CN109255341A - Extracting method, device, equipment and the medium of barrier perception wrong data - Google Patents
Extracting method, device, equipment and the medium of barrier perception wrong data Download PDFInfo
- Publication number
- CN109255341A CN109255341A CN201811273698.9A CN201811273698A CN109255341A CN 109255341 A CN109255341 A CN 109255341A CN 201811273698 A CN201811273698 A CN 201811273698A CN 109255341 A CN109255341 A CN 109255341A
- Authority
- CN
- China
- Prior art keywords
- data
- barrier
- instance
- sensing results
- pilot steering
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention discloses extracting method, device, equipment and the media of a kind of barrier perception wrong data.This method comprises: obtaining the drive test data collection of vehicle；The drive test data is concentrated comprising sensing data, barrier sensing results data and the pilot steering behavioral data for disturbance of perception object；By the comparative analysis to the barrier sensing results data and the pilot steering behavioral data, the instance data of acquired disturbance object misrecognition and/or barrier leakage identification；Wherein, instance data includes sensing data.A large amount of barrier leaks the extraction of the corresponding instance data such as identification and/or misrecognition in vehicle travel process when above-mentioned technical proposal is by drive test, retraining for barrier sensor model provides data supporting, and then barrier sensor model can be optimized, the accuracy rate and reliability for improving obstacle recognition, reduce the security risk of automatic driving vehicle.
Description
Technical field
The present embodiments relate to the extractions of unmanned technical field more particularly to a kind of barrier perception wrong data
Method, apparatus, equipment and medium.
Background technique
In unmanned sensory perceptual system, the image data of video camera output and the point cloud of laser radar output are relied primarily on
Data and the data of radio radar output carry out the perception identification of barrier.The prior art in the algorithm iteration development phase,
A large amount of data acquisition and standard are usually carried out by the open loop approach of pilot steering, for carrying out barrier sensor model instruction
Practice, so that barrier is identified and classified subsequently through barrier sensor model.
However, it is many kinds of due to barrier, cause the non-barriers such as green grafting branch leaf, excess surface water accidentally to be known
Not, the case where and/or low obstructions such as cone, A-frame or children pedestrian are identified by leakage happen occasionally, right
The traffic safety of automatic driving vehicle brings hidden danger.
Summary of the invention
The embodiment of the invention provides a kind of barrier perception wrong data extracting method, device, equipment and medium, with
Wrong data is perceived by the barrier of extraction and provides data supporting for the retraining of barrier sensor model, and then optimizes obstacle
Object sensor model improves obstacle recognition accuracy rate and reliability, reduces the security risk of automatic driving vehicle.
In a first aspect, the embodiment of the invention provides a kind of extracting methods of barrier perception wrong data, comprising:
Obtain the drive test data collection of vehicle；The drive test data concentrate comprising for disturbance of perception object sensing data,
Barrier sensing results data and pilot steering behavioral data；
Pass through the comparative analysis to the barrier sensing results data and the pilot steering behavioral data, acquired disturbance
Object misrecognition and/or the instance data of barrier leakage identification；Wherein, instance data includes sensing data.
Second aspect, the embodiment of the invention also provides a kind of extraction elements of barrier perception wrong data, comprising:
Drive test data collection obtains module, for obtaining the drive test data collection of vehicle；The drive test data is concentrated comprising being used for
Sensing data, barrier sensing results data and the pilot steering behavioral data of disturbance of perception object；
Instance data obtains module, for by the barrier sensing results data and the pilot steering behavior number
According to comparative analysis, acquired disturbance object misrecognition and/or barrier leakage identification instance data；Wherein, instance data includes passing
Sensor data.
The third aspect, the embodiment of the invention also provides a kind of electronic equipment, comprising:
One or more processors；
Memory, 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 processing
Device realizes a kind of extracting method of barrier perception wrong data as provided by first aspect embodiment.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program, realizing that a kind of barrier such as first aspect embodiment provided by perceives wrong data when which is executed by processor
Extracting method.
The embodiment of the present invention obtain vehicle for the sensing data of disturbance of perception object, barrier sensing results data and
Pilot steering behavioral data；And by the comparative analysis to barrier sensing results data and pilot steering behavioral data, obtain
Barrier misrecognition and/or the instance data of leakage identification；Wherein instance data includes sensing data.Above-mentioned technical proposal passes through
A large amount of barrier leaks the extraction of the corresponding instance data such as identification and/or misrecognition in vehicle travel process when to drive test, is
The retraining of barrier sensor model provides data supporting, and then can optimize barrier sensor model, improves barrier and knows
Other accuracy rate and reliability reduce the security risk of automatic driving vehicle.
