Method, apparatus, and computer readable storage medium for determining vehicle similarity

Abstract

The present disclosure relates to a method, device and computer readable storage medium for determining vehicle similarity. The method comprises: determining a target vehicle and a sample vehicle; determining at least one metric for measuring similarity between the target vehicle and the sample vehicle, the metric being represented by performance curves obtained by performance testing or simulation of the target vehicle and the sample vehicle; determining curve similarity of each performance curve of the sample vehicle to each corresponding performance curve of the target vehicle, determining similarity of the sample vehicle to the target vehicle on each metric based on the curve similarity; and determining similarity of the sample vehicle to the target vehicle based on the similarity of the sample vehicle to the target vehicle on each metric.

Description

Technical Field

[0001] This disclosure relates to the field of vehicles, and more particularly to methods, apparatus and computer-readable storage media for determining vehicle similarity. Background Technology

[0002] In the design of new vehicles and the improvement of existing vehicles, the design of new vehicles and the improvement of existing vehicles are generally based on existing vehicles that are most similar to the new vehicle model or the existing vehicle that needs improvement. When determining the existing vehicle that is most similar to the new vehicle model or other existing vehicles that are most similar to the existing vehicle that needs improvement, various performance metrics of the vehicle are generally considered, such as the vehicle's motion and flexibility (K&C) metrics.

[0003] However, the selection of which metrics to use and how to determine vehicle similarity based on those metrics heavily relies on the engineer's experience. Engineers with different levels of experience may focus on different metrics, and even when metrics are selected, there are currently no unified evaluation standards, nor are there quantitative analytical methods or tools to assess the similarity between different vehicles based on these metrics. Engineers often spend a significant amount of time performing repetitive or mechanical comparisons when evaluating vehicle similarity. Summary of the Invention

[0004] Based on the above, this disclosure provides a method, apparatus, and computer-readable storage medium capable of quantitatively determining vehicle similarity.

[0005] In one aspect of this disclosure, a method for determining vehicle similarity is provided, the method comprising: identifying a target vehicle and a sample vehicle; determining at least one metric for measuring the similarity between the target vehicle and the sample vehicle, the metric being represented by performance curves obtained by performance testing or simulation of the target vehicle and the sample vehicle; determining the curve similarity between the respective performance curves of the sample vehicle and the respective corresponding performance curves of the target vehicle, determining the similarity between the sample vehicle and the target vehicle in each metric based on the curve similarity; and determining the similarity between the sample vehicle and the target vehicle in each metric based on the similarity between the sample vehicle and the target vehicle.

[0006] In another aspect of this disclosure, an apparatus for determining vehicle similarity is provided, comprising: a memory having computer instructions stored thereon; and a processor, wherein the instructions, when executed by the processor, cause the processor to perform a method for determining vehicle similarity according to an embodiment of the fundamental disclosure.

[0007] In another aspect of this disclosure, a non-transitory computer-readable storage medium is provided that stores instructions that cause a processor to perform a method for determining vehicle similarity according to embodiments of the fundamental disclosure.

[0008] Furthermore, this disclosure also provides an apparatus for determining vehicle similarity, comprising: a vehicle determination module for determining a target vehicle and a sample vehicle; a metric determination module for determining at least one metric for measuring the similarity between the target vehicle and the sample vehicle, the metric being represented by performance curves obtained by performance testing or simulation of the target vehicle and the sample vehicle; a metric similarity determination module for determining the curve similarity between each performance curve of the sample vehicle and each corresponding performance curve of the target vehicle, and determining the similarity between the sample vehicle and the target vehicle in each metric based on the curve similarity; and a vehicle similarity determination module for determining the similarity between the sample vehicle and the target vehicle based on the similarity between the sample vehicle and the target vehicle in each metric.

[0009] Furthermore, this disclosure also provides a computer program product for determining vehicle similarity, comprising instructions that cause a processor to perform a method for determining vehicle similarity according to embodiments of the fundamental disclosure.

[0010] The methods, apparatus, computer-readable storage media, devices, and computer program products for determining vehicle similarity according to embodiments of the present disclosure provide a quantitative method for determining the similarity of different vehicles, which can quantitatively assess the similarity of different vehicles, thereby enabling engineers to quickly find the sample vehicle that best meets the design goals from existing vehicles, greatly improving design efficiency. Attached Figure Description

[0011] The above and other objects, features, and advantages of this disclosure will become more apparent from a more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to offer a further understanding of the embodiments of this disclosure and form part of the specification. The drawings, together with the embodiments of this disclosure, are used to explain this disclosure but do not constitute a limitation thereof. In the drawings, unless otherwise expressly indicated, the same reference numerals denote the same parts, steps, or elements. In the drawings,

[0012] Figure 1 This is an example flowchart of a method for determining vehicle similarity according to embodiments of the present disclosure;

[0013] Figure 2 This further demonstrates Figure 1 Example flowchart for determining the curve similarity of performance curves;

[0014] Figure 3 This is an example schematic diagram used to illustrate the determination of point similarity according to embodiments of the present disclosure;

[0015] Figure 4 This is another example schematic diagram used to illustrate the determination of point similarity according to embodiments of the present disclosure;

[0016] Figure 5 This is an example schematic diagram illustrating the determination of candidate points for interpolation according to embodiments of the present disclosure;

[0017] Figure 6 This further demonstrates Figure 2 Example flowchart for determining curve similarity based on point similarity;

[0018] Figure 7 This is an example schematic diagram illustrating segmentation of a performance curve according to an embodiment of the present disclosure;

[0019] Figure 8 This is another example schematic diagram illustrating the segmentation of a performance curve according to embodiments of the present disclosure;

[0020] Figure 9 This is an example schematic diagram illustrating the cutting of a curve according to an embodiment of the present disclosure;

[0021] Figure 10 This further demonstrates Figure 1 Example flowchart for determining the similarity between sample vehicles and target vehicles across various metrics based on curve similarity;

[0022] Figure 11 An example of the similarity results between a sample vehicle and a target vehicle determined using a method for determining vehicle similarity according to embodiments of the present disclosure is shown;

[0023] Figure 12 An example vehicle evaluation system according to an embodiment of the present disclosure is shown;

[0024] Figure 13 Another example vehicle evaluation system according to an embodiment of the present disclosure is shown;

[0025] Figure 14 This is an example block diagram of an apparatus for determining vehicle similarity according to embodiments of the present disclosure; and

[0026] Figure 15 This is an example block diagram of an apparatus for determining vehicle similarity according to embodiments of the present disclosure. Detailed Implementation

[0027] The technical solutions of this disclosure will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the protection scope of this disclosure.

[0028] In the description of this disclosure, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this disclosure. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Similarly, words such as "an," "a," or "the" do not indicate a quantity limitation, but rather indicate the presence of at least one. Words such as "comprising" or "including" mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0029] In the description of this disclosure, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this disclosure based on the specific circumstances.

[0030] Furthermore, the technical features involved in the different embodiments of this disclosure described below can be combined with each other as long as they do not conflict with each other.

[0031] Figure 1 This is an example flowchart of a method for determining vehicle similarity according to embodiments of the present disclosure. In this disclosure, a vehicle can be a specific model of vehicle or a representative vehicle representing a certain vehicle model. Therefore, the method for determining vehicle similarity according to embodiments of the present disclosure can be used to determine the similarity between different specific models of the same vehicle model, or it can be used to determine the similarity between different vehicle models. Figure 1 As shown, the method for determining vehicle similarity according to an embodiment of the present disclosure begins at step S100.

[0032] In step S100, the target vehicle and the sample vehicle are determined. In this disclosure, the target vehicle may include a target vehicle model obtained through simulation in the design of a new vehicle (model or vehicle variant under a model), and an existing vehicle requiring improvement or upgrade (e.g., an existing vehicle already on the market). The sample vehicle includes existing vehicles, such as vehicles already on the market or vehicles whose design has been completed but not yet on the market.

[0033] The method then proceeds to step S110. In step S110, at least one metric is determined to measure the similarity between the target vehicle and the sample vehicle. The metric can be represented by performance curves obtained from performance testing or simulation of the target vehicle and the sample vehicle. The performance curves can be two-dimensional or multi-dimensional. For example, Figure 3 The two-dimensional performance curve 300 in the figure. The metric can also be represented by points obtained from performance testing or simulation of the target vehicle and sample vehicles. A point can be a single point in multi-dimensional space, i.e., a single point represented by multiple coordinates. For example, a single point in two-dimensional space, which can be represented, for example, by abscissa and ordinate (e.g., ...). Figure 3 (e.g., point P5 in the diagram). Another example is a single point in one-dimensional space, such as a single numerical value. Furthermore, in this disclosure, curves also include straight lines.

[0034] In this disclosure, the metric can be any performance metric used to represent the performance of a vehicle, such as any specific performance metric in the vehicle's sportiness and flexibility (K&C) metrics, such as toe change, camber change, or wheel rate obtained at the wheel center, etc.

[0035] After determining at least one metric for measuring the similarity between the target vehicle and the sample vehicle, the method proceeds to step S120. In step S120, the curve similarity between the individual performance curves of the sample vehicle and the corresponding performance curves of the target vehicle is determined, which will be combined later. Figures 2 to 9 Detailed description.

[0036] The method then proceeds to step S130. At step S130, the similarity between the sample vehicle and the target vehicle across various metrics is determined based on curve similarity. In one embodiment, the curve similarity determined in step S120 can be directly used as the similarity between the sample vehicle and the target vehicle across the corresponding metrics. In another embodiment, the curve similarity determined in step S120 can be mathematically processed (e.g., normalized), and the result can be used as the similarity between the sample vehicle and the target vehicle across the corresponding metrics. In yet another embodiment, the similarity between the sample vehicle and the target vehicle across the corresponding metrics can be determined based on the curve similarity determined in step S120, combined with other factors (e.g., curve features characterizing the morphological changes of the performance curve), which will be combined later... Figure 10 Detailed description.

[0037] The method then proceeds to step S140. In step S140, the similarity between the sample vehicle and the target vehicle is determined based on their similarity across various metrics. In one embodiment, the minimum or maximum value of the similarity across the various metrics can be determined as the similarity between the sample vehicle and the target vehicle. In another embodiment, the root mean square error of the similarity between the sample vehicle and the target vehicle across various metrics can be calculated, thereby determining the similarity between the sample vehicle and the target vehicle. In yet another embodiment, a weighted sum of the similarity across the various metrics between the sample vehicle and the target vehicle can be used to determine the similarity between the sample vehicle and the target vehicle. In this case, regarding the determination of the weights of the various metrics, in one embodiment, the weights of the various metrics can be determined based on their importance using the Analytic Hierarchy Process (AHP), which will be combined later. Figure 12 Detailed description. In another embodiment, the weights of each metric can be determined using a neural network.

[0038] Combination Figure 1 The method for determining vehicle similarity according to embodiments of the present disclosure determines vehicle similarity based on the similarity of performance curves representing vehicle metrics. It provides a quantitative method for different vehicle similarities, which can quantitatively assess the similarity of different vehicles. This allows engineers to quickly find the sample vehicle that best meets the design goals from existing vehicles, greatly improving design efficiency.

[0039] Figure 2 This further demonstrates Figure 1 An example flowchart for determining the curve similarity of performance curves, starting from step S122.

[0040] In step S122, the point distance between a point on the performance curve of the target vehicle and a corresponding point on the corresponding performance curve of the sample vehicle is determined. The points on the performance curve of the target vehicle and the corresponding points on the corresponding performance curve of the sample vehicle can be points used to fit the performance curves, obtained through performance testing or simulation of the target vehicle and the sample vehicle. Specifically, assuming the points on the performance curve of the target vehicle and the corresponding points on the corresponding performance curve of the sample vehicle are represented as P(x1, y1) and Q(x2, y2) respectively, the distance between points P(x1, y1) and Q(x2, y2) can be Euclidean distances, for example, the two-dimensional Euclidean distance represented by equation (1):

[0041]

[0042] Regarding the corresponding points on the performance curve of the target vehicle and the corresponding points on the performance curve of the sample vehicle, in one embodiment, the corresponding points on the performance curve of the target vehicle and the corresponding points on the performance curve of the sample vehicle can be the points with the closest abscissas in the set of points used to fit the performance curves of the target vehicle and the corresponding performance curves of the sample vehicle. For example, as... Figure 3 As shown, assuming Figure 3 Curve 300 in the diagram represents the performance curve of the target vehicle for a certain metric, which is fitted by the black solid points on it; curve 310 represents the performance curve of the sample vehicle for the same metric, which is fitted by the black solid points on it. Therefore, for a point P5 on the performance curve 300 of the target vehicle, the corresponding point on the performance curve of the sample vehicle can be the point Q5 among the black solid points on curve 310 whose x-coordinate is closest to the x-coordinate of point P5.

[0043] In another embodiment, the points on the performance curve of the target vehicle and the corresponding points on the corresponding performance curve of the sample vehicle may have the same abscissa. In this case, the above equation (1) can be simplified to the following equation (2):

[0044] L=|y2-y1| (2)

[0045] Furthermore, in this embodiment, if for any point on the performance curve of the target vehicle, there is no corresponding point in the point set of the sample vehicle used to fit its performance curve, then a predetermined number of points in the point set whose x-coordinates are closest to the x-coordinates of that point are determined, and based on the predetermined number of points, interpolation processing (e.g., Lagrange interpolation) is performed to determine the corresponding point of that point.

[0046] For example, such as Figure 4 As shown, assuming Figure 4Curve 400 represents the performance curve of the target vehicle, which is fitted by the black solid points on it; curve 410 represents the performance curve of the sample vehicle, which is fitted by the black solid points on it. For point P5 on the performance curve 400 of the target vehicle, since there is no corresponding point in the set of points used to fit its performance curve (i.e., the black solid points on curve 410) of the sample vehicle whose x-coordinate is equal to that of point P5, a predetermined number of points, such as three points Q4, Q5 and Q6, need to be determined from the black solid points on curve 410 that are closest to the x-coordinate of point P5. Then, the point Q' with the same x-coordinate as point P5 is determined by Lagrange interpolation. After determining point Q', the point distance between point P5 and point Q' can be calculated by the above formula (2).

[0047] In one embodiment, determining a predetermined number of points in the point set whose x-coordinates are closest to the x-coordinates of any other point can be done by the following method: First, the ratio of the number of points in the point set of the performance curve of the sample vehicle to the number of points in the point set of the performance curve of the target vehicle is determined by equation (3);

[0048] ratio = Q N / P N (3)

[0049] Among them, Q N P represents the number of points in the point set representing the performance curves of the sample vehicles. N This represents the number of points in the point set of the target vehicle's performance curve.

[0050] Then, the ratio is compared with the sequential number of the point in its set of points (e.g., Figure 4 The sequential number of point P5 in the matrix is ​​multiplied by itself to obtain a base sequential number. Then, using this base sequential number as the center, a predetermined number of points are determined before and after it. For example, as shown in the figure... Figure 4 The situation is shown below.

[0051] In another embodiment, after determining the baseline sequence number as described above, a plurality of candidate points whose sequence number differs from the baseline sequence number by a predetermined threshold can be identified in the point set of the performance curve of the sample vehicle. Then, a predetermined number of points whose x-coordinates are closest to the x-coordinates of any of the candidate points are determined from the plurality of candidate points. For example, the plurality of candidate points can be determined by equation (4).

[0052] (n*ratio)±th (4)

[0053] Where (n*ratio)±th represents the range of the sequential numbers of the multiple candidate points in their point set (i.e., the point set of the performance curves of the sample vehicle), n represents the sequential number of any point in its point set (i.e., the point set of the performance curves of the target vehicle), ratio represents the ratio of the number of points in the point set of the performance curves of the sample vehicle to the number of points in the point set of the performance curves of the target vehicle, and th represents a predetermined threshold, which can be an integer greater than or equal to m / 2, where m represents the predetermined number. In one embodiment, th equals 2m. For example, in second-order Lagrange interpolation, the predetermined number is 3, in which case th can be 6. Of course, th can also be equal to other values, such as th = 5. Setting the predetermined threshold equal to the predetermined number multiplied by 2 can better ensure that the predetermined number of points determined in a search range that is not particularly large are the predetermined number of points whose x-coordinates on the entire performance curve are closest to the x-coordinates of any point, achieving a good trade-off between computational complexity and search accuracy.

[0054] For example, such as Figure 5 As shown, assuming Figure 5 Curve 500 represents the performance curve of the target vehicle, fitted by 6 solid black dots; curve 510 represents the performance curve of the sample vehicle, fitted by 24 solid black dots. Figure 5 As shown, for point P3 on the performance curve 500 of the target vehicle, there is no corresponding point Q' on the performance curve 510 of the sample vehicle whose abscissa is equal to that of point P3. In this case, assuming further that the corresponding point Q' is determined based on the three points on the performance curve 510 closest to the abscissa of point P3 by second-order Lagrange interpolation, the three points on the performance curve 510 closest to the abscissa of point P3 can be determined by the following method. First, the ratio of the number of points in the point set of the performance curve of the sample vehicle to the number of points in the point set of the performance curve of the target vehicle is determined to be 4 by the above equation (3). Then, the ratio 4 is multiplied by the sequence number 3 of point P3 in its point set to determine the baseline sequence number as 12. Then, multiple candidate points are determined by, for example, the above equation (4). Assuming that the predetermined threshold is 6 (i.e., the predetermined number multiplied by 2), the multiple candidate points can be determined as Figure 5 Points Q6, Q7, Q8, Q9, Q are shown. 10 Q 11 Q 12 Q 13 Q 14 Q 15 Q 16 Q 17 and Q 18 Finally, at candidate points Q6, Q7, Q8, Q9, and Q... 10 Q 11 Q12 Q 13 Q 14 Q 15 Q 16 Q 17 and Q 18 In the middle, determine the three points whose x-coordinates are closest to the x-coordinate of point P3, that is... Figure 5 Point Q in 12 Q 13 Q 14 .

[0055] As can be seen from the example above, point Q... 13 and Q 14 All are points Q with reference sequence number 12. 12 The point on the right, not point Q. 12 The points on both sides of the center. Therefore, compared to the method of determining a predetermined number of points before and after a point with a base numbered sequence as the center, the method of determining candidate points as described above, and then determining the predetermined number of points among the candidate points, can more accurately determine the predetermined number of points whose x-coordinates are closest to the x-coordinate of any given point.

[0056] Please note the above Figures 3 to 5 and the following Figures 7 to 9 These are merely schematic diagrams of performance curves intended to illustrate the concepts of this disclosure, and are not actual performance curves for any vehicle measurement, nor are they intended to limit this disclosure in any way.

[0057] Continue back to Figure 2 After determining the point distance, the method proceeds to step S124. In step S124, point similarity is determined based on the determined point distance, where the point similarity is negatively correlated with the point distance. In one implementation, point similarity can be determined directly based on the determined point distance; for example, the reciprocal of the point distance can be used as the point similarity.

[0058] In another embodiment, point similarity can be determined based on the determined point distances as follows: First, the ratio of the point distances to a predetermined value is calculated to determine the relative point distances, for example, by using equation (5):

[0059] Lr=L / V (5)

[0060] Where Lr represents the relative distance between points, L represents the determined distance between points, and V represents a predetermined value. For example, V = |y1|, which is the absolute value of the ordinate of a point on the performance curve of the target vehicle.

[0061] After determining the relative point distances, they can be normalized, and the normalized values ​​can be used as the point similarity. For example, the relative point distances can be normalized using equation (6):

[0062] S=1 / (Lr+1) (6)

[0063] Where S represents point similarity and Lr represents relative point distance.

[0064] It should be understood that the method for normalizing relative point distances shown in equation (6) above is merely an example and not a limitation of this disclosure. Those skilled in the art can choose a suitable normalization method as needed.

[0065] The method described above, which determines the relative point distance based on the point distance, then normalizes the relative point distance and uses the normalized value as the point similarity, can reduce or eliminate the influence caused by the different units of the horizontal and vertical coordinates of different curves. Furthermore, since the determined similarity is normalized and its range is between (0, 1), it can provide more intuitive results.

[0066] return Figure 2 After determining the point similarity between each point on the performance curve of the target vehicle and each corresponding point on the corresponding performance curve of the sample vehicle, the method proceeds to step S126. In step S126, the curve similarity is determined based on the determined point similarity between each point on the performance curve of the target vehicle and each corresponding point on the corresponding performance curve of the sample vehicle.

[0067] Figure 6 This further demonstrates Figure 2 Curve similarity is determined based on point similarity (i.e., Figure 2 The example flowchart for step 126) starts from step S126_2. In step S126_2, the performance curve of the target vehicle is segmented.

[0068] In one embodiment, segmentation can be based on the range of the horizontal axis of interest or focus. For example, as... Figure 7 Assume that for the performance curve 700 of the target vehicle, the range of the x-axis of interest is... B To X c Between, then it can be like Figure 7 As shown, the curve 700 is divided into segments: P A To P B P B To P C and P C To P D .

[0069] In another embodiment, segmentation can be performed by extracting curve features that characterize the morphological changes of the performance curve, such as slope, extreme points, stationary points, inflection points, or a specific point on the curve. Then, based on the extracted curve features, the performance curve is segmented. For example, as... Figure 8 As shown, for the performance curve 800 of the target vehicle, its stationary point M can be extracted, and then the curve 800 can be segmented into: P based on the stationary point M. A To P B and P B To P C .

[0070] The method then proceeds to step S126_4. In step S126_4, weights for each segment are determined, and the weight of each point on the performance curve is set as the weight of its respective segment. Correspondingly, the weight of the similarity between points related to each point on the performance curve is the weight of that point.

[0071] For example, such as Figure 7 As shown, the performance curve 700 can be segmented as follows: P A To P B P B To P C and P C To P D The weights of these three segments are set to w0, w1, and w2 respectively. Then, the weight of point P5 on the performance curve 700 can be set to the weight of its corresponding segment P. B To P C The weight of each segment is w1, and the weight of the similarity between points P5 and Q5 is also w1. For example, the weight of each segment can be determined based on the importance of each segment (i.e., the degree of interest or attention to each segment).

[0072] The method then proceeds to step S126_6. In step S126_6, the curve similarity is determined by weighted summation of the determined similarities of each point based on the weights.

[0073] Combination Figure 6 and Figure 7 The method for determining curve similarity based on point similarity describes that by segmenting the performance curve of the target vehicle and assigning corresponding weights to each segment, and then performing a weighted summation of the determined point similarities based on the weights of each segment to determine the curve similarity, the method can fully consider the influence of different segments of the performance curve on the curve similarity, so that the determined curve similarity better reflects the similarity of the vehicle.

[0074] As described above Figure 2The method for determining curve similarity of performance curves according to embodiments of the present disclosure, after determining the point distances between points on two curves, determines point similarity based on the determined point distances, and then determines curve similarity based on the point similarity. This method fully considers the characteristics of determining vehicle similarity; that is, in determining vehicle similarity, some metrics used to measure the similarity between the target vehicle and the sample vehicle can be represented by a single point, in addition to performance curves. Therefore, in determining curve similarity, the method according to the present disclosure can standardize the process of determining metric similarity, facilitating programming for engineers and reducing the processor load when running the method for determining vehicle similarity.

[0075] Furthermore, in practice, there may be situations where the horizontal axis range of the performance curve of the sample vehicle differs from that of the corresponding performance curve of the target vehicle. For example, ... Figure 9 This is the situation shown. In this case, before determining the curve similarity between the individual performance curves of the sample vehicle and the corresponding performance curves of the target vehicle, the curves can be segmented so that the segmented curves have the same range of horizontal coordinates. For example, as shown... Figure 9 As described above, curves 900 and 910 can be cut so that the cut curves have the same x-coordinate range. A To X B Then, calculate the x-coordinate range (X) of the sample vehicle and the target vehicle, which have the same x-coordinate range. A To X B Curve similarity.

[0076] Figure 10 This further demonstrates Figure 1The following is an example flowchart for determining the similarity between a sample vehicle and a target vehicle across various metrics based on curve similarity. It begins at step S132. At step S132, for any metric, curve features representing the morphological changes of the performance curves of the sample vehicle and the target vehicle representing that metric are extracted, such as slope, extreme points, stationary points, inflection points, or a specific point on the curve. The method then proceeds to step S134. At step S134, based on the extracted curve features of the sample vehicle and the target vehicle, the curve feature similarity between the respective curve features of the sample vehicle and the corresponding curve features of the target vehicle is determined. Curve features can be represented in various forms, and thus, curve feature similarity can be determined using appropriate methods based on the form of the curve features. For example, curve features can be represented by curves or points (e.g., slope can be represented by a curve, and extreme points can be represented by points). In this case, curve feature similarity can be determined, for example, by the methods for determining point similarity or curve similarity described above. Of course, other methods for determining point similarity or curve similarity are also feasible. After determining the curve feature similarity, the method proceeds to step S136. In step S136, the similarity between the sample vehicle and the target vehicle in the metric is determined based on the individual curve feature similarities and the curve similarity of the performance curve representing the metric. In one embodiment, the minimum or maximum value among the individual curve feature similarities and the curve similarity of the performance curve representing the metric can be determined as the similarity between the sample vehicle and the target vehicle in the metric. In another embodiment, the root mean square error of the individual curve feature similarities and the curve similarity of the performance curve representing the metric can be calculated, thereby determining the similarity between the sample vehicle and the target vehicle in the metric. In yet another embodiment, a weighted sum of the individual curve feature similarities and the curve similarity of the performance curve representing the metric can be used to determine the similarity between the sample vehicle and the target vehicle in the metric.

[0077] Combination Figure 10 The method described above for determining the similarity between sample vehicles and target vehicles on various metrics takes into account the curve characteristics of the performance curve, thus the determined metric similarity can better reflect the similarity of the vehicles.

[0078] In this disclosure, the similarity between the sample vehicle and the target vehicle is determined based on the similarity of the sample vehicle and the target vehicle across various metrics (i.e., Figure 1 Step S140 may include: determining the weight of each metric based on its importance using the Analytic Hierarchy Process (AHP). Then, based on the weights, a weighted summation of the similarity between the sample vehicle and the target vehicle on each metric is performed to determine the similarity between the sample vehicle and the target vehicle.

[0079] For example, it can be based on at least one metric used to measure the similarity between the target vehicle and the sample vehicles (e.g., Figure 12 The correlation between the following metrics (toe angle as a function of wheel bounce, camber angle as a function of wheel bounce, longitudinal displacement of wheel center as a function of wheel bounce, suspension stiffness obtained at the wheel center, and roll stiffness obtained at the tire contact point) is used to stratify at least one of the metrics. For example, as... Figure 12 The vehicle evaluation system is shown in layers. Each layer is then processed as follows to determine the weights of each parameter in that layer (e.g., ...). Figure 12 The system shown illustrates the weights of the third-layer metrics (parameters) for toe-in angle, camber angle, and wheel center longitudinal displacement as a function of wheel hop. It should be understood that the weights of single-layer parameters can also be determined using the method described below (e.g., in the case of only one layer).

[0080] The judgment matrix shown in Table 1 is determined based on the relative importance of each pair of parameters in each layer.

[0081] <![CDATA[R1]]> <![CDATA[R2]]> … <![CDATA[R n ]]> <![CDATA[R1]]> <![CDATA[b 11 ]]> <![CDATA[b 12 ]]> … <![CDATA[b 1n ]]> <![CDATA[R2]]> <![CDATA[b 21 ]]> <![CDATA[b 22 ]]> … <![CDATA[b 2n ]]> … … … … … <![CDATA[R n ]]> <![CDATA[b n1 ]]> <![CDATA[b n2 ]]> … <![CDATA[b nn ]]>

[0082] Table 1 Judgment Matrix

[0083] In Table 1, R i b represents the i-th parameter (metric). ij R represents i Relative to R j The relative importance of , where i = 0..n; j = 0..n, n represents the number of parameters in the judgment matrix, and b ij With b ji They are reciprocals of each other. The relative importance ranges from 1 to 9, and the corresponding values ​​from 1 / 9 to 1 (inclusive), where 1 indicates equal importance and 9 indicates extreme importance. The above judgment matrix can be set by engineers based on experience.

[0084] After determining the judgment matrix as shown in Table 1, the weights of each parameter can be determined using equation (6):

[0085]

[0086] Where, ω i Indicates parameter R i The weight, b ij R represents i Relative to R j The relative importance, where n represents the number of parameters in the judgment matrix.

[0087] After determining the weights of each parameter in each layer as described above, the final weight of each metric can be determined by multiplying the weights of each layer associated with a particular metric. For example, for Figure 12 The metric shown—the toe angle—varies with wheel hop, assuming it is in Figure 12 The weight of the third layer is ω1, and its weight in the second layer (i.e., the weight of parallel wheel jump motion) is ω2. Therefore, the final weight of the metric—the toe angle—as it changes with the wheel jump is ω1*ω2.

[0088] Furthermore, when determining the weights of each metric using AHP, the process can also include: performing a consistency check on the relative importance of each metric at each layer based on the determined weights of each metric at each layer, to determine whether there is a logical conflict between the relative importances of each metric, and adjusting the relative importance of each metric if a logical conflict is found. For example, the consistency check can be performed using the following method. First, the largest eigenvalue of the judgment matrix is ​​determined using equation (7):

[0089]

[0090] Where, λ max ω represents the largest eigenvalue. i Indicates parameter R i The weight, b ij R represents i Relative to R j The relative importance, where n represents the number of parameters in the judgment matrix.

[0091] After determining the largest eigenvalue λ max Then, the consistency index is determined using equation (8):

[0092]

[0093] Where CI represents a consistent index, λ max Let represent the largest eigenvalue parameter, and n represent the number of parameters in the judgment matrix. Then, consistency is determined using equation (9):

[0094] CR = CI / RI (9)

[0095] Here, CR represents consistency, CI represents consistency index, and RI represents random consistency index, which can be obtained, for example, from Table 2.

[0096] n 2 3 4 5 … RI 0 0.58 0.90 1.12 …

[0097] Table 2 Examples of Random Consistency Indices

[0098] After consistency is determined by equation (9), it can be compared with a predetermined value (e.g., 0.1). If the determined consistency is less than the predetermined value, the relative importance of each metric at that level is consistent (i.e., there is no logical conflict between the relative importance of each metric); otherwise, the relative importance of each metric at that level is inconsistent. In the case of inconsistent relative importance of each metric at that level, the relative importance of each metric at that level can be adjusted until the adjusted relative importance is consistent.

[0099] The AHP method has been described above; it should be understood that this method is merely an example for determining weights and not a limitation of this disclosure. Other methods, such as neural network methods, can also be used to determine the aforementioned weights.

[0100] After determining the similarity between the target vehicle and the sample vehicles using the method described above, the results can be ranked and displayed, for example... Figure 11 As shown.

[0101] In the foregoing, this disclosure combines Figures 1 to 11 A method for determining vehicle similarity according to embodiments of the present disclosure is described. Hereinafter, the present disclosure will be combined with... Figures 12 to 15 This disclosure describes a vehicle evaluation system and apparatus, non-transitory computer-readable storage medium, device, and computer program product for determining vehicle similarity according to embodiments of the present disclosure.

[0102] Figure 12 An example vehicle evaluation system according to an embodiment of the present disclosure is shown, which summarizes various metrics and performance data of the vehicle in a hierarchical manner, such as the performance curves described above. Furthermore, in Figure 12 In the vehicle evaluation system described above, the various metrics are arranged according to their importance (e.g., the order of their impact on vehicle similarity). Thus, as... Figure 12 The vehicle evaluation system described above makes it easier for engineers to select metrics for determining vehicle similarity during vehicle design, improving the accuracy and efficiency of metric selection.

[0103] In a further embodiment, the vehicle evaluation system may also include curve characteristics of the performance curves for each metric, such as... Figure 13 As shown. The curve characteristics can be determined, for example, by the following method: Performance testing or simulation is performed on the target vehicle and sample vehicles to obtain a set of points for fitting the performance curve. If the set of points is non-uniform, for example, the differences in the x-coordinates of the points in the set are not identical, the set of points can be homogenized using interpolation. Then, the curve characteristics of the performance curve are determined based on the homogenized set of points.

[0104] and Figure 12Compared to the aforementioned vehicle evaluation system, Figure 13 The vehicle evaluation system shown here facilitates the extraction of curve features by including the curve features of the performance curves, and also helps engineers determine which curve features to focus on.

[0105] Figure 14 This is an example block diagram of an apparatus for determining vehicle similarity according to embodiments of the present disclosure. Figure 14 As shown, the apparatus 1400 for determining vehicle similarity according to an embodiment of the present disclosure is manifested in the form of a general-purpose computing device, which may include, but is not limited to, one or more processors or processing units 1411, a memory 1412, and a bus 1413 connecting different system components (including processor 1411 and memory 1412).

[0106] Bus 1413 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0107] The apparatus 1400 for determining vehicle similarity according to embodiments of the present disclosure typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the apparatus 1400 for determining vehicle similarity according to embodiments of the present disclosure, including volatile and non-volatile media, removable and non-removable media.

[0108] Memory 1412 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 1412_1 and / or cache memory 1412_2. The apparatus 1400 for determining vehicle similarity according to embodiments of this disclosure may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 1412_3 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 14 (Not shown, usually referred to as a "hard drive"). Although Figure 14Not shown, but a disk drive for reading and writing to removable non-volatile disks (e.g., "floppy disks") and an optical disk drive for reading and writing to removable non-volatile optical discs (e.g., CD-ROMs, DVD-ROMs, or other optical media) may be provided. In these cases, each drive may be connected to bus 1413 via one or more data media interfaces. Memory 1412 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0109] A program / utility 1412_4 having at least one set of program modules 1412_4_1 can be stored, for example, in memory 1412. Such program modules 1412_4_1 include—but are not limited to—an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 1412_4_1 typically perform the functions and / or methods described in the embodiments of the present invention.

[0110] The device 1400 for determining vehicle similarity according to embodiments of the present disclosure can also communicate with one or more external devices 1430 (e.g., keyboard, pointing device, display 1420, etc.), one or more devices that enable a user to interact with the device 1400 for determining vehicle similarity according to embodiments of the present disclosure, and / or any device that enables the device 1400 for determining vehicle similarity according to embodiments of the present disclosure to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication can be performed via input / output (I / O) interface 1414. Furthermore, the device 1400 for determining vehicle similarity according to embodiments of the present disclosure can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 1415. As shown, network adapter 1415 communicates with other modules of the device 1400 for determining vehicle similarity according to embodiments of the present disclosure via bus 1413. It should be understood that, although not shown in the figures, other hardware and / or software modules may be used in conjunction with the device 1400 for determining vehicle similarity according to embodiments of the present disclosure, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0111] It should be understood that the device 1400 for determining vehicle similarity shown in FIG1400 is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0112] Furthermore, this disclosure also provides a non-transitory computer-readable storage medium for storing instructions that cause a processor to perform the method for determining vehicle similarity described above according to embodiments of this disclosure.

[0113] Figure 15 This is an example block diagram of a device for determining vehicle similarity according to embodiments of the present disclosure. Figure 15 As shown, the device 1500 for determining vehicle similarity according to embodiments of the present disclosure includes, but is not limited to: a vehicle determination module 1510 for determining a target vehicle and a sample vehicle; a metric determination module 1520 for determining at least one metric for measuring the similarity between the target vehicle and the sample vehicle, the metric being represented by performance curves obtained by performance testing or simulation of the target vehicle and the sample vehicle; a metric similarity determination module 1530, which further includes: a curve similarity determination module 1532 for determining the curve similarity between each performance curve of the sample vehicle and each corresponding performance curve of the target vehicle, the metric similarity determination module 1530 being used to determine the similarity between the sample vehicle and the target vehicle in each metric based on the curve similarity; and a vehicle similarity determination module 1540 for determining the similarity between the sample vehicle and the target vehicle in each metric based on the similarity between the sample vehicle and the target vehicle.

[0114] Optionally, the device 1500 for determining vehicle similarity described above may further include a raw data extraction module (not shown) and a weight management module (not shown). The raw data extraction module can be used to extract raw data, for example from... Figure 12 The raw data can be extracted from the vehicle evaluation system shown in Figure 13. The raw data extraction module may further include a curve feature extraction module for processing the raw curve data and finding curve features of interest. The weight management module may further include a weight determination module, which can determine the weights of each metric and other weights, for example, through the AHP described above.

[0115] Optionally, the similarity determination module 1530 may further include a point similarity determination module (not shown) and a missing data processing module (not shown). The point similarity determination module can determine the point similarity between a single performance point of the sample vehicle and a corresponding single performance point of the target vehicle. The missing data processing module can process the missing data, for example, by interpolating the missing measurement data.

[0116] Furthermore, this disclosure also provides a computer program product for a method of determining vehicle similarity, comprising instructions that cause a processor to perform the method for determining vehicle similarity described above according to embodiments of this disclosure.

[0117] Thus far, this disclosure has described, in conjunction with the accompanying drawings, methods, apparatus, computer-readable storage media, devices, computer program products, and vehicle evaluation systems for determining vehicle similarity.

[0118] It should be noted that the above description is merely an embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0119] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0120] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for determining vehicle similarity, the method comprising: Identify the target vehicle and the sample vehicle; Determine at least one metric for measuring the similarity between a target vehicle and a sample vehicle, the metric being represented by performance curves obtained from performance testing or simulation of the target vehicle and the sample vehicle; Determine the similarity between the performance curves of the sample vehicle and the corresponding performance curves of the target vehicle. The similarity between the sample vehicle and the target vehicle in each of the aforementioned metrics is determined based on the curve similarity. as well as The similarity between the sample vehicle and the target vehicle is determined based on their similarity across the various metrics.

2. The method according to claim 1, wherein, Determining the curve similarity includes: Determine the point distance between a point on the performance curve of the target vehicle and the corresponding point on the corresponding performance curve of the sample vehicle; Based on the determined point distances, point similarity is determined, and the point similarity is negatively correlated with the point distances; and The curve similarity is determined based on the point similarity between each point on the performance curve of the target vehicle and each corresponding point on the corresponding performance curve of the sample vehicle.

3. The method according to claim 2, wherein, Determining point similarity based on the determined point distances includes: Calculate the ratio of the point distance to a predetermined value to determine the relative point distance; The relative point distances are normalized, and the normalized values ​​are used as the point similarity.

4. The method according to claim 2, wherein, Determining the curve similarity based on the point similarity between each point on the performance curve of the target vehicle and each corresponding point on the corresponding performance curve of the sample vehicle includes: Segment the performance curve of the target vehicle; Determine the weights to be used for each segment; Set the weight of each point on the performance curve as the weight of its corresponding segment, and Based on the weights, the similarity of each determined point is summed using a weighted average to determine the curve similarity.

5. The method according to claim 4, wherein, Segmenting the performance curve of the target vehicle includes: Extract curve features that characterize the morphological changes of the performance curve; Based on the curve characteristics, the performance curve is segmented.

6. The method according to claim 1, wherein, Determining the similarity between the sample vehicle and the target vehicle on each of the aforementioned metrics based on the curve similarity further includes: For any metric: Extract curve features representing the morphological changes of the performance curves of the sample vehicle and the target vehicle that represent the metric. Based on the extracted curve features of the sample vehicle and the target vehicle, the similarity of the curve features of the sample vehicle with the corresponding curve features of the target vehicle is determined, and... Based on the similarity of the features of each curve and the similarity of the curves representing the performance curves of the metric, the similarity between the sample vehicle and the target vehicle in the metric is determined.

7. The method according to claim 1, wherein, Determining the similarity between the sample vehicle and the target vehicle based on their similarity across various metrics includes: Based on the importance of each metric, the weight of each metric is determined using the Analytic Hierarchy Process (AHP). Based on the weights, the similarity between the sample vehicle and the target vehicle on each of the aforementioned metrics is weighted and summed to determine the similarity between the sample vehicle and the target vehicle.

8. The method of claim 7, further comprising determining the weights of each metric based on its importance using the Analytic Hierarchy Process (AHP): Based on the determined weights of each metric, a consistency check is performed on the importance of each metric to determine whether there is a logical conflict between the importance of each metric, and if a logical conflict is determined, the importance of each metric is adjusted.

9. The method according to claim 1, wherein, Determining the curve similarity includes: When the performance curve of the sample vehicle and the corresponding performance curve of the target vehicle have different horizontal coordinate ranges, the curves are cut so that the cut curves have the same horizontal coordinate range. Calculate the curve similarity between the sample vehicle and the target vehicle for curves with the same abscissa range.

10. The method according to claim 2, wherein, Points on the performance curve of the target vehicle have the same x-coordinate as corresponding points on the performance curve of the sample vehicle. If, for any point on the performance curve of the target vehicle, there is no corresponding point in the set of points used to fit the performance curve of the sample vehicle, then the point distance between the point on the performance curve of the target vehicle and the corresponding point on the corresponding performance curve of the sample vehicle includes: In the set of points, determine a predetermined number of points whose x-coordinates are closest to the x-coordinates of any given point, and Based on the predetermined number of points, interpolation is performed to determine the corresponding point for any given point.

11. The method according to claim 10, wherein, Determining a predetermined number of points in the point set whose x-coordinates are closest to the x-coordinates of any given point further includes: Determine the ratio of the number of points in the performance curve set of the sample vehicle to the number of points in the performance curve set of the target vehicle. The ratio is multiplied by the sequential number of the point in its set of points to obtain a reference sequential number. Multiple candidate points whose sequential numbers differ from the reference sequential number by a predetermined threshold are determined in the set of points of the performance curve of the sample vehicle. From the plurality of candidate points, determine a predetermined number of points whose x-coordinates are closest to the x-coordinates of any one of the candidate points.

12. The method according to claim 1, wherein, The metrics include at least one of the vehicle's motion and flexibility (K&C) metrics.

13. An apparatus for determining vehicle similarity, comprising: Memory, on which computer instructions are stored; and processor, The instructions, when executed by the processor, cause the processor to perform the method according to any one of claims 1-12.

14. A non-transitory computer-readable storage medium storing instructions that cause a processor to perform the method according to any one of claims 1-12.

Images