CN117984773A

CN117984773A - Acceleration intention recognition method, device, electronic equipment and readable storage medium

Info

Publication number: CN117984773A
Application number: CN202211329175.8A
Authority: CN
Inventors: 蔡炜; 孙贤安
Original assignee: SAIC Motor Corp Ltd
Current assignee: SAIC Motor Corp Ltd
Priority date: 2022-10-27
Filing date: 2022-10-27
Publication date: 2024-05-07

Abstract

The application provides a method, a device, electronic equipment and a readable storage medium for identifying acceleration intention, which are used for acquiring accelerator signals of a vehicle according to a stipulated period, calculating an accelerator change rate based on the accelerator signals, respectively determining a first membership set of at least two accelerator categories and a second membership set of at least two accelerator change rate categories based on a preset accelerator membership curve and a preset accelerator change rate membership curve, determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set, determining a target value based on the third membership set and a preset acceleration intention membership curve, and determining a category corresponding to the acceleration grade range of the target value as the acceleration intention category of the current period based on the acceleration grade range corresponding to the at least two acceleration intention categories.

Description

Acceleration intention recognition method, device, electronic equipment and readable storage medium

Technical Field

The present application relates to the field of automobile control, and more particularly, to a method and apparatus for identifying an acceleration intention, an electronic device, and a readable storage medium.

Background

Accurate recognition of the driver's acceleration intention is very important for control of an automatic transmission. Under different acceleration intentions, through the correction to derailleur gear shifting diagram and engine output torque for whole car possesses better fuel economy when the driver accelerates slowly, possesses better dynamic nature under the condition of accelerating suddenly, thereby promotes whole car driving impression, improves the market competition of product.

Generally, the acceleration intention under the corresponding working condition is obtained through classified table look-up based on signals such as throttle, throttle change and the like.

First, a classification threshold value of the accelerator and a classification threshold value of the accelerator change rate are set, wherein the accelerator is classified into one of the following 5 types: the accelerator change rate is divided into one of the following 5 types: positive large (PB), positive Small (PS), none (Z), negative Small (NS), negative large (NB); then, according to a two-dimensional table look-up of a preset accelerator-accelerator change rate-accelerating intention, acquiring accelerating intention classification: rapid acceleration (Fast On), slow acceleration (Pedal On), etc.

However, in the aspect of recognition classification, only two types of rapid acceleration and slow acceleration are output, and the judgment on the acceleration intention is rough; and the algorithm is simpler, the robustness is insufficient, and for the condition of complex working conditions, the recognition accuracy is insufficient, so that the real intention of a driver can not be accurately obtained.

Disclosure of Invention

In view of this, the present application provides a method for identifying an acceleration intention, as follows:

A method of accelerating intent recognition, comprising:

acquiring an accelerator signal based on a contracted period;

calculating a throttle change rate based on the throttle signal;

determining a first membership set of at least two throttle categories to which the throttle signal belongs based on a preset throttle membership curve, wherein the throttle ranges of any two adjacent throttle categories in the preset throttle membership curve are partially overlapped;

Determining a second membership set of at least two throttle change rate categories based on a preset throttle change rate membership curve, wherein the throttle change rate ranges of any two adjacent throttle change rate categories in the preset throttle change rate membership curve are partially overlapped;

Determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set;

determining a target value based on the third membership set and a preset accelerating intention membership curve, wherein the target value represents the current accelerating grade;

And determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

An acceleration intention recognition device, comprising:

The acquisition module is used for acquiring an accelerator signal based on a contracted period;

the calculation module is used for calculating the throttle change rate based on the throttle signal;

The first class determining module is used for determining a first membership set of at least two accelerator classes of the accelerator signals based on a preset accelerator membership curve, and accelerator ranges of any two adjacent accelerator classes in the preset accelerator membership curve are partially overlapped;

the second class determining module is used for determining a second membership set of at least two throttle change rate classes based on a preset throttle change rate membership curve, wherein the throttle change rate ranges of any two adjacent throttle change rate classes in the preset throttle change rate membership curve are partially overlapped;

The set determining module is used for determining a third membership set corresponding to at least two accelerating intention categories based on the first membership set and the second membership set;

The target value determining module is used for determining a target value based on the third membership set and a preset accelerating intention membership curve, and the target value represents the current accelerating level;

And the third category determining module is used for determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

An electronic device, wherein the electronic device comprises: a memory, a processor;

Wherein the memory stores a processing program;

The processor is configured to load and execute the processing program stored in the memory, so as to implement the steps of the acceleration intention recognition method according to any one of the above.

A readable storage medium having stored thereon a computer program which is invoked and executed by a processor to implement the steps of the acceleration intention recognition method of any one of the above.

According to the technical scheme, the application provides an acceleration intention recognition method, accelerator signals of a vehicle are acquired according to a stipulated period, accelerator change rates are calculated based on the accelerator signals, first membership sets of at least two accelerator categories and second membership sets of at least two accelerator change rate categories are respectively determined based on a preset accelerator membership curve and a preset accelerator change rate membership curve, third membership sets corresponding to at least two acceleration intention categories are determined based on the first membership sets and the second membership sets, target values are determined based on the third membership sets and the preset acceleration intention membership curve, and categories corresponding to the acceleration grade ranges are determined as acceleration intention categories of the current period based on acceleration grade ranges corresponding to the at least two acceleration intention categories. In the scheme, the first membership degree and the second membership degree of the accelerator change rate of at least two accelerator categories in the current running process of the vehicle are respectively determined, the accelerator ranges of any two adjacent accelerator categories in the preset accelerator membership degree curve are partially overlapped, the accelerator change rate ranges of any two adjacent accelerator change rate categories in the preset accelerator change rate membership degree curve are partially overlapped, the process of determining the accelerator membership degree and the accelerator change rate membership degree is fuzzification processing of accelerator signals and accelerator change rate, the third membership degree and the preset accelerator intention membership degree curve are used for performing defuzzification processing after the acceleration intention fuzzification processing, a target value representing the current acceleration grade is obtained, the category of the acceleration intention can be accurately determined based on the target value, the acceleration intention of a driver is accurately identified, and the recognition process has stronger robustness due to the adoption of fuzzy recognition.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings may be obtained according to the provided drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of an embodiment 1 of a method for identifying an acceleration intention;

FIG. 2 is a schematic diagram of a threshold membership curve in example 1 of a method for identifying acceleration intention according to the present application;

FIG. 3 is a schematic diagram of a membership curve of the throttle change rate in embodiment 1 of a method for identifying acceleration intention according to the present application;

Fig. 4 is a schematic diagram of acceleration intention in embodiment 1 of a method for identifying acceleration intention provided by the present application;

FIG. 5 is a flowchart of an embodiment 2 of a method for identifying an acceleration intention;

FIG. 6 is a flowchart of an embodiment 3 of a method for identifying an acceleration intention;

FIG. 7 is a flowchart of an embodiment 4 of a method for identifying an acceleration intention;

FIG. 8 is a flowchart of an embodiment 5 of a method for identifying an acceleration intention;

FIG. 9 is a schematic diagram of a membership curve of the acceleration intention in embodiment 5 of a method for identifying acceleration intention;

FIG. 10 is a flowchart of an embodiment 6 of a method for identifying an acceleration intention;

FIG. 11 is a schematic diagram of a software architecture of a processor employing a method for identifying acceleration intent provided in the present application;

FIG. 12 is a schematic diagram of real vehicle verification data;

fig. 13 is a schematic structural diagram of an embodiment of an accelerating intention recognition device provided by the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

As shown in fig. 1, a flowchart of an embodiment 1 of a method for identifying an acceleration intention is provided, and the method is applied to a vehicle-mounted controller, and includes the following steps:

step S101: acquiring an accelerator signal based on a contracted period;

the throttle signal of the vehicle is acquired according to a contracted period, wherein the contracted period can be 10ms (millisecond), 20ms, 500ms and the like, the specific duration of the contracted period can be set according to actual conditions, and the specific duration of the period is not limited in the application.

