CN110069408A - Automatic driving vehicle sensory perceptual system test method and device - Google Patents
Automatic driving vehicle sensory perceptual system test method and device Download PDFInfo
- Publication number
- CN110069408A CN110069408A CN201910288926.8A CN201910288926A CN110069408A CN 110069408 A CN110069408 A CN 110069408A CN 201910288926 A CN201910288926 A CN 201910288926A CN 110069408 A CN110069408 A CN 110069408A
- Authority
- CN
- China
- Prior art keywords
- barrier
- data
- perception
- sub
- sensory perceptual
- 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.)
- Pending
Links
- 230000001953 sensory effect Effects 0.000 title claims abstract description 113
- 238000010998 test method Methods 0.000 title claims abstract description 22
- 230000004888 barrier function Effects 0.000 claims abstract description 314
- 230000008447 perception Effects 0.000 claims abstract description 199
- 238000012360 testing method Methods 0.000 claims abstract description 110
- 238000000034 method Methods 0.000 claims abstract description 45
- 230000008569 process Effects 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims description 20
- 230000015654 memory Effects 0.000 claims description 16
- 238000001914 filtration Methods 0.000 claims description 11
- 241001269238 Data Species 0.000 claims description 5
- 230000005055 memory storage Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 25
- 238000013461 design Methods 0.000 description 13
- 238000004891 communication Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000005194 fractionation Methods 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 208000010877 cognitive disease Diseases 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3692—Test management for test results analysis
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the present invention provides a kind of automatic driving vehicle sensory perceptual system test method and device, comprising: obtains barrier labeled data, barrier labeled data is the data that user carries out that barrier is marked according to drive test data packet;Barrier perception data is obtained, barrier perception data is the data that the sensory perceptual system of automatic driving vehicle is perceived according to drive test data packet;Matching treatment is carried out according to barrier labeled data and barrier perception data, obtains the test result of sensory perceptual system.Automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention and device, by obtaining drive test data packet, and then break the barriers labeled data and barrier perception data matching treatment obtain the test result of sensory perceptual system, determine quantitative criteria, the validity that sensory perceptual system is assessed according to test result saves manpower, while test process is run without vehicle actually road test, the failure rate for reducing automatic driving vehicle drive test, improves the reliability of sensory perceptual system.
Description
Technical field
The present embodiments relate to artificial intelligence field more particularly to a kind of automatic driving vehicle sensory perceptual system test methods
And device.
Background technique
Autonomous driving vehicle is also known as pilotless automobile, is a kind of comprehensive sensory perceptual system driving, and computer software control is real
Existing unpiloted intelligent vehicle.It is the important component of the following intelligent transportation, possesses very bright prospect, is to work as
One of the hot spot of vehicle in front engineering field research.
Automatic driving vehicle generally includes sensory perceptual system and vehicle control system, wherein sensory perceptual system is automatic Pilot system
Unite one of most basic composition system, can be perceived by sensory perceptual system, in vehicle travel process around locating for environment, including
Vehicle, personage, animal, road sign etc., collectively referred to here in as barrier.Information of the autonomous driving vehicle according to barrier on road, knot
Intelligent decision, navigation, planning, control system are closed, unpiloted purpose is reached.Therefore, on autonomous driving vehicle before road, it is necessary to
Test assessment is carried out to sensory perceptual system, road safety and the upper road testing efficiency of raising in guarantee.It is run currently, generalling use practical examination
Form the sensory perceptual system of automatic driving vehicle is tested, i.e., object of placing obstacles on closed test site makes automatic
It drives vehicle to travel on the experimental site for being provided with barrier, perception is measured according to the driving condition of automatic driving vehicle
Whether system is normal.
It is existing actually examination run by way of come measure automatic driving vehicle sensory perceptual system method, there are tests can
Not high, the time-consuming and laborious problem by property.
Summary of the invention
The embodiment of the present invention is to solve the above problems, provide a kind of automatic driving vehicle sensory perceptual system test method and dress
It sets.
In a first aspect, the embodiment of the present invention provides a kind of automatic driving vehicle sensory perceptual system test method, comprising: obtain barrier
Hinder object labeled data, the barrier labeled data is the data that user carries out that barrier is marked according to drive test data packet;
Barrier perception data is obtained, the barrier perception data is the sensory perceptual system of automatic driving vehicle according to
The data that drive test data packet perceives;
Matching treatment is carried out according to the barrier labeled data and the barrier perception data, obtains the perception system
The test result of system.
It is described to be carried out according to the barrier labeled data and the barrier perception data in a kind of possible design
Matching treatment obtains the test result of the sensory perceptual system, comprising:
Deconsolidation process is carried out to the barrier labeled data according to prefixed time interval, obtains markup information, the mark
Infusing information includes at least two frame labeled data and each sub- labeled data corresponding first time;
Deconsolidation process is carried out to the barrier perception data according to the prefixed time interval, obtains perception information, institute
Stating perception information includes at least two frame perception datas and each sub- perception data corresponding second time;
Matching treatment is carried out according to the markup information and the perception information, obtains the test knot of the sensory perceptual system
Fruit.
It is described that matching treatment is carried out according to the markup information and the perception information in a kind of possible design, it obtains
To the test result of the sensory perceptual system, comprising:
Processing is filtered to the markup information and the perception information, obtains the sub- labeled data and of the first quantity
The sub- perception data of two quantity;
In the sub- labeled data of first quantity and the sub- perception data of second quantity, obtain at the first time and
The sub- labeled data of second time identical target and the sub- perception data of target;
Matching treatment is carried out according to the sub- labeled data of the target and the sub- perception data of the target, obtains the perception system
The test result of system.
It is described to be carried out according to the sub- labeled data of the target and the sub- perception data of the target in a kind of possible design
Matching treatment obtains the test result of the sensory perceptual system, comprising:
Obtain the boundary point coordinate of the first barrier in the sub- labeled data of the target;
Obtain the boundary point coordinate of the second barrier in the sub- perception data of the target;
According to the boundary point coordinate of first barrier and the boundary point coordinate of second barrier, obstacle is obtained
Object matching result;
According to each barrier matching result, the test result of the sensory perceptual system is obtained.
In a kind of possible design, the boundary point coordinate and second obstacle according to first barrier
The boundary point coordinate of object obtains barrier matching result, comprising:
According to the boundary point coordinate of first barrier, the first covering frame is determined, described in the first covering frame covering
First barrier;
According to the boundary point coordinate of second barrier, the second covering frame is determined, described in the second covering frame covering
Second barrier;
If the registration of first covering frame and second covering frame is more than preset ratio, the first barrier and institute
The matching of the second barrier is stated, target obstacle is obtained.
It is described according to each barrier matching result in a kind of possible design, obtain the survey of the sensory perceptual system
Test result, comprising:
According to each barrier matching result, total target obstacle quantity is obtained;
According to the first barrier quantity in total target obstacle quantity and the barrier labeled data, obtain
The homing rate;
According to the second barrier quantity in total target obstacle quantity and the barrier perception data, obtain
The accuracy rate.
