CN113753063A - Vehicle driving instruction determination method, device, equipment and storage medium - Google Patents
Vehicle driving instruction determination method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113753063A CN113753063A CN202011322523.XA CN202011322523A CN113753063A CN 113753063 A CN113753063 A CN 113753063A CN 202011322523 A CN202011322523 A CN 202011322523A CN 113753063 A CN113753063 A CN 113753063A
- Authority
- CN
- China
- Prior art keywords
- instruction
- vehicle
- driving
- calibration
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
Abstract
The embodiment of the invention discloses a method, a device, equipment and a storage medium for determining a vehicle running instruction, wherein the method comprises the following steps: acquiring vehicle running characteristics; determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model; and sending the target instruction to a vehicle target execution mechanism so as to enable the target execution mechanism to control the vehicle to run according to the target instruction. According to the method provided by the embodiment of the invention, the instruction calibration table is generated based on the instruction calibration model trained in advance, so that the accuracy of the instruction calibration table is improved, and the accuracy of the target instruction determined based on the instruction calibration table is further improved.
Description
Technical Field
The embodiment of the invention relates to the field of vehicle control, in particular to a method, a device, equipment and a storage medium for determining a vehicle running instruction.
Background
With the development of unmanned driving technology, unmanned vehicles are also gradually widely used. In the control of the unmanned vehicle, the driving characteristics of the vehicle have an important influence on the control, and the vehicle driving characteristics which are as accurate as possible are obtained and applied to the control of the vehicle, so that the control precision is greatly improved.
The driving characteristics of the vehicle are generally determined in a calibrated manner. In the process of implementing the invention, the inventor finds that at least the following technical problems exist in the prior art: the traditional vehicle driving characteristic calibration method adopts a fixed program to enable a vehicle to drive along a fixed route, obtain the driving data of the vehicle, and then perform off-line processing on the data to obtain rough characteristic data. Compared with a mechanical method, the method has low efficiency, is difficult to realize the running characteristic calibration of the vehicle in a complex environment, and has poor adaptability.
Disclosure of Invention
The embodiment of the invention provides a method, a device, equipment and a storage medium for determining a vehicle running instruction, which are used for improving the accuracy of vehicle running characteristic calibration.
In a first aspect, an embodiment of the present invention provides a vehicle driving instruction determining method, including:
acquiring vehicle running characteristics;
determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and sending the target instruction to a vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
In a second aspect, an embodiment of the present invention further provides a vehicle travel instruction determining apparatus, including:
the vehicle driving characteristic module is used for acquiring vehicle driving characteristics;
the target instruction determining module is used for determining a target instruction according to the vehicle running characteristics and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and the vehicle running control module is used for sending the target instruction to the vehicle target execution mechanism so as to enable the target execution mechanism to control the vehicle to run according to the target instruction.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
storage means for storing one or more programs;
when the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle travel instruction determination method as provided by any of the embodiments of the present invention.
In a fourth aspect, the embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a vehicle travel instruction determination method as provided in any of the embodiments of the present invention.
The embodiment of the invention obtains the driving characteristics of the vehicle; determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model; the target instruction is sent to the vehicle target execution mechanism, so that the target execution mechanism controls the vehicle to run according to the target instruction, and the instruction calibration table is generated through the instruction calibration model based on pre-training, so that the accuracy of the instruction calibration table is improved, and the accuracy of the target instruction determined based on the instruction calibration table is further improved.
Drawings
Fig. 1 is a flowchart of a vehicle driving instruction determining method according to an embodiment of the present invention;
fig. 2 is a flowchart of a vehicle driving instruction determining method according to a second embodiment of the present invention;
fig. 3a is a schematic flowchart of a method for constructing an instruction calibration table according to a third embodiment of the present invention;
fig. 3b is a flowchart of a method for determining an action command according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a vehicle travel instruction determining apparatus according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of a computer device according to a fifth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a vehicle driving instruction determination method according to an embodiment of the present invention. The present embodiment is applicable to a case when the unmanned vehicle travel instruction is determined. The method may be performed by a vehicle travel instruction determining device, which may be implemented in software and/or hardware, for example, and may be configured in a computer device. As shown in fig. 1, the method includes:
and S110, acquiring the vehicle running characteristics.
