CN107450511B - Assess method, apparatus, equipment and the computer storage medium of vehicle control model - Google Patents
Assess method, apparatus, equipment and the computer storage medium of vehicle control model Download PDFInfo
- Publication number
- CN107450511B CN107450511B CN201710509075.6A CN201710509075A CN107450511B CN 107450511 B CN107450511 B CN 107450511B CN 201710509075 A CN201710509075 A CN 201710509075A CN 107450511 B CN107450511 B CN 107450511B
- Authority
- CN
- China
- Prior art keywords
- parameter value
- timestamp
- vehicle
- evaluation parameter
- value
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention provides a kind of method, apparatus, equipment and computer storage medium for assessing vehicle control model, wherein the method for assessment vehicle control model includes: acquisition vehicle operation data；The output data that vehicle control model is directed to the vehicle operation data is obtained, determines assessment parameter value as model evaluation parameter value according to output data；And assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data；Using the mean square deviation between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value, the vehicle control model is assessed.Provided technical solution through the invention, realization is objective, accurately assesses vehicle control model.
Description
[technical field]
The present invention relates to driving technology field more particularly to it is a kind of assess the method, apparatus of vehicle control model, equipment and
Computer storage medium.
[background technique]
In recent years, with the rapid development of deep learning, the further investigation of artificial intelligence, auto industry produces revolution
The variation of property.The automatic Pilot that vehicle is realized by deep learning is a main direction of studying in driving field, and how right
Carrying out assessment by deep learning vehicle control model obtained is the important guarantee for realizing vehicle safety automatic Pilot.
[summary of the invention]
In view of this, the present invention provides a kind of method, apparatus for assessing vehicle control model, equipment and computer storages
Medium, realization is objective, accurately assesses vehicle control model.
Used technical solution is to provide a kind of method for assessing vehicle control model to the present invention in order to solve the technical problem,
The described method includes: acquisition vehicle operation data；The output data that vehicle control model is directed to the vehicle operation data is obtained,
Determine assessment parameter value as model evaluation parameter value according to output data；And assessment ginseng is calculated according to vehicle operation data
Numerical value is as criterion evaluation parameter value；Using between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value
Mean square deviation, assess the vehicle control model.
According to one preferred embodiment of the present invention, the assessment parameter is traveling curvature, and the vehicle control model is vehicle
Lateral control model, the output data are steering wheel angle.
According to one preferred embodiment of the present invention, the vehicle operation data includes: vehicle position data and each vehicle position
Set the corresponding timestamp of data.
According to one preferred embodiment of the present invention, described to calculate assessment parameter value as criterion evaluation according to vehicle operation data
Parameter value includes: that it is corresponding to calculate each timestamp according to vehicle position data and the corresponding timestamp of each vehicle position data
Travel curvature value；Traveling curvature value corresponding to each timestamp carries out interpolation calculation respectively, obtains list where corresponding to each timestamp
Multiple traveling curvature values of position time interval；Respectively from multiple traveling curvature values of each timestamp unit one belongs to time interval of correspondence
In select a traveling curvature value as corresponding each timestamp；The traveling curvature value of preset threshold requirement will be met as correspondence
The criterion evaluation parameter value of each timestamp.
According to one preferred embodiment of the present invention, described to determine assessment parameter value as model evaluation parameter according to output data
Value includes: the traveling curvature value that the steering wheel angle of each timestamp of correspondence is converted to corresponding each timestamp using preset model；
The traveling curvature value of preset threshold requirement will be met as the model evaluation parameter value of corresponding each timestamp.
According to one preferred embodiment of the present invention, described to utilize the corresponding model evaluation parameter value of same timestamp and standard
Mean square deviation between assessment parameter value is assessed before the vehicle control model, further includes: judges the model evaluation parameter value
Number whether be greater than preset threshold, if more than then calculating the model evaluation parameter value based on the timestamp and standard being commented
Estimate the mean square deviation between parameter value, does not otherwise calculate.
The present invention in order to solve the technical problem and the technical solution adopted is that provide it is a kind of assess vehicle control model device,
Described device includes: acquisition unit, for acquiring vehicle operation data；Determination unit is directed to for obtaining vehicle control model
The output data of the vehicle operation data determines assessment parameter value as model evaluation parameter value according to output data；And
Assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data；Assessment unit, for utilizing same timestamp
Mean square deviation between corresponding model evaluation parameter value and criterion evaluation parameter value assesses the vehicle control model.
According to one preferred embodiment of the present invention, the assessment parameter is traveling curvature, and the vehicle control model is vehicle
Lateral control model, the output data are steering wheel angle.
According to one preferred embodiment of the present invention, acquisition unit vehicle operation data collected includes: vehicle location
Data and the corresponding timestamp of each vehicle position data.
According to one preferred embodiment of the present invention, the determination unit is for calculating assessment parameter according to vehicle operation data
It is specific to execute when value is as criterion evaluation parameter value: according to vehicle position data and each vehicle position data corresponding time
Stamp, calculates the corresponding traveling curvature value of each timestamp；Traveling curvature value corresponding to each timestamp carries out interpolation calculation respectively, obtains
To multiple traveling curvature values of each timestamp unit one belongs to time interval of correspondence；Respectively from correspondence each timestamp unit one belongs to time
A traveling curvature value as corresponding each timestamp is selected in multiple traveling curvature values in section；Preset threshold requirement will be met
Criterion evaluation parameter value of the traveling curvature value as each timestamp of correspondence.
According to one preferred embodiment of the present invention, the determination unit is for determining that assessment parameter value is made according to output data
It is specific to execute: the steering wheel angle of each timestamp of correspondence being converted into correspondence using preset model when for model evaluation parameter value
The traveling curvature value of each timestamp；The traveling curvature value of preset threshold requirement will be met as the model evaluation of corresponding each timestamp
Parameter value.
According to one preferred embodiment of the present invention, the assessment unit is utilizing the corresponding model evaluation parameter of same timestamp
Before mean square deviation between value and criterion evaluation parameter value assesses the vehicle control model, also executes: judging that the model is commented
Whether the number for estimating parameter value is greater than preset threshold, if more than then the model evaluation parameter value is calculated based on the timestamp
Mean square deviation between criterion evaluation parameter value, does not otherwise calculate.
As can be seen from the above technical solutions, the present invention is based on same timestamps, calculate criterion evaluation parameter value and mould
Type assesses the mean square deviation between parameter value, assesses vehicle control model using obtained mean square deviation, realization is objective, accurately comments
Estimate vehicle control model.
[Detailed description of the invention]
Fig. 1 is the method flow diagram for the assessment vehicle control model that one embodiment of the invention provides.
Fig. 2 is the structure drawing of device for the assessment vehicle control model that one embodiment of the invention provides.
Fig. 3 is the block diagram for the computer system/server that one embodiment of the invention provides.
[specific embodiment]
To make the objectives, technical solutions, and advantages of the present invention clearer, right in the following with reference to the drawings and specific embodiments
The present invention is described in detail.
The term used in embodiments of the present invention is only to be not intended to be limiting merely for for the purpose of describing particular embodiments
The present invention.In the embodiment of the present invention and the "an" of singular used in the attached claims, " described " and "the"
It is also intended to including most forms, unless the context clearly indicates other meaning.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, indicate
There may be three kinds of relationships, for example, A and/or B, can indicate: individualism A, exist simultaneously A and B, individualism B these three
Situation.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".
Depending on context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determination " or " in response to detection ".Similarly, depend on context, phrase " if it is determined that " or " if detection
(condition or event of statement) " can be construed to " when determining " or " in response to determination " or " when the detection (condition of statement
Or event) when " or " in response to detection (condition or event of statement) ".
In the present invention, on the one hand using it is selected assessment parameter normal data and model data between mean square deviation come
Assess vehicle control model, on the other hand, choose more pervasive, objective assessment parameter, thus realize it is objective, accurately assess
Vehicle control model.
If the assessment parameters such as direct use direction disk angle, vehicle shift center line distance carry out commenting for vehicle control model
When estimating, the influence of the intrinsic parameter of vehicle will receive.In embodiments of the present invention, preferably traveling curvature is as vehicle lateral control mould
The assessment parameter of type.Traveling curvature represents the angle of turn of vehicle in the process of moving, with turning radius in vehicle running path
Correlation, therefore by vehicle, inherently parameter is not influenced traveling curvature, chooses traveling curvature as vehicle lateral control model
Assessment parameter can be more objective, accurate.Traveling curvature is hereinafter chosen as assessment parameter, to vehicle lateral control model
Evaluation process be described.
Fig. 1 is the method flow diagram for the assessment vehicle control model that one embodiment of the invention provides.As shown in fig. 1, institute
The method of stating includes:
In 101, vehicle operation data is acquired.
In this step, vehicle operation data collected includes that vehicle position data and each vehicle position data are corresponding
Timestamp.
Wherein, vehicle position data is the longitude and latitude or coordinate of vehicle present position in the process of moving；Each vehicle location
At the time of the corresponding timestamp of data is that vehicle is corresponding when driving to each position.It is understood that vehicle is travelling
At the time of being able to record corresponding to passed through position and process each position in the process, therefore this step just can be according to vehicle
The data of record collect the vehicle row including vehicle position data and the corresponding timestamp of each vehicle position data
Sail data.
In 102, the output data that vehicle control model is directed to the vehicle operation data is obtained, it is true according to output data
Accepted opinion estimates parameter value as model evaluation parameter value；And assessment parameter value is calculated according to vehicle operation data and is commented as standard
Estimate parameter value.
Firstly, explaining to term involved in this step: vehicle control model is vehicle lateral control model,
For controlling the steering of vehicle；Output data is steering wheel angle；Assessing parameter is traveling curvature, and assessment parameter value is that traveling is bent
Rate value.
In this step, after obtaining vehicle control model for the output data of vehicle operation data, according to acquired
Output data determines assessment parameter value as model evaluation parameter value.
Specifically, after the steering wheel angle exported when obtaining vehicle driving to position by vehicle control model,
According to acquired steering wheel angle, the steering wheel angle of corresponding each position is converted into corresponding each position using preset model
Travel curvature value.It wherein, can be Ackermann model for the model of conversion direction disk corner and traveling curvature value, it can also
Think other switching networks.Since in acquired vehicle operation data, vehicle driving position is corresponding with the vehicle driving moment
, i.e., vehicle driving position is corresponding with timestamp, therefore the traveling curvature value for the correspondence each position being converted to is namely corresponding
The traveling curvature value of each timestamp.
After the traveling curvature value for obtaining each timestamp of correspondence exported by vehicle control model, default threshold will be unsatisfactory for
The traveling curvature value that value requires is filtered as exceptional value, and the traveling curvature value for meeting preset threshold requirement is commented as model
Estimate parameter value.
In this step, assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data.
Specifically, vehicle position data and the corresponding timestamp of each vehicle position data in foundation vehicle operation data,
Traveling curvature value corresponding to each timestamp is calculated using preset formula.Wherein, the preset formula of vehicle driving curvature value is calculated
It is as follows:
In above-mentioned formula: k represents traveling curvature value；X represents longitude, and y represents latitude, i.e. x, y is practical longitude and latitude warp
It obtains after crossing coordinate transform to indicate vehicle location；T represents timestamp.
After obtaining the corresponding traveling curvature value of each timestamp, respectively to the corresponding traveling curvature of obtained each timestamp
Value carries out interpolation calculation, obtains the multiple traveling curvature values for corresponding to each timestamp unit one belongs to time interval, then respectively from correspondence
One is selected in multiple traveling curvature values of each timestamp unit one belongs to time interval, the traveling curvature as each timestamp of correspondence
Value.It is understood that being chosen obtained by interpolation calculation since traveling curvature value changes less in unit time section
It is ok to any one in multiple traveling curvature values.In the present invention, the obtained multiple rows of interpolation calculation are preferentially chosen
Sail traveling curvature value of second traveling curvature value as corresponding timestamp (t) in curvature value.
For example, if it is k, benefit that traveling curvature value of the vehicle at certain timestamp (t), certain position (x, y), which is calculated,
Carry out interpolation calculation with obtained traveling curvature value k, with obtain [t, t+1) multiple traveling curvature value (k in this second1,
k2,k3....kn), then from by choosing one in the obtained multiple curvature of interpolation calculation as the corresponding timestamp (t)
Curvature value is travelled, such as chooses k2For the traveling curvature value of the corresponding timestamp (t).
Wherein, the traveling curvature value of preset threshold requirement will be met in the traveling curvature value of each timestamp of obtained correspondence
Filtering will only meet the traveling curvature value of preset threshold requirement as the standard precompensation parameter value of each timestamp.
For example, if calculate each timestamp of correspondence obtained traveling curvature value be respectively (15:36, -0.23),
(15:50,0.72) and (16:03,0.4), wherein previous item represents timestamp, latter represents traveling curvature value.If default
The requirement of threshold value is traveling curvature value less than 0.5, then 0.72 filtering in curvature value will be travelled, by -0.23 and 0.4 as mark
Quasi- assessment parameter value.
In 103, using square between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value
Difference assesses the vehicle control model.
Before this step, it is also necessary to judge whether the number of acquired model evaluation parameter value meets preset threshold
It is required that whether the number of i.e. judgment models assessment parameter value is greater than preset threshold.If the quantity of acquired model evaluation parameter value
It, can be because data volume be very few and influences the Evaluation accuracy of model, it is therefore desirable to which judgment models assess the number of parameter value when very few.
If the number of acquired model evaluation parameter value is greater than preset threshold, commented into the model is calculated based on the timestamp
The mean square deviation between parameter value and criterion evaluation parameter value is estimated, otherwise without calculating.Wherein, model evaluation parameter value is default
Threshold value can be set to the half of normal data number.
Since vehicle control model belongs to regression problem, the mean square deviation between observation and predicted value can be passed through
To measure the precision of prediction of vehicle control model.Therefore in this step, using criterion evaluation parameter value as observation, by model
Parameter value is assessed as predicted value, it is horizontal that the mean square deviation by calculating criterion evaluation parameter value and model evaluation parameter value assesses vehicle
To Controlling model.
In this step, the criterion evaluation parameter value and model evaluation parameter of corresponding each timestamp are first determined based on timestamp
Value, then judge criterion evaluation parameter value and model evaluation parameter value whether are existed simultaneously under each timestamp.If sometime stabbing
Under exist simultaneously criterion evaluation parameter value and model evaluation parameter value, then by the criterion evaluation parameter value and model evaluation parameter value
It is determined as corresponding to the criterion evaluation parameter value of the timestamp and model evaluation parameter value, if the sometime lower only criterion evaluation of stamp
Parameter value perhaps only model evaluation parameter value when then by the criterion evaluation parameter value or model evaluation parameter under the timestamp
Value is filtered.
For example, if sometime stamp is 15:36, the criterion evaluation parameter value of the corresponding timestamp is 0.3, and model is commented
Estimating parameter value is 0.28, then criterion evaluation parameter value corresponding to timestamp 15:36 is 0.3, and model evaluation parameter value is 0.28.
If under the timestamp, only existing criterion evaluation parameter value 0.3 or only existing model evaluation parameter value 0.28, then by the timestamp
Corresponding criterion evaluation parameter value or model evaluation parameter value is filtered.
Specifically, the calculation formula of the mean square deviation between criterion evaluation parameter value and model evaluation parameter value is being calculated such as
Under:
In formula: MSE represents mean square deviation, and n indicates data amount check, curvstandardCriterion evaluation parameter value is represented,
curvmodelRepresentative model assesses parameter value.
If showing vehicle when smaller according to criterion evaluation parameter value and the obtained mean square deviation of model evaluation parameter value calculation
Lateral control model is more accurate；If obtained according to criterion evaluation parameter value and model criteria assessment parameter value calculation
When variance is larger, then show that vehicle lateral control model accuracy is lower.
Fig. 2 is the structure drawing of device for the assessment vehicle control model that one embodiment of the invention provides, as shown in Figure 2, institute
Stating device includes: acquisition unit 21, determination unit 22 and assessment unit 23.
Acquisition unit 21, for acquiring vehicle operation data.
The vehicle operation data collected of acquisition unit 21 includes that vehicle position data and each vehicle position data are corresponding
Timestamp.
Wherein, vehicle position data is the longitude and latitude or coordinate of vehicle present position in the process of moving；Each vehicle location
At the time of the corresponding timestamp of data is that vehicle is corresponding when driving to each position.It is understood that vehicle is travelling
At the time of being able to record corresponding to passed through position and process each position in the process, therefore acquisition unit 21 just being capable of basis
The data of vehicle registration collect the vehicle including vehicle position data and the corresponding timestamp of each vehicle position data
Running data.
Determination unit 22, the output data for being directed to the vehicle operation data for obtaining vehicle control model, according to defeated
Data determine assessment parameter value as model evaluation parameter value out；And assessment parameter value is calculated according to vehicle operation data and is made
For criterion evaluation parameter value.
Firstly, explaining to term involved in determination unit 22: vehicle control model is vehicle lateral control mould
Type, for controlling the steering of vehicle；Output data is steering wheel angle；Assessing parameter is traveling curvature, and assessment parameter value is row
Sail curvature value.
After determination unit 22 obtains vehicle control model for the output data of vehicle operation data, according to acquired defeated
Data determine assessment parameter value as model evaluation parameter value out.
Specifically, it is determined that the direction that unit 22 is exported when obtaining vehicle driving to position by vehicle control model
After disk corner, according to acquired steering wheel angle, the steering wheel angle of corresponding each position is converted to pair using preset model
Answer the traveling curvature value of each position.Wherein it is determined that unit 22 for conversion direction disk corner and travel the model of curvature value can be with
For Ackermann model, or other switching networks.Due in acquired vehicle operation data, vehicle driving position
With the vehicle driving moment be it is corresponding, i.e., vehicle driving position is corresponding with timestamp, therefore the correspondence each position being converted to
Traveling curvature value namely corresponds to the traveling curvature value of each timestamp.
Determination unit 22, will not after the traveling curvature value for obtaining each timestamp of correspondence exported by vehicle control model
The traveling curvature value for meeting preset threshold requirement is filtered as exceptional value, will meet the traveling curvature value of preset threshold requirement
As model evaluation parameter value.
Determination unit 22 calculates assessment parameter value as criterion evaluation parameter value according to vehicle operation data.
Specifically, it is determined that unit 22 is corresponding according to vehicle position data in vehicle operation data and each vehicle position data
Timestamp, calculate traveling curvature value corresponding to each timestamp using preset formula.Wherein it is determined that unit 22 calculates vehicle row
The preset formula for sailing curvature value is as follows:
In above-mentioned formula: k represents traveling curvature value；X represents longitude, and y represents latitude, i.e. x, y is practical longitude and latitude warp
It obtains after crossing coordinate transform to indicate vehicle location；T represents timestamp.
Determination unit 22 is corresponding to obtained each timestamp respectively after obtaining the corresponding traveling curvature value of each timestamp
Traveling curvature value carry out interpolation calculation, obtain the multiple traveling curvature values for corresponding to each timestamp unit one belongs to time interval, then
One is selected from multiple traveling curvature values of each timestamp unit one belongs to time interval of correspondence respectively, as each timestamp of correspondence
Traveling curvature value.It is understood that since traveling curvature value changes in unit time section less, it is thus determined that unit
22 any one chosen in multiple traveling curvature values as obtained by interpolation calculation are ok.In the present invention, determination unit 22
Second traveling curvature value is as corresponding timestamp (t) in the preferential selection obtained multiple traveling curvature values of interpolation calculation
Travel curvature value.
For example, however, it is determined that it is bent that traveling of the vehicle at certain timestamp (t), certain position (x, y) is calculated in unit 22
Rate value is k, carries out interpolation calculation using obtained traveling curvature value k, with acquisition [t, t+1) multiple travelings in this second
Curvature value (k1,k2,k3....kn), then determination unit 22 is from by choosing one in the obtained multiple curvature of interpolation calculation
As the traveling curvature value of the corresponding timestamp (t), such as choose k2For the traveling curvature value of the corresponding timestamp (t).
Wherein it is determined that unit 22 will meet preset threshold requirement in the traveling curvature value of each timestamp of obtained correspondence
Curvature value filtering is travelled, will only meet the traveling curvature value of preset threshold requirement as the standard precompensation parameter value of each timestamp.
For example, however, it is determined that unit 22 calculate each timestamp of correspondence obtained traveling curvature value be respectively (15:
36, -0.23), (15:50,0.72) and (16:03,0.4), wherein previous item represents timestamp, latter represents traveling curvature
Value.If the requirement of preset threshold is traveling curvature value less than 0.5, it is determined that 0.72 filtering that unit 22 will travel in curvature value,
Criterion evaluation parameter value is used as by -0.23 and 0.4.
Assessment unit 23, for using between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value
Mean square deviation, assess the vehicle control model.
Assessment unit 23 is before being assessed, it is also necessary to judge whether the number of acquired model evaluation parameter value is full
Whether the number of the requirement of sufficient preset threshold, i.e. judgment models assessment parameter value is greater than preset threshold.If acquired model evaluation
It, can be because data volume be very few and influences the Evaluation accuracy of model, it is therefore desirable to which judgment models are assessed when the quantity of parameter value is very few
The number of parameter value.If the number of acquired model evaluation parameter value is greater than preset threshold, assessment unit 23 is based on institute
It states timestamp and calculates mean square deviation between the model evaluation parameter value and criterion evaluation parameter value, otherwise without calculating.Its
In, the preset threshold of model evaluation parameter value can be set to the half of normal data number.
Since vehicle control model belongs to regression problem, the mean square deviation between observation and predicted value can be passed through
To measure the precision of prediction of vehicle control model.Therefore assessment unit 23 is using criterion evaluation parameter value as observation, by model
Parameter value is assessed as predicted value, it is horizontal that the mean square deviation by calculating criterion evaluation parameter value and model evaluation parameter value assesses vehicle
To Controlling model.
Assessment unit 23 first determines the criterion evaluation parameter value and model evaluation parameter of corresponding each timestamp based on timestamp
Value, then judge criterion evaluation parameter value and model evaluation parameter value whether are existed simultaneously under each timestamp.If sometime stabbing
Under exist simultaneously criterion evaluation parameter value and model evaluation parameter value, then assessment unit 23 is by the criterion evaluation parameter value and model
Assessment parameter value is determined as corresponding to the criterion evaluation parameter value of the timestamp and model evaluation parameter value, if sometime stabbing lower
When having criterion evaluation parameter value or only model evaluation parameter value, then assessment unit 23 joins the criterion evaluation under the timestamp
Numerical value or model evaluation parameter value are filtered.
For example, if sometime stamp is 15:36, the criterion evaluation parameter value of the corresponding timestamp is 0.3, and model is commented
Estimating parameter value is 0.28, then criterion evaluation parameter value corresponding to timestamp 15:36 is 0.3, and model evaluation parameter value is 0.28.
If under the timestamp, only existing criterion evaluation parameter value 0.3 or only existing model evaluation parameter value 0.28, then by the timestamp
Corresponding criterion evaluation parameter value or model evaluation parameter value is filtered.
Specifically, assessment unit 23 is in the meter for calculating the mean square deviation between criterion evaluation parameter value and model evaluation parameter value
It is as follows to calculate formula:
In formula: MSE represents mean square deviation, and n indicates data amount check, curvstandardCriterion evaluation parameter value is represented,
curvmodelRepresentative model assesses parameter value.
If assessment unit 23 is smaller according to criterion evaluation parameter value and the obtained mean square deviation of model evaluation parameter value calculation
When, then show that vehicle lateral control model is more accurate；If assessment unit 23 is commented according to criterion evaluation parameter value with model criteria
Estimate the obtained mean square deviation of parameter value calculation it is larger when, then show that vehicle lateral control model accuracy is lower.
It is understood that according to the type of the vehicle control assessed model, choosing corresponding assessment parameter can be into
Row model evaluation.When to assessment vehicle longitudinal control model, then chooses car speed or vehicle acceleration is used as assessment parameter,
According to the mean square deviation between criterion evaluation parameter value and model evaluation parameter value, vehicle longitudinal control model is assessed.Also
It can be after assessment vehicle lateral control model respectively and vehicle longitudinal control model, using the mode of weighting to vehicle control
Model is assessed.The invention does not limit this.
Fig. 3 shows the frame for being suitable for the exemplary computer system/server 012 for being used to realize embodiment of the present invention
Figure.The computer system/server 012 that Fig. 3 is shown is only an example, should not function and use to the embodiment of the present invention
Range band carrys out any restrictions.
As shown in figure 3, computer system/server 012 is showed in the form of universal computing device.Computer system/clothes
The component of business device 012 can include but is not limited to: one or more processor or processing unit 016, system storage
028, connect the bus 018 of different system components (including system storage 028 and processing unit 016).
Bus 018 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Computer system/server 012 typically comprises a variety of computer system readable media.These media, which can be, appoints
The usable medium what can be accessed by computer system/server 012, including volatile and non-volatile media, movably
With immovable medium.
System storage 028 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 030 and/or cache memory 032.Computer system/server 012 may further include other
Removable/nonremovable, volatile/non-volatile computer system storage medium.Only as an example, storage system 034 can
For reading and writing immovable, non-volatile magnetic media (Fig. 3 do not show, commonly referred to as " hard disk drive ").Although in Fig. 3
It is not shown, the disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided, and to can
The CD drive of mobile anonvolatile optical disk (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these situations
Under, each driver can be connected by one or more data media interfaces with bus 018.Memory 028 may include
At least one program product, the program product have one group of (for example, at least one) program module, these program modules are configured
To execute the function of various embodiments of the present invention.
Program/utility 040 with one group of (at least one) program module 042, can store in such as memory
In 028, such program module 042 includes --- but being not limited to --- operating system, one or more application program, other
It may include the realization of network environment in program module and program data, each of these examples or certain combination.Journey
Sequence module 042 usually executes function and/or method in embodiment described in the invention.
Computer system/server 012 can also with one or more external equipments 014 (such as keyboard, sensing equipment,
Display 024 etc.) communication, in the present invention, computer system/server 012 is communicated with outside radar equipment, can also be with
One or more enable a user to the equipment interacted with the computer system/server 012 communication, and/or with make the meter
Any equipment (such as network interface card, the modulation that calculation machine systems/servers 012 can be communicated with one or more of the other calculating equipment
Demodulator etc.) communication.This communication can be carried out by input/output (I/O) interface 022.Also, computer system/clothes
Being engaged in device 012 can also be by network adapter 020 and one or more network (such as local area network (LAN), wide area network (WAN)
And/or public network, such as internet) communication.As shown, network adapter 020 by bus 018 and computer system/
Other modules of server 012 communicate.It should be understood that although not shown in the drawings, computer system/server 012 can be combined
Using other hardware and/or software module, including but not limited to: microcode, device driver, redundant processing unit, external magnetic
Dish driving array, RAID system, tape drive and data backup storage system etc..
Processing unit 016 by the program that is stored in system storage 028 of operation, thereby executing various function application with
And data processing, such as realize a kind of method for identifying website affinity, may include:
Acquire vehicle operation data；
The output data that vehicle control model is directed to the vehicle operation data is obtained, determines assessment ginseng according to output data
Numerical value is as model evaluation parameter value；And assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data；
Using the mean square deviation between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value, institute is assessed
State vehicle control model.
Above-mentioned computer program can be set in computer storage medium, i.e., the computer storage medium is encoded with
Computer program, the program by one or more computers when being executed, so that one or more computers execute in the present invention
State method flow shown in embodiment and/or device operation.For example, the method stream executed by said one or multiple processors
Journey may include:
Acquire vehicle operation data；
The output data that vehicle control model is directed to the vehicle operation data is obtained, determines assessment ginseng according to output data
Numerical value is as model evaluation parameter value；And assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data；
Using the mean square deviation between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value, institute is assessed
State vehicle control model.
With time, the development of technology, medium meaning is more and more extensive, and the route of transmission of computer program is no longer limited by
Tangible medium, can also be directly from network downloading etc..It can be using any combination of one or more computer-readable media.
Computer-readable medium can be computer-readable signal media or computer readable storage medium.Computer-readable storage medium
Matter for example may be-but not limited to-system, device or the device of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or
Any above combination of person.The more specific example (non exhaustive list) of computer readable storage medium includes: with one
Or the electrical connections of multiple conducting wires, portable computer diskette, hard disk, random access memory (RAM), read-only memory (ROM),
Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light
Memory device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer readable storage medium can
With to be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
Person is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium other than computer readable storage medium, which can send, propagate or
Transmission is for by the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In --- wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.It can be with one or more programmings
Language or combinations thereof writes the computer program code for executing operation of the present invention, described program design language include towards
The programming language-of object such as Java, Smalltalk, C++ further include that conventional procedural programming language-is all
Such as " C " language or similar programming language.Program code can execute fully on the user computer, partly with
It executes on the computer of family, executed as an independent software package, part is on the remote computer on the user computer for part
It executes or executes on a remote computer or server completely.In situations involving remote computers, remote computer can
To pass through the network of any kind --- it is connected to subscriber computer including local area network (LAN) or wide area network (WAN), alternatively, can
To be connected to outer computer (such as connecting using ISP by internet).
Using technical solution provided by the present invention, by calculating criterion evaluation parameter value and mould based on same timestamp
Type assess parameter value between mean square deviation, using obtained mean square deviation assess vehicle control model, thus realize it is objective, accurate
Assess vehicle control model in ground.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of hardware adds SFU software functional unit.
The above-mentioned integrated unit being realized in the form of SFU software functional unit can store and computer-readable deposit at one
In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the present invention
The part steps of embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read-
Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. it is various
It can store the medium of program code.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the present invention.
Claims (10)
1. a kind of method for assessing vehicle control model, which is characterized in that the described method includes:
Vehicle operation data is acquired, the vehicle operation data includes: that vehicle position data and each vehicle position data are corresponding
Timestamp；
The output data that vehicle control model is directed to the vehicle operation data is obtained, determines assessment parameter value according to output data
As model evaluation parameter value；And assessment parameter value is calculated as criterion evaluation parameter value according to vehicle operation data；
Using the mean square deviation between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value, the vehicle is assessed
Controlling model；
Wherein, described to include: as criterion evaluation parameter value according to vehicle operation data calculating assessment parameter value
According to vehicle position data and the corresponding timestamp of each vehicle position data, the corresponding traveling curvature of each timestamp is calculated
Value；
Traveling curvature value corresponding to each timestamp carries out interpolation calculation respectively, obtains corresponding to each timestamp unit one belongs to time zone
Between multiple traveling curvature values；
When one is selected from multiple traveling curvature values of each timestamp unit one belongs to time interval of correspondence respectively as corresponding to each
Between the traveling curvature value that stabs；
The traveling curvature value of preset threshold requirement will be met as the criterion evaluation parameter value of corresponding each timestamp.
2. the method according to claim 1, wherein the assessment parameter is traveling curvature, the vehicle control
Model is vehicle lateral control model, and the output data is steering wheel angle.
3. according to the method described in claim 2, it is characterized in that, described determine assessment parameter value as mould according to output data
Type assesses parameter value
The steering wheel angle of each timestamp of correspondence is converted to the traveling curvature value of corresponding each timestamp using preset model；
The traveling curvature value of preset threshold requirement will be met as the model evaluation parameter value of corresponding each timestamp.
4. the method according to claim 1, wherein described joined using the corresponding model evaluation of same timestamp
Mean square deviation between numerical value and criterion evaluation parameter value is assessed before the vehicle control model, further includes:
Judge whether the number of the model evaluation parameter value is greater than preset threshold, if more than then calculating based on the timestamp
Mean square deviation between the model evaluation parameter value and criterion evaluation parameter value, does not otherwise calculate.
5. a kind of device for assessing vehicle control model, which is characterized in that described device includes:
Acquisition unit, for acquiring vehicle operation data, the vehicle operation data includes: vehicle position data and each vehicle
The corresponding timestamp of position data；
Determination unit, the output data for being directed to the vehicle operation data for obtaining vehicle control model, according to output data
Determine assessment parameter value as model evaluation parameter value；And assessment parameter value is calculated as standard according to vehicle operation data
Assess parameter value；
Assessment unit, it is square between the corresponding model evaluation parameter value of same timestamp and criterion evaluation parameter value for utilizing
Difference assesses the vehicle control model；
Wherein, the determination unit is for calculating assessment parameter value as criterion evaluation parameter value according to vehicle operation data
When, it is specific to execute:
According to vehicle position data and the corresponding timestamp of each vehicle position data, the corresponding traveling curvature of each timestamp is calculated
Value；
Traveling curvature value corresponding to each timestamp carries out interpolation calculation respectively, obtains corresponding to each timestamp unit one belongs to time zone
Between multiple traveling curvature values；
When one is selected from multiple traveling curvature values of each timestamp unit one belongs to time interval of correspondence respectively as corresponding to each
Between the traveling curvature value that stabs；
The traveling curvature value of preset threshold requirement will be met as the criterion evaluation parameter value of corresponding each timestamp.
6. device according to claim 5, which is characterized in that the assessment parameter is traveling curvature, the vehicle control
Model is vehicle lateral control model, and the output data is steering wheel angle.
7. device according to claim 6, which is characterized in that the determination unit according to output data determination for commenting
It is specific to execute when estimating parameter value as model evaluation parameter value:
The steering wheel angle of each timestamp of correspondence is converted to the traveling curvature value of corresponding each timestamp using preset model；
The traveling curvature value of preset threshold requirement will be met as the model evaluation parameter value of corresponding each timestamp.
8. device according to claim 5, which is characterized in that the assessment unit is utilizing the corresponding mould of same timestamp
Before type assesses the mean square deviation assessment vehicle control model between parameter value and criterion evaluation parameter value, also execute:
Judge whether the number of the model evaluation parameter value is greater than preset threshold, if more than then calculating based on the timestamp
Mean square deviation between the model evaluation parameter value and criterion evaluation parameter value, does not otherwise calculate.
9. a kind of equipment, which is characterized in that the equipment includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-4.
10. a kind of storage medium comprising computer executable instructions, the computer executable instructions are by computer disposal
For executing the method as described in any in claim 1-4 when device executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509075.6A CN107450511B (en) | 2017-06-28 | 2017-06-28 | Assess method, apparatus, equipment and the computer storage medium of vehicle control model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509075.6A CN107450511B (en) | 2017-06-28 | 2017-06-28 | Assess method, apparatus, equipment and the computer storage medium of vehicle control model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107450511A CN107450511A (en) | 2017-12-08 |
| CN107450511B true CN107450511B (en) | 2019-11-26 |
Family
ID=60488286
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710509075.6A Active CN107450511B (en) | 2017-06-28 | 2017-06-28 | Assess method, apparatus, equipment and the computer storage medium of vehicle control model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107450511B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107783943A (en) * | 2017-09-05 | 2018-03-09 | 百度在线网络技术（北京）有限公司 | A kind of appraisal procedure and device of the longitudinally controlled model of end-to-end automated driving system |
| CN108152045B (en) * | 2017-12-13 | 2019-09-10 | 中国汽车工程研究院股份有限公司 | Vehicular data acquisition method, apparatus and system |
| CN110633596A (en) * | 2018-06-21 | 2019-12-31 | 北京京东尚科信息技术有限公司 | Method and device for predicting vehicle direction angle |
| CN109032116A (en) * | 2018-08-30 | 2018-12-18 | 百度在线网络技术（北京）有限公司 | Vehicle trouble processing method, device, equipment and storage medium |
| CN109598066B (en) * | 2018-12-05 | 2023-08-08 | 百度在线网络技术（北京）有限公司 | Effect evaluation method, apparatus, device and storage medium for prediction module |
| CN109739245A (en) * | 2019-02-19 | 2019-05-10 | 东软睿驰汽车技术（沈阳）有限公司 | One kind being based on unpiloted end to end model appraisal procedure and device |
| CN117217422B (en) * | 2023-11-07 | 2024-03-22 | 国汽(北京)智能网联汽车研究院有限公司 | Vehicle motion control capability assessment method, system, device and medium thereof |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102495631B (en) * | 2011-12-09 | 2013-08-21 | 中国科学院合肥物质科学研究院 | Intelligent control method of driverless vehicle tracking desired trajectory |
| CN105468888B (en) * | 2014-09-12 | 2018-04-20 | 广州汽车集团股份有限公司 | A kind of appraisal procedure and device of motor racing control performance |
| CN104571112B (en) * | 2015-01-14 | 2017-02-22 | 中国科学院合肥物质科学研究院 | Pilotless automobile lateral control method based on turning curvature estimation |
| CN104787091B (en) * | 2015-04-28 | 2016-08-03 | 北京交通大学 | A kind of train automatic Pilot control car effect monitoring system |
| CN106066644A (en) * | 2016-06-17 | 2016-11-02 | 百度在线网络技术（北京）有限公司 | Set up the method for intelligent vehicle control model, intelligent vehicle control method and device |
| CN106020203B (en) * | 2016-07-15 | 2020-09-29 | 百度在线网络技术（北京）有限公司 | Method and apparatus for controlling unmanned vehicle |
-
2017
- 2017-06-28 CN CN201710509075.6A patent/CN107450511B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107450511A (en) | 2017-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107450511B (en) | Assess method, apparatus, equipment and the computer storage medium of vehicle control model | |
| CN107817790B (en) | Method and device for calculating curvature of vehicle track | |
| JP2020530578A (en) | Driving behavior scoring method and equipment | |
| CN111383464B (en) | Vehicle lane change recognition method and device, electronic equipment and medium | |
| CN109738904A (en) | A kind of method, apparatus of detection of obstacles, equipment and computer storage medium | |
| US11361555B2 (en) | Road environment monitoring device, road environment monitoring system, and road environment monitoring program | |
| CN109116374A (en) | Determine the method, apparatus, equipment and storage medium of obstacle distance | |
| CN109035831A (en) | Recognition methods, device, equipment, storage medium and the vehicle of traffic light | |
| CN106918341B (en) | Method and apparatus for constructing map | |
| CN109975839B (en) | Joint filtering optimization method for vehicle satellite positioning data | |
| US10254123B2 (en) | Navigation system with vision augmentation mechanism and method of operation thereof | |
| CN110533940A (en) | Method, apparatus, equipment and the computer storage medium of abnormal traffic signal lamp identification | |
| CN109229109A (en) | Judge the method, apparatus, equipment and computer storage medium of vehicle heading | |
| CN112435469A (en) | Vehicle early warning control method and device, computer readable medium and electronic equipment | |
| CN110288154A (en) | Speed predicting method, device, equipment and medium | |
| JP2019212291A (en) | Indoor positioning system and method based on geomagnetic signals in combination with computer vision | |
| CN114274972A (en) | Scene recognition in an autonomous driving environment | |
| CN110232368A (en) | Method for detecting lane lines, device, electronic equipment and storage medium | |
| CN109635868B (en) | Method and device for determining obstacle type, electronic device and storage medium | |
| CN109739232A (en) | Barrier method for tracing, device, car-mounted terminal and storage medium | |
| CN115762139A (en) | Method, device and equipment for filtering predicted track of intersection and storage medium | |
| Ghose et al. | An enhanced automated system for evaluating harsh driving using smartphone sensors | |
| CN118298169B (en) | Image segmentation method based on large model guidance | |
| CN109215368A (en) | A kind of method, apparatus, equipment and computer storage medium that auxiliary drives | |
| CN109188419B (en) | Method and device for detecting speed of obstacle, computer equipment and storage medium |
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 |