CN118430224A - Method, device, electronic device and readable storage medium for predicting vehicle speed - Google Patents

Abstract

The application belongs to the field of intelligent automobiles, and particularly relates to a method and device for predicting a vehicle speed, electronic equipment and a readable storage medium. Because the method can determine the working condition state of the vehicle based on the driving data; the vehicle speed is predicted in the determined operating condition, so that the predicted vehicle speed in the determined operating condition is less likely than the predicted vehicle speed in the non-determined operating condition, that is, the predicted vehicle speed in the determined operating condition is less likely than the predicted vehicle speed in the non-determined operating condition, the numerical range of the predicted vehicle speed can be narrowed, the final predicted result can be determined in a smaller numerical range, and the accuracy of the predicted vehicle speed can be improved.

Description

Method, device, electronic device and readable storage medium for predicting vehicle speed

Technical Field

The present application relates to the field of smart cars, and in particular to a method, device, electronic device and readable storage medium for predicting vehicle speed.

Background technique

In the field of smart cars, the speed prediction algorithm can transmit the driving data of a period of time in the future to the vehicle decision-making body for analysis, so as to formulate the best driving strategy, which is an indispensable part of vehicle intelligence. However, the existing speed prediction algorithm does not accurately predict the speed of the vehicle.

Summary of the invention

The embodiments of the present application provide a method, device, electronic device and readable storage medium for predicting vehicle speed, thereby improving the accuracy of predicting vehicle speed.

In a first aspect, a method for predicting vehicle speed is provided, comprising:

Acquire driving data of the vehicle in a first time period;

Determining the operating state of the vehicle based on the driving data;

Based on the operating state of the vehicle, a vehicle speed of the vehicle at a first moment is determined, and the first moment is later than the first time period.

In the embodiment of the present application, since the operating state of the vehicle can be determined through driving data and the vehicle speed can be predicted under a determined operating state, the numerical range of the predicted vehicle speed can be narrowed compared to predicting the vehicle speed under an uncertain operating state, and the final prediction result can be determined within a smaller numerical range, thereby improving the accuracy of the predicted vehicle speed.

In one embodiment, determining the operating state of the vehicle based on the driving data includes:

Determine, based on the driving data, driving parameters of the vehicle in the first time period, the driving parameters comprising at least one of the following: maximum acceleration, minimum acceleration, average vehicle speed, parking time ratio, low speed time ratio, medium speed time ratio, and high speed time ratio;

The operating state of the vehicle is determined according to the driving parameters.

In the embodiment of the present application, since the driving parameters of the vehicle in the first time period can be determined through the driving data, the operating state of the vehicle can be determined based on the driving parameters, and the operating state of the vehicle can be accurately determined.

In one embodiment, determining the operating state of the vehicle according to the driving parameters includes:

The driving parameters are input into a working condition identification model to obtain the working condition state of the vehicle.

In the embodiment of the present application, since the driving parameters are input into the operating condition identification model to obtain the operating condition status of the vehicle, the accuracy of identifying the operating condition status can be improved.

In one embodiment, the operating condition identification model is trained based on the following method:

For each operating condition, a preset number of sets of training data are selected, each set of the training data includes a preset number of driving parameters and a benchmark analysis result;

The operating condition identification model is trained based on the training data until the operating condition identification model converges.

Through the training method of the embodiment of the present application, the trained operating condition recognition model can have a higher accuracy in identifying the operating condition state.

In one embodiment, determining the vehicle speed of the vehicle at a first moment based on the operating state of the vehicle includes:

Acquire a first speed and a first acceleration of the vehicle, wherein the first speed is the speed of the vehicle at a second moment, the first acceleration is the acceleration of the vehicle at the second moment, and the second moment is any moment within the first time period;

Based on the acceleration prediction model corresponding to the operating state, the first speed and the first acceleration, a second speed is determined, where the second speed is the speed of the vehicle at the first moment.

In the embodiment of the present application, since the vehicle speed is predicted based on a certain operating condition, and the acceleration can reflect the driver's driving intention, when predicting the vehicle speed, under a certain operating condition, the driver's future driving intention can be predicted based on the acceleration prediction model. Therefore, when predicting the vehicle speed, the embodiment of the present application can narrow the numerical range of the predicted vehicle speed, determine the final prediction result within a smaller numerical range, and take into account the driver's future driving intention, thereby improving the accuracy of the predicted vehicle speed.

In one embodiment, determining the second speed based on the acceleration prediction model corresponding to the operating state, the first speed and the first acceleration includes:

Processing the first acceleration by using an acceleration prediction model corresponding to the operating state to determine the second acceleration, where the second acceleration is the acceleration of the vehicle at the first moment;

Based on the first speed and the second acceleration, the second speed is determined.

In the embodiment of the present application, since acceleration can reflect the driver's driving intention, when predicting the vehicle speed, the second acceleration is first obtained based on the first acceleration and the acceleration prediction model corresponding to the operating condition, which can predict the driver's driving intention in the future. Then, based on the second acceleration and the first speed, the second speed is determined. Therefore, when predicting the vehicle speed, the embodiment of the present application takes into account the driver's driving intention in the future, thereby improving the accuracy of the predicted vehicle speed.

In one embodiment, the method further comprises:

Inputting the second acceleration into the acceleration prediction model corresponding to the operating state to obtain a third acceleration, where the third acceleration is the acceleration of the vehicle at a third moment, and the third moment is later than the first moment;

A third speed is determined based on the second speed and the third acceleration, where the third speed is the speed of the vehicle at the third moment.

In the embodiment of the present application, since acceleration can reflect the driver's driving intention, when predicting the vehicle speed, the third acceleration is first obtained based on the second acceleration and the acceleration prediction model corresponding to the operating condition, which can predict the driver's driving intention in the future. Then, based on the third acceleration and the second speed, the third speed is determined. Therefore, when predicting the vehicle speed, the embodiment of the present application takes into account the driver's driving intention in the future, thereby improving the accuracy of the predicted vehicle speed.

In one embodiment, the acceleration prediction model corresponding to the operating state is a Markov model.

In the embodiment of the present application, the acceleration prediction model corresponding to the working state is a Markov model. In the Markov model, the driving acceleration of the vehicle at a certain moment in the future has nothing to do with the historical driving data, but only with the current driving data. Therefore, when using the Markov model to predict the acceleration, it is not necessary to consider the historical driving data of the vehicle, which reduces the amount of calculation of the model when predicting the acceleration and improves the processing speed of the model.

In one embodiment, the acceleration prediction model is constructed by:

For each operating condition, based on the driving data in the training data, determine the state space of acceleration at time k and the state space of acceleration at time k+1;

Determine the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1;

The acceleration prediction model is constructed based on the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1.

