CN110018496A - Obstacle recognition method and device, electronic equipment, storage medium - Google Patents
Obstacle recognition method and device, electronic equipment, storage medium Download PDFInfo
- Publication number
- CN110018496A CN110018496A CN201810023434.1A CN201810023434A CN110018496A CN 110018496 A CN110018496 A CN 110018496A CN 201810023434 A CN201810023434 A CN 201810023434A CN 110018496 A CN110018496 A CN 110018496A
- Authority
- CN
- China
- Prior art keywords
- barrier
- model
- dynamic
- present frame
- obstacle recognition
- 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
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000004888 barrier function Effects 0.000 claims abstract description 306
- 239000000284 extract Substances 0.000 claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000010845 search algorithm Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 13
- 238000001514 detection method Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000000605 extraction Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 210000000352 storage cell Anatomy 0.000 description 1
- 230000003746 surface roughness Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
Abstract
The disclosure is directed to a kind of obstacle recognition method and device, electronic equipment, storage mediums, are related to technical field of data processing, this method comprises: the laser radar data of the present frame obtained by multi-line laser radar extracts barrier feature；The barrier model of present frame is established according to the barrier feature；The barrier model of present frame is compared with the barrier model of previous frame, to judge whether there is dynamic barrier；In judgement there are when the dynamic barrier, kinematics analysis is carried out to the barrier model, with the motion state of the determination dynamic barrier.The accuracy rate of obstacle recognition can be improved in the disclosure.
Description
Technical field
This disclosure relates to technical field of data processing, in particular to a kind of obstacle recognition method, obstacle recognition
Device, electronic equipment and computer readable storage medium.
Background technique
With the fast development of unmanned technology, automatic Pilot technical application is either used in logistics distribution
Other field becomes following development trend.In unmanned technology, the vehicle and pedestrian etc. of surrounding can be effectively identified
Dynamic barrier is extremely important.
In the related technology, dynamic barrier can be identified based on the detection method of geometrical characteristic, such as first can be to obtaining
The laser radar data taken carries out cluster and from feature is wherein extracted, and then carries out data correlation to feature, reuses Kalman
The motion state of filter forecasting dynamic barrier.In addition to this, it can also be based on multi-line laser radar, obtain pedestrian and vehicle
Frame model either point model, and then pedestrian is identified according to barrier contour feature and leg exercise feature.
But in the above method, based on geometry character detection identification pedestrian and vehicle it is possible that miss detection, thus
Cause recognition accuracy lower；Dynamic barrier is scanned by multi-line laser radar and pedestrian is identified according to barrier contour feature
It is larger by scanning angle and environment complexity effect with the method for vehicle, it is low so as to cause recognition efficiency.
It should be noted that information is only used for reinforcing the reason to the background of the disclosure disclosed in above-mentioned background technology part
Solution, therefore may include the information not constituted to the prior art known to persons of ordinary skill in the art.
Summary of the invention
The disclosure is designed to provide a kind of obstacle recognition method and device, electronic equipment, storage medium, Jin Erzhi
Obstacle recognition accuracy rate is low caused by overcoming the problems, such as the limitation and defect due to the relevant technologies to a certain extent less.
Other characteristics and advantages of the disclosure will be apparent from by the following detailed description, or partially by the disclosure
Practice and acquistion.
According to one aspect of the disclosure, a kind of obstacle recognition method is provided, comprising: obtain by multi-line laser radar
Present frame laser radar data extract barrier feature；The barrier mould of present frame is established according to the barrier feature
Type；The barrier model of present frame is compared with the barrier model of previous frame, is hindered with judging whether there is dynamic
Hinder object；In judgement there are when the dynamic barrier, kinematics analysis is carried out to the barrier model, with the determination dynamic
The motion state of barrier.
In a kind of exemplary embodiment of the disclosure, by the barrier of the barrier model and previous frame of present frame
Model is compared, comprising: is calculated separately between the barrier model of present frame and the barrier model of previous frame
Similarity；Maximum similarity is obtained from all similarities of calculating；By maximum similarity to the obstacle of present frame
Object model is matched with the barrier model of previous frame.
In a kind of exemplary embodiment of the disclosure, the calculation formula of the similarity are as follows:
Wherein, i, j are respectively barrier block coefficient, and (x, y) is the coordinate at barrier center, and l, w are barrier respectively
Long and wide, m is barrier laser echo pulse width mean value, and a, b, c, d are weighted value.
In a kind of exemplary embodiment of the disclosure, kinematics analysis is carried out to the barrier model, to determine
State the motion state of dynamic barrier, comprising: obtain the first location information of the barrier model of previous frame；It obtains current
The second location information of the barrier model in frame；By the first location information and the second location information,
Determine the direction of motion and movement velocity of the dynamic barrier.
In a kind of exemplary embodiment of the disclosure, pass through the laser radar number for the present frame that multi-line laser radar obtains
According to extraction barrier feature, comprising: judged in the laser radar data between adjacent laser point based on Neighborhood-region-search algorithm
Distance；Being determined according to the distance whether there is inflection point between adjacent laser point；Calculate there are two sides of laser point of inflection point it
Between first linearity；Calculate second linearity of the laser point there is no inflection point；According to first linearity fitting first
Line segment, and second line segment is fitted according to second linearity；By first line segment and the second line segment, described in extraction
The corner features and edge features of barrier.
In a kind of exemplary embodiment of the disclosure, in judgement there are when the dynamic barrier, the method is also wrapped
It includes: according to the barrier model foundation barrier block list, to be stored described in present frame by the barrier block list
Barrier model.
In a kind of exemplary embodiment of the disclosure, the barrier model includes frame model or point model.
In a kind of exemplary embodiment of the disclosure, the multi-line laser radar is set to vehicle front and vehicle top
Portion.
In a kind of exemplary embodiment of the disclosure, the method also includes: by described in the identification of deep learning method
The type of dynamic barrier.
According to one aspect of the disclosure, a kind of obstacle recognition system is provided, comprising: characteristic extracting module, for leading to
The laser radar data for crossing the present frame of multi-line laser radar acquisition extracts barrier feature；Model building module is used for basis
The barrier feature establishes the barrier model of present frame；Model comparison module, for by the barrier mould of present frame
Type is compared with the barrier model of previous frame, to judge whether there is dynamic barrier；State determining module, for sentencing
Break there are when the dynamic barrier, kinematics analysis is carried out to the barrier model, with the determination dynamic barrier
Motion state.
According to one aspect of the disclosure, a kind of electronic equipment is provided, comprising: processor；And memory, for storing
The executable instruction of the processor；Wherein, the processor is configured to above-mentioned to execute via the executable instruction is executed
Obstacle recognition method described in any one.
According to one aspect of the disclosure, a kind of computer readable storage medium is provided, computer program is stored thereon with,
The computer program realizes obstacle recognition method described in above-mentioned any one when being executed by processor.
A kind of obstacle recognition method, obstacle recognition system, the electronic equipment provided in disclosure exemplary embodiment
And in computer readable storage medium, the laser radar data of the present frame obtained by multi-line laser radar extracts barrier
Feature；The barrier model of present frame is established according to the barrier feature；By the barrier model and upper one of present frame
The barrier model of frame is compared, to judge whether there is dynamic barrier；It is right in judgement there are when the dynamic barrier
The barrier model carries out kinematics analysis, with the motion state of the determination dynamic barrier.On the one hand, pass through laser thunder
Barrier feature is extracted up to data, barrier model is established and is compared, can more accurately determine the fortune of dynamic barrier
Dynamic state, improves the accuracy rate of obstacle recognition；On the other hand, by establishing barrier model and to the barrier mould of different frame
Type is compared, and is avoided the influence of scanning angle and environment complexity, is improved recognition efficiency.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The disclosure can be limited.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.It should be evident that the accompanying drawings in the following description is only the disclosure
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 schematically shows a kind of obstacle recognition method schematic diagram in disclosure exemplary embodiment；
Fig. 2 schematically shows a kind of barrier model schematic in disclosure exemplary embodiment；
Fig. 3 schematically shows the flow chart of obstacle recognition method in disclosure exemplary embodiment；
Fig. 4 schematically shows a kind of block diagram of obstacle recognition system in disclosure exemplary embodiment；
Fig. 5 schematically shows the block diagram of a kind of electronic equipment in disclosure exemplary embodiment；
Fig. 6 schematically shows a kind of program product in disclosure exemplary embodiment.
Specific embodiment
Example embodiment is described more fully with reference to the drawings.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein；On the contrary, thesing embodiments are provided so that the disclosure will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.Described feature, knot
Structure or characteristic can be incorporated in any suitable manner in one or more embodiments.In the following description, it provides perhaps
More details fully understand embodiment of the present disclosure to provide.It will be appreciated, however, by one skilled in the art that can
It is omitted with technical solution of the disclosure one or more in the specific detail, or others side can be used
Method, constituent element, device, step etc..In other cases, be not shown in detail or describe known solution to avoid a presumptuous guest usurps the role of the host and
So that all aspects of this disclosure thicken.
In addition, attached drawing is only the schematic illustrations of the disclosure, it is not necessarily drawn to scale.Identical attached drawing mark in figure
Note indicates same or similar part, thus will omit repetition thereof.Some block diagrams shown in the drawings are function
Energy entity, not necessarily must be corresponding with physically or logically independent entity.These function can be realized using software form
Energy entity, or these functional entitys are realized in one or more hardware modules or integrated circuit, or at heterogeneous networks and/or place
These functional entitys are realized in reason device device and/or microcontroller device.
A kind of obstacle recognition method is provided firstly in this example embodiment, can be applied in unmanned field
The identification of dynamic barrier, perhaps also can be applied to the identification of dynamic barrier or volume in robot field can apply
In the scene of any required identification dynamic object type.Refering to what is shown in Fig. 1, the obstacle recognition method may include following step
It is rapid:
In step s 110, it is special that the laser radar data of the present frame obtained by multi-line laser radar extracts barrier
Sign；
In the step s 120, the barrier model of present frame is established according to the barrier feature；
In step s 130, the barrier model of present frame is compared with the barrier model of previous frame, with
Judge whether there is dynamic barrier；
In step S140, in judgement there are when the dynamic barrier, movement credit is carried out to the barrier model
Analysis, with the motion state of the determination dynamic barrier.
In the obstacle recognition method provided in the present example embodiment, on the one hand, extracted by laser radar data
Barrier feature is established barrier model and is compared, and can more accurately determine the motion state of dynamic barrier, improves
The accuracy rate of obstacle recognition；On the other hand, by establishing barrier model and being compared to the barrier model of different frame,
The influence for avoiding scanning angle and environment complexity, improves recognition efficiency.
Next explanation is further expalined to the obstacle recognition method in the present exemplary embodiment in conjunction with attached drawing.
In step s 110, it is special that the laser radar data of the present frame obtained by multi-line laser radar extracts barrier
Sign.
In this example, multi-line laser radar refers to emitting and receiving the laser rotary distance-measuring equipment of multiple laser, such as
It may include 16 lines, 32 lines, 64 lines etc..Multi-line laser radar can be set on vehicle and its installation site on vehicle
The detection range and detectability of barrier are determined with posture.Biggish detection range in order to obtain, can be by multi-thread laser
Radar is set to vehicle front and vehicle roof.For example, four line laser thunders can be installed in any position of vehicle front
It reaches, and by the front of four line laser radars insertion car body, to obtain the obstacle information in detection environment.Swash relative to three-dimensional
For optical radar, its scan period is shorter, and detecting distance is farther, it is easier to obtain the geometric profile of barrier.In addition to this,
Multi-line laser radar can also be installed in any position at the top of vehicle, more accurately to obtain laser radar data.It needs
It is noted that the multi-line laser radar of roof can be installed higher as far as possible in order to avoid the visual field is blocked.
Wherein, laser radar data can be for example the pedestrian in detection environment, automobile, bicycle, branch, fence, road
The point cloud data of road and all kinds of road affiliated facilities etc..Laser radar data can be arranged according to radar direction of rotation, and every
A point includes the information such as its angle, distance and echo impulse width, wherein the color of echo impulse width and barrier, material
Matter is related with surface roughness.In addition to this it is possible to the pretreatment such as it is filtered, samples to the laser radar data of acquisition,
To guarantee the accuracy rate of entire identification process.
After obtaining laser radar data, it can use feature extraction algorithm and extracted from the laser radar data of present frame
Barrier feature, and then it is corresponding can be extracted by circulation and program respectively from the laser radar data of all frames for each frame
Barrier feature.It should be noted that can include a variety of different types of obstacle object points in every frame laser radar data
Cloud data.Feature extraction algorithm for example may include that (Scale-invariant feature transform, scale is or not SIFT
Become Feature Conversion) algorithm, FAST (Features from Accelerated Segment Test) algorithm or other algorithms.
Specifically, the laser radar data of the present frame obtained by multi-line laser radar extract barrier feature can be with
It include: that the distance between adjacent laser point in the laser radar data is judged based on Neighborhood-region-search algorithm；According to the distance
It determines between adjacent laser point with the presence or absence of inflection point；There are first linearities between two sides of laser point of inflection point for calculating；Meter
Calculate second linearity of the laser point there is no inflection point；It is fitted the first line segment according to first linearity, and according to described the
Bilinear degree is fitted second line segment；By first line segment and the second line segment, the corner features of the barrier are extracted
And edge features.
In this example, the distance between every two adjacent laser point in present frame laser radar data can be calculated first,
Such as it can be calculated by Neighborhood-region-search algorithm either other algorithms.It is then possible to be determined according to the distance of calculating adjacent sharp
It whether there is inflection point between luminous point, for example, if distance is more than the empirical value being previously set, can be determined that this group of phase
There are inflection points between adjacent laser point；If distance is less than the empirical value being previously set, this group of adjacent laser can be determined that
Inflection point is not present between point, wherein empirical value can be configured according to actual needs.Next, presence can be calculated separately
First linearity between two sides of laser point of inflection point and second linearity there is no the laser point of inflection point, wherein line
Property degree be describe Sensor's Static characteristic an important indicator, the linearity is smaller, shows that linear characteristic is better.Further,
It can be fitted the first line segment according to first linearity, and second line segment is fitted according to second linearity, such as end group can be passed through
The methods of method, least square method matching line segment.Finally, the adjacent laser point if there is inflection point can be fitted to the first line segment,
There is no the laser points of inflection point can be fitted to second line segment, then can form barrier by the first line segment and second line segment
Corner features and edge features.
In the step s 120, the barrier model of present frame is established according to the barrier feature.
Present frame laser radar can be established according to the corner features and edge features of the barrier obtained in step S110
The corresponding barrier model of data.Refering to what is shown in Fig. 2, barrier model herein for example may include frame model and point model.
You need to add is that, since the object data point of distant place is sparse, being difficult to judge morphological feature, therefore can in laser radar data
To indicate barrier using point model.After object shifts near, data point becomes intensively, then can indicate obstacle using frame model
Object.Wherein, the length of frame model and wide array respectively indicate different barriers, and point model indicates barrier by point and radius.
It should be noted that the barrier model of previous frame and all frames can be established by the same method.
Next, in step s 130, the barrier model of the barrier model of present frame and previous frame is carried out
Compare, to judge whether there is dynamic barrier.
It, can be by the barrier model of present frame and the obstacle of previous frame in the laser radar data for getting present frame
Object model is compared and matches, by whether there is the side of identical barrier model in comparison different frame laser radar data
Formula judges whether there is dynamic barrier.Dynamic barrier can be for example various types of vehicles and pedestrian etc..Pass through
Dynamic barrier is judged whether there is, vehicle can be made to evade dynamic barrier in time in the process of moving, and then improve nobody
The safety in driving field.
Specifically, the barrier model of present frame is compared with the barrier model of previous frame can wrap
It includes: calculating separately the similarity between the barrier model of present frame and the barrier model of previous frame；From calculating
All similarities in obtain maximum similarity；By maximum similarity to the barrier model of present frame and previous frame
The barrier model is matched.
In this example, it may each comprise multiple barrier models in present frame and previous frame, it is therefore desirable to each barrier
Object model is hindered all to be compared and match, to determine whether there is dynamic barrier, this, which is sentenced, compares between two consecutive frames
It is illustrated for frame model.Firstly, passing through fusion position in barrier matching process in order to effectively track same barrier
Set, length, width and echo impulse width mean value construct a similarity, for calculating each barrier block in next frame
With the similarity between the present frame dynamic barrier that stores in barrier block list, or barrier block in present frame is calculated
It is realized with the similarity between the previous frame dynamic barrier that is stored in barrier block list to establish global similar matrix
By the purpose of similar maximum barrier Model Matching together.Specifically, the similarity as shown in formula (1) can be passed through
Calculation formula calculates separately the similarity between the barrier model of present frame and the barrier model of previous frame,
With determined by similarity whether belong between multiple barriers it is same.Wherein, formula (1) can indicate are as follows:
Wherein, i, j are respectively barrier block coefficient, and (x, y) is the coordinate at barrier center, and l, w are barrier respectively
Long and wide, m is barrier laser echo pulse width mean value, and a, b, c, d are weighted value.Pass through above-mentioned calculating formula of similarity pair
Phenomena such as barrier model is matched, can be to avoid misidentifying and leaking identification more accurately determines dynamic barrier.
Next, all similarities can be traversed, maximum similarity in all similarities is obtained by writing program；Most
It afterwards, can be by obtained maximum similarity to all barrier moulds in all barrier models and previous frame in present frame
Type is matched, such as it is considered that the maximum barrier model of matching degree belongs to the same barrier model.If continuous two
All there is the same barrier model in the either more than two frames of frame, hinders it may be considered that there is dynamic in current detection environment
Hinder object.
For example, the barrier model that present frame extracts includes A, B, C, if barrier model A and previous frame extract
All barrier model Ms, the similarity between N, Q be respectively 0.5,0.8,0.95, then can be according to maximum similarity 0.95
Determine that the barrier model A in barrier model Q and present frame in previous frame belongs to the same dynamic barrier.Next,
The similarity between barrier model M, N that barrier Model B in present frame, C and previous frame extract, example can be determined respectively
If barrier Model B and the similarity of M are 0.9, the similarity with N is 0.5, then can determine the barrier model in previous frame
Barrier Model B in M and present frame belongs to the same dynamic barrier；If barrier MODEL C and the similarity of N are 0.1,
It can then determine that the two is not belonging to same barrier, barrier MODEL C is emerging barrier in present frame.
In step S140, in judgement there are when the dynamic barrier, movement credit is carried out to the barrier model
Analysis, with the motion state of the determination dynamic barrier.
It, can be according to barrier model foundation barrier in judgement there are when dynamic barrier firstly the need of explanation
Block list, with the barrier model for the block list storage present frame that breaks the barriers.It is stored in this dynamic barrier list
Each barrier block may include barrier block number, the information such as acquisition time, location information, barrier moulded dimension.It is logical
The dynamic barrier that crossing will be present is stored in barrier block list, can the barrier model in real time to different frame compare
Compared with to determine that the barrier model extracted is that new barrier or original barrier have changed position.
Next, kinematics analysis can be carried out to barrier model according to the barrier model of extraction, to determine dynamic
The motion state of barrier.Kinematics analysis refers to the fortune such as the equation of motion, track, displacement, the speed, acceleration of research object
Dynamic feature, motion state for example may include the location information of dynamic barrier, the direction of motion, movement velocity etc..It is specific and
Speech, the first location information of the barrier model of available previous frame；And obtain the barrier mould in present frame
The second location information of type；And then by the first location information and the second location information, the dynamic barrier is determined
Hinder the direction of motion and movement velocity of object.
The location information of same barrier model in previous frame and present frame can be extracted respectively by position determination module.
For example, extracting the first location information X of the barrier model Q in previous frame, second of the barrier model A in present frame is extracted
Location information Y.Next, can be according to second location information relative to the orientation of first location information and obtaining for location information
The time is taken, determines the direction of motion and movement velocity of the dynamic barrier.For example, second of the barrier model A in present frame
3 o ' clock orientations of the first location information X of barrier model Q of the location information Y in previous frame can then determine that the dynamic hinders
Object is hindered to move towards 3 o ' clock orientations.In addition to this, either other filtering of Kalman filter, particle filter can also be passed through
Device predicts the motion state of dynamic barrier.
For example, the barrier model that previous frame extracts is round, the barrier mould that present frame extracts
Type is a bigger circle, if determining that two circles belong to same barrier by maximum similarity, it is determined that the barrier
For dynamic barrier, after being matched between present frame and the barrier model of previous frame, this available circle
The dynamic barrier result increasingly closer apart from vehicle.
For another example the barrier model that previous frame extracts is the frame model positioned at left side, what present frame extracted
Barrier model is the frame model positioned at right side, if determining that two frame models belong to same barrier by maximum similarity,
Then determine that the barrier is that dynamic barrier can after being matched between present frame and the barrier model of previous frame
The result on right side is moved to obtain this corresponding dynamic barrier of frame model from left side.By the location information essence for not having to frame
It really predicts the motion informations such as the direction of motion and the movement velocity of dynamic barrier, therefore can preferably avoid these dynamics and hinder
Hinder object, guarantees safety.
In addition to this it is possible to identify the type of the dynamic barrier by deep learning method.
In this example, it can be combined by different barrier moulded dimensions and respectively indicate different types of dynamic disorder
Object, such as when barrier model is frame model, oversize vehicle, bicycle can be uniquely indicated respectively with different long wide arrays
And the different types of dynamic barrier such as pedestrian.For example, a length of 10 can be used, the frame model that width is 8 indicates oversize vehicle, uses
A length of 5, the frame model that width is 4 indicates bicycle, and with a length of 1, the frame model that width is 2 indicates pedestrian.When the frame model got
A length of 5, when width is 4, can uniquely determine the dynamic barrier is pedestrian.
In addition to this it is possible to identify the type of dynamic barrier, by deep learning method to improve obstacle recognition
Accuracy rate and obstacle classification precision.Specifically, depth convolutional neural networks can be constructed, and to depth convolutional Neural
Network is trained to obtain obstacle recognition model；The image of barrier to be identified is input in obstacle recognition model and is obtained
Feature vector, and image classification is carried out to feature vector；It is obtained according to the class label of barrier to be identified and image classification result
To the type of barrier to be identified.
In conclusion establishing barrier model by extracting barrier feature from laser radar data in this example and going forward side by side
Row compares, and can more accurately predict the motion state of dynamic barrier, improve the accuracy rate of obstacle recognition, therefore can be more
These dynamic barriers are avoided well, guarantee safety；By establish barrier model and to the barrier model of different frame into
Row compares, and avoids the influence of scanning angle and environment complexity, improves recognition efficiency；By calculating formula of similarity to barrier
Object model is hindered to be compared to determine dynamic barrier, can be to avoid misidentifying and leaking identification phenomena such as, more accurately identification is dynamic
State barrier；By installing multi-line laser radar in vehicle front and top, it is easier to the geometric profile of barrier is obtained, thus
More accurately obtain laser radar data.
The specific flow chart of obstacle recognition method is diagrammatically illustrated in Fig. 3.Refering to what is shown in Fig. 3, obstacle recognition mistake
Journey specifically includes:
Step S310 establishes barrier model, such as can extract barrier feature, Jin Ergen from laser radar data
Frame model either point model is established according to barrier feature, is specifically included:
311, laser radar data is obtained, such as laser radar data is more accurately obtained by multi-line laser radar；
312, feature extraction is carried out to laser radar data, obtains barrier feature, such as pass through the features such as SIFT, FAST
Extraction algorithm extracts corner features and edge features of barrier etc.；
313, Fusion Features are carried out to barrier feature, to establish barrier model, such as establish frame model or point model.
Step S320 carries out detecting and tracking to dynamic barrier, specific to wrap to predict the motion state of dynamic barrier
It includes:
321, barrier model is matched, to determine whether there is dynamic barrier, such as calculates present frame first
In barrier model and previous frame in barrier model between similarity, then matched by maximum similarity,
To determine dynamic barrier；
322, obtained dynamic barrier is filtered, to establish barrier block list；
323, forecasting-obstacle is obtained as a result, for example according to present frame and previous frame barrier model from barrier list
Location information, predict the motion states such as the direction of motion, the movement velocity of the barrier.
The disclosure additionally provides a kind of obstacle recognition system.Refering to what is shown in Fig. 4, the obstacle recognition system 400 can be with
Include:
The laser radar data of characteristic extracting module 401, the present frame that can be used for obtaining by multi-line laser radar mentions
Take barrier feature；
Model building module 402 can be used for establishing the barrier model of present frame according to the barrier feature；
Model comparison module 403 can be used for the barrier model of present frame and the barrier model of previous frame
It is compared, to judge whether there is dynamic barrier；
State determining module 404, can be used in judgement there are when the dynamic barrier, to the barrier model into
Row kinematics analysis, with the motion state of the determination dynamic barrier.
It should be noted that the detail of each module is known in corresponding barrier in above-mentioned obstacle recognition system
It is described in detail in other method, therefore details are not described herein again.
It should be noted that although being referred to several modules or list for acting the equipment executed in the above detailed description
Member, but this division is not enforceable.In fact, according to embodiment of the present disclosure, it is above-described two or more
Module or the feature and function of unit can embody in a module or unit.Conversely, an above-described mould
The feature and function of block or unit can be to be embodied by multiple modules or unit with further division.
In addition, although describing each step of method in the disclosure in the accompanying drawings with particular order, this does not really want
These steps must be executed in this particular order by asking or implying, or having to carry out step shown in whole could realize
Desired result.Additional or alternative, it is convenient to omit multiple steps are merged into a step and executed by certain steps, and/
Or a step is decomposed into execution of multiple steps etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the disclosure
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, mobile terminal or network equipment etc.) is executed according to disclosure embodiment
Method.
In an exemplary embodiment of the disclosure, a kind of electronic equipment that can be realized the above method is additionally provided.
Person of ordinary skill in the field it is understood that various aspects of the invention can be implemented as system, method or
Program product.Therefore, various aspects of the invention can be embodied in the following forms, it may be assumed that complete hardware embodiment, complete
The embodiment combined in terms of full Software Implementation (including firmware, microcode etc.) or hardware and software, can unite here
Referred to as circuit, " module " or " system ".
The electronic equipment 500 of this embodiment according to the present invention is described referring to Fig. 5.The electronics that Fig. 5 is shown
Equipment 500 is only an example, should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in figure 5, electronic equipment 500 is showed in the form of universal computing device.The component of electronic equipment 500 can wrap
It includes but is not limited to: at least one above-mentioned processing unit 510, at least one above-mentioned storage unit 520, the different system components of connection
The bus 530 of (including storage unit 520 and processing unit 510).
Wherein, the storage unit is stored with program code, and said program code can be held by the processing unit 510
Row, so that various according to the present invention described in the execution of the processing unit 510 above-mentioned " illustrative methods " part of this specification
The step of illustrative embodiments.For example, the processing unit 510 can execute step as shown in fig. 1: in step S110
In, the laser radar data of the present frame obtained by multi-line laser radar extracts barrier feature；In the step s 120, according to
The barrier feature establishes the barrier model of present frame；In step s 130, by the barrier model of present frame with
The barrier model of previous frame is compared, to judge whether there is dynamic barrier；In step S140, in judgement, there are institutes
When stating dynamic barrier, kinematics analysis is carried out to the barrier model, with the motion state of the determination dynamic barrier.
Storage unit 520 may include the readable medium of volatile memory cell form, such as Random Access Storage Unit
(RAM) 5201 and/or cache memory unit 5202, it can further include read-only memory unit (ROM) 5203.
Storage unit 520 can also include program/utility with one group of (at least one) program module 5205
5204, such program module 5205 includes but is not limited to: operating system, one or more application program, other program moulds
It may include the realization of network environment in block and program data, each of these examples or certain combination.
Bus 530 can be to indicate one of a few class bus structures or a variety of, including storage unit bus or storage
Cell controller, peripheral bus, graphics acceleration port, processing unit use any bus structures in a variety of bus structures
Local bus.
Electronic equipment 500 can also be with one or more external equipments 600 (such as keyboard, sensing equipment, bluetooth equipment
Deng) communication, can also be enabled a user to one or more equipment interact with the electronic equipment 500 communicate, and/or with make
Any equipment (such as the router, modulation /demodulation that the electronic equipment 500 can be communicated with one or more of the other calculating equipment
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 550.Also, electronic equipment 500 can be with
By network adapter 560 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network,
Such as internet) communication.As shown, network adapter 560 is communicated by bus 530 with other modules of electronic equipment 500.
It should be understood that although not shown in the drawings, other hardware and/or software module can not used in conjunction with electronic equipment 500, including but not
Be limited to: microcode, device driver, redundant processing unit, external disk drive array, RAID system, tape drive and
Data backup storage system etc..
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can also be realized by software realization in such a way that software is in conjunction with necessary hardware.Therefore, according to the disclosure
The technical solution of embodiment can be embodied in the form of software products, which can store non-volatile at one
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are so that a calculating
Equipment (can be personal computer, server, terminal installation or network equipment etc.) is executed according to disclosure embodiment
Method.
In an exemplary embodiment of the disclosure, a kind of computer readable storage medium is additionally provided, energy is stored thereon with
Enough realize the program product of this specification above method.In some possible embodiments, various aspects of the invention may be used also
In the form of being embodied as a kind of program product comprising program code, when described program product is run on the terminal device, institute
Program code is stated for executing the terminal device described in above-mentioned " illustrative methods " part of this specification according to this hair
The step of bright various illustrative embodiments.
Refering to what is shown in Fig. 6, describing the program product for realizing the above method of embodiment according to the present invention
700, can using portable compact disc read only memory (CD-ROM) and including program code, and can in terminal device,
Such as it is run on PC.However, program product of the invention is without being limited thereto, in this document, readable storage medium storing program for executing can be with
To be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
It is in connection.
Described program product can be using any combination of one or more readable mediums.Readable medium can be readable letter
Number medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can be but be not limited to electricity, magnetic, optical, electromagnetic, infrared ray or
System, device or the device of semiconductor, or any above combination.The more specific example of readable storage medium storing program for executing is (non exhaustive
List) include: electrical connection with one or more conducting wires, portable disc, hard disk, random access memory (RAM), read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read only memory
(CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry readable program code.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Readable signal medium can also be any readable Jie other than readable storage medium storing program for executing
Matter, the readable medium can send, propagate or transmit for by instruction execution system, device or device use or and its
The program of combined use.
The program code for including on readable medium can transmit with any suitable medium, including but not limited to wirelessly, have
Line, optical cable, RF etc. or above-mentioned any appropriate combination.
The program for executing operation of the present invention can be write with any combination of one or more programming languages
Code, described program design language include object oriented program language-Java, C++ etc., further include conventional
Procedural programming language-such as " C " language or similar programming language.Program code can be fully in user
It calculates and executes in equipment, partly executes on a user device, being executed as an independent software package, partially in user's calculating
Upper side point is executed on a remote computing or is executed in remote computing device or server completely.It is being related to far
Journey calculates in the situation of equipment, and remote computing device can pass through the network of any kind, including local area network (LAN) or wide area network
(WAN), it is connected to user calculating equipment, or, it may be connected to external computing device (such as utilize ISP
To be connected by internet).
In addition, above-mentioned attached drawing is only the schematic theory of processing included by method according to an exemplary embodiment of the present invention
It is bright, rather than limit purpose.It can be readily appreciated that the time that above-mentioned processing shown in the drawings did not indicated or limited these processing is suitable
Sequence.In addition, be also easy to understand, these processing, which can be, for example either synchronously or asynchronously to be executed in multiple modules.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
His embodiment.This application is intended to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Adaptive change follow the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure or
Conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by claim
It points out.
Claims (12)
1. a kind of obstacle recognition method characterized by comprising
The laser radar data of the present frame obtained by multi-line laser radar extracts barrier feature；
The barrier model of present frame is established according to the barrier feature；
The barrier model of present frame is compared with the barrier model of previous frame, is hindered with judging whether there is dynamic
Hinder object；
In judgement there are when the dynamic barrier, kinematics analysis is carried out to the barrier model, with the determination dynamic
The motion state of barrier.
2. obstacle recognition method according to claim 1, which is characterized in that by the barrier model of present frame with
The barrier model of previous frame is compared, comprising:
Calculate separately the similarity between the barrier model of present frame and the barrier model of previous frame；
Maximum similarity is obtained from all similarities of calculating；
The barrier model of present frame is matched with the barrier model of previous frame by maximum similarity.
3. obstacle recognition method according to claim 2, which is characterized in that the calculation formula of the similarity are as follows:
Wherein, i, j are respectively barrier block coefficient, and (x, y) is the coordinate at barrier center, l, w be respectively barrier length and
Width, m are barrier laser echo pulse width mean value, and a, b, c, d are weighted value.
4. obstacle recognition method according to claim 1, which is characterized in that carry out kinematics to the barrier model
Analysis, with the motion state of the determination dynamic barrier, comprising:
Obtain the first location information of the barrier model of previous frame；
Obtain the second location information of the barrier model in present frame；
By the first location information and the second location information, determine the dynamic barrier the direction of motion and
Movement velocity.
5. obstacle recognition method according to claim 1, which is characterized in that obtained by multi-line laser radar current
The laser radar data of frame extracts barrier feature, comprising:
The distance between adjacent laser point in the laser radar data is judged based on Neighborhood-region-search algorithm；
Being determined according to the distance whether there is inflection point between adjacent laser point；
There are first linearities between two sides of laser point of inflection point for calculating；
Calculate second linearity of the laser point there is no inflection point；
It is fitted the first line segment according to first linearity, and second line segment is fitted according to second linearity；
By first line segment and the second line segment, the corner features and edge features of the barrier are extracted.
6. obstacle recognition method according to claim 1, which is characterized in that in judgement, there are the dynamic barriers
When, the method also includes:
According to the barrier model foundation barrier block list, to be stored described in present frame by the barrier block list
Barrier model.
7. obstacle recognition method according to claim 1, which is characterized in that the barrier model include frame model or
Point model.
8. obstacle recognition method according to claim 1, which is characterized in that before the multi-line laser radar is set to vehicle
Side and vehicle roof.
9. obstacle recognition method according to claim 1, which is characterized in that the method also includes:
The type of the dynamic barrier is identified by deep learning method.
10. a kind of obstacle recognition system characterized by comprising
Characteristic extracting module, it is special that the laser radar data of the present frame for being obtained by multi-line laser radar extracts barrier
Sign；
Model building module, for establishing the barrier model of present frame according to the barrier feature；
Model comparison module, for the barrier model of present frame to be compared with the barrier model of previous frame, with
Judge whether there is dynamic barrier；
State determining module, for, there are when the dynamic barrier, carrying out movement credit to the barrier model in judgement
Analysis, with the motion state of the determination dynamic barrier.
11. a kind of electronic equipment characterized by comprising
Processor；And
Memory, for storing the executable instruction of the processor；
Wherein, the processor is configured to require 1-9 described in any item via executing the executable instruction and carry out perform claim
Obstacle recognition method.
12. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
Claim 1-9 described in any item obstacle recognition methods are realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810023434.1A CN110018496A (en) | 2018-01-10 | 2018-01-10 | Obstacle recognition method and device, electronic equipment, storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810023434.1A CN110018496A (en) | 2018-01-10 | 2018-01-10 | Obstacle recognition method and device, electronic equipment, storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110018496A true CN110018496A (en) | 2019-07-16 |
Family
ID=67188108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810023434.1A Pending CN110018496A (en) | 2018-01-10 | 2018-01-10 | Obstacle recognition method and device, electronic equipment, storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110018496A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458055A (en) * | 2019-07-29 | 2019-11-15 | 江苏必得科技股份有限公司 | A kind of obstacle detection method and system |
| CN110568861A (en) * | 2019-09-19 | 2019-12-13 | 中国电子科技集团公司电子科学研究院 | Man-machine movement obstacle monitoring method, readable storage medium and unmanned machine |
| CN110580045A (en) * | 2019-08-30 | 2019-12-17 | 安徽科微智能科技有限公司 | obstacle avoidance method for unmanned ship |
| CN111338336A (en) * | 2020-02-11 | 2020-06-26 | 腾讯科技（深圳）有限公司 | Automatic driving method and device |
| CN111507278A (en) * | 2020-04-21 | 2020-08-07 | 浙江大华技术股份有限公司 | Method and device for detecting roadblock and computer equipment |
| CN111765882A (en) * | 2020-06-18 | 2020-10-13 | 浙江大华技术股份有限公司 | Laser radar positioning method and related device thereof |
| CN111967374A (en) * | 2020-08-14 | 2020-11-20 | 安徽海博智能科技有限责任公司 | Mine obstacle identification method, system and equipment based on image processing |
| CN112329552A (en) * | 2020-10-16 | 2021-02-05 | 爱驰汽车（上海）有限公司 | Obstacle detection method and device based on automobile |
| CN112445209A (en) * | 2019-08-15 | 2021-03-05 | 纳恩博（北京）科技有限公司 | Robot control method, robot, storage medium, and electronic apparatus |
| CN112596074A (en) * | 2020-12-09 | 2021-04-02 | 南京农业大学 | Navigation vehicle obstacle detection method based on laser radar |
| CN112633101A (en) * | 2020-12-14 | 2021-04-09 | 深兰人工智能(深圳)有限公司 | Obstacle speed detection method and device |
| CN112649813A (en) * | 2020-12-15 | 2021-04-13 | 北京星天地信息科技有限公司 | Method for indoor safety inspection of important place, inspection equipment, robot and terminal |
| CN112731447A (en) * | 2021-04-06 | 2021-04-30 | 北京三快在线科技有限公司 | Obstacle tracking method and device, storage medium and electronic equipment |
| CN112734811A (en) * | 2021-01-21 | 2021-04-30 | 清华大学 | Obstacle tracking method, obstacle tracking device and chip |
| CN113311443A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Anti-pinch detection method and system for bus shielding door |
| CN113759331A (en) * | 2021-02-07 | 2021-12-07 | 贵州京邦达供应链科技有限公司 | Radar positioning method, device, equipment and storage medium |
| CN113771846A (en) * | 2020-10-28 | 2021-12-10 | 北京京东乾石科技有限公司 | Automatic control method, apparatus, system, and medium for vehicle |
| CN114282597A (en) * | 2020-05-11 | 2022-04-05 | 华为技术有限公司 | Method and system for detecting vehicle travelable area and automatic driving vehicle adopting system |
| 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 |
| CN114743181A (en) * | 2022-04-29 | 2022-07-12 | 重庆长安汽车股份有限公司 | Road vehicle target detection method and system, electronic device and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104614733A (en) * | 2015-01-30 | 2015-05-13 | 福州华鹰重工机械有限公司 | Dynamic obstacle detection method |
-
2018
- 2018-01-10 CN CN201810023434.1A patent/CN110018496A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104614733A (en) * | 2015-01-30 | 2015-05-13 | 福州华鹰重工机械有限公司 | Dynamic obstacle detection method |
Non-Patent Citations (1)
| Title |
|---|
| 黄如林等: "基于激光雷达的无人驾驶汽车动态障碍物检测、跟踪与识别方法", 《机器人》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458055B (en) * | 2019-07-29 | 2021-10-15 | 江苏必得科技股份有限公司 | Obstacle detection method and system |
| CN110458055A (en) * | 2019-07-29 | 2019-11-15 | 江苏必得科技股份有限公司 | A kind of obstacle detection method and system |
| CN112445209A (en) * | 2019-08-15 | 2021-03-05 | 纳恩博（北京）科技有限公司 | Robot control method, robot, storage medium, and electronic apparatus |
| CN110580045A (en) * | 2019-08-30 | 2019-12-17 | 安徽科微智能科技有限公司 | obstacle avoidance method for unmanned ship |
| CN110568861B (en) * | 2019-09-19 | 2022-09-16 | 中国电子科技集团公司电子科学研究院 | Man-machine movement obstacle monitoring method, readable storage medium and unmanned machine |
| CN110568861A (en) * | 2019-09-19 | 2019-12-13 | 中国电子科技集团公司电子科学研究院 | Man-machine movement obstacle monitoring method, readable storage medium and unmanned machine |
| CN111338336A (en) * | 2020-02-11 | 2020-06-26 | 腾讯科技（深圳）有限公司 | Automatic driving method and device |
| CN111338336B (en) * | 2020-02-11 | 2021-07-13 | 腾讯科技（深圳）有限公司 | Automatic driving method and device |
| CN113311443A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Anti-pinch detection method and system for bus shielding door |
| CN111507278A (en) * | 2020-04-21 | 2020-08-07 | 浙江大华技术股份有限公司 | Method and device for detecting roadblock and computer equipment |
| CN111507278B (en) * | 2020-04-21 | 2023-05-16 | 浙江大华技术股份有限公司 | Method and device for detecting roadblock and computer equipment |
| CN114282597A (en) * | 2020-05-11 | 2022-04-05 | 华为技术有限公司 | Method and system for detecting vehicle travelable area and automatic driving vehicle adopting system |
| CN111765882A (en) * | 2020-06-18 | 2020-10-13 | 浙江大华技术股份有限公司 | Laser radar positioning method and related device thereof |
| CN111967374A (en) * | 2020-08-14 | 2020-11-20 | 安徽海博智能科技有限责任公司 | Mine obstacle identification method, system and equipment based on image processing |
| CN112329552B (en) * | 2020-10-16 | 2023-07-14 | 爱驰汽车（上海）有限公司 | Obstacle detection method and device based on automobile |
| CN112329552A (en) * | 2020-10-16 | 2021-02-05 | 爱驰汽车（上海）有限公司 | Obstacle detection method and device based on automobile |
| CN113771846A (en) * | 2020-10-28 | 2021-12-10 | 北京京东乾石科技有限公司 | Automatic control method, apparatus, system, and medium for vehicle |
| CN113771846B (en) * | 2020-10-28 | 2023-08-04 | 北京京东乾石科技有限公司 | Automatic control method, device, system and medium for vehicle |
| CN112596074A (en) * | 2020-12-09 | 2021-04-02 | 南京农业大学 | Navigation vehicle obstacle detection method based on laser radar |
| CN112633101A (en) * | 2020-12-14 | 2021-04-09 | 深兰人工智能(深圳)有限公司 | Obstacle speed detection method and device |
| CN112649813A (en) * | 2020-12-15 | 2021-04-13 | 北京星天地信息科技有限公司 | Method for indoor safety inspection of important place, inspection equipment, robot and terminal |
| CN112734811A (en) * | 2021-01-21 | 2021-04-30 | 清华大学 | Obstacle tracking method, obstacle tracking device and chip |
| CN113759331A (en) * | 2021-02-07 | 2021-12-07 | 贵州京邦达供应链科技有限公司 | Radar positioning method, device, equipment and storage medium |
| CN113759331B (en) * | 2021-02-07 | 2024-04-12 | 贵州京邦达供应链科技有限公司 | Radar positioning method, device, equipment and storage medium |
| CN112731447A (en) * | 2021-04-06 | 2021-04-30 | 北京三快在线科技有限公司 | Obstacle tracking method and device, storage medium and electronic equipment |
| 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 |
| CN114743181A (en) * | 2022-04-29 | 2022-07-12 | 重庆长安汽车股份有限公司 | Road vehicle target detection method and system, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110018496A (en) | Obstacle recognition method and device, electronic equipment, storage medium | |
| CN109927719B (en) | Auxiliary driving method and system based on obstacle trajectory prediction | |
| Meng et al. | HYDRO-3D: Hybrid object detection and tracking for cooperative perception using 3D LiDAR | |
| CN113366495B (en) | Searching an autonomous vehicle sensor data store | |
| Sivaraman et al. | Looking at vehicles on the road: A survey of vision-based vehicle detection, tracking, and behavior analysis | |
| US11783568B2 (en) | Object classification using extra-regional context | |
| Barth et al. | Estimating the driving state of oncoming vehicles from a moving platform using stereo vision | |
| KR20200125731A (en) | Neural networks for object detection and characterization | |
| EP4296721A2 (en) | Generating 3-dimensional maps of a scene using passive and active measurements | |
| CN109229109B (en) | Method, device, equipment and computer storage medium for judging vehicle driving direction | |
| US11755917B2 (en) | Generating depth from camera images and known depth data using neural networks | |
| EP4181091A1 (en) | Pedestrian behavior prediction with 3d human keypoints | |
| WO2017139516A1 (en) | System and method for achieving fast and reliable time-to-contact estimation using vision and range sensor data for autonomous navigation | |
| US11967103B2 (en) | Multi-modal 3-D pose estimation | |
| CN114061581A (en) | Ranking agents in proximity to autonomous vehicles by mutual importance | |
| Zelener et al. | Cnn-based object segmentation in urban lidar with missing points | |
| US11727690B2 (en) | Behavior prediction of surrounding agents | |
| CN115273002A (en) | Image processing method, device, storage medium and computer program product | |
| EP4060626A1 (en) | Agent trajectory prediction using context-sensitive fusion | |
| Jia et al. | Front-view vehicle detection by Markov chain Monte Carlo method | |
| KR20200058272A (en) | Method and system for providing road driving situation through preprocessing of road driving image | |
| US20220164350A1 (en) | Searching an autonomous vehicle sensor data repository based on context embedding | |
| Rahman et al. | Predicting driver behaviour at intersections based on driver gaze and traffic light recognition | |
| CN116868239A (en) | Static occupancy tracking | |
| Gao et al. | Deep learning‐based hybrid model for the behaviour prediction of surrounding vehicles over long‐time periods |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210304 Address after: 101, 1st floor, building 2, yard 20, Suzhou street, Haidian District, Beijing 100080 Applicant after: Beijing Jingbangda Trading Co.,Ltd. Address before: 100195 Beijing Haidian Xingshikou Road 65 West Cedar Creative Garden 4 District 11 Building East 1-4 Floor West 1-4 Floor Applicant before: BEIJING JINGDONG SHANGKE INFORMATION TECHNOLOGY Co.,Ltd. Applicant before: BEIJING JINGDONG CENTURY TRADING Co.,Ltd. Effective date of registration: 20210304 Address after: Room a1905, 19 / F, building 2, No. 18, Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Beijing Jingdong Qianshi Technology Co.,Ltd. Address before: 101, 1st floor, building 2, yard 20, Suzhou street, Haidian District, Beijing 100080 Applicant before: Beijing Jingbangda Trading Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190716 |
|
| RJ01 | Rejection of invention patent application after publication |