Specifically, the accelerator signal is an accelerator opening of the vehicle, and the accelerator opening is generally expressed by a percentage, such as 35%, 70%, or the like.

In a specific implementation, the throttle signal is obtained from a controller CAN (Controller Area Network ) bus, and the throttle signal is an original signal and is not subjected to any processing.

In particular, the vehicle-mounted controller to which the method of the present application is applied may be used in conventional automobiles and hybrid automobiles equipped with various types of automatic transmissions, which may employ CVT (Continuously Variable Transmission, mechanical continuously variable automatic transmission)/AT (Automatic Transmission, hydrodynamic automatic transmission)/DCT (Dual Clutch Transmission, dual clutch automatic transmission)/AMT (Automated Mechanical Transmission, electrically controlled mechanical automatic transmission), or the like.

The vehicle-mounted controller in the present application is applicable to all vehicles equipped with an automatic transmission, including conventional internal combustion engine vehicles, hybrid vehicles, and the like.

Step S102: calculating a throttle change rate based on the throttle signal;

the accelerator change rate is the accelerator change rate in the present cycle.

Specifically, the accelerator change rate in the present period is calculated based on the accelerator signal, and the accelerator change rate in the present period is calculated by calculating the slope based on the accelerator signal acquired in the previous period and the accelerator signal acquired in the present period.

Wherein, the calculation formula is as follows:

Throttle change rate in this cycle= (throttle signal in this cycle-throttle signal in the upper cycle)/cycle duration.

Step S103: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

The throttle membership curve is preset in the controller, wherein the abscissa is the throttle opening, and the ordinate is the membership, and each throttle category corresponds to one throttle range.

As one example, when the throttle categories are 4, the throttle categories include: the values of the accelerator ranges corresponding to the 4 accelerator categories are sequentially increased without accelerator, small accelerator, medium accelerator and large accelerator.

In the specific implementation, the number of the throttle categories can be set according to actual conditions, and the number of the throttle categories is not limited in the application.

Wherein, a fuzzy classification mode is adopted in the throttle membership curve.

Specifically, in the preset throttle membership curve, the throttle ranges of any two adjacent throttle categories are partially overlapped.

And the throttle categories adopt triangular membership functions, and finally a membership function curve of each set throttle category is obtained.

As shown in fig. 2, there are 4 throttle categories in the schematic diagram, which are no throttle, small throttle, medium throttle and large throttle. In the throttle membership curve, the x-axis represents the throttle, the y-axis represents the membership, and the membership is 0-1. Wherein, the accelerator range without accelerator is 0-Kz, the accelerator range with small accelerator is Ks 1-Ks 2, the middle value of the value range with small accelerator is Ksc, the accelerator range with middle accelerator is Km 1-Km 2, the middle value of the value range with middle accelerator is Kmc, and the accelerator range with large accelerator is Kb-100%. Wherein Ks1 is more than 0 and less than Ksc, km1 is more than Kmc, kb is more than Km2 and less than 100%.

And searching a corresponding y value in the throttle membership curve by taking the throttle signal as an x value, wherein the y value corresponding to the x value is the membership of the throttle signal, and determining the membership of the throttle signal in the plurality of throttle categories.

As an example, the range of oil free throttle is 0-10%, the range of small throttle is 0-30%, the range of medium throttle is 25-60%, and the range of large throttle is 50-100%.

For example, the throttle opening is 55%, the value range of the middle throttle is 25% -60%, the value range of the large throttle is 50% -100%, in the throttle membership curve, the y-axis value corresponding to the throttle opening of 55% is searched, the membership corresponding to the middle throttle is 0.5, the membership corresponding to the large throttle is 0.2, the membership corresponding to the oil-free throttle and the small throttle is 0, and the obtained first membership set is {0,0,0.5,0.2}.

For example, the accelerator opening is 80%, the value range of the middle throttle is 25% -60%, the value range of the large throttle is 50% -100%, in the accelerator membership curve, the y-axis value corresponding to 55% of the accelerator opening is searched, the membership corresponding to the large throttle is 0.8, the membership corresponding to the no-accelerator, small-accelerator and middle throttle is 0, and the obtained first membership set is {0,0,0,0.8}.

Step S104: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

The throttle change rate membership curve is preset in the controller, wherein the abscissa is the throttle change rate, and the ordinate is the membership, and each throttle change rate category corresponds to one throttle change rate range.

As one example, when the accelerator change rate category is 5, the accelerator change rate category includes: the values of the accelerator change rate ranges corresponding to the 5 accelerator change rate categories are sequentially increased.

In the specific implementation, the number of the throttle change rate categories can be set according to actual conditions, and the number of the throttle change rate categories is not limited in the application.

Wherein, a fuzzy classification mode is adopted in the throttle change rate membership curve.

Specifically, in the preset throttle change rate membership curve, the throttle change rate ranges of any two adjacent throttle change rate categories are partially overlapped.

Wherein, the unit of the throttle change rate is Pct/ms.

It should be noted that, the actual vehicle test finds that the variation range of the throttle in one period of 10ms is far from reaching the theoretical value. Through the analysis of the real vehicle test results of a plurality of drivers, the throttle change rate of 1.5Pct/ms can be rarely achieved, so that the membership of more than 1.5Pct/ms is considered to be set to 1, and the membership functions of the reverse large throttle change rate and the forward large throttle change rate are trapezoidal functions. And adopting triangular membership functions for other 3 classes in the middle, and finally setting membership function curves of all throttle change rate classes.

As shown in fig. 3, the throttle change rate membership curve is a schematic diagram, in which the throttle change rate categories are 5, and are respectively a large reverse throttle change rate, a small reverse throttle change rate, an oil-free throttle change rate, a small forward throttle change rate and a large forward throttle change rate. In the throttle change rate membership curve, the x-axis represents the throttle change rate, the y-axis represents the membership, and the membership is 0-1. The membership curves corresponding to the reverse large throttle change rate and the forward large throttle change rate are trapezoids, and the membership curves corresponding to the reverse small throttle change rate, the no-throttle change rate and the forward small throttle change rate are triangles.

Wherein, the throttle change rate range corresponding to the reverse large throttle change rate is-10-Knb, and the membership degrees corresponding to-10-Knb 0 are all 1; the range of the throttle change rate corresponding to the reverse small throttle change rate is Kns < 1 > -Kns < 2 >, and the middle value of the value range of the reverse small throttle change rate is Knsc; the range of the throttle change rate corresponding to the small throttle change rate is Kz 1-Kz 2, and the middle value of the value range of the small throttle change rate is Kzc; the range of the throttle change rate corresponding to the forward small throttle change rate is Kps 1-Kps 2, and the intermediate value of the value range of the forward small throttle change rate is Kpsc; the range of the accelerator change rate corresponding to the forward large accelerator change rate is Kpb-10, wherein the membership degrees corresponding to Kpb0-10 are all 1. Wherein Knb0 < Kns1 < Knb < Knsc < Kz1 < Kns < Kzc < Kps1 < Kz2 < Kpsc < Kbp < Kps2 < Kpb0.