By using the acceleration prediction model provided in the embodiment of the present application, when predicting the acceleration at time k+1, it is independent of the historical acceleration and is only related to the acceleration at time k. The vehicle's historical driving data does not need to be considered, which reduces the amount of calculation of the model when predicting acceleration and improves the processing speed of the model.

In one embodiment, the operating state is any one of the following: a congested operating state, an urban operating state, a suburban operating state, and a high-speed operating state.

In a second aspect, a device for predicting vehicle speed is provided, comprising: a processing unit, wherein the processing unit is configured to:

Acquire driving data of the vehicle in a first time period;

Determining the operating state of the vehicle based on the driving data;

Based on the operating state of the vehicle, a vehicle speed of the vehicle at a first moment is determined, and the first moment is later than the first time period.

In a third aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for predicting vehicle speed as described in any one of the first aspects is implemented.

In a fourth aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting vehicle speed as described in any one of the first aspects is implemented.

In a fifth aspect, a vehicle is provided, comprising the electronic device described in the fourth aspect.

It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

FIG1 is a schematic diagram of an application scenario of a method for predicting vehicle speed provided in an embodiment of the present application.

FIG. 2 is a schematic flowchart of a method for predicting vehicle speed provided in an embodiment of the present application.

FIG3 is a schematic flowchart of a method for training a working condition recognition model provided in an embodiment of the present application.

FIG4 is a schematic flowchart of a method for constructing an acceleration prediction model provided in an embodiment of the present application.

FIG5 is a schematic diagram of the structure of a device for predicting vehicle speed provided in an embodiment of the present application.

FIG. 6 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

Detailed ways

In the following description, specific details such as specific system structures, technologies, etc. are proposed for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application. In other cases, the specific technical details of each embodiment may be referenced to each other, and a specific system not described in one embodiment may be referenced to other embodiments.

It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or combinations thereof.

It should also be understood that the term “and/or” used in the specification and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

References to "an embodiment of the present application" or "some embodiments" described in the specification of the present application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in other embodiments", "an embodiment of the present application", "other embodiments of the present application", etc. that appear in different places in the specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

In addition, in the description of the present application specification and the appended claims, the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

As described in the background technology, the existing vehicle speed prediction algorithm predicts the vehicle speed inaccurately. The specific reason is that the existing vehicle speed prediction algorithm predicts the vehicle speed under uncertain working conditions, which results in the final prediction result needing to be determined within a larger vehicle speed value range, resulting in inaccurate final prediction result.

Based on this, the inventor of the present application discovered through research that when predicting the vehicle speed, the operating condition is first determined based on the vehicle's driving data, and the vehicle speed is predicted under the determined operating condition. In this way, compared to predicting the vehicle speed under an uncertain operating condition, the numerical range of the predicted vehicle speed can be narrowed, and the final prediction result can be determined within a smaller numerical range, thereby improving the accuracy of the predicted vehicle speed.

In order to illustrate the technical solution of the present application, a specific embodiment is provided below for illustration.

Please refer to Figure 1, which is a schematic diagram of an application scenario of a method for predicting vehicle speed provided by an embodiment of the present application. For the convenience of explanation, only the part related to the present application is shown. The application scenario includes:

Vehicle 10 , vehicle 10 may be a fuel vehicle, a gas vehicle or a new energy vehicle, the new energy vehicle may be a pure electric vehicle, a hybrid vehicle or an extended-range vehicle, etc., and the embodiment of the present application does not limit the type of vehicle 10 .

The vehicle 10 is equipped with a vehicle control unit (VCU) 11. The VCU 11 is the central control unit of the vehicle and the core of the vehicle control system. The VCU 11 communicates with various electronic devices such as the engine, transmission, accelerator pedal, brake pedal, and body controller of the vehicle 10 through a serial communication bus (Controller Area Network, CAN), reads the working status of each control unit, and controls each control unit when necessary.

Exemplarily, the VCU 11 is used to read the driving data of the vehicle 10 , such as the vehicle speed, acceleration, etc., determine the operating state of the vehicle 10 based on the driving data of the vehicle 10 , and predict the vehicle speed under the determined operating state.

In some embodiments, the vehicle 10 may also collect driving data through configured sensors. For example, the vehicle 10 collects the speed of the vehicle 10 through a configured speed sensor, and the vehicle 10 collects the acceleration of the vehicle 10 through a configured acceleration sensor.

In some embodiments, the vehicle 10 transmits the driving data of the vehicle 10 read by the VCU 11 or the driving data collected by the sensor to the electronic device 20 by wired or wireless means. The wired means include wired Ethernet (RJ45 line, optical fiber, etc.), industrial serial bus (RS485 bus, RS232 bus, serial communication bus (Controller Area Network, CAN), etc.). The wireless means include General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), wireless gateway, etc. The embodiment of the present application does not limit the communication method between the vehicle 10 and the electronic device 20.

In some embodiments, the electronic device 20 may also be referred to as a big data background, and the electronic device 20 includes but is not limited to a server and a server cluster. Exemplarily, the electronic device 20 is used to obtain driving data of the vehicle 10, determine the operating state of the vehicle 10 based on the driving data, and predict the vehicle speed under the determined operating state.

The embodiments of the present application do not limit the specific composition of the application scenario. The application scenario may include more or fewer components than the example shown in FIG1, or a combination of certain components, or different components. FIG1 is only an exemplary description and cannot be interpreted as a specific limitation of the present application. For example, it may also include a body control unit and an electronic power steering system, etc.

Please refer to FIG. 2, which is a schematic flow chart of a method 200 for predicting vehicle speed provided in an embodiment of the present application. Exemplarily, the execution subject of the method 200 may be the VCU 11 or the electronic device 20 in FIG. 1, or the execution subject of the method 200 may be the chip or processor in the VCU 11, or the execution subject of the method 200 may be the chip or processor in the electronic device 20. For ease of description, the method 200 is described in detail by taking the execution subject as the VCU as an example.

S201. Acquire driving data of a vehicle in a first time period.

In an embodiment of the present application, the first time period may be the current time period. For example, when obtaining driving data of the current time period during driving of the vehicle, the current time period may be calculated from the current time, which is 11:25. The duration of the first time period is pre-set in the vehicle. For example, the duration of the first time period is 5 minutes, then the current time period is from 11:20 to 11:25.

In an embodiment of the present application, the first time period may also be any time period before the current time. For example, the current time is 11:25 and the length of the first time period is 5 minutes. Then the first time period may be from 11:00 to 11:05, or from 10:35 to 10:40, etc.