Detailed description of the invention
Fig. 1 is the flow chart of the extracting method of one of the embodiment of the present invention one barrier perception wrong data；
Fig. 2 is the flow chart of the extracting method of one of the embodiment of the present invention two barrier perception wrong data；
Fig. 3 is the structure chart of the extraction element of one of the embodiment of the present invention three barrier perception wrong data；
Fig. 4 is the structural schematic diagram of one of the embodiment of the present invention four electronic equipment.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart of the extracting method of one of the embodiment of the present invention one barrier perception wrong data.This hair
Bright embodiment is suitable for carrying out in the algorithm iteration development phase of unmanned sensory perceptual system by the open loop approach of pilot steering
The case where acquisition of the training sample of barrier sensor model, this method can be perceived the extraction element of wrong data by barrier
It executes, the device is by software and or hardware realization, and concrete configuration is in automatic driving vehicle.
The extracting method of barrier perception wrong data as shown in Figure 1, comprising:
S110, the drive test data collection for obtaining vehicle；The drive test data is concentrated comprising the sensor for disturbance of perception object
Data, barrier sensing results data and pilot steering behavioral data.
Wherein, sensing data includes the image data of video camera output in vehicle, the point that laser radar exports in vehicle
The data that radio radar exports in cloud data and vehicle.Wherein, vehicle can be automatic driving vehicle, or can be with
It is other vehicles for being provided with unmanned sensory perceptual system.Wherein, barrier sensing results data can be understood as vehicle driving
The sensing results that perception algorithm module is obtained according to the sensing data that sensor is exported in the process, or can also be basis
The sensing data generated when vehicle driving, the sensing results obtained offline by perception algorithm module, refers to do not have to offline
Test of getting on the bus is carried out, under computer hardware environment identical on vehicle, the sensing data being collected is issued into sense
Know algoritic module, perception algorithm module can correspond to output sensing results.
Wherein, pilot steering behavioral data can be understood as the data of characterization pilot steering behavior, such as can be direction
The behavioural analyses such as the rotational angle of disk, the working condition of Vehicular brake device, the starting state of vehicle and/or vehicle movement parameter
Data can also be driver with the presence or absence of parking and/or with the presence or absence of the intuitive behavioral data for the traveling that detours.Wherein it is possible to
With the presence or absence of the driving behavior for traveling of stopping and/or detour when obtaining driver's pilot steering by behavioural analysis data.Wherein,
Vehicle movement parameter includes: Velicle motion velocity, angular speed and/or acceleration.
Illustratively, sensing data is carried out in vehicle travel process and the real-time of pilot steering behavioral data acquires simultaneously
Storage, and carry out the real-time storage for the barrier sensing results data that barrier real-time perception obtains；Correspondingly, can pass
When sensor data, barrier sensing results data and pilot steering behavioral data real-time storage, by above-mentioned each data synchronous transfer
To barrier perception wrong data extraction element in, and by each sensing data, each barrier sensing results data and each one
Work driving behavior data generate moment corresponding storage according to data and obtain drive test data collection.It is, of course, also possible in vehicle driving mistake
After journey terminates, by the artificial triggering of technical staff, each sensing data stored in driving process, each barrier sense are transferred
Know result data and each artificial driving behavior data, and generates moment corresponding storage according to data and obtain drive test data collection.
S120, by the comparative analysis to the barrier sensing results data and the pilot steering behavioral data, obtain
Obtain the instance data of barrier misrecognition and/or barrier leakage identification；Wherein, instance data includes sensing data.
Specifically, being hindered by the comparative analysis to the sensing results data and the pilot steering behavioral data
The instance data for hindering object to misidentify, comprising: for every barrier sensing results data, if the barrier sensing results data are
Barrier is perceived, then determines pilot steering corresponding at the time of perceiving barrier according to the pilot steering behavioral data
Behavior judges whether the artificial driving behavior is default pilot steering behavior；If not, it is determined that the barrier sensing results number
It is the instance data of barrier misrecognition according to corresponding sensing data；Wherein, the default pilot steering behavior includes: parking
With the traveling that detours.
Illustratively, when pilot steering behavioral data includes the rotational angle of steering wheel, if rotational angle meets setting
Rotational angle turn threshold value, then show that driver has the driving behavior of traveling of detouring；When pilot steering behavioral data includes
When the working condition of Vehicular brake device, if the working condition of Vehicular brake device is non-idle state, show that driver deposits
In the driving behavior of parking；When pilot steering behavioral data includes vehicle launch state, if the vehicle in the first preset time period
State be " starting-non-start up-starting ", then show driver exist parking driving behavior；When pilot steering behavior number
When according to including Velicle motion velocity, if running velocity reduces amplitude greater than larger movement velocity in the second preset time period
Preset percentage, then show driver exist parking driving behavior；When pilot steering behavioral data includes vehicle movement angle
When speed, the threshold value if angular speed that the difference of vehicle operation angular speed is greater than setting in third preset time period is turned, table
There is the driving behavior for the traveling that detours in bright driver；When pilot steering behavioral data includes vehicle movement acceleration, if
Vehicle runs acceleration continuously less than zero in four preset time periods, then shows that driver has the driving behavior of parking.Wherein, turn
It is dynamic angle turning threshold value, the first preset time period, the second preset time period, third preset time period, the 4th preset time period, pre-
If percentage and angular speed turning threshold value etc. can be set by technical staff according to experiment value or empirical value.