And searching a corresponding y value in the throttle change rate membership curve by taking the throttle change rate as an x value, wherein the corresponding y value is the membership of the throttle change rate, and determining the membership of the throttle change rate in the throttle change rate categories.

In practice, the throttle change rate is almost rarely above 1.5Pct/ms, so the range of throttle change rate used in the schematic is (-10, 10).

As an example, the accelerator change rate range corresponding to the reverse large accelerator change rate is-10 to-1.5 to-0.3, the accelerator change rate range corresponding to the reverse small accelerator change rate is-0.6 to-0.225 to 0, the accelerator change rate range corresponding to the oilless accelerator change rate is-0.075 to 0.075, the accelerator change rate range corresponding to the forward small accelerator change rate is 0 to 0.225 to 0.6, and the accelerator change rate range corresponding to the forward large accelerator change rate is 0.3 to 1.5 to 10.

For example, the throttle change rate is 0.4, the throttle change rate range corresponding to the forward small throttle change rate is 0-0.225-0.6, the throttle change rate range corresponding to the forward large throttle change rate is 0.3-1.5-10, in the throttle change rate membership curve, the y-axis value corresponding to the throttle change rate of 0.4 is searched, the membership corresponding to the forward small throttle change rate is 0.3, the membership corresponding to the forward large throttle change rate is 0.1, the membership corresponding to the reverse large throttle change rate, the reverse small throttle change rate and the membership corresponding to the oil-free throttle change rate are 0, and the obtained second membership set is {0,0,0,0.3,0.2}.

For example, the accelerator change rate is-1.0, the accelerator change rate range corresponding to the reverse large accelerator change rate is-10 to-1.5 to-0.3, the y-axis value corresponding to the accelerator change rate-1.0 is searched in the accelerator change rate membership curve, the membership corresponding to the reverse large accelerator change rate is 0.6, the membership corresponding to other types is 0, and the obtained second membership set is {0.6,0,0,0,0}.

Step S105: determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set;

And determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set based on the agreed determination rule.

The third membership set comprises a third membership set corresponding to various acceleration intention categories corresponding to the accelerator information and the accelerator change rate.

The following embodiments will be described in detail with respect to this process, and will not be described in detail in this embodiment.

Step S106: determining a target value based on the third membership set and a preset accelerating intention membership curve;

wherein the target value characterizes the current acceleration level.

The acceleration intention membership curve is preset in the controller, wherein the abscissa is an acceleration grade, and the ordinate is a membership grade, and each acceleration intention type corresponds to an acceleration grade range.

As one example, the acceleration intent is in 4 categories, including: the values of the acceleration level ranges corresponding to the 4 acceleration intention types are sequentially increased without acceleration intention, slow acceleration intention, medium acceleration intention and rapid acceleration intention.

In specific implementation, the number of types of the acceleration intention can be set according to actual conditions, and the number of types of the acceleration intention is not limited in the application.

Wherein the target value characterizes the intensity of the acceleration intention, and the larger the target value is, the stronger the acceleration intention is.

As an example, the acceleration level is expressed by a floating point number in the range of 0 to 7, the acceleration level corresponding to no acceleration intention is 0 to 1, the acceleration level corresponding to slow acceleration intention is 1 to 3, the acceleration level corresponding to medium acceleration intention is 3 to 5, and the acceleration level corresponding to rapid acceleration intention is 5 to 7.

It should be noted that, the process of determining the target value based on the third membership set and the preset accelerating intention membership curve is a process of performing the fuzzification processing on the accelerating intention and then performing the defuzzification processing to obtain an accurate value.

It should be noted that, the acceleration levels corresponding to the acceleration intentions are not overlapped, the type of the acceleration intention to which the endpoint value belongs can be set, and the type of the acceleration intention to which each endpoint belongs is not limited in the present application.

The maximum value 7 of the acceleration level corresponding to the sudden acceleration intention is obtained based on a real vehicle test.

Step S107: and determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

Wherein the target value is a strong degree characterizing the intention of acceleration, in particular the level of acceleration.

The acceleration level range corresponding to each acceleration intention type is known, and only needs to determine which acceleration level range the target value belongs to, and the acceleration intention type corresponding to the acceleration level range is the acceleration intention type of the current period.

As shown in fig. 4, there are 4 types of acceleration intention, namely, no acceleration intention, slow acceleration intention, medium acceleration intention, and rapid acceleration intention. Wherein, the acceleration grade corresponding to no acceleration intention is 0-Gz; the acceleration level corresponding to the slow acceleration intention is Gs 1-Gs 2, and the middle value of the value range is Gsc; the acceleration grade corresponding to the middle acceleration intention is Gm 1-Gm 2, and the middle value of the value range is Gmc; the acceleration level corresponding to the rapid acceleration intention is Gb to 7. Wherein Gz is more than Gs1 and Gs2, gm1 and Gm2, gb. In the schematic diagram of fig. 4, 7 is used as the maximum value of the acceleration level.

For example, the target value is 6, the acceleration level range to which the target value belongs is determined to be 5 to 7, the corresponding acceleration intention type is a sudden acceleration intention, and the acceleration intention in the current period is determined to be a sudden acceleration intention.

For example, the target value is 2, it is determined that the acceleration level range to which the target value belongs is 1 to 3, the corresponding acceleration intention class is a slow acceleration intention, and it is determined that the acceleration intention of the current period is a slow acceleration intention.

In summary, according to the method for identifying an acceleration intention provided in the embodiment, an accelerator signal for a vehicle is acquired according to a stipulated period, an accelerator change rate is calculated based on the accelerator signal, a first membership set of at least two accelerator categories to which the accelerator signal belongs and a second membership set of at least two accelerator change rate categories to which the accelerator change rate belongs are respectively determined based on a preset accelerator membership curve and a preset accelerator change rate membership curve, a third membership set corresponding to at least two acceleration intention categories is determined based on the first membership set and the second membership set, a target value is determined based on the third membership set and the preset acceleration intention membership curve, and a category corresponding to the acceleration level range of the target value is determined as the acceleration intention category of the current period based on the acceleration level range corresponding to the at least two acceleration intention categories. In the scheme, first membership of at least two accelerator categories in accelerator information and second membership of at least two accelerator change rate categories in the current running process of the vehicle are determined respectively, accelerator ranges of any two adjacent accelerator categories in a preset accelerator membership curve are partially overlapped, accelerator change rate ranges of any two adjacent accelerator change rate categories in a preset accelerator change rate membership curve are partially overlapped, the process of determining accelerator membership and accelerator change rate membership is fuzzification processing on accelerator signals and accelerator change rates, defuzzification processing is carried out after fuzzification processing on acceleration intention is carried out on a third membership and preset acceleration intention membership curve, a target value representing the current acceleration grade is obtained, the category of acceleration intention can be accurately determined based on the target value, acceleration intention of a driver is accurately identified, and the recognition process has stronger robustness due to the adoption of fuzzy recognition.