In the present embodiment, the vehicle running characteristic is a characteristic of different dimensions during running of the vehicle. Such as at least two characteristics of the vehicle's speed, vehicle acceleration, vehicle weight, front wheel steering angle, vehicle pitch angle, road adhesion coefficient. The acceleration of the vehicle can be calculated by an upper controller according to the longitudinal deviation, and the speed, the weight, the front wheel rotation angle and the vehicle pitch angle of the vehicle can be determined by sensors on the vehicle; the road adhesion coefficient of the vehicle can be calculated from the slip ratio of the vehicle tires.
It is understood that the greater the number of vehicle travel characteristics, the more accurate the wood handle command determined based on the vehicle travel characteristics. Preferably, the vehicle speed, the vehicle acceleration, the vehicle weight, the front wheel rotation angle, the vehicle pitch angle and the road adhesion coefficient of the vehicle can be collected as the vehicle running characteristics, so that the determination of the target command is more accurate.
And S120, determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model.
In this embodiment, a calibration instruction table including a corresponding relationship between the calibration running characteristic and the calibration instruction is calibrated in advance, and after the vehicle running characteristic is obtained, the calibration instruction corresponding to the current vehicle running characteristic is determined as the target instruction by searching the pre-constructed instruction calibration table.
In one embodiment of the invention, the determining the target instruction according to the vehicle running characteristic and the pre-constructed instruction calibration table comprises the following steps: and acquiring a pre-constructed instruction calibration table, and performing interpolation processing on the preset associated characteristics to obtain a target instruction. Optionally, when searching for the pre-calibrated instruction calibration table, the calibrated running characteristics in the instruction calibration table may not be in one-to-one correspondence with the vehicle running characteristics of the current vehicle. The preset associated characteristics can be interpolated by adopting an interpolation processing mode, and the target instruction is finally obtained by adjusting according to other characteristics. The relevant characteristics can be characteristics which have a large influence relation with the target command, such as two characteristics of speed and acceleration. For example, assuming that the speed and the acceleration are used as the correlation characteristics, in the table look-up process, two-dimensional interpolation processing needs to be performed on the two factors of the speed and the acceleration, and data obtained by interpolating the speed and the acceleration is adjusted according to the vehicle weight, the front wheel rotation angle, the pitch angle and the slip ratio, so that the target instruction is finally obtained. The two-dimensional interpolation processing method may refer to a two-dimensional interpolation processing method in the prior art, and is not described herein again. And interpolation processing is carried out on the correlation characteristics to obtain the target instruction, so that the determination efficiency of the target instruction is improved.
Optionally, in order to obtain the calibrated driving characteristics in the complex scene as much as possible, the calibrated driving characteristics and the control commands corresponding to the calibrated driving characteristics may be obtained by a deep learning method. Specifically, still include: acquiring a calibration driving characteristic, and inputting the calibration driving characteristic into a pre-trained instruction calibration model to obtain a calibration instruction output by the instruction calibration model; and constructing an instruction calibration table based on the calibration driving characteristics and the calibration instruction. Optionally, the instruction calibration model is pre-constructed and trained, the trained instruction calibration model is used for performing instruction calibration on different calibrated driving characteristics to obtain corresponding calibration instructions, and a constructed instruction calibration table is obtained according to the calibrated driving characteristics and the corresponding calibration instructions. The calibrated driving characteristics can be obtained according to the acquired driving characteristics in an expanding mode, and the instruction calibration model can be constructed based on the existing convolutional neural network. For example, the instruction calibration model may be a deep learning model including three layers of convolutional neural networks. The instruction calibration table is built in a deep learning mode, calibration instructions corresponding to the calibration driving characteristics under a large number of complex scenes can be obtained without real measurement, and the technical problem of inaccurate instruction calibration caused by insufficient data acquisition is solved.
