples. Isolated short driving samples are short driving samples that deviate from the characteristics of other short driving samples in the same short driving sample set. i is a positive integer.

[0189] The overall speed range is divided into multiple speed categories, and the speed difference between short driving samples in different speed categories is greater than the speed category threshold.

[0190] The overall speed distribution is obtained according to the proportion of the number of times each speed category appears in multiple short driving samples to the total number of speed categories generated by the multiple short driving samples.

[0191] According to the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution, the exclusion strategy obtained by the i-th update is updated to obtain the exclusion strategy obtained by the (i+1)-th update.

[0192] The set determination unit 203 is specifically used for:

[0193] According to the speed sub-interval in which the average speed of the short driving sample is located, an initial short driving sample set corresponding to each speed sub-interval is determined.

[0194] Based on the exclusion strategy obtained by the (i+1)th update, isolated short driving samples corresponding to each speed subinterval are deleted from the short driving sample sets corresponding to each speed subinterval, so as to obtain second short driving sample sets corresponding to each speed subinterval.

[0195] The second short driving sample sets corresponding to the speed sub-intervals are determined as the short driving sample sets corresponding to the speed sub-intervals.

[0196] As a possible implementation manner, the exclusion strategy in the updating exclusion strategy unit is used to exclude isolated short driving samples, and includes excluding the proportion of isolated short driving samples in each initial short driving sample set.

[0197] The embodiment of the present application also provides a vehicle road spectrum generating device, including a processor and a memory:

[0198] The memory is used to store the program code and transmit the program code to the processor;

[0199] The processor is used to execute the method of the above method embodiment according to the instructions in the program code.

[0200] The embodiment of the present application also provides a computer-readable storage medium, which is used to store a computer program, and the computer program is used to execute the method of the above method embodiment.

[0201] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0202] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0203] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0204] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle road spectrum generation method, characterized in that: The method comprises: Acquire an overall speed interval and a plurality of short driving samples, wherein the short driving sample is a relationship between a driving speed and a driving duration from when the vehicle starts to move to when the vehicle stops, and the overall speed interval is a driving speed coverage range of the plurality of short driving samples; Dividing the overall speed interval into multiple speed sub-intervals, wherein the speed sub-intervals are parts of the overall speed interval; Determine, according to the speed subinterval in which the average speed of the short driving sample is located, a short driving sample set corresponding to each of the speed subintervals, wherein the short driving sample set includes at least one short driving sample located in the corresponding speed subinterval; For a target speed subinterval among the plurality of speed subintervals, according to a short driving sample set corresponding to the target speed subinterval, a sub-vehicle road spectrum corresponding to the target speed subinterval is obtained through a relationship between a driving speed described by short driving samples included in the short driving sample set and a driving duration, wherein the sub-vehicle road spectrum is used to indicate a driving speed of a vehicle that changes with a driving duration in the target speed subinterval; Taking each of the speed sub-intervals as the target speed sub-interval, and obtaining a sub-vehicle road spectrum corresponding to each of the speed sub-intervals; The sub-vehicle road spectra corresponding to each of the speed sub-intervals are spliced ​​to obtain a vehicle road spectrum, and the vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time.

2. The method according to claim 1, characterized in that The sub-vehicle road spectrum corresponding to the target speed sub-interval is obtained according to the short driving sample set corresponding to the target speed sub-interval by using the relationship between the driving speed and the driving time described by the short driving samples included in the short driving sample set, including: Obtaining a plurality of short driving sample categories according to the short driving sample set corresponding to the target speed subinterval, determining representative short driving samples corresponding to each of the short driving sample categories, wherein the similarity between the short driving samples in different short driving sample categories is less than a similarity threshold; The relationship between the driving speed and the driving time described by the multiple representative short driving samples is spliced ​​together to obtain the sub-vehicle road spectrum corresponding to the target speed sub-interval.

3. The method according to claim 2, characterized in that The determining of representative short driving samples corresponding to the respective short driving sample categories comprises: It is determined that each of the short driving sample categories corresponds to a representative short driving sample, and the representative short driving samples correspond to the short driving sample categories one by one.

4. The method according to claim 3, characterized in that The step of obtaining a plurality of short driving sample categories according to the short driving sample set corresponding to the target speed subinterval and determining representative short driving samples corresponding to each of the short driving sample categories comprises: According to the short driving sample set corresponding to the target speed sub-interval, a plurality of clusters are obtained through a clustering operation, wherein the cluster includes a cluster center, and each cluster corresponds to a short driving sample category; If the cluster center of the cluster corresponding to the short driving sample category has a corresponding representative short driving sample, the short driving sample corresponding to the cluster center is determined as the representative short driving sample corresponding to the short driving sample category; If the cluster center of the cluster corresponding to the short driving sample category does not have a corresponding representative short driving sample, the short driving sample with the smallest feature difference from the cluster center of the cluster corresponding to the short driving sample category is determined as the representative short driving sample corresponding to the short driving sample category.