As shown in fig. 5, a flowchart of an embodiment 2 of a method for identifying an acceleration intention according to the present application includes the following steps:

step S501: acquiring an accelerator signal based on a contracted period;

Step S501 is identical to step S101 in embodiment 1, and is not described in detail in this embodiment.

Step S502: preprocessing the throttle signal;

in an actual use scene, the throttle signal can jump accidentally due to the influence of signal interference in a vehicle, vibration of a sensor, shake of the whole vehicle and other multiparty factors. At this time, the accelerator change rate signal calculated from the original accelerator signal cannot truly reflect the acceleration intention of the driver.

Therefore, the throttle signal needs to be preprocessed.

Specifically, an inertial filter is adopted to perform real-time pretreatment, specifically filtering treatment, on the accelerator signal.

In specific implementation, a first-order, a second-order and a third-order inertial low-pass filter can be used for filtering the accelerator signal.

The maximum time delay of the three inertial low-pass filters is almost equal to 90ms, and the performance is almost equal to the maximum time delay of the three inertial low-pass filters.

But the third-order inertia low-pass filter has better attenuation performance for signals larger than 5Hz (hertz), can filter signals above 5Hz, and has no low-frequency overshoot of the second-order low-pass filter, so the third-order inertia low-pass filter is preferably used for processing.

In practice, signals above 5Hz are typically high frequency noise due to electrical signal infection or other factors, and need to be filtered out.

Step S503: calculating a throttle change rate based on the throttle signal;

step S504: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

Step S505: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

step S506: determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set;

Step S507: determining a target value based on the third membership set and a preset accelerating intention membership curve;

Step S508: and determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

Steps S503-508 are identical to steps S102-107 in embodiment 1, and are not described in detail in this embodiment.

In summary, the method for identifying an acceleration intention provided in this embodiment further includes: the obtained accelerator signal is preprocessed to filter clutter, and the preprocessed accelerator signal can truly reflect the acceleration intention of a driver, so that the basis of the subsequent analysis process is accurate.

As shown in fig. 6, a flowchart of an embodiment 3 of a method for identifying an acceleration intention according to the present application includes the following steps:

Step S601: acquiring an accelerator signal based on a contracted period;

step S602: calculating a throttle change rate based on the throttle signal;

step S603: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

Step S604: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

steps S601 to 604 are identical to steps S101 to 104 in embodiment 1, and are not described in detail in this embodiment.

Step S605: obtaining at least two preset matrixes corresponding to at least two acceleration intents;

corresponding preset matrixes are correspondingly arranged aiming at different acceleration intents.

Wherein the number of elements of the matrix is related to the throttle category and the throttle change rate category.

The number of rows/columns (or columns/rows) in the preset matrix corresponds to the number of throttle categories/throttle change rate categories in sequence.

For example, if the throttle categories are 4 and the throttle change rate categories are 5, the matrix is a4×5 or 5×4 matrix.

Specifically, the membership degrees defining that a certain throttle signal belongs to the 4 throttle categories of no throttle, small throttle, medium throttle and large throttle are a, b, c, d, and the membership degrees defining that the corresponding throttle change rate belongs to the 5 throttle change rate categories of reverse large throttle change rate, reverse small throttle change rate, no throttle change rate, forward small throttle change rate and forward large throttle change rate are a ', b ', c ', d ' and e '.

In this embodiment, the rows in the preset matrix correspond to the accelerator categories, and the columns correspond to the accelerator change rate categories, resulting in a 4×5 matrix.

For example, a matrix of slow acceleration intents is as follows:

Step S606: substituting the first membership degree set and the second membership degree set into the at least two preset matrixes respectively to obtain at least two target matrixes;

After the preset matrix is obtained, substituting each data in the first membership set and the second membership set into the preset matrix corresponding to each acceleration intention type, so that a plurality of target matrices with values can be obtained, wherein each target matrix corresponds to one acceleration intention type.

Wherein, there are multiple elements in the target matrix, there are data of 0 and non-0 in the multiple elements.

Step S607: sequentially selecting a numerical value meeting a preset condition from a target matrix corresponding to each accelerating intention as a third membership corresponding to the accelerating intention, and obtaining a third membership set corresponding to the at least two accelerating intentions;

And sequentially selecting values meeting preset conditions from the target matrixes corresponding to the acceleration intents as third membership degrees corresponding to the corresponding acceleration intents, and collecting the third membership degrees corresponding to the acceleration intents together to obtain a third membership degree set.

The preset condition may be that the maximum values are respectively taken according to rows and columns in the matrix, and the finally obtained value is the value of the third membership degree.

It should be noted that the preset matrix is determined based on a fuzzy rule, and the fuzzy rule refers to a corresponding relationship between an input and an output of the fuzzy control.

The preset condition is a specific relationship between input and output determined based on the fuzzy rule.

The following 4 kinds of fuzzy logic are generally involved in the operation of the fuzzy rule. Through simulation and real vehicle debugging, the finally selectable fuzzy logic algorithm is as follows: "And" logic-prod (a. Times.b), "Or" logic-max, "Implication" implies logic-min, "Aggregation logic-max. Since software used by the vehicle-mounted controller is generally MATLAB/Simulink as a modeling tool, aggregation logic of max is used in this embodiment for simplicity of modeling.

Of course, in the specific implementation, other calculation rules can be selected, and only the corresponding calibration parameters need to be adjusted, so that the corresponding results can be obtained through debugging.

The process of generating the preset matrix will be described in detail in the following embodiments, which will not be described in detail in this embodiment.

Step S608: determining a target value based on the third membership set and a preset accelerating intention membership curve;

Step S609: and determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

Steps S608-609 are identical to steps S106-107 in embodiment 1, and are not described in detail in this embodiment.

In summary, in the method for identifying an acceleration intention provided in this embodiment, after at least two preset matrices corresponding to at least two acceleration intentions are obtained, numerical values in a first membership set and a second membership set are substituted into the at least two preset matrices respectively to obtain at least two target matrices, target numerical values meeting preset conditions are selected from the at least two target matrices respectively as third membership corresponding to the corresponding acceleration intentions, and the third membership corresponding to each acceleration intention is collected together to obtain a third membership set.

As shown in fig. 7, a flowchart of an embodiment 4 of a method for identifying an acceleration intention according to the present application includes the following steps:

Step S701: obtaining an initial matrix based on the combination of at least two throttle categories and at least two throttle change rate categories;

Wherein each element of the initial matrix corresponds to any acceleration intention category, and each acceleration intention category corresponds to at least one accelerator category and accelerator change rate category combination;

in this embodiment, fuzzy logic is used to calculate and determine membership degrees of various acceleration intentions.

Note that the fuzzy rule refers to a correspondence relationship between input and output of fuzzy control.

The following 4 kinds of fuzzy logic are generally involved in the operation of the fuzzy rule. Through simulation and real vehicle debugging, the finally selectable fuzzy logic algorithm is as follows: "And" logic-prod (a. Times.b), "Or" logic-max, "Implication" implies logic-min, "Aggregation logic-max.

Specifically, in this embodiment, an initial matrix is obtained by using a fuzzy logic algorithm of "And logic.