Specifically, being hindered by the comparative analysis to the sensing results data and the pilot steering behavioral data
Hinder the instance data of object leakage identification, comprising: for the default pilot steering behavior in the pilot steering behavioral data, according to institute
It states barrier sensing results data and determines barrier sensing results number corresponding at the time of generating the default pilot steering behavior
According to, judge the barrier sensing results data whether be perceive barrier, if not, it is determined that the barrier sensing results number
It is the instance data of barrier leakage identification according to corresponding sensing data；Wherein, the default pilot steering behavior includes: parking
With the traveling that detours.
The embodiment of the present invention obtain vehicle for the sensing data of disturbance of perception object, barrier sensing results data and
Pilot steering behavioral data；And by the comparative analysis to barrier sensing results data and pilot steering behavioral data, obtain
Barrier misrecognition and/or the instance data of leakage identification；Wherein instance data includes sensing data.Above-mentioned technical proposal passes through
A large amount of barrier leaks the extraction of the corresponding instance data such as identification and/or misrecognition in vehicle travel process when to drive test, is
The retraining of barrier sensor model provides data supporting, and then can optimize barrier sensor model, improves obstacle recognition
Accuracy rate and reliability, reduce the security risk of automatic driving vehicle.
Further, after acquired disturbance object misrecognition and/or the instance data of barrier leakage identification, further includes: will
Barrier misrecognition and/or the instance data of barrier leakage identification carry out barrier sensor model as training sample data
Training；And/or
Using barrier misrecognition and/or the instance data of barrier leakage identification, the identification of barrier sensor model is imitated
Fruit is tested.
Wherein, instance data can also include that barrier misidentifies and/or leak corresponding actual perceived number of results when identification
According to.Wherein, actual perceived result data when barrier misidentifies is opposite with barrier sensing results data content；Barrier leakage
Actual perceived result data when identification is opposite with barrier sensing results data content.
Illustratively, when the result using " 0 " as clear identifies, and " 1 " is as the result mark for having barrier,
When barrier misrecognition, corresponding barrier sensing results data are " 1 ", and actual perceived result data is " 0 "；Work as barrier
When leakage identification, corresponding barrier sensing results data are " 0 ", and actual perceived result data is " 1 ".
Illustratively, using barrier misrecognition and/or the instance data of barrier leakage identification, to barrier sensor model
Recognition effect tested, may is that and be input in barrier sensor model using the sensing data in instance data, obtained
To corresponding prediction sensing results, according to the actual perceived result data in prediction sensing results and instance data, to barrier
Sensor model is evaluated.
Specifically, according to formulaBarrier sensor model is evaluated；
Wherein, TP indicates prediction sensing results and actual perceived result data is to have barrier；TN indicates prediction perception
It as a result is clear with actual perceived result data；FP indicates that prediction sensing results are to have barrier, but actual perceived knot
Fruit data are clear；FN indicates that prediction sensing results are clear, but actual perceived result data is to have barrier；
Accuracy is the accuracy rate of barrier sensor model；Precision is the accurate rate of barrier sensor model；Recall is barrier
Hinder the recall rate of object sensor model.
The embodiment of the present invention using barrier by misidentifying and/or leaking the instance data of identification as training sample data pair
Barrier sensor model is trained, and to optimize barrier sensor model, improves the accuracy rate of barrier sensor model and reliable
Property；By the way that barrier is misidentified and/or leaks the instance data identified as test sample data to barrier sensor model
It is tested, is evaluated with the recognition effect to barrier sensor model.
Embodiment two
Fig. 2 is the flow chart of the extracting method of one of the embodiment of the present invention two barrier perception wrong data.This hair
Bright embodiment has carried out additional optimization on the basis of the technical solution of the various embodiments described above.
Further, will after operation " in the instance data of acquired disturbance object leakage identification ", it is additional " in conjunction with road net data and
Location data screens the instance data of barrier leakage identification ", with non-mistake included in the instance data to leakage identification
Accidentally data are rejected, and improve the purity of barrier perception wrong data.
The extracting method of barrier perception wrong data as shown in Figure 2, comprising:
S210, the drive test data collection for obtaining vehicle；The drive test data is concentrated comprising the sensor for disturbance of perception object
Data, barrier sensing results data and pilot steering behavioral data.
S220, for the default pilot steering behavior in the pilot steering behavioral data, perceived according to the barrier
Result data determines barrier sensing results data corresponding at the time of generating the default pilot steering behavior.
S230, judge the barrier sensing results data whether be perceive barrier, if not, it is determined that the barrier
The corresponding sensing data of sensing results data is the instance data of barrier leakage identification.
Wherein, the default pilot steering behavior includes: parking and the traveling that detours.
Specifically, obtaining barrier sensing results data when in pilot steering behavioral data including parking driving behavior
In, driver generates the current time corresponding barrier sensing results data of parking driving behavior；If barrier sensing results
Data are the result mark that characterization does not perceive barrier, it is determined that the corresponding sensing data of barrier sensing results data is
The instance data of barrier leakage identification.It is understood that the corresponding sensing data of barrier sensing results data is to produce
The current time corresponding sensing data of raw parking driving behavior.
Specifically, obtaining barrier sensing results data when including detouring to travel behavior in pilot steering behavioral data
In, driver generates the current time corresponding barrier sensing results data for the behavior of travelling that detours；If barrier sensing results
Data are the result mark that characterization does not perceive barrier, it is determined that the corresponding sensing data of barrier sensing results data is
The instance data of barrier leakage identification.It is understood that the corresponding sensing data of barrier sensing results data is to produce
The current time corresponding sensing data of the raw behavior of travelling that detours.
Wherein, corresponding actual perceived result data when instance data can also include barrier leakage identification.Wherein, obstacle
Actual perceived result data when object leakage identification is opposite with barrier sensing results data content.
S240, in conjunction with road net data and location data, the instance data of barrier leakage identification is screened.
Specifically, being generated for the instance data of every barrier leakage identification according to location data and road net data judgement
The traffic lights information of vehicle present position at the time of the default pilot steering behavior, if traffic lights information instruction forbids straight line logical
Row, then the instance data by the leakage identification of this barrier is deleted.
Specifically, obtaining the generation moment corresponding location data of sensing data included in every instance data；
Vehicle present position is determined according to location data, and determines traffic lights color letter of the vehicle at present position in conjunction with road net data
Breath and traffic lights transmits information；If traffic lights colouring information is that red or traffic lights transmits information is turning,
The instance data of this barrier leakage identification is deleted.
The embodiment of the present invention will generate parking or the driving behavior for the traveling that detours due to traffic lights information instruction driver, and
When obstacle recognition result does not perceive barrier, barrier is correctly identified in the case where misjudging as barrier leakage identification really
Fixed instance data is rejected, and the purity of barrier perception wrong data is improved,
Good data basis is provided for the optimization of barrier sensor model.
Embodiment three
Fig. 3 is the structural schematic diagram of the extraction element of one of the embodiment of the present invention three barrier perception wrong data.
The embodiment of the present invention is suitable for passing through the open loop approach of pilot steering in the algorithm iteration development phase of unmanned sensory perceptual system
The case where carrying out the acquisition of the training sample of barrier sensor model, the device is by software and or hardware realization, and concrete configuration
In automatic driving vehicle.The extraction element of barrier perception wrong data shown in Fig. 3, comprising: drive test data collection obtains mould
Block 310 and instance data obtain module 320.
Drive test data collection obtains module 310, for obtaining the drive test data collection of vehicle；The drive test data is concentrated comprising using
In the sensing data of disturbance of perception object, barrier sensing results data and pilot steering behavioral data；
Instance data obtains module 320, for by the barrier sensing results data and the pilot steering row
For the comparative analysis of data, the instance data of acquired disturbance object misrecognition and/or barrier leakage identification；Wherein, instance data packet
Include sensing data.
The embodiment of the present invention obtains the sensor number for disturbance of perception object that module obtains vehicle using drive test data collection
According to, barrier sensing results data and pilot steering behavioral data；And module is obtained using instance data and is passed through to barrier sense
Know the instance data of the comparative analysis of result data and pilot steering behavioral data, acquired disturbance object misrecognition and/or leakage identification；
Wherein instance data includes sensing data.A large amount of barrier in vehicle travel process when above-mentioned technical proposal is by drive test
The extraction of the corresponding instance data such as leakage identification and/or misrecognition, the retraining for barrier sensor model provide data supporting,
And then barrier sensor model can be optimized, the accuracy rate and reliability of obstacle recognition are improved, automatic driving vehicle is reduced
Security risk.
Further, the instance data obtains module 320, comprising:
Barrier misidentifies judging unit, is used for for every barrier sensing results data, if the barrier
Sensing results data are to perceive barrier, then perceive barrier according to pilot steering behavioral data determination
At the time of corresponding pilot steering behavior, judge whether the artificial driving behavior is default pilot steering behavior；If it is not, then
Determine that the corresponding sensing data of barrier sensing results data is the instance data of barrier misrecognition.