The calibrated running characteristics can be at least one of vehicle speed, vehicle acceleration, vehicle weight, front wheel rotation angle, vehicle pitch angle and road adhesion coefficient, and the more the number of the calibrated running characteristics is, the more accurate the instruction calibration result is. All the characteristics can be used as the calibrated driving characteristics, and the deep learning network can realize learning based on the speed, the acceleration, the weight, the corner of the front wheel, the pitch angle and the road adhesion coefficient of the vehicle and can learn more accurate associated characteristics.
And S130, sending the target instruction to the vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
In the embodiment, after the target command is determined, the target command is sent to the target execution mechanism to control the vehicle to run. Optionally, the target instruction is a throttle control instruction or a brake control instruction, and the name of the target execution mechanism is associated with the name of the target instruction. The target instruction may specifically include an instruction name and instruction execution parameters. For example, assuming the target command may be a fuel door, 10%, then the target actuator is a throttle control device. Assuming that the target command may be braking, 10%, the target actuator is the brake control device.
The embodiment of the invention obtains the driving characteristics of the vehicle; determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model; the target instruction is sent to the vehicle target execution mechanism, so that the target execution mechanism controls the vehicle to run according to the target instruction, and the instruction calibration table is generated through the instruction calibration model based on pre-training, so that the accuracy of the instruction calibration table is improved, and the accuracy of the target instruction determined based on the instruction calibration table is further improved.
Example two
Fig. 2 is a flowchart of a vehicle driving instruction determining method according to a second embodiment of the present invention. The embodiment is further optimized on the basis of the scheme. As shown in fig. 2, the method includes:
s210, collecting multiple groups of running sample information, and generating training sample data based on the running sample information, wherein each group of running sample information is vehicle running information under a set running scene in a set time period.
In this embodiment, real driving data of a plurality of unmanned vehicles is collected as driving sample information, training sample data is generated based on the driving sample information, a pre-constructed instruction calibration model is trained to obtain a trained instruction calibration model, and then an instruction calibration table is constructed based on the trained instruction calibration model.
Specifically, each set of travel sample information includes a plurality of travel sample characteristics and travel sample instructions. Specifically, the plurality of driving sample characteristics may be a vehicle speed, a vehicle acceleration, a vehicle weight, a front wheel rotation angle, a vehicle pitch angle, and a road adhesion coefficient, and the driving sample instruction may be an accelerator instruction or a brake instruction. In order to ensure the availability of the collected data, the average value of the numerical values of the driving sample characteristics collected at certain time intervals is used as the calibration value of the driving sample characteristics. In order to ensure the diversity of the collected data, the driving data of the vehicle in different driving scenes, such as driving data in driving scenes of straight lines, curves, broken roads, etc., needs to be collected.
In one embodiment of the present invention, generating training sample data based on travel sample information includes: for each driving sample characteristic in each group of driving sample information, preprocessing the driving sample characteristic based on the mean value of the driving sample characteristic to obtain a driving processing characteristic; and generating training sample data based on the driving processing characteristics and the driving sample instructions corresponding to each group of driving sample information. Optionally, in the collected driving sample information, noise may be generated in the collected driving sample information due to interference of a sensor, a line and the outside, and although mean value processing is subsequently performed, accuracy of data cannot be guaranteed. Therefore, the collected driving sample information needs to be denoised. For each driving sample feature, the data is not recorded for a single frame, but rather is averaged over a period of time. And setting the sampling time period as t, acquiring the average value of data in the period, taking the data with the difference value between two adjacent frames of data larger than 0.5 times the average value as suspicious data, and filtering the data with the larger difference value between the two frames of data and the average value so as to ensure the accuracy of the data. And generating training sample data by using the driving processing characteristics and the driving sample instructions obtained after the processing.
S220, training the pre-constructed instruction calibration model by using the training sample data to obtain the trained instruction calibration model.
And after obtaining training sample data, training the pre-constructed specified calibration model to obtain the trained specified calibration model. The structure of the designated calibration model can refer to the structure of a convolutional neural network in the prior art, the input is the driving processing characteristic, the labeled data is a driving sample instruction, a supervised learning method is adopted for training, a gradient descent method is adopted for parameter optimization in the training process, and finally the trained instruction calibration model is obtained.