The process of combining the throttle category and the throttle change rate category to obtain the initial matrix is to obtain the initial matrix by a fuzzy rule that membership corresponding to the throttle category and membership corresponding to the throttle change rate category are subjected to a fuzzy algorithm.

The throttle categories are 4 types, and the throttle change rate is 5 types, so that the throttle (4) ×the throttle change rate (5) =fuzzy rule (20) is shared, and correspondingly, the initial matrix is provided with 20 elements.

Specifically, according to the general knowledge of people and engineering driving experience, each type of throttle and throttle change rate are classified correspondingly.

The classes defining the throttle are no throttle (Z), a small throttle (S), a medium throttle (M) and a large throttle (B) respectively; the categories defining the throttle change rate are a reverse large throttle change rate (NB), a reverse small throttle change rate (NS), an oil-free throttle change rate (Z), a forward small throttle change rate (PS) and a forward large throttle change rate (PB), respectively; there are 4 categories of defined acceleration intents, including: no acceleration intention (Z), slow acceleration intention (S), medium acceleration intention (M), and rapid acceleration intention (B).

The resulting fuzzy rules are shown in table 1 below:

TABLE 1

For example, in the present embodiment, the slow acceleration intention, there are 6 cases in total, which may correspond to the slow acceleration intention: ① Throttle (Z) and throttle rate (PB), ② throttle (S) and throttle rate (NS), ③ throttle (S) and throttle rate (Z), ④ throttle (S) and throttle rate (PS), ⑤ throttle (M) and throttle rate (NS), ⑥ throttle (B) and throttle rate (NB).

The membership degree of the 4 throttle categories of defining that a certain throttle signal belongs to no throttle, small throttle, medium throttle and large throttle is a, b, c, d, and the membership degrees of the 5 throttle change rate categories of corresponding throttle change rate belongs to reverse large throttle change rate, reverse small throttle change rate, no throttle change rate, forward small throttle change rate and forward large throttle change rate are a ', b ', c ', d ' and e '.

Correspondingly, the initial matrix obtained is as follows:

Step S702: assigning values to at least two types of acceleration intents in the initial matrix to obtain a first matrix;

Wherein different categories are assigned different values;

wherein a value is assigned to each of the categories of acceleration intents.

In order to modify fuzzy rules in real time, the universality of the algorithm is enhanced, and the fuzzy rules are represented by matrixes.

Specifically, each acceleration intention category is assigned a different value.

Wherein, no acceleration intention (Z) corresponds to a value 1, a slow acceleration intention (S) corresponds to a value 2, a medium acceleration intention (M) corresponds to a value 3, and a rapid acceleration intention (B) corresponds to a value 4.

Table 1 above can be expressed as:

the first matrix is used for representing a fuzzy rule corresponding to the acceleration intention.

Step S703: obtaining at least two preset matrixes based on the assignment of each of the at least two classes of acceleration intention, the first matrix and the initial matrix;

Firstly, comparing assignment of each category in acceleration intention with a first matrix to obtain at least one combination condition of accelerator and accelerator change rate corresponding to the category; then, a preset matrix corresponding to the acceleration intention is determined based on the combination situation and the initial matrix.

Specifically, 1 is assigned to the non-accelerating intention (Z), 2 is assigned to the slow accelerating intention (S), 3 is assigned to the medium accelerating intention (M), 4 is assigned to the fast accelerating intention (B), the values corresponding to the accelerating intentions are respectively compared with the above (3), wherein Aij is the element of the ith row and the jth column, if the values are equal to the values corresponding to the accelerating intentions, 1 is obtained, if the values are not equal to the values to obtain 0, the obtained values replace the Aij, after all the elements in the first matrix are compared, a new matrix is obtained, only 0 and 1 values in the matrix are obtained, and the new matrix is subjected to the bit multiplication operation of the matrix with the above (2), so that the preset matrix is obtained.

As an example, taking the slow acceleration intention as an example, the slow acceleration intention corresponds to a value of 2 compared with the first matrix as follows:

Performing bit multiplication operation on the matrix (4) and the matrix (2), and obtaining a preset matrix corresponding to the slow acceleration intention as follows:

step S704: acquiring an accelerator signal based on a contracted period;

step S705: calculating a throttle change rate based on the throttle signal;

Step S706: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

Step S707: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

step S708: obtaining at least two preset matrixes corresponding to at least two acceleration intents;

step S709: substituting the first membership degree set and the second membership degree set into the at least two preset matrixes respectively to obtain at least two target matrixes;

Step S710: sequentially selecting a numerical value meeting a preset condition from a target matrix corresponding to each accelerating intention as a third membership corresponding to the accelerating intention, and obtaining a third membership set corresponding to the at least two accelerating intentions;

Step S711: determining a target value based on the third membership set and a preset accelerating intention membership curve;

Step S712: and determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

Steps S703 to 712 are identical to steps S601 to 609 in embodiment 6, and are not described in detail in this embodiment.

In summary, the method for identifying an acceleration intention provided in this embodiment includes: obtaining an initial matrix based on at least two accelerator categories and at least two accelerator change rate category combinations, wherein each element of the initial matrix corresponds to any acceleration intention category, and each acceleration intention category corresponds to at least one accelerator category and accelerator change rate category combination; assigning values to at least two types of acceleration intents in the initial matrix to obtain a first matrix, and assigning different values to different types; and comparing the assignment of each of the at least two classes of acceleration intention with the first matrix and the initial matrix respectively to obtain at least two preset matrices. In this embodiment, an initial matrix is obtained based on a combination of an accelerator category and an accelerator change rate category, each element in the initial matrix is determined to correspond to any acceleration intention category based on fuzzy logic calculation, a first matrix is assigned to each acceleration intention category in the initial matrix based on the initial matrix, then a preset matrix corresponding to each acceleration intention category is determined based on different values corresponding to the acceleration intention category and the first matrix and the initial matrix, a process of generating the preset matrix is realized, and a basis is provided for identifying the acceleration intention category subsequently.

As shown in fig. 8, a flowchart of an embodiment 5 of a method for identifying an acceleration intention according to the present application includes the following steps:

step S801: acquiring an accelerator signal based on a contracted period;

Step S802: calculating a throttle change rate based on the throttle signal;

Step S803: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

Step S804: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

step S805: determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set;

Steps S801 to 805 are identical to steps S101 to 105 in embodiment 1, and are not described in detail in this embodiment.

Step S806: determining a target graph area formed by the third membership set, a target coordinate axis and the preset accelerating intention membership curve;

The target coordinate axis is the coordinate axis where the acceleration level range is located.

The third membership degree set comprises a third membership degree corresponding to each accelerating intention category.

And determining the third membership set and a target graphic area formed by the target coordinate axis and the preset accelerating intention membership curve in the preset accelerating intention membership curve.

Wherein the target graphic region is composed of a plurality of regions.

As shown in fig. 9, there is a schematic diagram of the membership of the acceleration intention, in which the number of acceleration intention is 4, and there are no acceleration intention, slow acceleration intention, medium acceleration intention, and rapid acceleration intention, respectively. In the accelerating intention membership curve, the x-axis represents the accelerating grade, the y-axis represents the membership grade, and the accelerating grade corresponding to no accelerating intention is 0-Gz; the acceleration level corresponding to the slow acceleration intention is Gs 1-Gs 2, and the middle value of the value range is Gsc; the acceleration grade corresponding to the middle acceleration intention is Gm 1-Gm 2, and the middle value of the value range is Gmc; the acceleration level corresponding to the rapid acceleration intention is Gb to 7. Wherein Gz is more than Gs1 and Gs2, gm1 and Gm2, gb. In the schematic diagram of fig. 4, 7 is used as the maximum value of the acceleration level.