In the embodiment of the present application, the driving data includes vehicle speed, acceleration, etc.

In the embodiment of the present application, since the VCU can read the vehicle speed in real time, the VCU can obtain the vehicle speed according to the real-time read vehicle speed. For example, when the first time period is from 11:20 to 11:25, the VCU reads the vehicle speed once every 1 second, and the VCU can obtain 300 vehicle speed data. The embodiment of the present application does not limit the interval time.

In the embodiment of the present application, the vehicle speed sensor can collect the vehicle speed in real time, and the VCU obtains the vehicle speed data in the first time period through the vehicle speed sensor. For example, when the first time period is from 11:20 to 11:25, the vehicle speed sensor collects the vehicle speed every 1 second, and the VCU can obtain 300 vehicle speed data. The embodiment of the present application does not limit the interval time.

In the embodiment of the present application, the acceleration sensor can collect the acceleration of the vehicle in real time, and the VCU obtains the acceleration data within the first time period through the acceleration sensor.

In the embodiment of the present application, the VCU can calculate the acceleration of the vehicle according to the acquired vehicle speed data according to the following formula:

Wherein, a _m represents the acceleration corresponding to time m, and time m is any time in the first time period. It represents the speed corresponding to time n, which is the next time after time m. represents the speed corresponding to time m, t _n represents time n, and t _m represents time m.

S202: Determine the operating status of the vehicle based on the driving data.

It can be understood that the working condition of a vehicle refers to the working condition of the vehicle during driving. The so-called working condition actually refers to the road condition.

Exemplarily, the operating condition is any one of the following: a congested operating condition, an urban operating condition, a suburban operating condition, and a high-speed operating condition, etc.

The congestion condition refers to the working state of the vehicle when it is driving on congested roads. The urban condition refers to the working state of the vehicle when it is driving on urban roads. The suburban condition refers to the working state of the vehicle when it is driving on suburban roads. The high-speed condition refers to the working state of the vehicle when it is driving on high-speed roads. These four conditions can basically cover all the working states of the vehicle during driving.

The driving data of the vehicle in each operating state is very different. For example, the vehicle speed in a congested operating state may range from (0, 30 km/h], while the vehicle speed in a high-speed operating state may range from (30 km/h, 60 km/h]. The embodiment of the present application can determine the operating state of the vehicle based on this difference in driving data.

In order to prevent frequent determination of the vehicle's operating status in the embodiment of the present application, the VCU will re-determine the vehicle's operating status at a preset interval after determining the vehicle's operating status. The preset interval duration can be 1 hour, 2 hours, etc., and the embodiment of the present application does not limit the preset interval duration.

It should be noted that the above four operating conditions are divided according to the actual measured driving data. In some embodiments, the operating condition can also be a typical cycle condition, such as: the urban dynamometer driving schedule (UDDS) in the United States, the New York City Cycle (NYCC), the Federal Test Procedure (FTP) used by the US Environmental Protection Agency to certify vehicle emissions, the US06 condition (Supplemental FTP Driving Schedule, US06), the European urban cycle condition (UDC), the extra urban driving cycle (EUDC), the new European automobile regulation cycle (NEDC), the Japanese 10-15 (J10-15), the world-wide harmonized light duty test cycle (WLTC), etc. The embodiment of the present application does not limit the operating condition.

S203: Determine the vehicle speed at a first moment based on the operating state of the vehicle, where the first moment is later than the first time period.

It can be understood that the first moment refers to a certain moment in the future, and the vehicle speed corresponding to the certain moment in the future is the vehicle speed predicted by VCU. For example: the current moment is 16:44 on October 1, 2020, and the vehicle speed corresponding to the current moment is 40km/h. When VCU predicts the vehicle speed at 16:44 on October 2, 2020, 16:44 on October 2, 2020 is the first moment.

In the embodiment of the present application, the first moment is later than the first time period. It is understood that the vehicle speed in the first time period can be directly obtained through the vehicle speed sensor or VCU, and is the historical data that has been generated. The vehicle speed at the first moment is the data that needs to be predicted that has not yet been generated, so the first moment needs to be later than the first time period. For example, when the first time period is from 11:20 to 11:25, the first moment is the moment later than 11:25.

In the embodiment of the present application, the VCU can determine the vehicle speed at the first moment, i.e., the predicted vehicle speed, according to the determined working condition. In the embodiment of the present application, the vehicle speed can be predicted based on a convolutional neural network speed prediction method, a Markov-based speed prediction method, or the like.

The vehicle speed prediction method in the prior art predicts the vehicle speed under an uncertain working condition. For example, the vehicle's driving data is directly input into a vehicle speed prediction model to predict the vehicle speed. The working condition corresponding to the driving data is not determined in advance in the vehicle speed prediction model. The VCU can give a predicted vehicle speed according to each working condition. For example, 16 vehicle speeds are predicted based on 4 working conditions, and 4 vehicle speeds are predicted for each working condition. The final prediction result is determined from these 16 predicted vehicle speeds. The real-time example of the present application predicts the vehicle speed under a certain working condition. For example, 4 vehicle speeds are predicted under a certain working condition, and the final prediction result is determined from these 4 predicted vehicle speeds.

It can be seen from this that in the embodiments of the present application, since the operating state of the vehicle can be determined through driving data and the vehicle speed can be predicted under a determined operating state, the numerical range of the predicted vehicle speed can be narrowed compared to predicting the vehicle speed under an uncertain operating state, and the final prediction result can be determined within a smaller numerical range, thereby improving the accuracy of the predicted vehicle speed.

In some embodiments, S202 also includes: the VCU determines the driving parameters of the vehicle in the first time period based on the driving data; and determines the operating state of the vehicle according to the driving parameters.

It can be understood that the driving parameter is a parameter used to reflect the working state of the vehicle. The driving parameter includes at least one of the following: maximum acceleration, minimum acceleration, average vehicle speed, parking time ratio, low speed time ratio, medium speed time ratio and high speed time ratio.

In the embodiment of the present application, the driving data is the data of the vehicle in the first time period. For example, the duration of the first time period is T, the vehicle speed set in the first time period is {V _i |i=1, 2, ..., i-1, i}; the acceleration set in the first time period is {a _i |i=1, 2, ..., i-1, i}.

The VCU may select the maximum value in the acceleration set as the maximum acceleration, which is recorded as a _max , and select the minimum value as the minimum acceleration, which is recorded as a _min .

The VCU can calculate the average vehicle speed according to the following formula:

in, Represents the average vehicle speed.