And S230, acquiring the calibrated driving characteristics, and inputting the calibrated driving characteristics into a pre-trained instruction calibration model to obtain a calibration instruction output by the instruction calibration model.
And S240, constructing a command calibration table based on the calibration running characteristics and the calibration command.
And S250, acquiring the vehicle running characteristics.
And S260, determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table.
And S270, sending the target instruction to the vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
The embodiment of the invention generates training sample data based on the running sample information by collecting a plurality of groups of running sample information; training a pre-constructed instruction calibration model by using training sample data to obtain a trained instruction calibration model, reasonably and comprehensively determining a target instruction according to a plurality of vehicle running characteristics influencing vehicle running by constructing and training the instruction calibration model based on deep learning, continuously iterating the deep learning, continuously enriching information of an instruction calibration table, enabling the instruction calibration table to cover more scenes and enhancing the expansibility of the instruction calibration table.
EXAMPLE III
The present embodiment provides a preferred embodiment based on the above-described embodiments. In the embodiment, the core of the vehicle running instruction determination method lies in the construction of the instruction calibration table. Fig. 3a is a schematic flowchart of a method for constructing an instruction calibration table according to a third embodiment of the present invention. As shown in FIG. 3a, the instruction calibration table construction mainly comprises four parts of data acquisition, data drying, deep learning and calibration table construction.
In data acquisition, data acquired by a vehicle comprise information such as an accelerator, a brake, a vehicle speed, a vehicle acceleration, a vehicle weight, a front wheel rotation angle, a vehicle pitch angle and a road adhesion coefficient. The road surface adhesion coefficient can be adjusted
Where ω is the wheel angular velocity, r is the wheel radius. To ensure the availability of the collected data, the average of the collected values at certain time intervals is used as a calibration value for the vehicle characteristics. At t1The individual data over time can be expressed as:
in addition, the vehicle is required to run under different road conditions, including straight lines, curves, broken roads and the like, and data collected for multiple times are required to ensure the validity of the data. It should be noted that the data collected in this embodiment may be only real driving data of the unmanned vehicle, and specifically may be data of autonomous driving of the unmanned vehicle, or data of artificially controlling driving of the unmanned vehicle.
In order to make the instruction calibration model trained based on sample data more accurate, the acquired data needs to be denoised. For a specific data denoising method, reference may be made to the above embodiments, which are not described herein again.
In the deep learning stage, the ideal result is obtained by learning the internal rules and the representation levels of the sample data and continuously fitting a multi-level network. In the embodiment, the input of the command calibration model is collected data of an accelerator, a brake, a vehicle weight, a vehicle speed, a vehicle acceleration, a front wheel rotation angle, a vehicle pitch angle, a tire slip rate and the like, and the output is an accelerator and brake control command. In this embodiment, a supervised learning manner is adopted, a three-layer convolutional neural network is adopted, and a gradient descent method is adopted for optimization, so as to obtain weights and offsets in the network of each layer, and obtain a trained instruction calibration model.
For example, the linear units in the command calibration model can be represented as:
y＝wx+b
wherein y is a mark data matrix, w is a characteristic weight matrix, x is a characteristic data matrix, and b is a bias matrix, which can be specifically expressed as
The output of the deep learning is
The value of the mark is y, and the degree of proximity of deep learning to the mark is expressed as
In the learning process of data, a plurality of sample data exist, and the objective function of the whole model can be expressed as
And after the objective function is determined, optimizing by adopting a gradient descent method to obtain the weight and the offset in the network of each layer, and obtaining the trained instruction calibration model.
After the command calibration model is obtained through training, the driving sample characteristics are expanded to obtain calibrated driving characteristics, the calibrated driving characteristics are input into the command calibration model to obtain calibration commands corresponding to the calibrated driving characteristics, and a final command calibration table of the vehicle driving characteristics is obtained based on the results. The form is shown in table 1.