The third membership set includes membership degrees corresponding to the 4 acceleration intention categories, namely Lz (no acceleration intention), ls (slow acceleration intention), lm (medium acceleration intention) and Lb (rapid acceleration intention). The intersection point of Lz and the non-acceleration intention curve is Gz1, the intersection point of Ls and the slow acceleration intention curve is Gls1 and Gls2, the intersection point of Lm and the medium acceleration intention curve is Glm and Glm, and the intersection point of Lb and the fast acceleration intention curve is Gb1. In fig. 9, the diagonally filled region is a target pattern region, and the target pattern region is composed of two right trapezoid and two isosceles trapezoid.

Step S807: determining a target centroid of the target graphic region, the target centroid comprising a centroid of at least one graphic in the target graphic region;

wherein, the area gravity center method is adopted to calculate the target value representing the intensity degree of the acceleration intention.

Specifically, the center of gravity of the target graphic region, specifically, the center of gravity of each right triangle or trapezoid in the target graphic region is determined.

Step S808: determining a target value based on the target center of gravity;

and determining the area of the target graphic area and the abscissa of the gravity center, and calculating to obtain a target numerical value.

The principle of the area gravity center method is as follows:

Where k is the target value.

In the above formula (5), the ∈f (x) dx is the area of the target pattern region, and the ∈x·f (x) dx can be obtained by multiplying the abscissa of the center of gravity of the target by the area.

The process of calculating ∈f (x) dx in conjunction with fig. 9 described above is specifically as follows:

in the formula (6) [ Gz1, gls2, glm1, glm, gb1] was calculated by analysis of the curve:

wherein Gsc and Gmc are the abscissa of the triangle blur curve center point of the slow acceleration intention and the medium acceleration intention, respectively.

The process of calculating ∈x·f (x) dx in conjunction with fig. 9 described above is specifically as follows:

in summary, equation (5) can be obtained from equations (6) and (8), and the final output target value is obtained.

In specific implementation, the specific numerical values represented by the characters are substituted into the above formulas (5), (6) and (8), so as to obtain the target numerical value.

Step S809: and determining a first accelerating intention category corresponding to a first accelerating intention category to which the target value belongs as the accelerating intention category of the current period based on the accelerating intention ranges corresponding to the at least two accelerating intention categories.

Step S809 is identical to steps S101-107 in embodiment 1, and details are not described in this embodiment.

In summary, in the method for identifying an acceleration intention provided in this embodiment, a target graphic area formed by the third membership set, a target coordinate axis and the preset acceleration intention membership curve is determined, where the target coordinate axis is a coordinate axis in which an acceleration level range is located; determining a target centroid of the target graphic region, the target centroid comprising a centroid of at least one graphic in the target graphic region; a target value is determined based on the target center of gravity. In the embodiment, the target graphic area is determined based on the target graphic area formed by the third membership set, the coordinate axis where the acceleration level range is located and the preset acceleration intention membership curve, the target numerical value is determined based on the gravity center of the target graphic area, the specific process of determining the target numerical value is clarified, the target numerical value is determined by performing anti-fuzzy processing after the third membership set corresponding to at least two acceleration intention categories determined by fuzzy logic calculation is realized,

As shown in fig. 10, a flowchart of an embodiment 6 of a method for identifying an acceleration intention according to the present application includes the following steps:

Step S1001: acquiring an accelerator signal based on a contracted period;

Step S1002: calculating a throttle change rate based on the throttle signal;

Step S1003: determining a first membership set of at least two accelerator categories to which the accelerator signal belongs based on a preset accelerator membership curve;

step S1004: determining a second membership set of at least two throttle change rate categories according to a preset throttle change rate membership curve;

Step S1005: determining a third membership set corresponding to at least two acceleration intention categories based on the first membership set and the second membership set;

step S1006: determining a target value based on the third membership set and a preset accelerating intention membership curve;

step S1007: determining a first accelerating intention category corresponding to a first accelerating intention category to which the target numerical value belongs as an accelerating intention category of the current period based on accelerating intention ranges corresponding to the at least two accelerating intention categories;

Steps S1001-1007 are identical to steps S101-107 in embodiment 1, and detailed description is omitted in this embodiment.

Step S1008: if the first accelerating intention type is different from the second accelerating intention type of the previous period, judging whether the target value belongs to a fluctuation interval range corresponding to the second accelerating intention type, and obtaining a judging result;

In actual operation, since continuous output inevitably has a certain fluctuation, some jumps will be generated at the place of accelerating the jump of the intention category, and because of the jitter and processing of the throttle and the throttle change rate and the characteristics of the algorithm itself, there may be value jumps when the intention changes.

Therefore, in order to eliminate these hops as much as possible, it is necessary to determine whether or not it is a hop when it is determined that the acceleration intention category is changed.

Specifically, in this embodiment, the hysteresis condition is used for determining. Specifically, the acceleration intention category does not jump within a certain fluctuation interval range, and the acceleration intention category of the previous period is maintained.

After the first acceleration intention category corresponding to the first acceleration level range to which the target value belongs is based, determining the first acceleration intention category as the acceleration intention category of the current period, and then judging which of the acceleration intention categories of the previous period is, if the first acceleration intention category is also the first acceleration intention category, directly executing subsequent responses, such as gear shifting of a transmission, correction of the output torque of an engine and the like, without processing; if the acceleration intention category of the previous cycle is a second acceleration intention category different from the first acceleration intention category, step S1008 is performed.

Wherein, for each acceleration intention category, a fluctuation interval range is set, which is slightly larger than the corresponding acceleration level range, and any two adjacent fluctuation interval ranges of the acceleration intention are available to overlap.

Step S1009: if the judging result represents that the target value belongs to the fluctuation interval range corresponding to the second accelerating intention category, determining the second accelerating intention category as the accelerating intention category of the current period;

And if the target value belongs to a fluctuation interval range corresponding to the second accelerating intention, taking the second accelerating intention type as the accelerating intention type of the current period.

For example, the determined target value is 3.2, the medium acceleration intention is 3 to 5, the slow acceleration intention is 1 to 3, the acceleration intention of the current period is determined to be the medium acceleration intention, the acceleration intention determined in the previous period is the slow acceleration intention, the slow acceleration intention is different from the medium acceleration intention, the fluctuation interval range of the slow acceleration intention is 0.7 to 3.3, the target value 3.2 is in the fluctuation interval range of the slow acceleration intention, and the slow acceleration intention is finally determined as the acceleration intention of the current period.

Step S1010: and if the judging result represents that the target value does not belong to the fluctuation interval range corresponding to the second accelerating intention category, determining the first accelerating intention category as the accelerating intention category of the current period.

If the target value does not belong to the fluctuation interval range corresponding to the second acceleration intention, the target value only belongs to the acceleration level range corresponding to the first acceleration intention and does not belong to the fluctuation interval range corresponding to the second acceleration intention, and the first acceleration intention type is taken as the acceleration intention type of the current period.