It is described that processing is filtered to the markup information and the perception information in a kind of possible design, it obtains
The sub- labeled data of first quantity and the sub- perception data of the second quantity, comprising:
According to the markup information and the perception information, invalid frame data are obtained, wherein the invalid frame data packet
It includes: the obstacle of the sub- labeled data and sub- perception data frame data non-shared on the barrier time, opposite lane
Object data and beyond perception effective range barrier data;
According to the invalid frame data, processing is filtered to the markup information and the perception information, obtains first
The sub- labeled data of quantity and the sub- perception data of the second quantity, first quantity are identical as second quantity.
Second aspect, the embodiment of the present invention provide a kind of automatic driving vehicle sensory perceptual system integrated test set, comprising:
First obtains module, and for obtaining barrier labeled data, the barrier labeled data is user according to drive test
Data packet carries out the data that barrier marks;
Second obtains module, and for obtaining barrier perception data, the barrier perception data is automatic driving vehicle
The data that are perceived according to the drive test data packet of sensory perceptual system;
Matching module, for carrying out matching treatment according to the barrier labeled data and the barrier perception data,
Obtain the test result of the sensory perceptual system.
In alternatively possible embodiment, the matching module is specifically used for:
Deconsolidation process is carried out to the barrier labeled data according to prefixed time interval, obtains markup information, the mark
Infusing information includes at least two frame labeled data and each sub- labeled data corresponding first time;
Deconsolidation process is carried out to the barrier perception data according to the prefixed time interval, obtains perception information, institute
Stating perception information includes at least two frame perception datas and each sub- perception data corresponding second time;
Matching treatment is carried out according to the markup information and the perception information, obtains the test knot of the sensory perceptual system
Fruit.
In alternatively possible embodiment, the matching module is specifically used for:
Processing is filtered to the markup information and the perception information, obtains the sub- labeled data and of the first quantity
The sub- perception data of two quantity;
In the sub- labeled data of first quantity and the sub- perception data of second quantity, obtain at the first time and
The sub- labeled data of second time identical target and the sub- perception data of target;
Matching treatment is carried out according to the sub- labeled data of the target and the sub- perception data of the target, obtains the perception system
The test result of system.
In alternatively possible embodiment, the matching module is specifically used for:
Obtain the boundary point coordinate of the first barrier in the sub- labeled data of the target;
Obtain the boundary point coordinate of the second barrier in the sub- perception data of the target;
According to the boundary point coordinate of first barrier and the boundary point coordinate of second barrier, obstacle is obtained
Object matching result;
According to each barrier matching result, the test result of the sensory perceptual system is obtained.
In alternatively possible embodiment, the matching module is specifically used for:
According to the boundary point coordinate of first barrier, the first covering frame is determined, described in the first covering frame covering
First barrier;
According to the boundary point coordinate of second barrier, the second covering frame is determined, described in the second covering frame covering
Second barrier;
If the registration of first covering frame and second covering frame is more than preset ratio, the first barrier and institute
The matching of the second barrier is stated, target obstacle is obtained.
In alternatively possible embodiment, the matching module is specifically used for:
According to each barrier matching result, total target obstacle quantity is obtained;
According to the first barrier quantity in total target obstacle quantity and the barrier labeled data, obtain
Obstacle recognition homing rate;
According to the second barrier quantity in total target obstacle quantity and the barrier perception data, obtain
Obstacle recognition accuracy rate.
In alternatively possible embodiment, the matching module is specifically used for:
According to the markup information and the perception information, invalid frame data are obtained, wherein the invalid frame data packet
It includes: the obstacle of the sub- labeled data and sub- perception data frame data non-shared on the barrier time, opposite lane
Object data and beyond perception effective range barrier data;
According to the invalid frame data, processing is filtered to the markup information and the perception information, obtains first
The sub- labeled data of quantity and the sub- perception data of the second quantity, first quantity are identical as second quantity.
The third aspect, the embodiment of the present invention provide a kind of automatic driving vehicle sensory perceptual system test equipment, comprising: processing
Device, the processor are coupled with memory;
The memory is used for, and stores computer program;
At least one described processor executes the computer executed instructions of memory storage so that it is described at least one
Processor executes the above-mentioned described in any item automatic driving vehicle sensory perceptual system test methods of first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of readable storage medium storing program for executing, deposit in the computer readable storage medium
Computer executed instructions are contained, when processor executes the computer executed instructions, realize that any one of above-mentioned first aspect is appointed
Automatic driving vehicle sensory perceptual system test method described in one.
A kind of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention and device, pass through drive test data
Packet is to obtain barrier labeled data and barrier perception data, while break the barriers labeled data and barrier perception data
Matching treatment obtains the test result of sensory perceptual system, determines quantitative criteria, assesses having for sensory perceptual system according to test result
Effect property effectively saves manpower, while test process is run without vehicle actually road test, also reduces automatic driving vehicle drive test
Failure rate improves the reliability of sensory perceptual system.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is automatic driving vehicle sensory perceptual system test macro schematic diagram provided in an embodiment of the present invention;
Fig. 2 is the flow diagram of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention;
Fig. 3 is the flow diagram of the test result provided in an embodiment of the present invention for obtaining sensory perceptual system;
Fig. 4 is fractionation barrier labeled data schematic diagram provided in an embodiment of the present invention;
Fig. 5 is fractionation barrier perception data schematic diagram provided in an embodiment of the present invention;
Fig. 6 a is filtering schematic diagram data one provided in an embodiment of the present invention;
Fig. 6 b is filtering schematic diagram data two provided in an embodiment of the present invention;
Fig. 7 obtains sensory perceptual system according to the sub- labeled data of target and the sub- perception data of target to be provided in an embodiment of the present invention
The flow diagram of test result;
Fig. 8 is the first covering frame schematic diagram provided in an embodiment of the present invention;
Fig. 9 is the second covering frame schematic diagram provided in an embodiment of the present invention;
Figure 10 is that the first covering frame provided in an embodiment of the present invention and the second covering frame are overlapped schematic diagram;
Figure 11 is the structural schematic diagram of automatic driving vehicle sensory perceptual system integrated test set provided in an embodiment of the present invention;
Figure 12 is the hardware structural diagram of automatic driving vehicle sensory perceptual system test equipment provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is automatic driving vehicle sensory perceptual system test macro schematic diagram provided in an embodiment of the present invention, as shown in Figure 1,
The system includes client 11, server 12 and sensory perceptual system 13, which can be computer, plate or other terminals
Equipment is mainly used for obtaining user according to the barrier labeled data obtained after drive test data packet mark, is sent to server 12.
Sensory perceptual system 13 is mainly used for that barrier perception data is calculated according to drive test data packet, and barrier perception data is sent
To server 12.Server 12 performs corresponding processing after receiving barrier labeled data and barrier perception data, obtains
To sensory perceptual system test result, for evaluating the reliability of sensory perceptual system.
Wherein, drive test data packet (i.e. bag) is that automatic driving vehicle actually drive test tries, obtained test data.The road
Automatic driving vehicle is had recorded in measured data packet in detail and travels all data on the way recorded, including but not limited to localization message
Data, sensory perceptual system sensor information data and radar point cloud data etc..
In some scenes, when client 11 is the equipment with enough processing capacities such as computer, sensory perceptual system 13
Barrier perception data can be sent to client 11, number is perceived according to barrier labeled data and barrier by client 11
According to testing sensory perceptual system, i.e., client 11 is integrated with the function of server 12.
How to be solved with technical solution of the specific embodiment to technical solution of the present invention and the application below above-mentioned
Technical problem is described in detail.These specific embodiments can be combined with each other below, for the same or similar concept
Or process may repeat no more in certain embodiments.Below in conjunction with attached drawing, the embodiment of the present invention is described.
Fig. 2 is the flow diagram of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention, such as Fig. 2
It is shown, comprising:
S21, obtains barrier labeled data, and the barrier labeled data is user according to drive test data packet progress obstacle
The data that object marks;
Wherein, the radar point cloud data in drive test data packet can transfer to user to carry out artificial mark barrier, to obtain
Barrier labeled data.What barrier here was primarily referred to as that automatic driving vehicle encounters in the process of moving may be to traveling
The object impacted, including but not limited to other vehicles, people, animal and road sign etc. need to feel when encountering above-mentioned barrier
Know that system can be perceived accurately, to smoothly avoid.User manually marks radar point cloud data by client, and with
This result as standard barrier.Client is sent to obtained barrier labeled data is manually marked at server
Reason.
S22, obtain barrier perception data, the barrier perception data be automatic driving vehicle sensory perceptual system according to
The data that the drive test data packet perceives;
Sensory perceptual system is one of most basic component part of automatic driving vehicle, and vehicle perceives surrounding by sensory perceptual system
The barriers such as locating environment, including vehicle, road sign, then vehicle is according to the obstacle information on road, in conjunction with intelligent decision,
Navigation, planning and control system, to reach unpiloted purpose.Actually in the traveling of road, sensory perceptual system is by multiple
Sensor perceives the information of surrounding, and in embodiments of the present invention, when obtaining barrier perception data, vehicle does not need reality
The road Ji Shang, sensory perceptual system is according to the available barrier perception data of drive test data packet, for example, the sensory perceptual system can pass through road
Barrier perception data is calculated in sensor information data in measured data packet.The sensory perceptual system is obtaining barrier perception
After data, barrier perception data is occurred to server.
S23 carries out matching treatment according to the barrier labeled data and the barrier perception data, obtains the sense
Know the test result of system.
After getting barrier labeled data and barrier perception data, needs to carry out matching treatment, be marked with barrier
Result of the data as standard barrier.For example, by mutually in the same time by the barrier perceived in barrier perception data believe
Breath matched with the obstacle information in barrier labeled data, matched content include but is not limited to barrier size,
Position, shape and area etc..According to matched as a result, barrier and barrier in available barrier perception data mark
The coincidence ratio of barrier in data, if the ratio of coincidence is bigger, then it represents that the perceptual performance of the sensory perceptual system is more reliable, on the contrary
It is then more unreliable.
A kind of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention, is obtained by drive test data packet
Barrier labeled data and barrier perception data are taken, while at break the barriers labeled data and barrier perception data matching
Reason determines quantitative criteria to obtain the test result of sensory perceptual system, and the validity of sensory perceptual system is assessed according to test result, is had
Effect saves manpower, while test process is run without vehicle actually road test, also reduces the failure rate of automatic driving vehicle drive test,
Improve the reliability of sensory perceptual system.
Below with reference to Fig. 3, and specific embodiment is used, the process of the matching treatment in Fig. 2 embodiment is carried out detailed
Explanation.Fig. 3 is the flow diagram of the test result provided in an embodiment of the present invention for obtaining sensory perceptual system, as shown in figure 3, packet
It includes:
S31 carries out deconsolidation process to the barrier labeled data according to prefixed time interval, obtains markup information, institute
Stating markup information includes at least two frame labeled data and each sub- labeled data corresponding first time.
S32 carries out deconsolidation process to the barrier perception data according to the prefixed time interval, obtains perception letter
Breath, the perception information include at least two frame perception datas and each sub- perception data corresponding second time.
After obtaining barrier labeled data and barrier perception data, that due to the description of two parts of data is all a Duan Chixu
Process, the position that barrier occurs at different times is also inconsistent, and in order to match two parts of data, the present invention is real
It applies example and deconsolidation process is carried out to barrier labeled data according to prefixed time interval, obtain markup information, wherein between preset time
It every that can set according to actual needs, does not limit specifically herein, under normal circumstances, prefixed time interval can be set in
Between 50ms-1s.
By taking prefixed time interval is 100ms as an example, Fig. 4 is that fractionation barrier labeled data provided in an embodiment of the present invention shows
It is intended to, Fig. 4 shows the barrier 42 marked on each frame labeled data 41 and sub- labeled data after splitting.By barrier
Labeled data is split according to time sequencing, and every 100ms is 1 frame labeled data, i.e. the rectangular graphic data of each of figure,
Object in rectangle as manually marks out the barrier come.Meanwhile obtain each frame labeled data it is corresponding first when
Between, for example, the second frame labeled data corresponding first time is t when the sub- labeled data of first frame corresponding first time is t
+ 100ms, third frame labeled data corresponding first time are t+200ms, and so on.
Fig. 5 is fractionation barrier perception data schematic diagram provided in an embodiment of the present invention, as shown in figure 5, including after splitting
Each frame perception data 51 and sub- perception data on the barrier 52 that perceives.Likewise, according to prefixed time interval pair
Barrier perception data is split, and the sub- perception data of multiframe and each sub- perception data corresponding second time are obtained.It will mark
The sub- labeled data of multiframe in information and the sub- perception data of the multiframe in perception information carry out matching treatment, obtain sensory perceptual system
Test result.S33 is filtered processing to the markup information and the perception information, obtains the son mark number of the first quantity
According to the sub- perception data with the second quantity.
During specific implementation, processing can be filtered to the invalid data in markup information and perception information, from
And matching treatment is carried out by valid data.
In one possible implementation, invalid frame data can be obtained according to markup information and perception information.At this
In embodiment, which for example may include: that sub- labeled data and sub- perception data are non-shared on the barrier time
Frame data, opposite lane barrier data and barrier data beyond perception effective range;Then, according to invalid frame
Data are filtered processing to markup information and the perception information, i.e., weed out the nothing in markup information and perception information
Frame data are imitated, the sub- labeled data of the first quantity and the sub- perception data of the second quantity are obtained.
According to above-mentioned invalid frame data to markup information and perception information filtration treatment after, obtain the first quantity son mark
The sub- perception data of data and the second quantity, since filtered sub- labeled data and sub- perception data are an a pair in time
It answers, therefore the first quantity and the second quantity are equal.
Below with reference to Fig. 6 a and Fig. 6 b, it is described in detail with process of the specific example to filtration treatment.
Fig. 6 a be filtering schematic diagram data one provided in an embodiment of the present invention, as Fig. 6 a showed filter labeled data with
The process of sub- perception data frame data non-shared on the barrier time, including sub- labeled data 61,62 and of sub- perception data
Non-shared frame data 63 on time.Since markup information and perception information are not necessarily completely corresponding in time, for example, mark
The information corresponding time is 0s-10s, and the perception information corresponding time is 0.2s-9.8s, and perception information is in 0s-0.2s at this time
And be not no data between 9.8s-10s, and markup information and perception information must could mutually matched in the same time
Processing, therefore 0s-0.2s and 9.8s-10s sections of markup information are invalid frame data, need to filter out.Assuming that with 100ms number
According to for a frame, then need to remove front cross frame data and last two frame data in markup information.
Fig. 6 b is filtering schematic diagram data two provided in an embodiment of the present invention, and Fig. 6 b shows the obstacle for filtering opposite lane
The process of object data and the barrier data beyond perception effective range, as shown in Figure 6 b, the sense including automatic driving vehicle
Know range, i.e. circle in Fig. 6 b further includes the barrier 64 outside sensing range, the lane obstructions object 65 in the same direction in sensing range
With the opposite lane 66 in sensing range.When artificial mark barrier, since the barrier in opposite lane will not influence driving vehicle
Traveling, therefore manually will not by the barrier in opposite lane mark enter.If sensory perceptual system perceives the barrier in opposite lane
Hinder object, then needs to go this frame perception data and time upper sub- labeled data corresponding with the sub- perception data at this time
It removes.And being limited in scope due to sensory perceptual system perception, more than the barrier of its sensing range, sensory perceptual system can not be perceived, because
This be also required to by a corresponding frame labeled data of the barrier of the mark other than sensory perceptual system and on the time with the sub- mark
The corresponding sub- perception data removal of data.
S34, in the sub- labeled data of first quantity and the sub- perception data of second quantity, when obtaining first
Between the sub- labeled data of identical with the second time target and the sub- perception data of target.
S35 carries out matching treatment according to the sub- labeled data of the target and the sub- perception data of the target, obtains the sense
Know the test result of system.
After filtration treatment, the sub- labeled data of the corresponding target of acquisition time and the sub- perception data of target, i.e., when first
Between the sub- labeled data of identical with the second time target and the sub- perception data of target.Since automatic driving vehicle is in driving scene
In traveling, the boundary point position that different moments difference barrier perceives is all different, when first time and the second time
When identical, corresponding each sub- labeled data of frame target and each frame target sub- perception data at the time of be all it is identical, at this time
Matching treatment is carried out to the sub- labeled data of target and the sub- perception data of target, obtains the test result of sensory perceptual system.
A kind of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention, is obtained by drive test data packet
Barrier labeled data and barrier perception data are taken, while by splitting barrier labeled data and barrier perception data
And matching treatment is carried out after filtering out invalid frame data, to obtain the test result of sensory perceptual system, due to filtered target
The test result that labeled data and the sub- perception data of target are corresponded in time, therefore obtained can more effectively be come
System is assessed, and effectively saves manpower, while test process is run without vehicle actually road test, also reduces automatic driving vehicle
The failure rate of drive test improves the reliability of sensory perceptual system.
Below with reference to Fig. 7, and specific embodiment is used, the process of the matching treatment in Fig. 3 embodiment is carried out detailed
Explanation.Fig. 7 obtains sensory perceptual system survey according to the sub- labeled data of target and the sub- perception data of target to be provided in an embodiment of the present invention
The flow diagram of test result, as shown in fig. 7, comprises:
S71 obtains the boundary point coordinate of the first barrier in the sub- labeled data of the target.
S72 obtains the boundary point coordinate of the second barrier in the sub- perception data of the target.
When obtaining the boundary point coordinate of the first barrier and the second barrier, need two parts of numbers related with barrier
Barrier labeled data and barrier perception number can be being got according to the step of all going under the same coordinate system, converting coordinate system
According to progress, it can also carry out after getting the sub- labeled data of target and the sub- perception data of target, be not especially limited herein.
After determining coordinate system, labeled data for each frame target and each sub- perception data of frame target be can get corresponding
The boundary point coordinate of first barrier and the second barrier.
S73 determines the first covering frame, the first covering frame covering according to the boundary point coordinate of first barrier
First barrier.
S74 determines the second covering frame, the second covering frame covering according to the boundary point coordinate of second barrier
Second barrier.
Specifically, since the shape of the first barrier and the second barrier may be irregular shape, in order to improve
Matched efficiency, the second covering that the present embodiment passes through the second barrier of the first covering frame and covering of the first barrier of covering
Frame matches to realize.
For example, the first covering frame can be determined, according to the side of the second barrier according to the boundary point coordinate of the first barrier
Boundary's point coordinate, determines the second covering frame.By the size of the available barrier of boundary point coordinate, so that it is determined that covering frame.It should
Boundary point coordinate can be the coordinate of left margin, right margin, coboundary and lower boundary.Reality of the present embodiment to boundary point coordinate
Existing mode is not particularly limited, as long as the covering frame of covering barrier can be obtained.
Fig. 8 is the first covering frame schematic diagram provided in an embodiment of the present invention, as shown in figure 8, including 81 He of the first barrier
The form of corresponding first covering frame 82, specific covering frame can be set according to actual needs, only be arranged in the embodiment of the present invention
Lift one of rectangular covering frame, the design method of the covering frame be find out the left margin of the first barrier, right margin,
The coordinate of coboundary and lower boundary makes a rectangle, as the first covering frame with this.
Fig. 9 is the second covering frame schematic diagram provided in an embodiment of the present invention, as shown in figure 9, including 91 He of the second barrier
Corresponding second covering frame 92.Likewise, finding out the seat of the left margin of the second barrier, right margin, coboundary and lower boundary
Mark, makes a rectangle, as the second covering frame with this.After obtaining the first covering frame and the second covering frame, compared accordingly
Compared with.
S75, if the registration of first covering frame and second covering frame is more than preset ratio, the first barrier
It is matched with second barrier, obtains target obstacle.
In the present embodiment, the first barrier is matched with second barrier is covered by the first covering frame and second
For the registration of frame come what is determined, registration is higher, i.e., the two more matches.Therefore, if registration is more than preset ratio, then it is assumed that two
Person's matching.The preset ratio is, for example, 0.4 or 0.6 etc., and the present embodiment does not do special limit to the specific implementation of preset ratio
System.Wherein, when target obstacle refers to that the first barrier is matched with second barrier, i.e. the first barrier and the second obstacle
When object is identical, the corresponding barrier of the two.
Figure 10 is that the first covering frame provided in an embodiment of the present invention and the second covering frame are overlapped schematic diagram, as shown in Figure 10,
Including the first covering frame 101 and the second covering frame 102.If the first covering frame 101 and the second covering frame 102 have intersection, and
Registration is more than preset ratio, then the first barrier is matched with the second barrier, wherein and there are many algorithms of registration, below
Illustrate wherein possible one kind:
If the area of the first covering frame is S1, the area of the second covering frame is S2, the first covering frame and the second covering frame weight
The area for closing part is S, then, registration is defined as follows:
N=S/ (S1+S2-S),
N is registration.
When N is more than preset ratio, show that the registration of the first covering frame 101 and the second covering frame 102 is higher, at this time
Match, obtain target obstacle, which is the first barrier or the second barrier after matching, wherein preset ratio
It may be set according to actual conditions, be not specifically limited herein.If N is less than preset ratio, show the first covering frame 101 and
Two covering frames, 102 registration is lower, and the first barrier and the second barrier mismatch at this time.For example, if preset ratio is 0.4,
The area of first covering frame 101 is 36 square centimeters, and the area of the second covering frame 102 is 42 square centimeters, the face of intersection
Product is 36 square centimeters, then registration are as follows:
N=S/ (S1+S2-S)=36/ (36+42-36)=0.86 > 0.4,
Registration is more than preset ratio, the first barrier and the matching of the second barrier at this time.
S76 obtains the test result of the sensory perceptual system according to each barrier matching result.
According to the above method, available all target obstacles, to obtain barrier matching result, barrier
With in result include barrier labeled data in the first barrier quantity and the second barrier in barrier perception data
Quantity.
In one possible implementation, according to each barrier matching result, total target obstacle number is obtained
Amount;According to the first barrier quantity in total target obstacle quantity and the barrier labeled data, obstacle is obtained
Object identifies homing rate;According to the second barrier number in total target obstacle quantity and the barrier perception data
Amount, obtains obstacle recognition accuracy rate.According to the sub- labeled data of each frame target and time upper corresponding each frame target sub- sense
Primary data is matched, and the target obstacle quantity on each frame is obtained, and addition obtains total target obstacle quantity.
Obstacle recognition homing rate refers to that total target obstacle quantity accounts for the first barrier in barrier labeled data
The ratio of quantity, and obstacle recognition accuracy rate refers to that total target obstacle quantity accounts for the second obstacle in barrier perception data
The ratio of the quantity of object indicates obstacle recognition accuracy rate and obstacle recognition homing rate respectively with precision and recall,
Then have:
Precision=(match_total/perception_total) * 100%,
Recall=(match_total/labeled_total) * 100%,
Wherein, match_total indicates total matching barrier quantity, and perception_total indicates barrier perception
The quantity of second barrier in data, labeled_total indicate the quantity of the first barrier in barrier labeled data.Work as barrier
When hindering the quantity of the second barrier in the quantity of the first barrier and barrier perception data in object labeled data constant, if total
Target obstacle quantity increases, i.e. barrier labeled data and barrier perception data matching degree is higher, at this time obstacle recognition
Accuracy rate and obstacle recognition homing rate are all increased, therefore obstacle recognition accuracy rate and obstacle recognition homing rate conduct
The test result of sensory perceptual system can carry out comprehensive assessment to sensory perceptual system.It further, can be by the obstacle recognition of acquisition
Accuracy rate and obstacle recognition homing rate, obstacle recognition accuracy rate corresponding with old version and obstacle recognition homing rate into
Row compares, if obstacle recognition accuracy rate and obstacle recognition homing rate are all larger than the corresponding data of old version, shows this
The sensory perceptual system comprehensive assessment of version is more excellent, better reliability, conversely, then showing the sensory perceptual system of the version not as good as history version
This, then can choose and not issue.
A kind of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention, is obtained by drive test data packet
Barrier labeled data and barrier perception data are taken, while by splitting barrier labeled data and barrier perception data
And matching treatment is carried out after filtering out invalid frame data, to obtain the obstacle recognition accuracy rate and homing rate of sensory perceptual system, more
It is intuitive that sensory perceptual system is assessed, manpower is effectively saved, while test process is run without vehicle actually road test,
The failure rate for reducing automatic driving vehicle drive test, improves the reliability of sensory perceptual system.
The scheme of the application is described in detail with a specific embodiment below.
Step 101, continuous automatic driving vehicle drive test data packet is chosen, i.e. bag, the inside has recorded automatic Pilot in detail
All data that vehicle driving on the way records, including radar point cloud data, localization message data and sensor information data etc.,
All data inside bag is mainly to pass through camera and radar signal acquisition, is by the inside when eventually forming drive test data packet
All data be converted into binary form and saved;
Step 102, the radar point cloud data in selecting step 101 transfers to mark personnel to mark road obstacle data.It will
The barrier labeled data got, the mode according to the time sort and split, and every 100ms is 1 frame data, interior to contain the moment
Barrier on radar points cloud.After mark personnel are labeled, each barrier has corresponding mark on each frame data
ID, mark ID are used to distinguish and cognitive disorders object;
Step 103, integration testing main program is enabled, and subscribes to the topic information of needs, is believed by subscribing to corresponding topic
Breath can cross the information result for obtaining sensor, and subscribing to not only includes sensor information in topic, calculate there are also aware software
Barrier topic information;
Step 104, the software that starting automatic driving vehicle sensory perceptual system relies on, and ensure that these softwares can operate normally,
For simulating the scene of automatic driving vehicle traveling on the way.
Step 105, start to play back the bag that automatic driving vehicle drive test data records, playback here is primarily referred to as bag
In binary data convert back the data that camera and radar obtain, wherein including sensor information data, automatic driving vehicle
Sensory perceptual system by sensor information data, barrier perception data can be got;
Step 106, integration testing main program collects correlation topic information, after bag is finished, closes aware software
And the program accordingly relied on, start to calculate the disturbance of perception object data being collected into.
The process for calculating matching barrier data is as follows:
(1) in disturbance of perception object data, each topic information has the record temporal information that barrier occurs at that time.
The barrier data that the barrier data either marked still perceive, their corresponding temporal informations are all consistent.According to
This, the data perceived are sorted sequentially in time.
(2) according to the sequence of time, perception information and all shared frame of markup information is selected, extra frame is filtered out.Than
The data frame having in such as sub- labeled data, and the frame of sub- perception data not.
(3) the useless frame data in data frame, such as the vehicle in opposite lane are filtered out, or beyond perception effective range
Barrier data etc..
(4) left data are directed to, are sorted sequentially in time.Compare frame by frame, according to the algorithm for comparing barrier, sieve
Select matched target obstacle.The quantity for recording the barrier of every frame in mark and perception respectively, matches the quantity of barrier.
(5) data for counting all data frames, calculate target obstacle quantity total in data packet, are denoted as match_
total.The first barrier quantity in barrier labeled data is calculated for labeled_total, in barrier perception data
Second barrier quantity is perception_total, is calculated:
Recall=(match_total/labeled_total) * 100%,
Precision=(match_total/perception_total) * 100%.
(6) recall calculated in (5) and precision are compared with historical data, if recall is improved,
The probability for illustrating that manually mark barrier data are perceived is bigger.If precision is improved, illustrate the perception system on vehicle
Uniting, the redundancy barrier perceived is few, and accuracy rate just improves naturally.
A kind of automatic driving vehicle sensory perceptual system test method provided in an embodiment of the present invention, is obtained by drive test data packet
Barrier labeled data and barrier perception data are taken, while at break the barriers labeled data and barrier perception data matching
Reason determines quantitative criteria to obtain the test result of sensory perceptual system, and barrier labeled data and barrier perception data are carried out
Matching treatment obtains obstacle recognition accuracy rate and homing rate, and assesses the validity of sensory perceptual system on this basis, measures
The reliability of sensory perceptual system, effectively saves manpower.Meanwhile test method provided in an embodiment of the present invention, it can be as publication
The reference of new aware software version, before issuing new version every time, the obstacle recognition for testing new version and legacy version is accurate
Rate and homing rate are unsatisfactory for the aware software version of standard, directly not issue, to reduce the event of automatic driving vehicle drive test
Barrier rate improves reliability.
Figure 11 is the structural schematic diagram of automatic driving vehicle sensory perceptual system integrated test set provided in an embodiment of the present invention,
As shown in figure 11, including first obtains the acquisition module 112 of module 111, second and matching module 113, in which:
For first acquisition module 111 for obtaining barrier labeled data, the barrier labeled data is user according to road
Measured data packet carries out the data that barrier marks;
For second acquisition module 112 for obtaining barrier perception data, the barrier perception data is automatic Pilot vehicle
The data that are perceived according to the drive test data packet of sensory perceptual system;
Matching module 113 according to the barrier labeled data and the barrier perception data for carrying out at matching
Reason, obtains the test result of the sensory perceptual system.
In a kind of possible design, matching module 113 is specifically used for:
Deconsolidation process is carried out to the barrier labeled data according to prefixed time interval, obtains markup information, the mark
Infusing information includes at least two frame labeled data and each sub- labeled data corresponding first time;
Deconsolidation process is carried out to the barrier perception data according to the prefixed time interval, obtains perception information, institute
Stating perception information includes at least two frame perception datas and each sub- perception data corresponding second time;
Matching treatment is carried out according to the markup information and the perception information, obtains the test knot of the sensory perceptual system
Fruit.
In a kind of possible design, matching module 113 also particularly useful for:
Processing is filtered to the markup information and the perception information, obtains the sub- labeled data and of the first quantity
The sub- perception data of two quantity;
In the sub- labeled data of first quantity and the sub- perception data of second quantity, obtain at the first time and
The sub- labeled data of second time identical target and the sub- perception data of target;
Matching treatment is carried out according to the sub- labeled data of the target and the sub- perception data of the target, obtains the perception system
The test result of system.
In a kind of possible design, matching module 113 also particularly useful for:
Obtain the boundary point coordinate of the first barrier in the sub- labeled data of the target;
Obtain the boundary point coordinate of the second barrier in the sub- perception data of the target;
According to the boundary point coordinate of first barrier and the boundary point coordinate of second barrier, obstacle is obtained
Object matching result;
According to each barrier matching result, the test result of the sensory perceptual system is obtained.
In a kind of possible design, matching module 113 also particularly useful for:
According to the boundary point coordinate of first barrier, the first covering frame is determined, described in the first covering frame covering
First barrier;
According to the boundary point coordinate of second barrier, the second covering frame is determined, described in the second covering frame covering
Second barrier;
If the registration of first covering frame and second covering frame is more than preset ratio, the first barrier and institute
The matching of the second barrier is stated, target obstacle is obtained.
In a kind of possible design, matching module 113 also particularly useful for:
According to each barrier matching result, total target obstacle quantity is obtained;
According to the first barrier quantity in total target obstacle quantity and the barrier labeled data, obtain
Obstacle recognition homing rate;
According to the second barrier quantity in total target obstacle quantity and the barrier perception data, obtain
Obstacle recognition accuracy rate.
In a kind of possible design, matching module 113 also particularly useful for:
According to the markup information and the perception information, invalid frame data are obtained, wherein the invalid frame data packet
It includes: the obstacle of the sub- labeled data and sub- perception data frame data non-shared on the barrier time, opposite lane
Object data and beyond perception effective range barrier data;
According to the invalid frame data, processing is filtered to the markup information and the perception information, obtains first
The sub- labeled data of quantity and the sub- perception data of the second quantity, first quantity are identical as second quantity.
Device provided in an embodiment of the present invention can be used for executing the technical solution of above method embodiment, realization principle
Similar with technical effect, details are not described herein again.
Figure 12 is the hardware structural diagram of automatic driving vehicle sensory perceptual system test equipment provided in an embodiment of the present invention,
As shown in figure 12, which includes: at least one processor 121 and memory 122.Its
In, processor 121 and memory 122 are connected by bus 123.
Optionally, model determination further includes communication component.For example, communication component may include receiver and/or transmission
Device.
During specific implementation, at least one processor 121 executes the computer execution that the memory 122 stores and refers to
It enables, so that at least one processor 121 executes automatic driving vehicle sensory perceptual system test method as above.
The specific implementation process of processor 121 can be found in above method embodiment, and it is similar that the realization principle and technical effect are similar,
Details are not described herein again for the present embodiment.
In the embodiment shown in above-mentioned Figure 12, it should be appreciated that processor can be central processing unit (English: Central
Processing Unit, referred to as: CPU), it can also be other general processors, digital signal processor (English: Digital
Signal Processor, referred to as: DSP), specific integrated circuit (English: Application Specific Integrated
Circuit, referred to as: ASIC) etc..General processor can be microprocessor or the processor is also possible to any conventional place
Manage device etc..Hardware processor can be embodied directly in conjunction with the step of invention disclosed method and executes completion, or with handling
Hardware and software module combination in device execute completion.
Memory may include high speed RAM memory, it is also possible to and it further include non-volatile memories NVM, for example, at least one
Magnetic disk storage.
Bus can be industry standard architecture (Industry Standard Architecture, ISA) bus, outer
Portion's apparatus interconnection (Peripheral Component, PCI) bus or extended industry-standard architecture (Extended
Industry Standard Architecture, EISA) bus etc..Bus can be divided into address bus, data/address bus, control
Bus etc..For convenient for indicating, the bus in illustrations does not limit only a bus or a type of bus.
The application also provides a kind of computer readable storage medium, and calculating is stored in the computer readable storage medium
Machine executes instruction, and when processor executes the computer executed instructions, realizes automatic driving vehicle perception as described above system
Unified test method for testing.
Above-mentioned computer readable storage medium, above-mentioned readable storage medium storing program for executing can be by any kind of volatibility or non-
Volatile storage devices or their combination realize that, such as static random access memory (SRAM), electrically erasable is only
It reads memory (EEPROM), Erasable Programmable Read Only Memory EPROM (EPROM), programmable read only memory (PROM) is read-only to deposit
Reservoir (ROM), magnetic memory, flash memory, disk or CD.Readable storage medium storing program for executing can be general or specialized computer capacity
Any usable medium enough accessed.
A kind of illustrative readable storage medium storing program for executing is coupled to processor, to enable a processor to from the readable storage medium storing program for executing
Information is read, and information can be written to the readable storage medium storing program for executing.Certainly, readable storage medium storing program for executing is also possible to the composition portion of processor
Point.Processor and readable storage medium storing program for executing can be located at specific integrated circuit (Application Specific Integrated
Circuits, referred to as: ASIC) in.Certainly, processor and readable storage medium storing program for executing can also be used as discrete assembly and be present in equipment
In.
The division of the unit, only a kind of logical function partition, there may be another division manner in actual implementation,
Such as multiple units or components can be combined or can be integrated into another system, or some features can be ignored, or not hold
Row.Another point, shown or discussed mutual coupling, direct-coupling or communication connection can be through some interfaces,
The indirect coupling or communication connection of device or unit can be electrical property, mechanical or other forms.
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.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can
To be done through the relevant hardware of the program instructions.Program above-mentioned can be stored in a computer readable storage medium.
When being executed, execution includes the steps that above-mentioned each method embodiment to the program;And storage medium above-mentioned includes: ROM, RAM, magnetic
The various media that can store program code such as dish or CD.
Claims (10)
1. a kind of automatic driving vehicle sensory perceptual system test method characterized by comprising
Barrier labeled data is obtained, the barrier labeled data is that user marks according to drive test data packet progress barrier
The data arrived;
Barrier perception data is obtained, the barrier perception data is the sensory perceptual system of automatic driving vehicle according to the drive test
The data that data packet perceives;
Matching treatment is carried out according to the barrier labeled data and the barrier perception data, obtains the sensory perceptual system
Test result.
2. the method according to claim 1, wherein described according to the barrier labeled data and the obstacle
Object perception data carries out matching treatment, obtains the test result of the sensory perceptual system, comprising:
Deconsolidation process is carried out to the barrier labeled data according to prefixed time interval, obtains markup information, the mark letter
Breath includes at least two frame labeled data and each sub- labeled data corresponding first time;
Deconsolidation process is carried out to the barrier perception data according to the prefixed time interval, obtains perception information, the sense
Know that information includes at least two frame perception datas and each sub- perception data corresponding second time;
Matching treatment is carried out according to the markup information and the perception information, obtains the test result of the sensory perceptual system.
3. according to the method described in claim 2, it is characterized in that, it is described according to the markup information and the perception information into
Row matching treatment obtains the test result of the sensory perceptual system, comprising:
Processing is filtered to the markup information and the perception information, obtains the sub- labeled data and the second number of the first quantity
The sub- perception data of amount;
In the sub- labeled data of first quantity and the sub- perception data of second quantity, first time and second are obtained
The sub- labeled data of time identical target and the sub- perception data of target;
Matching treatment is carried out according to the sub- labeled data of the target and the sub- perception data of the target, obtains the sensory perceptual system
Test result.
4. according to the method described in claim 3, it is characterized in that, described according to the sub- labeled data of the target and the target
Sub- perception data carries out matching treatment, obtains the test result of the sensory perceptual system, comprising:
Obtain the boundary point coordinate of the first barrier in the sub- labeled data of the target;
Obtain the boundary point coordinate of the second barrier in the sub- perception data of the target;
According to the boundary point coordinate of first barrier and the boundary point coordinate of second barrier, barrier is obtained
With result;
According to each barrier matching result, the test result of the sensory perceptual system is obtained.
5. according to the method described in claim 4, it is characterized in that, the boundary point coordinate according to first barrier with
And the boundary point coordinate of second barrier, obtain barrier matching result, comprising:
According to the boundary point coordinate of first barrier, the first covering frame, the first covering frame covering described first are determined
Barrier;
According to the boundary point coordinate of second barrier, the second covering frame, the second covering frame covering described second are determined
Barrier;
If the registration of first covering frame and second covering frame is more than preset ratio, the first barrier and described the
The matching of two barriers, obtains target obstacle.
6. according to the method described in claim 5, obtaining institute it is characterized in that, described according to each barrier matching result
State the test result of sensory perceptual system, comprising:
According to each barrier matching result, total target obstacle quantity is obtained;
According to the first barrier quantity in total target obstacle quantity and the barrier labeled data, obstacle is obtained
Object identifies homing rate;
According to the second barrier quantity in total target obstacle quantity and the barrier perception data, obstacle is obtained
Object recognition accuracy.
7. according to the method described in claim 3, it is characterized in that, described carry out the markup information and the perception information
Filtration treatment obtains the sub- labeled data of the first quantity and the sub- perception data of the second quantity, comprising:
According to the markup information and the perception information, invalid frame data are obtained, wherein the invalid frame data include: institute
State the barrier data of sub- labeled data and sub- perception data frame data non-shared on the barrier time, opposite lane
And the barrier data beyond perception effective range;
According to the invalid frame data, processing is filtered to the markup information and the perception information, obtains the first quantity
Sub- labeled data and the second quantity sub- perception data, first quantity is identical as second quantity.
8. a kind of automatic driving vehicle sensory perceptual system integrated test set characterized by comprising
First obtains module, and for obtaining barrier labeled data, the barrier labeled data is user according to drive test data
Packet carries out the data that barrier marks;
Second obtains module, and for obtaining barrier perception data, the barrier perception data is the sense of automatic driving vehicle
Know the data that system is perceived according to the drive test data packet;
Matching module is obtained for carrying out matching treatment according to the barrier labeled data and the barrier perception data
The test result of the sensory perceptual system.
9. a kind of automatic driving vehicle sensory perceptual system test equipment characterized by comprising at least one processor and storage
Device;
The memory stores computer executed instructions;
At least one described processor executes the computer executed instructions of the memory storage, so that at least one described processing
Device executes automatic driving vehicle sensory perceptual system test method as described in any one of claim 1 to 7.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
It executes instruction, when processor executes the computer executed instructions, realizes as described in any one of claim 1 to 7 automatic
Drive vehicle sensory perceptual system test method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910288926.8A CN110069408A (en) | 2019-04-11 | 2019-04-11 | Automatic driving vehicle sensory perceptual system test method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910288926.8A CN110069408A (en) | 2019-04-11 | 2019-04-11 | Automatic driving vehicle sensory perceptual system test method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110069408A true CN110069408A (en) | 2019-07-30 |
Family
ID=67367471
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910288926.8A Pending CN110069408A (en) | 2019-04-11 | 2019-04-11 | Automatic driving vehicle sensory perceptual system test method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110069408A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108762226A (en) * | 2018-05-14 | 2018-11-06 | 济南浪潮高新科技投资发展有限公司 | A kind of automatic driving vehicle test method, apparatus and system |
CN111753765A (en) * | 2020-06-29 | 2020-10-09 | 北京百度网讯科技有限公司 | Detection method, device and equipment of sensing equipment and storage medium |
CN112541475A (en) * | 2020-12-24 | 2021-03-23 | 北京百度网讯科技有限公司 | Sensing data detection method and device |
CN112598314A (en) * | 2020-12-29 | 2021-04-02 | 潍柴动力股份有限公司 | Method, device, equipment and medium for determining perception confidence of intelligent driving automobile |
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 |
CN112697188A (en) * | 2020-12-08 | 2021-04-23 | 北京百度网讯科技有限公司 | Detection system test method and apparatus, computer device, medium, and program product |
CN112764013A (en) * | 2020-12-25 | 2021-05-07 | 北京百度网讯科技有限公司 | Method, device and equipment for testing automatic driving vehicle perception system and storage medium |
CN112798008A (en) * | 2021-04-06 | 2021-05-14 | 新石器慧通(北京)科技有限公司 | Unmanned vehicle path planning method and device, readable storage medium and electronic equipment |
CN112817873A (en) * | 2021-03-02 | 2021-05-18 | 上海初斟智能科技有限公司 | Millimeter wave radar alarm automatic testing method |
CN112991735A (en) * | 2021-03-05 | 2021-06-18 | 北京百度网讯科技有限公司 | Test method, device and equipment of traffic flow monitoring system |
CN113176097A (en) * | 2021-03-15 | 2021-07-27 | 北京汽车研究总院有限公司 | Detection method of perception module, computer readable storage medium and vehicle |
CN113591888A (en) * | 2020-04-30 | 2021-11-02 | 上海禾赛科技有限公司 | Point cloud data labeling network system and method for laser radar |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107063713A (en) * | 2017-04-27 | 2017-08-18 | 百度在线网络技术(北京)有限公司 | Method of testing and device applied to pilotless automobile |
CN107991898A (en) * | 2016-10-26 | 2018-05-04 | 法乐第(北京)网络科技有限公司 | A kind of automatic driving vehicle simulating test device and electronic equipment |
CN107992016A (en) * | 2016-10-26 | 2018-05-04 | 法乐第(北京)网络科技有限公司 | A kind of automatic driving vehicle analog detection method |
CN108762226A (en) * | 2018-05-14 | 2018-11-06 | 济南浪潮高新科技投资发展有限公司 | A kind of automatic driving vehicle test method, apparatus and system |
US20180349713A1 (en) * | 2017-05-31 | 2018-12-06 | Baidu Usa Llc | Planning control in response to a driving obstruction during operation of an autonomous driving vehicle (adv) |
CN109297725A (en) * | 2018-10-29 | 2019-02-01 | 百度在线网络技术(北京)有限公司 | Test method, device, equipment, medium and the vehicle of vehicle boundary ability |
-
2019
- 2019-04-11 CN CN201910288926.8A patent/CN110069408A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107991898A (en) * | 2016-10-26 | 2018-05-04 | 法乐第(北京)网络科技有限公司 | A kind of automatic driving vehicle simulating test device and electronic equipment |
CN107992016A (en) * | 2016-10-26 | 2018-05-04 | 法乐第(北京)网络科技有限公司 | A kind of automatic driving vehicle analog detection method |
CN107063713A (en) * | 2017-04-27 | 2017-08-18 | 百度在线网络技术(北京)有限公司 | Method of testing and device applied to pilotless automobile |
US20180349713A1 (en) * | 2017-05-31 | 2018-12-06 | Baidu Usa Llc | Planning control in response to a driving obstruction during operation of an autonomous driving vehicle (adv) |
CN108762226A (en) * | 2018-05-14 | 2018-11-06 | 济南浪潮高新科技投资发展有限公司 | A kind of automatic driving vehicle test method, apparatus and system |
CN109297725A (en) * | 2018-10-29 | 2019-02-01 | 百度在线网络技术(北京)有限公司 | Test method, device, equipment, medium and the vehicle of vehicle boundary ability |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108762226A (en) * | 2018-05-14 | 2018-11-06 | 济南浪潮高新科技投资发展有限公司 | A kind of automatic driving vehicle test method, apparatus and system |
CN113591888A (en) * | 2020-04-30 | 2021-11-02 | 上海禾赛科技有限公司 | Point cloud data labeling network system and method for laser radar |
CN111753765A (en) * | 2020-06-29 | 2020-10-09 | 北京百度网讯科技有限公司 | Detection method, device and equipment of sensing equipment and storage medium |
CN111753765B (en) * | 2020-06-29 | 2024-05-31 | 北京百度网讯科技有限公司 | Sensing device detection method, sensing device detection apparatus, sensing device detection device and storage medium |
CN112697188A (en) * | 2020-12-08 | 2021-04-23 | 北京百度网讯科技有限公司 | Detection system test method and apparatus, computer device, medium, and program product |
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 |
CN112541475B (en) * | 2020-12-24 | 2024-01-19 | 北京百度网讯科技有限公司 | Sensing data detection method and device |
CN112541475A (en) * | 2020-12-24 | 2021-03-23 | 北京百度网讯科技有限公司 | Sensing data detection method and device |
US11869247B2 (en) | 2020-12-24 | 2024-01-09 | Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. | Perception data detection method and apparatus |
CN112764013B (en) * | 2020-12-25 | 2024-03-01 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for testing sensing system of automatic driving vehicle |
CN112764013A (en) * | 2020-12-25 | 2021-05-07 | 北京百度网讯科技有限公司 | Method, device and equipment for testing automatic driving vehicle perception system and storage medium |
CN112598314B (en) * | 2020-12-29 | 2022-08-23 | 潍柴动力股份有限公司 | Method, device, equipment and medium for determining perception confidence of intelligent driving automobile |
CN112598314A (en) * | 2020-12-29 | 2021-04-02 | 潍柴动力股份有限公司 | Method, device, equipment and medium for determining perception confidence of intelligent driving automobile |
CN112817873A (en) * | 2021-03-02 | 2021-05-18 | 上海初斟智能科技有限公司 | Millimeter wave radar alarm automatic testing method |
US20210398419A1 (en) * | 2021-03-05 | 2021-12-23 | Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. | Method, apparatus, and device for testing traffic flow monitoring system |
CN112991735A (en) * | 2021-03-05 | 2021-06-18 | 北京百度网讯科技有限公司 | Test method, device and equipment of traffic flow monitoring system |
CN113176097B (en) * | 2021-03-15 | 2023-03-14 | 北京汽车研究总院有限公司 | Detection method of sensing module, computer readable storage medium and vehicle |
CN113176097A (en) * | 2021-03-15 | 2021-07-27 | 北京汽车研究总院有限公司 | Detection method of perception module, computer readable storage medium and vehicle |
CN112798008A (en) * | 2021-04-06 | 2021-05-14 | 新石器慧通(北京)科技有限公司 | Unmanned vehicle path planning method and device, readable storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110069408A (en) | Automatic driving vehicle sensory perceptual system test method and device | |
CN110287276A (en) | High-precision map updating method, device and storage medium | |
CN109813328A (en) | A kind of driving path planing method, device and vehicle | |
CN104737214B (en) | United reaper and combining machine management system | |
CN108303103A (en) | The determination method and apparatus in target track | |
CN109164785B (en) | Emulation test method, device, equipment and the storage medium of automatic driving vehicle | |
CN106097755A (en) | For identifying the method parked place and/or vacate place | |
CN109101861A (en) | Obstacle identity recognition methods, device, equipment and storage medium | |
CN106485233A (en) | Drivable region detection method, device and electronic equipment | |
CN107223261A (en) | Man-machine hybrid decision method and device | |
CN106909886A (en) | A kind of high accuracy method for traffic sign detection and system based on deep learning | |
CN112528773B (en) | Obstacle information fusion method and device, electronic equipment and storage medium | |
CN106960210A (en) | The method and apparatus of target detection | |
CN107422736A (en) | A kind of unmanned boat independently makes a return voyage system and its method of work | |
CN105575155B (en) | Method and apparatus for determining vehicle traveling information | |
KR20200102378A (en) | Information processing method and device and storage medium | |
CN106803083A (en) | The method and device of pedestrian detection | |
CN111483464B (en) | Dynamic automatic driving lane changing method, equipment and storage medium based on road side unit | |
CN108268831A (en) | The robustness test method and system of a kind of unmanned vehicle vision-based detection | |
CN109871016A (en) | A kind of steered reference line generation method, device, vehicle and server | |
CN106845324A (en) | The treating method and apparatus of guideboard information | |
CN107861992A (en) | A kind of running route processing method and apparatus | |
CN110175654A (en) | A kind of update method and system of track road sign | |
CN109886198B (en) | Information processing method, device and storage medium | |
CN110375736A (en) | Paths planning method, system, equipment and the readable storage medium storing program for executing of smart machine |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190730 |
|
RJ01 | Rejection of invention patent application after publication |