TABLE 1
Fig. 3b is a flowchart of a method for determining an action command according to a third embodiment of the present invention. As shown in fig. 3b, when a new controller outputs acceleration in combination with the state information of the vehicle, the action commands of the vehicle, i.e. the accelerator and the brake of the vehicle, can be obtained through the deep learning network.
The upper layer controller calculates the acceleration required by the vehicle according to the longitudinal deviation, then searches a response control command in a command calibration table according to the running state information of the vehicle, such as the vehicle weight, the vehicle speed, the vehicle rotation angle, the vehicle pitch angle, the tire slip rate and the like, and sends the corresponding command to an executing mechanism of the vehicle to complete the corresponding action.
In the table look-up process, partial information needs to be interpolated. In order to improve the efficiency of the algorithm, two-dimensional interpolation processing is carried out on two factors of the maximum speed v and the acceleration a influenced by the control command, and the data for carrying out interpolation on the speed and the acceleration are adjusted according to the vehicle weight, the front wheel rotating angle, the pitch angle and the slip ratio. The principle of taking approximate values is adopted in the adjusting process. The corresponding section of the vehicle weight g in the calibration table is g1And g2Judging g and g1 and g2The difference value of the accelerator and the brake can be obtained in a calibration table by adopting the reference value of g which is the smaller difference value, so that the effective control is realized on the basis of simplifying the algorithm. With the increase of deep learning network training data, the values in the calibration table are more and more, so that the final adjustment is more and more accurate.
The embodiment of the invention adopts a vehicle multi-dimensional driving characteristic calibration method, and takes the factors of vehicle speed, vehicle acceleration, vehicle weight, front wheel rotation angle, pitch angle, tire rotation rate and the like which influence the driving of the vehicle into consideration, thereby greatly enhancing the accuracy of calibration; the information of the calibration table can be enriched continuously through continuous iteration of deep learning, so that the calibration table can cover more scenes, and the expansibility of the calibration table is enhanced; in the application process of the calibration table, the efficiency of the algorithm and the calibration accuracy are considered, and the method of combining interpolation and approximation is adopted to search the response control instruction, so that the availability of the calibration table is greatly improved.
Example four
Fig. 4 is a schematic structural diagram of a vehicle travel instruction determining device according to a fourth embodiment of the present invention. The vehicle travel instruction determination device may be implemented in software and/or hardware, and may be configured in a computer device, for example. As shown in fig. 4, the apparatus includes a vehicle travel characteristic module 410, a target instruction determination module 420, and a vehicle travel control module 430, wherein:
a vehicle driving characteristics module 410 for obtaining vehicle driving characteristics;
a target instruction determining module 420, configured to determine a target instruction according to the vehicle driving characteristics and a pre-constructed instruction calibration table, where the instruction calibration table is generated based on a pre-trained instruction calibration model;
and the vehicle running control module 430 is used for sending the target instruction to the vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
According to the embodiment of the invention, the vehicle running characteristics are obtained through a vehicle running characteristic module; the target instruction determining module determines a target instruction according to the vehicle running characteristics and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model; the vehicle running control module sends the target instruction to the vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction, and the instruction calibration table is generated through the instruction calibration model based on pre-training, so that the accuracy of the instruction calibration table is improved, and the accuracy of the target instruction determined based on the instruction calibration table is further improved.
Optionally, on the basis of the foregoing scheme, the target instruction determining module 420 is specifically configured to:
and acquiring a pre-constructed instruction calibration table, and performing interpolation processing on the preset associated characteristics to obtain a target instruction.
Optionally, on the basis of the above scheme, the apparatus further includes an instruction calibration table building module, configured to:
acquiring a calibration driving characteristic, and inputting the calibration driving characteristic into a pre-trained instruction calibration model to obtain a calibration instruction output by the instruction calibration model;
and constructing an instruction calibration table based on the calibration driving characteristics and the calibration instruction.