Further, the instance data obtains module 320, comprising:
Barrier leakage identification judging unit, for for the default pilot steering row in the pilot steering behavioral data
To determine barrier sense corresponding at the time of generating the default pilot steering behavior according to the barrier sensing results data
Know result data, judge the barrier sensing results data whether be perceive barrier, if not, it is determined that the barrier sense
Know that the corresponding sensing data of result data is the instance data of barrier leakage identification.
Further, the default pilot steering behavior includes: parking and the traveling that detours.
Further, the instance data obtains module 320, further includes instance data screening unit, is specifically used for:
After the instance data of barrier leakage identification judging unit acquired disturbance object leakage identification, in conjunction with road net data and determine
Position data screen the instance data of barrier leakage identification.
Further, the instance data screening unit, comprising:
Instance data deletes subelement, for the instance data for the leakage identification of every barrier, according to location data and
The traffic lights information of vehicle present position at the time of road net data judgement generates the default pilot steering behavior, if the traffic lights are believed
Traveled straight is forbidden in breath instruction, then the instance data by the leakage identification of this barrier is deleted.
Further, described device, further includes:
Training module, for acquired disturbance object misrecognition and/or barrier leakage identification instance data after, by obstacle
Object misrecognition and/or the instance data of barrier leakage identification are trained barrier sensor model as training sample data；
And/or
Test module, for using barrier after acquired disturbance object misrecognition and/or the instance data of barrier leakage identification
Hinder the instance data that object misidentifies and/or barrier leakage identifies, the recognition effect of barrier sensor model is tested.
Barrier provided by the embodiment of the present invention perceives any implementation of the executable present invention of extraction element of wrong data
The extracting method of the perception wrong data of barrier provided by example, has the extracting method phase for executing barrier perception wrong data
The functional module and beneficial effect answered.
Example IV
Fig. 4 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention four provides.Fig. 4, which is shown, to be suitable for being used in fact
The block diagram of the example electronic device 412 of existing embodiment of the present invention.The electronic equipment 412 that Fig. 4 is shown is only an example,
Should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in figure 4, electronic equipment 412 is showed in the form of universal computing device.The component of electronic equipment 412 can wrap
Include but be not limited to: one or more processor or processing unit 416, system storage 428 connect different system components
The bus 418 of (including system storage 428 and processing unit 416).
Bus 418 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.
Electronic equipment 412 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that electronic equipment 412 accesses, including volatile and non-volatile media, moveable and immovable medium.
System storage 428 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 430 and/or cache memory 432.Electronic equipment 412 may further include it is other it is removable/no
Movably, volatile/non-volatile computer system storage medium.Only as an example, storage system 434 can be used for reading and writing
Immovable, non-volatile magnetic media (Fig. 4 do not show, commonly referred to as " hard disk drive ").Although not shown in fig 4, may be used
To provide the disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk "), and it is non-volatile to moving
Property CD (such as CD-ROM, DVD-ROM or other optical mediums) read and write CD drive.In these cases, each drive
Dynamic device can be connected by one or more data media interfaces with bus 418.Memory 428 may include at least one journey
Sequence product, the program product have one group of (for example, at least one) program module, these program modules are configured to perform this hair
The function of bright each embodiment.
Program/utility 440 with one group of (at least one) program module 442, can store in such as memory
In 428, such program module 442 includes but is not limited to operating system, one or more application program, other program modules
And program data, it may include the realization of network environment in each of these examples or certain combination.Program module 442
Usually execute the function and/or method in embodiment described in the invention.
Electronic equipment 412 can also be with one or more external equipments 414 (such as keyboard, sensing equipment, display 424
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 412 communicate, and/or with make
Any equipment (such as network interface card, the modem that the electronic equipment 412 can be communicated with one or more of the other calculating equipment
Etc.) communication.This communication can be carried out by input/output (I/O) interface 422.Also, electronic equipment 412 can also lead to
Cross network adapter 420 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network, example
Such as internet) communication.As shown, network adapter 420 is communicated by bus 418 with other modules of electronic equipment 412.It answers
When understanding, although not shown in the drawings, other hardware and/or software module can be used in conjunction with electronic equipment 412, including but unlimited
In: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and number
According to backup storage system etc..
Processing unit 416 passes through at least one program in multiple programs that operation is stored in system storage 428, from
And application and data processing are performed various functions, such as realize a kind of barrier perception mistake provided by the embodiment of the present invention
The extracting method of data.
The embodiment of the invention also provides a kind of vehicles, including car body, further include above-mentioned electronic equipment.
Embodiment five
The embodiment of the present invention five provides a kind of computer readable storage medium, is stored thereon with computer program, the journey
A kind of extracting method of barrier perception wrong data provided by any embodiment of the present invention is realized when sequence is executed by processor,
It include: the drive test data collection for obtaining vehicle；The drive test data is concentrated comprising sensing data, the obstacle for disturbance of perception object
Object sensing results data and pilot steering behavioral data；By to the barrier sensing results data and the pilot steering row
For the comparative analysis of data, the instance data of acquired disturbance object misrecognition and/or barrier leakage identification；Wherein, instance data packet
Include sensing data.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage
Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device
Using or it 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 unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used 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.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on a remote computer or server completely on the remote computer on the user computer.?
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as mentioned using Internet service
It is connected for quotient by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of extracting method of barrier perception wrong data characterized by comprising
Obtain the drive test data collection of vehicle；The drive test data is concentrated comprising sensing data, the obstacle for disturbance of perception object
Object sensing results data and pilot steering behavioral data；
By the comparative analysis to the barrier sensing results data and the pilot steering behavioral data, acquired disturbance object is missed
The instance data of identification and/or barrier leakage identification；Wherein, instance data includes sensing data.
2. the method according to claim 1, wherein described by sensing results data and described artificial
The comparative analysis of driving behavior data, the instance data of acquired disturbance object misrecognition, comprising:
For every barrier sensing results data, if the barrier sensing results data are to perceive barrier, according to institute
It states the determination of pilot steering behavioral data and perceives pilot steering behavior corresponding at the time of barrier, judge the pilot steering row
Whether to be default pilot steering behavior；If not, it is determined that the corresponding sensing data of barrier sensing results data is
The instance data of barrier misrecognition.
3. the method according to claim 1, wherein described by sensing results data and described artificial
The comparative analysis of driving behavior data, the instance data of acquired disturbance object leakage identification, comprising:
It is true according to the barrier sensing results data for the default pilot steering behavior in the pilot steering behavioral data
Raw this of fixed output quota presets barrier sensing results data corresponding at the time of pilot steering behavior, judges the barrier sensing results
Data whether be perceive barrier, if not, it is determined that the corresponding sensing data of barrier sensing results data be barrier
Hinder the instance data of object leakage identification.
4. according to the method in claim 2 or 3, which is characterized in that the default pilot steering behavior include: parking and around
Road traveling.
5. according to the method described in claim 3, it is characterized in that, being gone back after the instance data of acquired disturbance object leakage identification
Include:
In conjunction with road net data and location data, the instance data of barrier leakage identification is screened.
6. according to the method described in claim 5, it is characterized in that, the combination road net data and location data, to barrier
The instance data of leakage identification is screened, comprising:
For the instance data of every barrier leakage identification, this is generated according to location data and road net data judgement and default is manually driven
The traffic lights information of vehicle present position at the time of sailing behavior, if traveled straight is forbidden in traffic lights information instruction, by this
The instance data of barrier leakage identification is deleted.
7. the method according to claim 1, wherein in acquired disturbance object misrecognition and/or barrier leakage identification
Instance data after, further includes:
Using barrier misrecognition and/or the instance data of barrier leakage identification as training sample data, mould is perceived to barrier
Type is trained；And/or
Using barrier misrecognition and/or the instance data of barrier leakage identification, to the recognition effect of barrier sensor model into
Row test.
8. a kind of extraction element of barrier perception wrong data characterized by comprising
Drive test data collection obtains module, for obtaining the drive test data collection of vehicle；The drive test data is concentrated comprising for perceiving
Sensing data, barrier sensing results data and the pilot steering behavioral data of barrier；
Instance data obtains module, for by the barrier sensing results data and the pilot steering behavioral data
The instance data of comparative analysis, acquired disturbance object misrecognition and/or barrier leakage identification；Wherein, instance data includes sensor
Data.
9. a kind of electronic equipment characterized by comprising
One or more processors；
Memory, 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
Now such as a kind of described in any item extracting methods of barrier perception wrong data of claim 1-7.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
It is realized when execution such as a kind of described in any item extracting methods of barrier perception wrong data of claim 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811273698.9A CN109255341B (en) | 2018-10-30 | 2018-10-30 | Method, device, equipment and medium for extracting obstacle perception error data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811273698.9A CN109255341B (en) | 2018-10-30 | 2018-10-30 | Method, device, equipment and medium for extracting obstacle perception error data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109255341A true CN109255341A (en) | 2019-01-22 |
| CN109255341B CN109255341B (en) | 2021-08-10 |
Family
ID=65042913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811273698.9A Active CN109255341B (en) | 2018-10-30 | 2018-10-30 | Method, device, equipment and medium for extracting obstacle perception error data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109255341B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110046569A (en) * | 2019-04-12 | 2019-07-23 | 北京百度网讯科技有限公司 | A kind of data processing method, device and electronic equipment |
| CN110287832A (en) * | 2019-06-13 | 2019-09-27 | 北京百度网讯科技有限公司 | High-Speed Automatic Driving Scene barrier perception evaluating method and device |
| CN110853389A (en) * | 2019-11-21 | 2020-02-28 | 白犀牛智达(北京)科技有限公司 | Drive test monitoring system suitable for unmanned commodity circulation car |
| CN111177869A (en) * | 2020-01-02 | 2020-05-19 | 北京百度网讯科技有限公司 | Method, device and equipment for determining sensor layout scheme |
| CN112380942A (en) * | 2020-11-06 | 2021-02-19 | 北京石头世纪科技股份有限公司 | Method, device, medium and electronic equipment for identifying obstacle |
| CN112633518A (en) * | 2021-01-25 | 2021-04-09 | 国汽智控(北京)科技有限公司 | Automatic driving model training method and system based on multi-subject mutual learning |
| CN112698421A (en) * | 2020-12-11 | 2021-04-23 | 北京百度网讯科技有限公司 | Evaluation method, device, equipment and storage medium for obstacle detection |
| CN113095125A (en) * | 2019-12-19 | 2021-07-09 | 通用汽车环球科技运作有限责任公司 | Diagnosing a perception system based on scene continuity |
| CN114357814A (en) * | 2022-03-21 | 2022-04-15 | 禾多科技（北京）有限公司 | Automated driving simulation test method, apparatus, device and computer readable medium |
| CN114563007A (en) * | 2022-04-28 | 2022-05-31 | 新石器慧通（北京）科技有限公司 | Obstacle motion state prediction method, obstacle motion state prediction device, electronic device, and storage medium |
| CN116434041A (en) * | 2022-12-05 | 2023-07-14 | 北京百度网讯科技有限公司 | Mining method, device and equipment for error perception data and automatic driving vehicle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170004345A1 (en) * | 2015-07-02 | 2017-01-05 | Topcon Corporation | Survey data processing device, survey data processing method, and survey data processing program |
| US20170314930A1 (en) * | 2015-04-06 | 2017-11-02 | Hrl Laboratories, Llc | System and method for achieving fast and reliable time-to-contact estimation using vision and range sensor data for autonomous navigation |
| CN107506733A (en) * | 2017-08-28 | 2017-12-22 | 济南浪潮高新科技投资发展有限公司 | A kind of obstacle recognition method, mist node and system |
| CN107571864A (en) * | 2017-09-05 | 2018-01-12 | 百度在线网络技术（北京）有限公司 | The collecting method and device of automatic driving vehicle |
| CN108162973A (en) * | 2016-12-07 | 2018-06-15 | 法乐第(北京)网络科技有限公司 | Improve the device and automated driving system of automatic Pilot reliability |
| CN108508881A (en) * | 2017-02-27 | 2018-09-07 | 北京百度网讯科技有限公司 | Automatic Pilot control strategy method of adjustment, device, equipment and storage medium |
| CN108549366A (en) * | 2018-05-04 | 2018-09-18 | 同济大学 | Intelligent automobile road driving mapping experiment method parallel with virtual test |
-
2018
- 2018-10-30 CN CN201811273698.9A patent/CN109255341B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170314930A1 (en) * | 2015-04-06 | 2017-11-02 | Hrl Laboratories, Llc | System and method for achieving fast and reliable time-to-contact estimation using vision and range sensor data for autonomous navigation |
| US20170004345A1 (en) * | 2015-07-02 | 2017-01-05 | Topcon Corporation | Survey data processing device, survey data processing method, and survey data processing program |
| CN108162973A (en) * | 2016-12-07 | 2018-06-15 | 法乐第(北京)网络科技有限公司 | Improve the device and automated driving system of automatic Pilot reliability |
| CN108508881A (en) * | 2017-02-27 | 2018-09-07 | 北京百度网讯科技有限公司 | Automatic Pilot control strategy method of adjustment, device, equipment and storage medium |
| CN107506733A (en) * | 2017-08-28 | 2017-12-22 | 济南浪潮高新科技投资发展有限公司 | A kind of obstacle recognition method, mist node and system |
| CN107571864A (en) * | 2017-09-05 | 2018-01-12 | 百度在线网络技术（北京）有限公司 | The collecting method and device of automatic driving vehicle |
| CN108549366A (en) * | 2018-05-04 | 2018-09-18 | 同济大学 | Intelligent automobile road driving mapping experiment method parallel with virtual test |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110046569A (en) * | 2019-04-12 | 2019-07-23 | 北京百度网讯科技有限公司 | A kind of data processing method, device and electronic equipment |
| CN110046569B (en) * | 2019-04-12 | 2022-04-12 | 北京百度网讯科技有限公司 | Unmanned driving data processing method and device and electronic equipment |
| CN110287832A (en) * | 2019-06-13 | 2019-09-27 | 北京百度网讯科技有限公司 | High-Speed Automatic Driving Scene barrier perception evaluating method and device |
| CN110853389A (en) * | 2019-11-21 | 2020-02-28 | 白犀牛智达(北京)科技有限公司 | Drive test monitoring system suitable for unmanned commodity circulation car |
| CN113095125A (en) * | 2019-12-19 | 2021-07-09 | 通用汽车环球科技运作有限责任公司 | Diagnosing a perception system based on scene continuity |
| CN113095125B (en) * | 2019-12-19 | 2023-11-24 | 通用汽车环球科技运作有限责任公司 | Diagnosing a perception system based on scene continuity |
| CN111177869B (en) * | 2020-01-02 | 2023-09-01 | 北京百度网讯科技有限公司 | Method, device and equipment for determining sensor layout scheme |
| CN111177869A (en) * | 2020-01-02 | 2020-05-19 | 北京百度网讯科技有限公司 | Method, device and equipment for determining sensor layout scheme |
| CN112380942A (en) * | 2020-11-06 | 2021-02-19 | 北京石头世纪科技股份有限公司 | Method, device, medium and electronic equipment for identifying obstacle |
| CN112698421A (en) * | 2020-12-11 | 2021-04-23 | 北京百度网讯科技有限公司 | Evaluation method, device, equipment and storage medium for obstacle detection |
| CN112633518A (en) * | 2021-01-25 | 2021-04-09 | 国汽智控(北京)科技有限公司 | Automatic driving model training method and system based on multi-subject mutual learning |
| CN112633518B (en) * | 2021-01-25 | 2024-03-01 | 国汽智控(北京)科技有限公司 | Automatic driving model training method and system based on multi-subject mutual learning |
| CN114357814A (en) * | 2022-03-21 | 2022-04-15 | 禾多科技（北京）有限公司 | Automated driving simulation test method, apparatus, device and computer readable medium |
| CN114563007A (en) * | 2022-04-28 | 2022-05-31 | 新石器慧通（北京）科技有限公司 | Obstacle motion state prediction method, obstacle motion state prediction device, electronic device, and storage medium |
| CN116434041A (en) * | 2022-12-05 | 2023-07-14 | 北京百度网讯科技有限公司 | Mining method, device and equipment for error perception data and automatic driving vehicle |
| CN116434041B (en) * | 2022-12-05 | 2024-06-21 | 北京百度网讯科技有限公司 | Mining method, device and equipment for error perception data and automatic driving vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109255341B (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255341A (en) | Extracting method, device, equipment and the medium of barrier perception wrong data | |
| US10429841B2 (en) | Vehicle control method and apparatus and method and apparatus for acquiring decision-making model | |
| CN106218639B (en) | Automatic driving vehicle, the method and apparatus for controlling automatic driving vehicle | |
| US11561540B2 (en) | Augmenting autonomous driving with remote viewer recommendation | |
| CN112180921B (en) | Automatic driving algorithm training system and method | |
| CN109064763A (en) | Test method, device, test equipment and the storage medium of automatic driving vehicle | |
| CN111386563B (en) | Teacher data generation device | |
| CN109032103B (en) | Method, device and equipment for testing unmanned vehicle and storage medium | |
| US9304515B2 (en) | Regional operation modes for autonomous vehicles | |
| CN107571864B (en) | The collecting method and device of automatic driving vehicle | |
| CN108508881B (en) | Automatic driving control strategy adjusting method, device, equipment and storage medium | |
| CN112789619A (en) | Simulation scene construction method, simulation method and device | |
| CN109358627A (en) | Based on unpiloted driving assistance method, device, equipment, medium and vehicle | |
| US11076022B2 (en) | Systems and methods for implementing robotics frameworks | |
| CN110163176A (en) | The recognition methods of lane line change location, device, equipment and medium | |
| CN109213134A (en) | The method and apparatus for generating automatic Pilot strategy | |
| CN108399752A (en) | A kind of driving infractions pre-judging method, device, server and medium | |
| JP2020123333A (en) | Method and device for detecting travel scenario generated in middle of travel and providing information for evaluating travel havit of driver | |
| CN105388021A (en) | ADAS virtual development and test system | |
| CN109583776A (en) | A kind of vehicle body-sensing evaluating method, device, electronic equipment, medium and vehicle | |
| CN109117691A (en) | Drivable region detection method, device, equipment and storage medium | |
| CN109472884A (en) | Unmanned vehicle date storage method, device, equipment and storage medium | |
| WO2022246860A1 (en) | Performance test method for autonomous driving system | |
| CN109117690A (en) | Drivable region detection method, device, equipment and storage medium | |
| CN110533940A (en) | Method, apparatus, equipment and the computer storage medium of abnormal traffic signal lamp identification |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211019 Address after: 105 / F, building 1, No. 10, Shangdi 10th Street, Haidian District, Beijing 100085 Patentee after: Apollo Intelligent Technology (Beijing) Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TR01 | Transfer of patent right |