The VCU may determine the number of zero vehicle speeds A in the vehicle speed set, and determine the parking time ratio based on A and the total number of vehicle speeds i in the vehicle speed set based on the following formula:

For example, the low speed range is (0,, 30km/h], and the VCU may determine the number B of vehicle speeds within the low speed range in the vehicle speed set, and determine the low speed time ratio based on B and the total number i of vehicle speeds in the vehicle speed set based on the following formula:

For example, the medium speed range is (30 km/h, 60 km/h], and the VCU may determine the number C of vehicle speeds within the medium speed range in the vehicle speed set, and determine the medium speed time ratio based on C and the total number i of vehicle speeds in the vehicle speed set based on the following formula:

For example, the range of high speed is (60 km/h, +∞), and the VCU may determine the number D of vehicle speeds within the range of high speed in the vehicle speed set, and determine the high speed time ratio based on D and the total number i of vehicle speeds in the vehicle speed set based on the following formula:

In the embodiment of the present application, after the VCU determines the driving parameters that can reflect the operating status of the vehicle, the operating status can be determined based on the driving parameters.

In some embodiments, the method for the VCU to determine the working condition based on the driving parameters includes: processing the driving parameters through a working condition recognition model constructed based on measured data to determine the working condition; processing the driving parameters through a working condition recognition model constructed based on historical data to determine the working condition, etc.

In the embodiment of the present application, since the driving parameters of the vehicle in the first time period can be determined through the driving data, the operating state of the vehicle can be determined based on the driving parameters, and the operating state of the vehicle can be accurately determined.

In some embodiments, the VCU determines the operating state of the vehicle based on the driving parameters, including:

The driving parameters are input into the operating condition identification model to obtain the vehicle's operating condition.

In an embodiment of the present application, the operating condition recognition model is a trained classification model. Exemplarily, the operating condition recognition model is a multi-classification model. For example, it can be a four-classification model. The four categories are congested operating condition state, urban operating condition state, suburban operating condition state and highway operating condition state. The operating condition state of the vehicle can be obtained by processing the driving parameters through the operating condition recognition model.

In some embodiments, the operating condition recognition model is obtained by training a classification model of a bagged tree.

In other embodiments, the operating condition identification model may also be other types of operating condition identification models such as an operating condition identification model based on a decision tree, an operating condition identification model based on a classification tree, and an operating condition identification model based on a random forest, and the embodiments of the present application do not impose any limitations thereto.

Since the operating condition identification model based on the bagged tree has higher accuracy in identifying the operating condition state than other operating condition identification models, the operating condition identification model based on the bagged tree is selected to process the driving parameters to obtain the operating condition state of the vehicle.

In the embodiment of the present application, since the driving parameters are input into the operating condition identification model to obtain the operating condition status of the vehicle, the accuracy of identifying the operating condition status can be improved.

Please refer to FIG3, which is a schematic flow chart of a training method 300 for a working condition recognition model provided in an embodiment of the present application. Exemplarily, the execution subject of the method 300 may be the VCU 11 or the electronic device 20 in FIG1, or the execution subject of the method 300 may be the chip or processor in the VCU 11, or the execution subject of the method 300 may be the chip or processor in the electronic device 20. For ease of description, the method 300 is described in detail by taking the execution subject as the VCU as an example.

S301. For each operating condition, select a preset number of sets of training data, each set of training data including a preset number of driving parameters and a benchmark analysis result.

It should be understood that in some embodiments, the training data may be obtained based on measured data.

Exemplarily, when the test vehicle is traveling in different operating conditions (e.g., congested operating conditions, urban operating conditions, suburban operating conditions, and high-speed operating conditions), first, the driving data of the test vehicle in different operating conditions are collected respectively, and a driving data set corresponding to each operating condition can be obtained. When collecting driving data, the collection can be performed at intervals of a preset time length, for example, the driving data is collected once every 1 second. For example, the driving data set can be {V _i |i＝1, 2, ..., i-1, i}. The driving data set can also be {a _i |i＝1, 2, ..., i-1, i}, where the interval between _Vi-1 and _Vi is 1 second, and the interval between a _i-1 and a _i is 1 second.

Secondly, in the driving data set corresponding to each operating state, a preset number of groups of driving data are selected, and each group of training data includes a preset number of driving data.

In the embodiment of the present application, the preset number of driving data in each set of training data is less than or equal to the total number of data in the driving data set. In other words, the duration corresponding to the preset number of driving data in each set of training data is less than or equal to the total duration corresponding to the driving data set. For example, if the driving data set is {V _i |i=1, 2, 3, 4, 5}, and the total duration corresponding to the driving data set is 5 seconds, then the duration corresponding to the preset number of driving data in each set of training data is less than or equal to 5 seconds.

When selecting a set of driving data from a driving data set, a driving data can be randomly selected from the driving data set as the starting data, for example, V ₃ is selected from the vehicle speed set as the starting data. After the starting data, n-1 driving data are selected continuously. For example, if the preset number is 4, V ₄ , V ₅ , and V ₆ are selected continuously after V _3. However, a special case may occur, that is, when n-1 driving data are selected continuously after the starting data, there are not enough n-1 driving data after the starting data in the driving data set. For example, the driving data set is {V _i |i＝1, 2, 3, 4, 5}, and there are only V ₄ and V ₅ after V _3. In this case, it is necessary to reselect a driving data as the starting data. According to this method, a preset number of driving data are selected for each working condition. For example, for 4 working conditions, 10 sets of training data are selected for each working condition, and each set of training data includes 4 driving data.

Then, based on a preset number of groups of driving data corresponding to each operating state, driving parameters are determined respectively, and a benchmark analysis result is marked for the driving parameter.

In the embodiment of the present application, the preset number of driving data corresponding to each operating state can be expressed as:

Wherein, n represents the preset number of groups, and m represents the number of driving data contained in each group.

In the embodiment of the present application, the driving parameters can be determined according to each set of driving data. For example, the driving parameters in the embodiment of the present application may include: maximum acceleration a _max , minimum acceleration a _min , average speed The parking time ratio is A/i, the low speed time ratio is B/i, the medium speed time ratio is C/i, and the high speed time ratio is D/i. The determined driving parameters can be expressed as:

Since the embodiments of the present application calculate driving parameters based on driving data under each operating condition, in a certain operating condition, the vehicle speed in the vehicle's driving data fluctuates within a small range. For example, in a congested operating condition, the vehicle speed fluctuates within the range of (0, 30 km/h], and in a high-speed operating condition, the vehicle speed fluctuates within the range of (60 km/h, +∞). Therefore, the values of the medium-speed time ratio and the high-speed time ratio calculated based on the driving data under the congested operating condition may be very small. Similarly, the values of the low-speed time ratio and the medium-speed time ratio calculated based on the driving data under the high-speed operating condition may be very small. These driving parameters with very small values have little effect on the subsequent judgment of the operating condition. Therefore, in some embodiments, in order to reduce the amount of calculation, the low-speed time ratio B/i, the medium-speed time ratio C/i, and the high-speed time ratio D/i are not calculated.

For example, the driving parameters in the embodiment of the present application may also include a maximum acceleration a _max , a minimum acceleration a _min , an average speed The parking time ratio A/i. The determined driving parameters can be expressed as:

Since the driving data in the embodiments of the present application are collected when the vehicle is driving in different operating conditions, in some operating conditions, such as high-speed operating conditions, urban operating conditions, etc., the vehicle will continue to drive and will not stop, so the parking time ratio A/i has little effect on the subsequent judgment of the operating condition. Therefore, in some embodiments, in order to reduce the amount of calculation, the parking time ratio A/i is not calculated.

For example, the driving parameters in the embodiment of the present application may also include a maximum acceleration a _max , a minimum acceleration a _min , an average speed The determined driving parameters can be expressed as:

It should be understood that the benchmark analysis result may represent the operating state corresponding to the driving parameter. For example, if the driving parameter is obtained based on the driving data of the congested operating state, the driving parameter is marked as "1", and "1" may represent the congested operating state. For example, in the congested operating state, the selected training data may be represented as:

For another example, if the driving parameter is obtained based on the driving data of the high-speed working condition, the driving parameter is marked as "2", and "2" can represent the high-speed working condition. For example, in the high-speed working condition, the selected training data can be represented as:

Of course, the training data selected based on the same operating condition can be labeled with different labels as long as they are distinguished from the training data selected based on other operating conditions.

Finally, based on the above method, training data under various working conditions can be obtained.

It should be understood that in some embodiments, the training data may also be obtained based on historical data.

For example, in typical cyclic operating conditions, such as NYCC, FTP, EUDC, etc., the data in these cyclic operating conditions can be quantified into a time-vehicle speed curve or a time-acceleration curve, and different time periods in the curve correspond to different operating conditions. For example, the horizontal axis of a curve can be divided into a first time period, a second time period, a third time period and a fourth time period, the first time period corresponds to a first operating condition, the second time period corresponds to a second operating condition, the third time period corresponds to a third operating condition, and the fourth time period corresponds to a fourth operating condition.

The VCU can select a preset number of driving data for each operating state based on the existing data in the international standard cycle operating conditions (the existing data is referred to as historical data in the embodiment of the present application). The selection of a preset number of training data for each operating state has been described in other embodiments and will not be repeated here.

It can be understood that the larger the number of preset groups of training data is, the better the training effect of the working condition recognition model will be. Therefore, in the embodiment of the present application, as much training data as possible is selected.

S302: Train the operating condition identification model based on the training data until the operating condition identification model converges.

It should be understood that when training the operating condition identification model, firstly, the training data obtained based on various operating conditions are respectively input into the initial operating condition identification model to obtain the training analysis results of the initial operating condition identification model.

Since the working condition identification model has not been trained yet at the initial stage, there will be certain deviations and errors between the training analysis results output at this time and the benchmark analysis results.

Secondly, the global error of this round of training is calculated based on the training analysis results and the benchmark analysis results.

It should be understood that after obtaining each training analysis result, the VCU can calculate the global error of this round of training based on each training analysis result and the corresponding benchmark analysis result, and determine whether the global error meets the preset conditions, such as whether the global error is less than 5%. Here, the preset conditions can be determined when training the working condition recognition model. For example, the preset condition can be set to be that the global error is less than a specific threshold, and the specific threshold can be a percentage value. The smaller the specific threshold, the more stable the working condition recognition model obtained after the final training is completed, and the higher the accuracy of the predicted working condition state will be.

In the embodiment of the present application, the global error refers to the loss function, which may include mean square error loss, mean absolute error loss, cross entropy loss function, etc. The embodiment of the present application does not limit the type of loss function.

Then, if the global error does not meet the preset conditions, the model parameters of the operating condition identification model are adjusted, and the operating condition identification model after the model parameters are adjusted is determined as the initial operating condition identification model.

It should be understood that when the global error of this round of training does not meet the preset conditions, for example, when the global error of this round of training is 10%, the model parameters of the operating condition identification model can be adjusted, and the operating condition identification model after the model parameters are adjusted is determined as the initial operating condition identification model, and then re-trained with the training data to repeatedly adjust the model parameters of the operating condition identification model so as to minimize the global error subsequently calculated based on the training analysis results and the corresponding benchmark analysis results until the final global error meets the preset conditions.

Finally, if the global error meets the preset conditions, it is determined that the operating condition identification model has converged.

It should be understood that when the global error of this round of training meets a preset condition, for example, when the global error of this round of training is less than 5%, it can be determined that the operating condition identification model has converged.

Through the training method of the embodiment of the present application, the trained operating condition recognition model can have a higher accuracy in identifying the operating condition state.

The embodiment of the present application can complete the training of the operating condition recognition model by using method 300, and the operating condition status of the vehicle can be determined by using the trained operating condition recognition model.

In the related art, when predicting the vehicle speed, the driver's driving intention is not taken into account, which may result in the prediction result that the driver should be in an accelerating state, but the prediction result is a deceleration, which reduces the accuracy of the prediction. When predicting the vehicle speed, the embodiment of the present application first predicts the acceleration based on the acceleration prediction model, and then predicts the vehicle speed based on the predicted acceleration. The acceleration can reflect the driver's driving intention.

In some embodiments, S203 further includes: obtaining a first speed and a first acceleration of the vehicle, the first speed being the speed of the vehicle at a second moment, the first acceleration being the acceleration of the vehicle at the second moment, and the second moment being any moment within the first time period;

Based on the acceleration prediction model corresponding to the operating condition, the first speed and the first acceleration, a second speed is determined, where the second speed is the speed of the vehicle at the first moment.

It should be understood that the first speed and the first acceleration belong to the driving data of the vehicle in the first time period, and the second moment is any moment in the first time period. Exemplarily, the vehicle speed set collected within 1 second to 5 seconds is {V ₁ V ₂ V ₃ V ₄ V ₅ }, and the acceleration set collected or calculated based on the vehicle speed set is {a ₁ a ₂ a ₃ a ₄ a ₅ }, then the first speed can be any speed in the vehicle speed set, the first acceleration can be any acceleration in the acceleration set, and the second moment is any moment in 1 second to 5 seconds.

It should be understood that the acceleration prediction model corresponding to the operating state means that each operating state corresponds to one acceleration prediction model. For example, if the operating state includes only 4 types, there are 4 acceleration prediction models, and if the acceleration prediction model includes 5 types, there are 5 acceleration prediction models.

The acceleration prediction model corresponding to the working state in the embodiment of the present application is used to predict the acceleration at a future moment. Since the acceleration of the vehicle during driving is generated based on the driver's actions such as shifting, braking, and stepping on the accelerator, the embodiment of the present application uses these actions of the driver to characterize the driver's driving intention, and therefore, the acceleration can reflect the driver's driving intention.

The first moment in the embodiment of the present application has been described in other embodiments and will not be repeated here. The second acceleration is the acceleration of the vehicle at the first moment, and the second speed is the speed of the vehicle at the first moment.

In an embodiment of the present application, the driving data of the vehicle in a first time period is input into a working condition identification model to obtain the working condition state of the vehicle, and then the driving data of the vehicle in the first time period is input into an acceleration prediction model corresponding to the working condition state to predict the vehicle speed.

Specifically, when predicting the vehicle, first, based on the first acceleration (the first acceleration can represent the current driver's driving intention) and the acceleration prediction model, the future driver's driving intention is predicted to obtain the second acceleration. Secondly, based on the first speed and the second acceleration, the second speed is obtained.

In the embodiment of the present application, since the vehicle speed is predicted based on a certain operating condition, and the acceleration can reflect the driver's driving intention, when predicting the vehicle speed, under a certain operating condition, the second acceleration is first obtained based on the first acceleration and the acceleration prediction model, and the driver's future driving intention can be predicted. Then, based on the second acceleration and the first speed, the second speed is determined. Therefore, when predicting the vehicle speed, the embodiment of the present application can narrow the numerical range of the predicted vehicle speed, determine the final prediction result within a smaller numerical range, and take into account the driver's future driving intention, thereby improving the accuracy of the predicted vehicle speed.

In some embodiments, the VCU determines the second speed based on the acceleration prediction model corresponding to the operating state, the first speed and the first acceleration, including:

The VCU processes the first acceleration through an acceleration prediction model corresponding to the operating condition to determine the second acceleration; and determines the second speed based on the first speed and the second acceleration.

In the embodiment of the present application, the acceleration prediction model corresponding to the working state is a Markov model. In the Markov model, the driving acceleration of the vehicle at a certain moment in the future has nothing to do with the historical driving data, but only with the current driving data. Therefore, when using the Markov model to predict the acceleration, it is not necessary to consider the historical driving data of the vehicle, which reduces the amount of calculation of the model when predicting the acceleration and improves the processing speed of the model.

In the embodiment of the present application, since acceleration can reflect the driver's driving intention, when predicting the vehicle speed, the second acceleration is first obtained based on the first acceleration and the acceleration prediction model corresponding to the working state, which can predict the driver's driving intention in the future, and then the second speed is determined based on the second acceleration and the first speed. Therefore, when predicting the vehicle speed, the embodiment of the present application takes into account the driver's driving intention in the future, thereby improving the accuracy of the predicted vehicle speed. In addition, the acceleration prediction model of the embodiment of the present application is a Markov model, and when predicting the second acceleration, the historical driving data of the vehicle does not need to be considered, which reduces the amount of calculation of the model when predicting the second acceleration and improves the processing speed of the model.

Before processing the first acceleration through the acceleration prediction model corresponding to the operating condition, it is necessary to first construct an acceleration prediction model corresponding to the operating condition. Please refer to Figure 4 for the method of constructing the acceleration prediction model. Figure 4 is a schematic flow chart of a method 400 for constructing an acceleration prediction model provided in an embodiment of the present application. The execution subject of method 400 may be the VCU11 or the electronic device 20 in Figure 1, or the execution subject of method 400 may be a chip or processor in VCU11, or the execution subject of method 400 may be a chip or processor in the electronic device 20. For the sake of ease of description, taking the execution subject as VCU as an example, method 400 is described in detail.

S401. For each operating state, based on the driving data in the training data, determine the state space of acceleration at time k and the state space of acceleration at time k+1.

It should be understood that the set of all possible acceleration values of the vehicle at time k is called the state space of acceleration at time k. The set of all possible acceleration values of the vehicle at time k+1 is called the state space of acceleration at time k+1. Each acceleration is called a state of the vehicle. The k moment and k+1 moment in the embodiments of the present application do not specifically refer to a certain moment, but only emphasize that the k+1 moment is the next moment of the k moment.

In the embodiment of the present application, the method for obtaining training data has been stated in S301 and will not be repeated here.

For example, the acceleration of a preset number of groups corresponding to an acquired working condition can be expressed as:

There are n*m accelerations in the matrix. For example, there are 3*4 accelerations in the matrix. For example, these 3*4 accelerations can be expressed as:

Among them, E, F, G, H and I all represent a state of acceleration.

It can be seen from the above matrix that under a certain working condition, all possible acceleration values of the vehicle at time k are E, F, G, H and I. Therefore, the set consisting of E, F, G, H and I is called the state space of acceleration at time K. Similarly, in the embodiment of the present application, the set consisting of E, F, G, H and I can also be called the state space of acceleration at time k+1.

Based on the above method, the state space of acceleration at time k and the state space of acceleration at time k+1 under each working condition can be determined.

S402, determining the probability of any state in the state space of acceleration at time k transferring to any state in the state space of acceleration at time k+1.

In the embodiment of the present application, the VCU may calculate the probability P _ij according to the following formula:

Where n represents the number of states in the state space.

_Fri represents the number of times the acceleration is transferred from a certain state when it is transferred from time k to time k+1. For example, in the following matrix:

When the acceleration transfers from time k to time k+1, the number of times it transfers from the E state is 5 times, namely, the acceleration transfers from the moment corresponding to the first row and first column to the moment corresponding to the first row and second column, which is the transfer from the E state; the acceleration transfers from the moment corresponding to the second row and second column to the moment corresponding to the second row and third column, which is the transfer from the E state; the acceleration transfers from the moment corresponding to the second row and third column to the moment corresponding to the second row and fourth column, which is the transfer from the E state; the acceleration transfers from the moment corresponding to the third row and second column to the moment corresponding to the third row and third column, which is the transfer from the E state; and the acceleration transfers from the moment corresponding to the third row and third column to the moment corresponding to the third row and fourth column, which is the transfer from the E state.

Wherein, _Frij represents the number of times the acceleration is transferred from one state to another state when it is transferred from time k to time k+1. For example, the number of times the acceleration is transferred from state E to state F when it is transferred from time k to time k+1 is 1, the number of times the acceleration is transferred from state E to state E when it is transferred from time k to time k+1 is 3, and the number of times the acceleration is transferred from state E to state G when it is transferred from time k to time k+1 is 3.

Wherein, _Pij represents the probability of any state in the state space of acceleration at time k transferring to any state in the state space of acceleration at time k+1. For example, the probability of the E state in the state space of acceleration at time k transferring to the F state in the state space of acceleration at time k+1 is 1/5, the probability of the E state in the state space of acceleration at time k transferring to the E state in the state space of acceleration at time k+1 is 3/5, and the probability of the E state in the state space of acceleration at time k transferring to the G state in the state space of acceleration at time k+1 is 1/5.

S403: Construct an acceleration prediction model based on the probability of any state in the state space of acceleration at time k transferring to any state in the state space of acceleration at time k+1.

It should be understood that based on S402, an acceleration transfer matrix can be constructed, and the constructed acceleration transfer matrix can be expressed as:

Among them, n is the number of states of acceleration, _P11 can represent the probability of acceleration transferring from the first state to the first state, _P1n can represent the probability of acceleration transferring from the first state to the nth state, _Pn1 can represent the probability of acceleration transferring from the nth state to the first state, and _Pnn can represent the probability of acceleration transferring from the nth state to the nth state.

By using the acceleration prediction model provided in the embodiment of the present application, when predicting the acceleration at time k+1, it is independent of the historical acceleration and is only related to the acceleration at time k. The vehicle's historical driving data does not need to be considered, which reduces the amount of calculation of the model when predicting acceleration and improves the processing speed of the model.

In the embodiment of the present application, the VCU processes the first acceleration input into the acceleration prediction model constructed by the method 400 to determine the second acceleration.

It should be understood that the processing process of VCU in the acceleration prediction model is: based on the acceleration transfer matrix and the first acceleration, determine the probability that the first acceleration corresponding to the second moment is transferred to other accelerations corresponding to the first moment, and determine the acceleration corresponding to the maximum probability as the second acceleration.

Exemplarily, assuming that the first acceleration is E, based on the acceleration transfer matrix and E, the probability that E corresponding to the first moment transfers to F corresponding to the second moment is 1/5, the probability that E transfers to E is 3/5, and the probability that E transfers to G is 1/5, and the acceleration E corresponding to the maximum probability 3/5 is determined as the second acceleration.

In the embodiment of the present application, after determining the second acceleration, the VCU may determine the second speed according to the following formula:

v ^k+1 =v ^k +ak ⁺¹ *Δt.

Wherein, v ^k+1 represents the second speed, v ^k represents the first speed, a ^k+1 represents the second acceleration, and Δt represents the interval between the first speed and the second speed. It should be understood that the interval here is the same as the interval of the driving data in the first time period in S201.

In some embodiments, the VCU predicts the second acceleration through the first acceleration; based on the first speed and the second acceleration, the second speed is predicted. It should be understood that in this embodiment, the first speed and the first acceleration correspond to the second moment, the second speed and the second acceleration correspond to the first moment, and the first moment and the second moment are adjacent moments. The embodiment of the present application uses the first acceleration of the second moment to predict the second acceleration of the adjacent moment (first moment), and then uses the speed of the second moment and the acceleration of the first moment to predict the second speed of the adjacent moment (first moment) for illustration. It is not limited to only using the first acceleration and the first speed to predict the speed of the next moment, and the first acceleration and the first speed can also be used to predict the speed of a future moment that is not adjacent to the second moment.

In some embodiments, the method 200 further includes:

The VCU inputs the second acceleration into the acceleration prediction model corresponding to the working condition to obtain a third acceleration, where the third acceleration is the acceleration of the vehicle at a third moment, which is later than the first moment;

A third speed is determined based on the second speed and the third acceleration, where the third speed is the speed of the vehicle at a third moment.

It should be understood that in the embodiment of the present application, the next moment of the second moment is the first moment, the next moment of the first moment is the third moment, and the second moment is not adjacent to the first moment. When determining the third acceleration, it is determined based on the second acceleration corresponding to the first moment. The specific method for determining the third acceleration is the same as the method for determining the second acceleration, which will not be repeated here.

When determining the third speed, it is determined based on the second speed and the third acceleration corresponding to the first moment. The specific method for determining the third speed is the same as the method for determining the second speed, which will not be repeated here.

It should be understood that the embodiment of the present application only uses three moments for illustration. In some embodiments, the VCU can predict the speed and acceleration of more than two moments later than the second moment based on the speed and acceleration at the second moment according to the prediction method of the embodiment of the present application.

In the embodiment of the present application, since acceleration can reflect the driver's driving intention, when predicting the vehicle speed, the third acceleration is first obtained based on the acceleration prediction model corresponding to the second acceleration and the operating condition, which can predict the driver's driving intention in the future, and then the third speed is determined based on the third acceleration and the second speed. Therefore, when predicting the vehicle speed, the embodiment of the present application takes into account the driver's driving intention in the future, thereby improving the accuracy of the predicted vehicle speed. In addition, the acceleration prediction model of the embodiment of the present application is a Markov model. When predicting the third acceleration, the vehicle's historical driving data does not need to be considered, thereby reducing the amount of calculation of the model when predicting the third acceleration and improving the processing speed of the model.

In some embodiments, S203 also includes: obtaining a first speed of the vehicle; and determining a second speed based on a vehicle speed prediction model corresponding to the operating state, the second speed being the vehicle speed at the first moment.

In the embodiment of the present application, the vehicle speed prediction model is a Markov model. The method of constructing a vehicle speed prediction model based on the Markov model is the same as the method of constructing an acceleration prediction model based on the Markov model, which will not be described in detail here.

The processing process of the VCU in the vehicle speed prediction model is: based on the vehicle speed transfer matrix and the first speed, determine the probability that the first speed at the second moment is transferred to other speeds corresponding to the first moment, and determine the speed corresponding to the maximum probability as the second speed.

It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

Please refer to FIG. 5 , which is a schematic diagram of the structure of a device 50 for predicting vehicle speed provided in an embodiment of the present application, the device comprising:

The processing unit 51 obtains the driving data of the vehicle in a first time period;

Determine the vehicle's operating status based on driving data;

Based on the operating state of the vehicle, a speed of the vehicle at a first moment is determined, where the first moment is later than the first time period.

The processing unit 51 is further used to determine a driving parameter of the vehicle in a first time period based on the driving data, the driving parameter comprising at least one of the following: a maximum acceleration, a minimum acceleration, an average vehicle speed, a parking time ratio, a low speed time ratio, a medium speed time ratio, and a high speed time ratio;

Determine the vehicle's operating status based on driving parameters.

The processing unit 51 is also used to input the driving parameters into the working condition identification model to obtain the working condition status of the vehicle.

The processing unit 51 is also used to select a preset number of sets of training data for each operating state, each set of training data including a preset number of driving parameters and a benchmark analysis result;

The operating condition identification model is trained based on the training data until the operating condition identification model converges.

The processing unit 51 is further used to obtain a first speed and a first acceleration of the vehicle, where the first speed is the speed of the vehicle at a second moment, the first acceleration is the acceleration of the vehicle at the second moment, and the second moment is any moment within the first time period;

Based on the acceleration prediction model corresponding to the operating condition, the first speed and the first acceleration, a second speed is determined, where the second speed is the speed of the vehicle at the first moment.

The processing unit 51 is further used to process the first acceleration by using an acceleration prediction model corresponding to the operating state to determine a second acceleration, where the second acceleration is the acceleration of the vehicle at the first moment;

Based on the first velocity and the second acceleration, a second velocity is determined.

The processing unit 51 is further used to input the second acceleration into the acceleration prediction model corresponding to the working condition to obtain a third acceleration, where the third acceleration is the acceleration of the vehicle at a third moment, and the third moment is later than the first moment;

A third speed is determined based on the second speed and the third acceleration, where the third speed is the speed of the vehicle at a third moment.

The acceleration prediction model corresponding to the operating state in the processing unit 51 is a Markov model.

The processing unit 51 is further used to determine, for each operating state, a state space of acceleration at time k and a state space of acceleration at time k+1 based on driving data in the training data;

Determine the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1;

The acceleration prediction model is constructed based on the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1.

The operating state in the processing unit 51 is any one of the following: a congested operating state, an urban operating state, a suburban operating state, and a high-speed operating state.

The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

As shown in FIG6 , an embodiment of the present application further provides an electronic device 60, including a memory 61, a processor 62, and a computer program 63 stored in the memory 61 and executable on the processor 62. When the processor 62 executes the computer program 63, the method for predicting vehicle speed in the above-mentioned embodiments is implemented.

The processor 62 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

The memory 61 may be an internal storage unit of the electronic device 60. The memory 61 may also be an external storage device of the electronic device 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 60. Further, the memory 61 may include both an internal storage unit of the electronic device 60 and an external storage device. The memory 61 is used to store computer programs and other programs and data required by the electronic device 60. The memory 61 may also be used to temporarily store data that has been output or is to be output.

The embodiment of the present application further provides a vehicle, including the electronic device 60 shown in FIG. 6 .

An embodiment of the present application provides a chip, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for predicting vehicle speed of the above-mentioned embodiments is implemented.

An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method for predicting vehicle speed in the above embodiments is implemented.

An embodiment of the present application provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal implements the method for predicting vehicle speed of the above embodiments when executing the computer program product.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may at least include: any entity or device that can carry the computer program code to the camera/electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk.

In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

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

Claims (14)

1. A method for predicting vehicle speed, comprising: Acquire driving data of the vehicle in a first time period; Determining the operating state of the vehicle based on the driving data; Based on the operating state of the vehicle, a vehicle speed of the vehicle at a first moment is determined, and the first moment is later than the first time period. 2. The method according to claim 1, characterized in that the determining the operating state of the vehicle based on the driving data comprises: Determine, based on the driving data, driving parameters of the vehicle in the first time period, the driving parameters comprising at least one of the following: maximum acceleration, minimum acceleration, average vehicle speed, parking time ratio, low speed time ratio, medium speed time ratio, and high speed time ratio; The operating state of the vehicle is determined according to the driving parameters. 3. The method according to claim 2, characterized in that the determining the operating state of the vehicle according to the driving parameters comprises: The driving parameters are input into a working condition identification model to obtain the working condition state of the vehicle. 4. The method according to claim 3, characterized in that the operating condition identification model is trained based on the following method: For each operating condition, a preset number of sets of training data are selected, each set of the training data includes a preset number of driving parameters and a benchmark analysis result; The operating condition identification model is trained based on the training data until the operating condition identification model converges. 5. The method according to any one of claims 1 to 4, characterized in that determining the speed of the vehicle at the first moment based on the operating state of the vehicle comprises: Acquire a first speed and a first acceleration of the vehicle, wherein the first speed is the speed of the vehicle at a second moment, the first acceleration is the acceleration of the vehicle at the second moment, and the second moment is any moment within the first time period; Based on the acceleration prediction model corresponding to the operating state, the first speed and the first acceleration, a second speed is determined, where the second speed is the speed of the vehicle at the first moment. 6. The method according to claim 5, characterized in that the determining the second speed based on the acceleration prediction model corresponding to the operating state, the first speed and the first acceleration comprises: Processing the first acceleration by using an acceleration prediction model corresponding to the operating state to determine the second acceleration, where the second acceleration is the acceleration of the vehicle at the first moment; Based on the first speed and the second acceleration, the second speed is determined. 7. The method according to claim 5, characterized in that the method further comprises: Inputting the second acceleration into an acceleration prediction model corresponding to the operating state to obtain a third acceleration, where the third acceleration is the acceleration of the vehicle at a third moment, and the third moment is later than the first moment; A third speed is determined based on the second speed and the third acceleration, where the third speed is the speed of the vehicle at the third moment. 8. The method according to claim 5, characterized in that the acceleration prediction model corresponding to the operating state is a Markov model. 9. The method according to claim 8, characterized in that the acceleration prediction model is constructed in a manner comprising: For each operating condition, based on the driving data in the training data, determine the state space of acceleration at time k and the state space of acceleration at time k+1; Determine the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1; The acceleration prediction model is constructed based on the probability of any state in the state space of the acceleration at time k transferring to any state in the state space of the acceleration at time k+1. 10. The method according to any one of claims 1 to 4 and 6 to 9, characterized in that the operating state is any one of the following: a congested operating state, an urban operating state, a suburban operating state and a high-speed operating state. 11. A device for predicting vehicle speed, comprising: a processing unit, the processing unit being used for: Acquire driving data of the vehicle in a first time period; Determining the operating state of the vehicle based on the driving data; Based on the operating state of the vehicle, a vehicle speed of the vehicle at a first moment is determined, and the first moment is later than the first time period. 12. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for predicting vehicle speed according to any one of claims 1 to 10. 13. An electronic device, characterized in that it comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting vehicle speed as claimed in any one of claims 1 to 10 is implemented. 14. A vehicle, comprising the electronic device according to claim 13.