Optionally, on the basis of the above scheme, the apparatus further includes an instruction calibration model training module, configured to:
acquiring multiple groups of running sample information, and generating training sample data based on the running sample information, wherein each group of running sample information is vehicle running information under a set running scene in a set time period;
and training the pre-constructed instruction calibration model by using the training sample data to obtain the trained instruction calibration model.
Optionally, on the basis of the above scheme, the driving sample information includes a plurality of driving sample characteristics and driving sample instructions, and the instruction calibration model training is specifically configured to:
for each driving sample characteristic in each group of driving sample information, preprocessing the driving sample characteristic based on the mean value of the driving sample characteristic to obtain a driving processing characteristic;
and generating training sample data based on the driving processing characteristics and the driving sample instructions corresponding to each group of driving sample information.
Optionally, on the basis of the above scheme, the target command is an accelerator control command or a brake control command.
Optionally, on the basis of the above scheme, the vehicle running characteristic includes at least two of a vehicle speed, an acceleration, a front wheel rotation angle, a vehicle weight, a vehicle pitch angle, and a slip ratio.
The vehicle running instruction determining device provided by the embodiment of the invention can execute the vehicle running instruction determining method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.
EXAMPLE five
Fig. 5 is a schematic structural diagram of a computer device according to a fifth embodiment of the present invention. FIG. 5 illustrates a block diagram of an exemplary computer device 512 suitable for use in implementing embodiments of the present invention. The computer device 512 shown in FIG. 5 is only an example and should not bring any limitations to the functionality or scope of use of embodiments of the present invention.
As shown in FIG. 5, computer device 512 is in the form of a general purpose computing device. Components of computer device 512 may include, but are not limited to: one or more processors 516, a system memory 528, and a bus 518 that couples the various system components including the system memory 528 and the processors 516.
The system memory 528 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)530 and/or cache memory 532. The computer device 512 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage 534 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 5, and commonly referred to as a "hard drive"). Although not shown in FIG. 5, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 518 through one or more data media interfaces. Memory 528 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 540 having a set (at least one) of program modules 542, including but not limited to an operating system, one or more application programs, other program modules, and program data, may be stored in, for example, the memory 528, each of which examples or some combination may include an implementation of a network environment. The program modules 542 generally perform the functions and/or methods of the described embodiments of the invention.
The computer device 512 may also communicate with one or more external devices 514 (e.g., keyboard, pointing device, display 524, etc.), with one or more devices that enable a user to interact with the computer device 512, and/or with any devices (e.g., network card, modem, etc.) that enable the computer device 512 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 522. Also, computer device 512 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via network adapter 520. As shown, the network adapter 520 communicates with the other modules of the computer device 512 via the bus 518. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the computer device 512, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processor 516 executes various functional applications and data processing by executing programs stored in the system memory 528, for example, implementing a vehicle travel instruction determination method provided by an embodiment of the present invention, the method includes:
acquiring vehicle running characteristics;
determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and sending the target instruction to a vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
Of course, those skilled in the art can understand that the processor may also implement the technical solution of the vehicle driving instruction determination method provided in any embodiment of the present invention.
EXAMPLE six
The sixth embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, where the computer program, when executed by a processor, implements a method for determining a vehicle travel instruction, the method including:
acquiring vehicle running characteristics;
determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and sending the target instruction to a vehicle target execution mechanism so that the target execution mechanism controls the vehicle to run according to the target instruction.
Of course, the computer-readable storage medium provided by the embodiments of the present invention, on which the computer program is stored, is not limited to the above method operations, and may also perform operations related to the vehicle travel instruction determination method provided by any embodiment of the present invention.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, or the like, as well as conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments illustrated herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A vehicle travel instruction determination method characterized by comprising:
acquiring vehicle running characteristics;
determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and sending the target instruction to a vehicle target execution mechanism so as to enable the target execution mechanism to control the vehicle to run according to the target instruction.
2. The method of claim 1, wherein determining a target command based on the vehicle travel characteristics and a pre-built command calibration table comprises:
and acquiring a pre-constructed instruction calibration table, and performing interpolation processing on the preset associated characteristics to obtain the target instruction.