For example, the target value is determined to be 4, the acceleration level range of the medium acceleration intention is 3 to 5, the acceleration level range of the slow acceleration intention is 1 to 3, the acceleration intention of the current period is determined to be the medium acceleration intention, the acceleration intention determined in the previous period is the slow acceleration intention, the fluctuation interval range of the slow acceleration intention is different from the medium acceleration intention, the fluctuation interval range of the slow acceleration intention is 0.7 to 3.3, the target value 4 is in the acceleration level range of the medium acceleration intention but is not in the fluctuation interval range of the slow acceleration intention, and the medium acceleration intention is finally determined as the acceleration intention of the current period.

In summary, the method for identifying an acceleration intention provided in this embodiment further includes: based on the fact that the first accelerating intention type is different from the second accelerating intention type of the previous period, whether the target value belongs to a fluctuation interval range corresponding to the second accelerating intention type is further determined, and if the target value belongs to the fluctuation interval range corresponding to the second accelerating intention type, the second accelerating intention type is determined to be used as the accelerating intention type of the current period; if the target value does not belong to the fluctuation interval range corresponding to the second accelerating intention category, determining the first accelerating intention category as the accelerating intention category of the current period. In this embodiment, in a certain fluctuation range, the acceleration intention type does not jump, and the acceleration intention type in the previous period is maintained, so that the jump of the acceleration intention type is prevented.

Corresponding to the embodiment of the method for identifying the accelerating intention provided by the application, the application also provides a scene applying the method for identifying the accelerating intention.

The present application provides a method for identifying an acceleration intention, and a method for classifying and looking up a table in the prior art.

As shown in fig. 11, a software architecture diagram of a processor employing a method for identifying an acceleration intention provided in the present application includes: the signal preprocessing 1101, blurring processing 1102, a matrix of fuzzy rules 1103, fuzzy logic calculation 1104, anti-blurring processing 1105 and signal post-processing 1106 have 6 components corresponding to different recognition stages.

The signal preprocessing stage is used for carrying out filtering processing on the throttle signal obtained from the CAN and determining the throttle change rate;

And in the fuzzification processing stage, fuzzification processing is carried out on the filtered throttle signal by adopting a preset throttle membership curve to obtain throttle membership, and fuzzification processing is carried out on the throttle change rate by adopting a preset throttle change rate membership curve to obtain throttle change rate membership.

And determining preset matrixes corresponding to different accelerating intention categories based on the set fuzzy rules.

The fuzzy logic calculation stage is used for determining the membership degree of each acceleration intention based on the throttle membership degree and the throttle change rate membership degree;

Wherein, in the anti-fuzzy processing stage, a numerical value (target numerical value) representing the intensity degree of the accelerating intention is determined based on the accelerating intention membership curve and the membership degree of each accelerating intention;

in the signal post-processing stage, an initial accelerating intention is firstly determined based on the value, hysteresis conditions are introduced, the accelerating intention type does not jump within a certain fluctuation interval range, the accelerating intention type of the last period is maintained, and a final accelerating intention is determined.

Fig. 12 is a schematic diagram of real vehicle verification data, which includes an accelerator opening degree (dPct), an accelerator change rate (dPcts), an acceleration category obtained by using the acceleration intention recognition method (abbreviated as fuzzy control) and an acceleration category obtained by using the classified lookup method.

The road condition verified by the real vehicle is complex, and the driving style of the driver is slightly excited, so that the change of the throttle is more and faster.

Analysis of this fig. 12 can yield the following results:

(1) The acceleration intention obtained by the classified table look-up algorithm is slow acceleration (Pedal On), and under many working conditions, the algorithm does not effectively judge that the accuracy is problematic under the condition that the acceleration intention is actually needed.

(2) The acceleration intention obtained by the fuzzy control algorithm is slow acceleration and medium acceleration, and compared with the situation that the classification table look-up algorithm only recognizes the slow acceleration, the classification table look-up algorithm recognizes more intention types and is closer to the idea of an actual driver; on the other hand, the corresponding acceleration intention can be identified almost every time the accelerator is changed, and the accuracy is better.

Statistical analysis is performed on the real vehicle verification data to obtain the results shown in the table 2 of the real vehicle verification data acceleration intention recognition expression summary table:

TABLE 2

	Classification look-up table	Fuzzy control	Results
				Total number of intents to be identified	39	39
With the sum of recognition results	17	39	Up to 100% identifiable
				Accurately identifying the total number	17 (Containing more accurate)	37 (Without more accurate)
Accurate duty cycle identification	43.6％	94.9％	The accuracy is greatly improved
				Average identification time (ms)	117	19	The time is shortened by 83.8 percent

The data shown in the table can be intuitively seen that in the test data, whether the accuracy or the rapidity of the intention recognition is accelerated, the fuzzy control is obviously superior to the existing classified table look-up algorithm:

(1) Rapidity: according to actual measurement data statistics, compared with the traditional classified table look-up algorithm, the algorithm strategy of the application shortens the recognition time by 83.8% on average, and greatly improves the recognition speed;

(2) Accuracy 1: under complex working conditions, the acceleration intention is changeable, the average recognition accuracy of the traditional classification table lookup algorithm is about 43.6%, the algorithm strategy recognition accuracy of the application is 94.9%, and the recognition accuracy is greatly improved by about 50%;

(3) Accuracy 2: on the basis of improving the recognition accuracy, the recognized acceleration intention is more classified (2 to 4), and the real intention of a driver can be more accurately represented subjectively.

Corresponding to the embodiment of the method for identifying the accelerating intention provided by the application, the application also provides an embodiment of a device applying the method for identifying the accelerating intention.

Fig. 13 is a schematic structural diagram of an embodiment of an acceleration intention recognition device according to the present application, where the device includes the following structures: an acquisition module 1301, a calculation module 1302, a first category determination module 1303, a second category determination module 1304, a set determination module 1305, a target value determination module 1306, and a third category determination module 1307;

The acquiring module 1301 is configured to acquire an accelerator signal based on a default period;

wherein, the calculating module 1302 is configured to calculate a throttle change rate based on the throttle signal;

The first class determining module 1303 is configured to determine, based on a preset accelerator membership curve, a first membership set of at least two accelerator classes to which the accelerator signal belongs, where accelerator ranges of any two adjacent accelerator classes in the preset accelerator membership curve are partially overlapped;

The second class determining module 1304 is configured to determine, based on a preset accelerator change rate membership curve, that the accelerator change rate belongs to a second membership set of at least two accelerator change rate classes, where accelerator change rate ranges of any two adjacent accelerator change rate classes in the preset accelerator change rate membership curve are partially overlapped;

The set determining module 1305 is configured to determine a third membership set corresponding to at least two types of acceleration intention based on the first membership set and the second membership set;

The target value determining module 1306 is configured to determine a target value based on the third membership set and a preset acceleration intention membership curve, where the target value characterizes the current acceleration level;

The third category determining module 1307 is configured to determine, based on the acceleration level ranges corresponding to the at least two acceleration intention categories, a first acceleration intention category corresponding to a first acceleration level range to which the target value belongs as an acceleration intention category of the current period.

Optionally, the method further comprises:

And the filtering module is used for preprocessing the throttle signal.

Optionally, the set determining module includes:

the acquisition unit is used for acquiring at least two preset matrixes corresponding to at least two acceleration intents;

the substituting unit is used for substituting the first membership degree set and the second membership degree set into the at least two preset matrixes respectively to obtain at least two target matrixes;

the selection unit is used for sequentially selecting a numerical value meeting a preset condition from the target matrix corresponding to each accelerating intention as a third membership corresponding to the accelerating intention, and obtaining a third membership set corresponding to the at least two accelerating intentions.

Optionally, further comprising:

The system comprises a preset matrix generation module, a control module and a control module, wherein the preset matrix generation module is used for obtaining an initial matrix based on at least two accelerator categories and at least two accelerator change rate category combinations, each element of the initial matrix corresponds to any acceleration intention category, and each acceleration intention category corresponds to at least one accelerator category and accelerator change rate category combination; assigning values to at least two types of acceleration intents in the initial matrix to obtain a first matrix, and assigning different values to different types; and obtaining at least two preset matrixes based on the assignment of each of the at least two classes of acceleration intention, the first matrix and the initial matrix.

Optionally, the target value determining module includes:

The first determining unit is used for determining a target graph area formed by the third membership set, a target coordinate axis and the preset accelerating intention membership curve, wherein the target coordinate axis is the coordinate axis where the accelerating grade range is located;

a second determining unit configured to determine a target barycenter of the target graphic region, the target barycenter including a barycenter of at least one graphic in the target graphic region;

and a third determination unit configured to determine a target value based on the target center of gravity.

Optionally, the method further comprises:

The judging module is used for judging whether the target value belongs to a fluctuation interval range corresponding to the second accelerating intention category or not if the first accelerating intention category is different from the second accelerating intention category of the previous period, so as to obtain a judging result;

A fourth category determining unit, configured to determine, if the determination result indicates that the target value belongs to a range of a fluctuation interval corresponding to the second acceleration intention category, the second acceleration intention category as an acceleration intention category of the current period; and if the judging result represents that the target value does not belong to the fluctuation interval range corresponding to the second accelerating intention category, determining the first accelerating intention category as the accelerating intention category of the current period.

It should be noted that, for the functional explanation of each component structure in the accelerating intention recognition device provided in this embodiment, please refer to the explanation in the foregoing method embodiment, and details are not described in this embodiment.

In summary, in the method for identifying an acceleration intention provided in this embodiment, it is determined that, in a current running process of a vehicle, accelerator information belongs to a first membership degree of at least two accelerator categories and accelerator change rate belongs to a second membership degree of at least two accelerator change rate categories, because accelerator ranges of any two adjacent accelerator categories in a preset accelerator membership degree curve are partially overlapped and accelerator change rate ranges of any two adjacent accelerator change rate categories in a preset accelerator change rate membership degree curve are partially overlapped, a process of determining accelerator membership degree and accelerator change rate membership degree is a fuzzification process on accelerator signals and accelerator change rate, and a third membership degree and a preset accelerator intention membership degree curve implement an anti-fuzzification process after the fuzzification process on the acceleration intention is performed, so as to obtain a target value representing a current acceleration level, and based on the target value, the category of the acceleration intention can be accurately determined, so that the acceleration intention of a driver is accurately identified.

Corresponding to the embodiment of the method for identifying the accelerating intention provided by the application, the application also provides the electronic equipment and the readable storage medium corresponding to the method for identifying the accelerating intention.

Wherein, this electronic equipment includes: a memory, a processor;

Wherein the memory stores a processing program;

In a specific implementation, the electronic device is specifically a vehicle-mounted controller.

The method for identifying the acceleration intention of the electronic device is only needed to refer to the embodiment of the method for identifying the acceleration intention.

Wherein the readable storage medium has stored thereon a computer program which is invoked and executed by a processor to implement the steps of the acceleration intention recognition method according to any one of the above.

The computer program stored in the readable storage medium may be executed to implement the method for identifying the acceleration intention, and the method may be executed with reference to the foregoing embodiment of the method for identifying the acceleration intention.

In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. The device provided in the embodiment corresponds to the method provided in the embodiment, so that the description is simpler, and the relevant points refer to the description of the method.

The previous description of the provided embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features provided herein.

Claims

1. A method for identifying an acceleration intention, comprising:

acquiring an accelerator signal based on a contracted period;

calculating a throttle change rate based on the throttle signal;

2. The method of claim 1, wherein after the acquiring the throttle signal based on the contracted period, before the calculating the throttle change rate based on the throttle signal, further comprises:

and filtering the throttle signal.

3. The method of claim 1, wherein the determining a third set of membership corresponding to at least two acceleration intents based on the first set of membership and the second set of membership comprises:

Obtaining at least two preset matrixes corresponding to at least two acceleration intents;

substituting the first membership degree set and the second membership degree set into the at least two preset matrixes respectively to obtain at least two target matrixes;

And sequentially selecting a numerical value meeting a preset condition from the target matrix corresponding to each accelerating intention as a third membership corresponding to the accelerating intention, and obtaining a third membership set corresponding to the at least two accelerating intentions.

4. The method of claim 3, further comprising, prior to said acquiring the throttle signal based on the contracted period:

Obtaining an initial matrix based on at least two accelerator categories and at least two accelerator change rate category combinations, wherein each element of the initial matrix corresponds to any acceleration intention category, and each acceleration intention category corresponds to at least one accelerator category and accelerator change rate category combination;

Assigning values to at least two types of acceleration intents in the initial matrix to obtain a first matrix, and assigning different values to different types;

And obtaining at least two preset matrixes based on the assignment of each of the at least two classes of acceleration intention, the first matrix and the initial matrix.

5. The method of claim 1, wherein the determining the target value based on the third set of membership and a preset acceleration intention membership curve comprises:

Determining a target graphic area formed by the third membership set, a target coordinate axis and the preset acceleration intention membership curve, wherein the target coordinate axis is the coordinate axis in which the acceleration grade range is located;

determining a target centroid of the target graphic region, the target centroid comprising a centroid of at least one graphic in the target graphic region;

a target value is determined based on the target center of gravity.

6. The method according to claim 1, wherein after determining, based on the acceleration level ranges corresponding to the at least two acceleration intention categories, a first acceleration intention category corresponding to a first acceleration level range to which the target value belongs as the acceleration intention category of the current period, further includes:

If the first accelerating intention type is different from the second accelerating intention type of the previous period, judging whether the target value belongs to a fluctuation interval range corresponding to the second accelerating intention type, and obtaining a judging result;

If the judging result represents that the target value belongs to the fluctuation interval range corresponding to the second accelerating intention category, determining the second accelerating intention category as the accelerating intention category of the current period;

and if the judging result represents that the target value does not belong to the fluctuation interval range corresponding to the second accelerating intention category, determining the first accelerating intention category as the accelerating intention category of the current period.

7. An acceleration intention recognition device, characterized by comprising:

8. The apparatus as recited in claim 7, further comprising:

And the filtering module is used for preprocessing the throttle signal.

9. An electronic device, characterized in that,

Wherein, this electronic equipment includes: a memory, a processor;

Wherein the memory stores a processing program;

The processor is configured to load and execute the processing program stored in the memory to implement the steps of the acceleration intention recognition method according to any one of claims 1 to 7.

10. A readable storage medium, having stored thereon a computer program, the computer program being invoked and executed by a processor, implementing the steps of the acceleration intention recognition method of any one of claims 1-7.