5. The method according to claim 1, characterized in that The method further comprises: Obtain speed threshold and driving time threshold; The determining, according to the speed subinterval in which the average speed of the short driving samples is located, a set of short driving samples corresponding to each speed subinterval respectively comprises: Determine a high-speed sub-interval according to the lowest speed of each speed sub-interval, wherein the high-speed sub-interval is a speed sub-interval whose lowest speed is greater than the speed threshold; Determining initial short driving sample sets corresponding to the high-speed subintervals respectively according to the speed subintervals in which the average speed of the short driving samples is located; From the initial short driving sample sets corresponding to the high-speed sub-intervals, short driving samples whose driving duration is greater than the driving duration threshold are deleted to obtain first short driving sample sets corresponding to the high-speed sub-intervals; The first short driving sample sets respectively corresponding to the high-speed sub-intervals are determined as the short driving sample sets respectively corresponding to the high-speed sub-intervals.

6. The method according to claim 1, characterized in that The method further comprises: Obtaining the exclusion strategy obtained by the i-th update, wherein the exclusion strategy is used to exclude isolated short driving samples, wherein the isolated short driving samples are short driving samples that deviate from the characteristics of other short driving samples in the same short driving sample set, and i is a positive integer; The overall speed interval is divided into a plurality of speed categories, and the speed difference between short driving samples in different speed categories is greater than a speed category threshold; Obtaining an overall speed distribution according to a proportion of the number of times each of the speed categories appears in the multiple short driving samples to the total number of times the speed categories are generated in the multiple short driving samples; According to the difference between the speed distribution corresponding to the vehicle road spectrum and the overall speed distribution, updating the exclusion strategy obtained by the i-th update to obtain the exclusion strategy obtained by the i+1-th update; The determining, according to the speed subinterval in which the average speed of the short driving samples is located, a set of short driving samples corresponding to each speed subinterval respectively comprises: According to the speed subinterval in which the average speed of the short driving sample is located, determining the initial short driving sample sets corresponding to the speed subintervals respectively; Based on the exclusion strategy obtained by the (i+1)th update, the isolated short driving samples corresponding to each speed subinterval are deleted from the short driving sample sets corresponding to each speed subinterval, so as to obtain the second short driving sample sets corresponding to each speed subinterval; The second short driving sample sets corresponding to the speed sub-intervals are determined as short driving sample sets corresponding to the speed sub-intervals.

7. The method according to claim 6, characterized in that The exclusion strategy is used to exclude the isolated short driving samples, and includes excluding a proportion of the isolated short driving samples in each of the initial short driving sample sets.

8. A vehicle road spectrum generating device, characterized in that: The device comprises: an acquisition unit, a division unit, a set determination unit, a target sub-road spectrum unit, a sub-road spectrum unit and a road spectrum splicing unit; The acquisition unit is used to acquire an overall speed interval and a plurality of short driving samples, wherein the short driving sample is a relationship between a driving speed and a driving duration from the start of movement of the vehicle to the stop of the vehicle, and the overall speed interval is a driving speed coverage range of the plurality of short driving samples; The dividing unit is used to divide the overall speed interval into multiple speed sub-intervals, where the speed sub-intervals are parts of the overall speed interval; The set determination unit is used to determine the short driving sample sets corresponding to the speed subintervals respectively according to the speed subintervals in which the average speed of the short driving samples is located, wherein the short driving sample sets include at least one short driving sample located in the corresponding speed subinterval; The target sub-road spectrum unit is used to obtain, for a target speed sub-interval in the plurality of speed sub-intervals, a sub-vehicle road spectrum corresponding to the target speed sub-interval according to a short driving sample set corresponding to the target speed sub-interval and a relationship between a driving speed and a driving duration described by the short driving samples included in the short driving sample set, wherein the sub-vehicle road spectrum is used to indicate a driving speed of a vehicle that changes with a driving duration in the target speed sub-interval; The sub-road spectrum unit is used to use each of the speed sub-intervals as the target speed sub-interval to obtain the sub-vehicle road spectrum corresponding to each of the speed sub-intervals; The road spectrum splicing unit is used to splice the sub-vehicle road spectra corresponding to each of the sub-intervals to obtain a vehicle road spectrum, and the vehicle road spectrum is used to indicate the driving speed of the vehicle that changes with the driving time.

9. A vehicle road spectrum generating device, characterized in that: The device comprises a memory and a processor: The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the method according to any one of claims 1 to 7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 7.

Images