3. The method of claim 1, further comprising:
acquiring a calibration driving characteristic, and inputting the calibration driving characteristic into a pre-trained instruction calibration model to obtain a calibration instruction output by the instruction calibration model;
and constructing the instruction calibration table based on the calibration running characteristics and the calibration instruction.
4. The method of claim 3, further comprising, prior to inputting the calibrated driving characteristics into a pre-trained commanded calibration model:
acquiring multiple groups of running sample information, and generating training sample data based on the running sample information, wherein each group of running sample information is vehicle running information under a set running scene in a set time period;
and training a pre-constructed instruction calibration model by using the training sample data to obtain a trained instruction calibration model.
5. The method according to claim 4, wherein the driving sample information includes a plurality of driving sample characteristics and driving sample instructions, and the generating training sample data based on the driving sample information includes:
for each driving sample feature in each set of driving sample information, preprocessing the driving sample feature based on a mean value of the driving sample feature to obtain a driving processing feature;
and generating training sample data based on the driving processing characteristics corresponding to each group of driving sample information and the driving sample instructions.
6. The method of any one of claims 1-5, wherein the target command is a throttle control command or a brake control command.
7. The method according to any one of claims 1-5, wherein the vehicle driving characteristics include at least two of vehicle speed, acceleration, front wheel steering angle, vehicle weight, vehicle pitch angle, and slip rate.
8. A vehicle travel instruction determining device characterized by comprising:
the vehicle driving characteristic module is used for acquiring vehicle driving characteristics;
the target instruction determining module is used for determining a target instruction according to the vehicle running characteristic and a pre-constructed instruction calibration table, wherein the instruction calibration table is generated based on a pre-trained instruction calibration model;
and the vehicle running control module is used for sending the target instruction to a vehicle target execution mechanism so as to enable the target execution mechanism to control the vehicle to run according to the target instruction.
9. A computer device, the device comprising:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the vehicle travel instruction determination method of any one of claims 1-7.
10. A computer-readable storage medium on which a computer program is stored, the program, when being executed by a processor, implementing a vehicle travel instruction determination method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322523.XA CN113753063B (en) | 2020-11-23 | 2020-11-23 | Method, device, equipment and storage medium for determining vehicle running instruction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322523.XA CN113753063B (en) | 2020-11-23 | 2020-11-23 | Method, device, equipment and storage medium for determining vehicle running instruction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113753063A true CN113753063A (en) | 2021-12-07 |
| CN113753063B CN113753063B (en) | 2023-05-09 |
Family
ID=78786079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011322523.XA Active CN113753063B (en) | 2020-11-23 | 2020-11-23 | Method, device, equipment and storage medium for determining vehicle running instruction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113753063B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104299236A (en) * | 2014-10-20 | 2015-01-21 | 中国科学技术大学先进技术研究院 | Target locating method based on scene calibration and interpolation combination |
| CN106990714A (en) * | 2017-06-05 | 2017-07-28 | 李德毅 | Adaptive Control Method and device based on deep learning |
| US20180113458A1 (en) * | 2016-10-24 | 2018-04-26 | International Business Machines Corporation | Method of effective driving behavior extraction using deep learning |
| CN108549911A (en) * | 2018-04-18 | 2018-09-18 | 清华大学 | Driver based on neural network turns to intervention recognition methods |
| CN109606383A (en) * | 2018-12-29 | 2019-04-12 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating model |
| CN109747659A (en) * | 2018-11-26 | 2019-05-14 | 北京汽车集团有限公司 | The control method and device of vehicle drive |
| CN110939726A (en) * | 2019-12-31 | 2020-03-31 | 潍柴动力股份有限公司 | Vehicle gear shifting control method and device |
| CN110967991A (en) * | 2018-09-30 | 2020-04-07 | 百度（美国）有限责任公司 | Method and device for determining vehicle control parameters, vehicle-mounted controller and unmanned vehicle |
| US20200172117A1 (en) * | 2018-11-29 | 2020-06-04 | Baidu Usa Llc | Predetermined calibration table-based vehicle throttle/brake assist system for l2 autonomous driving |
-
2020
- 2020-11-23 CN CN202011322523.XA patent/CN113753063B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104299236A (en) * | 2014-10-20 | 2015-01-21 | 中国科学技术大学先进技术研究院 | Target locating method based on scene calibration and interpolation combination |
| US20180113458A1 (en) * | 2016-10-24 | 2018-04-26 | International Business Machines Corporation | Method of effective driving behavior extraction using deep learning |
| CN106990714A (en) * | 2017-06-05 | 2017-07-28 | 李德毅 | Adaptive Control Method and device based on deep learning |
| CN108549911A (en) * | 2018-04-18 | 2018-09-18 | 清华大学 | Driver based on neural network turns to intervention recognition methods |
| CN110967991A (en) * | 2018-09-30 | 2020-04-07 | 百度（美国）有限责任公司 | Method and device for determining vehicle control parameters, vehicle-mounted controller and unmanned vehicle |
| CN109747659A (en) * | 2018-11-26 | 2019-05-14 | 北京汽车集团有限公司 | The control method and device of vehicle drive |
| US20200172117A1 (en) * | 2018-11-29 | 2020-06-04 | Baidu Usa Llc | Predetermined calibration table-based vehicle throttle/brake assist system for l2 autonomous driving |
| CN109606383A (en) * | 2018-12-29 | 2019-04-12 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating model |
| CN110939726A (en) * | 2019-12-31 | 2020-03-31 | 潍柴动力股份有限公司 | Vehicle gear shifting control method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113753063B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7086111B2 (en) | Feature extraction method based on deep learning used for LIDAR positioning of autonomous vehicles | |
| KR102292277B1 (en) | LIDAR localization inferring solutions using 3D CNN networks in autonomous vehicles | |
| JP7256758B2 (en) | LIDAR positioning with time smoothing using RNN and LSTM in autonomous vehicles | |
| US11704554B2 (en) | Automated training data extraction method for dynamic models for autonomous driving vehicles | |
| CN108995657A (en) | Operate the method and data processing system of automatic driving vehicle | |
| CN112651456B (en) | Unmanned vehicle control method based on RBF neural network | |
| CN110509916A (en) | A kind of body gesture antihunt means and system based on deep neural network | |
| WO2022247203A1 (en) | Method and apparatus for controlling autonomous vehicle, device, and storage medium | |
| WO2022252457A1 (en) | Autonomous driving control method, apparatus and device, and readable storage medium | |
| CN116560223A (en) | Intelligent vehicle dynamics model, ILQR control algorithm and track tracking controller based on physical information neural network | |
| CN113076988B (en) | Mobile robot vision SLAM key frame self-adaptive screening method based on neural network | |
| CN114021080A (en) | Trajectory prediction model training and trajectory prediction method, apparatus, device and medium | |
| WO2023045791A1 (en) | Lane keeping method and apparatus, device, medium, and system | |
| CN110779526B (en) | Path planning method, device and storage medium | |
| CN113753063B (en) | Method, device, equipment and storage medium for determining vehicle running instruction | |
| CN112758109B (en) | Transverse tracking steady state deviation compensation method and device | |
| CN113822593A (en) | Security situation assessment method and device, storage medium and electronic equipment | |
| Aghli et al. | Online system identification and calibration of dynamic models for autonomous ground vehicles | |
| Yamauchi et al. | Adaptive identification method of vehicle modeling according to the fluctuation of road and running situation in autonomous driving | |
| Schäfke et al. | Transformer Neural Networks for Maximum Friction Coefficient Estimation of Tire-Road Contact using Onboard Vehicle Sensors | |
| CN115503501A (en) | Hub motor torque distribution method and device based on artificial intelligence | |
| CN117455004A (en) | Intelligent learning method and device for automatic driving model of vehicle | |
| CN116430852A (en) | Motion control method, system and device for four-wheel differential mobile robot | |
| CN117369266A (en) | Path tracking control system and method based on pre-steering NMPC | |
| CN118011824A (en) | Self-adaptive method for automatic driving vehicle path tracking |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |