CN107018522A - A kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition - Google Patents
A kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition Download PDFInfo
- Publication number
- CN107018522A CN107018522A CN201710109786.4A CN201710109786A CN107018522A CN 107018522 A CN107018522 A CN 107018522A CN 201710109786 A CN201710109786 A CN 201710109786A CN 107018522 A CN107018522 A CN 107018522A
- Authority
- CN
- China
- Prior art keywords
- base station
- unmanned plane
- ultrasonic
- time
- zero crossing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000004807 localization Effects 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 230000003993 interaction Effects 0.000 claims abstract description 13
- 230000005693 optoelectronics Effects 0.000 claims abstract description 11
- 238000013135 deep learning Methods 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 23
- 238000007493 shaping process Methods 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000007423 decrease Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- XCWPUUGSGHNIDZ-UHFFFAOYSA-N Oxypertine Chemical compound C1=2C=C(OC)C(OC)=CC=2NC(C)=C1CCN(CC1)CCN1C1=CC=CC=C1 XCWPUUGSGHNIDZ-UHFFFAOYSA-N 0.000 claims description 3
- 230000000903 blocking effect Effects 0.000 claims description 3
- 238000003032 molecular docking Methods 0.000 claims description 3
- 238000002604 ultrasonography Methods 0.000 claims description 3
- 102000002274 Matrix Metalloproteinases Human genes 0.000 claims 1
- 108010000684 Matrix Metalloproteinases Proteins 0.000 claims 1
- 238000013461 design Methods 0.000 abstract description 3
- 238000000691 measurement method Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 238000013459 approach Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000005622 photoelectricity Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000002463 transducing effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- 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
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/18—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using ultrasonic, sonic, or infrasonic waves
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The present invention relates to a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition, its realization positioned need to carry out particular design to ground base station and unmanned plane and require that unmanned plane cooperates with ground base station, specifically include following steps：Positioning step based on GPS realizes the substantially positioning in ground base station region；Ultrasonic wave positioning step based on precision distance measurement method realizes that unmanned plane hovers in ground base station overhead centre；Landing step based on image procossing and graviational interaction realizes unmanned plane landing in high precision；Detection and localization step based on optoelectronic switch returns positioning to put for detecting whether unmanned plane has accurately been parked in.The present invention enables to unmanned plane automatic identification ground base station level point, and is precisely landed.
Description
Technical field
The present invention relates to unmanned plane (i.e. unmanned aerial vehicle) field of locating technology, more particularly to one
Plant the localization method of the unmanned aerial vehicle base station based on Multi-information acquisition.
Background technology
With the rapid development of science and technology, research of the people to aircraft is increasingly deep, and all kinds of aircraft are increasingly
Many occasions are applied.Unmanned plane is compared with other aircraft, and its mechanical structure is simply compact, more flexible, landing of taking action
Environmental requirement is relatively low, with good operating characteristics, can be realized in small range and take off, hovers, landing.Due to these features,
Be widely used in taking photo by plane, monitored, investigate, searching and rescuing, the numerous areas such as control of agricultural pest.
Nowadays, domestic and foreign scholars have delivered substantial amounts of related article and achievement in research, continue to bring out out new application neck
Domain, has been obviously improved the application value of unmanned plane.For these applications, unmanned plane needs autonomous for a long time in some area
Work.Because the flight time of unmanned plane is limited, after the performing a period of time in the air of the task, it is necessary to return to ground base station
Charged and carry out information interchange with associated mechanisms.Therefore, accurate efficient ground base station alignment system becomes increasingly to weigh
Will.
The content of the invention
The technical problem to be solved in the present invention is：Unmanned plane is enabled to be automatically positioned ground base station level point and carry out
Precisely landing.
In order to solve the above-mentioned technical problem, the technical scheme is that there is provided a kind of nothing based on Multi-information acquisition
The localization method of man-machine ground base station, unmanned plane includes N number of universal wheel for being used to land, N >=3, it is characterised in that describedly
4 angles of face base station are provided with ultrasonic receiver, and ultrasonic receiver is connected with ultrasonic ranging system, ground base station
Two cornerwise intersection points are base station center, and the heart has 1 image recognition object small icon, N number of image recognition object in a base station
Large icons, which is surrounded, 1 positioning calibration cartridge in image recognition object small icon, the region of each image recognition object large icons
Put, the orientation of the direction of N number of locating calibration device respectively with N number of universal wheel of unmanned plane matches；
Electronic compass, the spherical all-around ultrasonic wave transducing for omnidirectional emission ultrasonic wave are also equipped with unmanned plane
Device and cradle head camera, cradle head camera vertically ground；
The localization method comprises the following steps：
The first step, unmanned plane are calculated using airborne navigation positioning module and obtain present position, and earthward base station is sent out
Wireless signal and data are sent, ground base station is received after signal, it is allowed to which unmanned plane lands, and the azimuth information of base station location is sent out
Unmanned plane is sent to, unmanned aerial vehicle (UAV) control module flies to the approximate location of base station according to azimuth information, control unmanned plane；
By spherical omnidirectional transducer in the same time interval, earthward base station sends two band frequencies for second step, unmanned plane
Respectively f1、f2Ultrasonic signal, ultrasonic ranging system determines 4 ultrasonic receivers respectively using bifrequency telemetry
Ultrasonic propagation time TOF, randomly choose 3 ultrasonic receivers ultrasonic propagation time TOF as one group, calculate
Go out unmanned plane coordinate, using base station center as the origin of coordinates, the unmanned plane coordinate obtained according to calculating, control unmanned plane is flown to
Base station center, and hovered in setting height, in this step：
Determine the ultrasonic propagation time TOF of any one ultrasonic receiver tool respectively using bifrequency telemetry
Body step is：The frequency respectively f that ultrasonic ranging system is received using current ultrasonic receiver1、 f2Ultrasonic wave letter
The relative time difference of related zero crossing determines ultrasonic propagation time between number, wherein, frequency is respectively f1、f2Ultrasonic wave
The related zero crossing of one group of confidence level highest is judged by deep learning algorithm between signal；
After 3rd step, unmanned plane are hovered over above base station center, start slow decline, during decline, pass through electronics
Compass adjusts the orientation of unmanned plane universal wheel so that direction setting orientation in body front during unmanned plane land, meanwhile, by head
Camera obtains image recognition object small icon and image recognition object large icons, so that base station center is obtained, according in base station
Relative position of the heart in the image that cradle head camera is obtained in real time obtains the control parameter of unmanned plane, and control unmanned plane causes
The centre movement for the image that base station center is obtained in real time to cradle head camera, final unmanned plane is dropped on ground base station, nothing
Man-machine N number of universal wheel is fallen under graviational interaction in N number of locating calibration device.
Preferably, described image identification object small icon is small circle ring icon；Described image identification object large icons is big
Annulus icon.
Preferably, the top half of the locating calibration device is the circular arc concave surface that a Ge Xiang centers are sunk, with unmanned plane
Universal wheel be engaged, and by graviational interaction realize be accurately positioned；The latter half of the locating calibration device is shaped as
Cylinder, is matched with the size of universal wheel, for blocking universal wheel.
Preferably, the upper optoelectronic switch equipped with docking form of facing the wall and meditating of the latter half of the locating calibration device, is used for
Positioning carries out detection and localization after terminating, in the 3rd step, after unmanned plane drops to ground base station, if optoelectronic switch is complete
Portion is closed, then is positioned successfully, otherwise, positioning failure.
Preferably, in the second step, determine the ultrasonic propagation time TOF's of any one ultrasonic receiver
Method comprises the following steps：
The frequency respectively f that step 2.1, ultrasonic ranging system are received using current ultrasonic receiver1、f2It is super
After acoustic signals, at the time of extracting the approximate zero crossing after two sections of ultrasonic signals, corresponding two groups of time data P are obtained1、
P2, P1={ t11, t12, t13..., t1m... }, P2={ t21, t22, t23..., t2n... }, in formula, t1mFor frequency f1It is super
The approximate zero crossing moment that m-th is extracted in acoustic signals, itself and frequency f1Ultrasonic signal in m-th of zero crossing
The correspondence momentBetween relation it is unknown；t2nFor frequency f2Ultrasonic signal in be extracted to for n-th approximate zero crossing when
Carve, itself and frequency f2Ultrasonic signal in n-th zero crossing correspondence momentBetween relation it is unknown.Then, P is extracted1、P2
The time data that calibrates for error, obtain Q1、Q2, Q1={ t '11, t '12, t '13..., t '1m... }, Q2={ t '21, t '22, t
′23..., t '2n... }, in formula, t '1m、t′2nRespectively t1m、t2nCalibrate for error the time；
Step 2.2, by P1With P2In time data match to form multigroup time data group two-by-two, with reference to Q1、 Q2, pass through
Deep learning algorithm obtains one group of time data group of confidence level highest in time data group.If during one group of confidence level highest
Between γ groups in all related approximate zero crossing groups of data correspondence, then the related approximate zero crossing correspondence moment be expressed asWith
The ultrasonic propagation time such as step 2.3, meter TOF：
In formula,WithIt is the related approximate zero passage of γ groups respectively
Frequency is f in point1And f2The acoustic signals corresponding time,ForIt is corresponding to calibrate for error the time,ForCorresponding mistake
Poor time, Δ t is the time interval of twice emitting,It is that relative time between the related approximate zero crossing of γ groups is poor, γ
It is the group number of related zero crossing.Preferably, deep learning algorithm described in step 2.2 is used as depth using BP neural network
Practise network model.
Preferably, in the step 2.1, the approximate mistake after two sections of ultrasonic signals of Schmidt's integer circuit extraction is passed through
At the time of zero point；Pass through reverse comparator integer circuit extraction P1、P2The time data that calibrates for error.
Preferably, the training sample acquisition methods of the deep learning network model are：
By testing two groups of time data P are obtained using with step 2.1 identical method1、P2, P1={ t11, t12,
t13..., t1m... }, P2={ t21, t22, t23..., t2n... }, by two groups of time data P1And P2Combination of two is carried out, P=is obtained
{t11t21；t11t22；t11t23；...；t1mt2n... }；Then, P, Q are utilized1And Q2Constitute matrix
T in formula12-mn=1/ (1+e-α(k-β))、α is that empirical parameter, β are to receive signal amplitude maximum correspondence
Periodicity, t1-m=| t '1m-t1m-0.5/f1|/2, t2-n=| t '2n-t2n-0.5/f2|/2；It can thus be concluded that each of X is classified as
One input feature value, its first row is that the characteristic value of data, the second row and the third line are Schmidt's shaping respectively to two
Acoustic signals each zero crossing time-bands come error influence；Finally, the corresponding 1 dimension output of each characteristic vector passes through
Artificial experience setting, its value is between 0 and 1.
Preferably, the realization of the deep learning network model is as follows：First, input data is normalized；
Secondly, build network and carry out the initial work of network, it is that 3, hidden layer is 10, output to set input layer number
Layer is 1, and weight w, threshold value b, learning rate, training objective minimal error and maximum allowable train epochs are initialized；
Then, start the training of deep learning network, substantial amounts of input and output training sample is put into network, carry out in network
The training of weights and threshold value, finally when error is limited in tolerance interval or beyond frequency of training, obtains new with each
The deep learning network model of weights and threshold value；Finally, network test is carried out, test result is contrasted with desired value, with
Just the training result of the network model is assessed, and network parameter is modified.
As a result of above-mentioned technical scheme, the present invention compared with prior art, has the following advantages that and actively imitated
Really：
(1) traditional ultrasonic ranging method is single threshold detection method and phase method.But for the former, receive signal jump
The time for crossing threshold value can be because of signal amplitude it is different and different, so as to cause the uncertainty in measurement；And for the latter,
Its finding range very little, when finding range is more than ultrasonic wavelength, can produce range ambiguity.In patent of the present invention
Ultrasonic ranging method be a kind of distance-finding method based on bifrequency, can cleverly avoid disadvantages mentioned above, and essence can be realized
True ranging.In addition, location algorithm with BP neural network be combined can from the related zero crossing of fuzzy angle-determining so that
Problem be simplified with it is extensive.And the time error of BP neural network fusions measurement, so as to improve the robust of ranging
Property.Method for ultrasonic locating based on the precision distance measurement realizes the locating effect of high-precision high robust.
(2) patent of the present invention ground base station 4 locating calibration devices of Center, and by 4 wheels of unmanned plane
Son is designed to universal wheel.Thus, unmanned plane by graviational interaction just can simple realization position, it is to avoid complicated image procossing,
So as to accelerate the processing speed of processor and improve the accuracy of landing.
(3) multiple system globe areas are got up to carry out the positioning of unmanned aerial vehicle base station by the present invention, can further increase fixed
The swiftness of position, accuracy and robustness.
Brief description of the drawings
Fig. 1 is the ground base station structure diagram in the embodiment of the present invention；
Fig. 2 is the ground base station planar structure sketch in the embodiment of the present invention；
Fig. 3 is the bifrequency telemetry schematic diagram in the embodiment of the present invention；
Fig. 4 is Schmidt's shaping method error analysis and processing figure in the embodiment of the present invention；
Fig. 5 is the BP neural network prediction output comparison diagram in the embodiment of the present invention；
Fig. 6 is the locating calibration device structure diagram in the embodiment of the present invention；
Fig. 7 and Fig. 8 is that the universal wheel in the embodiment of the present invention blocks figure.
Embodiment
With reference to specific embodiment, the present invention is expanded on further.It should be understood that these embodiments are merely to illustrate this hair
Bright rather than limitation the scope of the present invention.In addition, it is to be understood that after the content of the invention lectured has been read, this area skill
Art personnel can make various changes or modifications to the present invention, and these equivalent form of values equally fall within the application appended claims
Limited range.
Embodiments of the present invention are related to a kind of fusion and positioning method of unmanned aerial vehicle base station.Its positioning step is main
Including 4 big steps：Positioning step based on GPS, the ultrasonic wave positioning step based on precision distance measurement method, based on image procossing and
The landing step of graviational interaction and the detection and localization step based on optoelectronic switch.These positioning steps are realized, need to be to ground base station
Particular design is carried out with unmanned plane and requires that unmanned plane cooperates with ground base station.
As shown in figure 1, the ground base station profile being related in the present invention is length, width and height be respectively 1m, 1m, 0.2m cube
Body, inside there is specific control system.Ultrasonic receiver 1 built in 4 angles of ground base station, constitutes ultrasonic receiver together
Array.There are 4 large circle icons 4 and a small circle ring icon 3 in the middle of ground base station.Have in the region of large circle icon 4
4 locating calibration devices.Wherein, ultrasonic receiver 1 is torus receiver, can be with comprehensive reception ultrasonic wave.4 great circles
Ring icon 4 and small circle ring icon 3 are the objects of image recognition, for determining ground base station center.4 locating calibration devices
Towards the respectively southeast, northeast, southwest, northwest, the orientation for 4 universal wheels with unmanned plane is matched, as shown in Figure 2.It is fixed
The top half 2 of position calibrating installation is the circular arc concave surface that a Ge Xiang centers are sunk, and is engaged with the universal wheel of unmanned plane, and according to
Realize and be accurately positioned by graviational interaction；The latter half 5 of the locating calibration device is shaped as cylinder, with universal wheel
Size is matched, for blocking universal wheel.The upper photoelectricity equipped with docking form of facing the wall and meditating of the latter half 5 of locating calibration device is opened
Close, carry out detection and localization after terminating for positioning, as illustrated in figs. 7 and 8.The concrete shape of the locating calibration device such as Fig. 7
It is shown.
Unmanned plane except substantially control sensor in addition to, be also equipped with electronic compass sensor, spherical all-around ultrasonic wave transducer,
Cradle head camera.Wherein, described electronic compass sensor is located at internal body；Described cradle head camera is vertically
Face, will be moved into the underface of body when using；Described spherical ultrasonic sensor is located at organism bottom center, can be entirely square
Position transmitting ultrasonic wave.
Positioning step based on GPS
GPS location step, main purpose be realize body from perform approximate region where the ground base station that flies back, energy
Substantially navigate to around ground base station in 2m square range.After unmanned plane receives return ground base station signal, its is airborne
INS/GPS integrated navigation and locations module will calculate current unmanned plane position, and earthward base station sends wireless signal sum
According to.Ground base station is received after signal, it is allowed to which unmanned plane lands, and base station location is sent into unmanned plane.Unmanned aerial vehicle (UAV) control module
According to azimuth information, control unmanned plane flies to the approximate location of base station.
Ultrasonic wave positioning step based on precision distance measurement method
Described ultrasonic wave positioning step is core positioning step, and main purpose is so that unmanned plane hovers in ground base
Stand the centre in overhead, consequently facilitating follow-up exact localization operation.Unmanned plane is carried out after coarse localization by GPS, will be slowed down
Flying speed, and ultrasonic wave is periodically launched downwards by the spherical omnidirectional transducer in body center.Now, ground base station 4
Ultrasonic receiver array on individual angle receives priority the ultrasonic wave of unmanned plane transmission, so that when being propagated accordingly
Between TOF (time of flight).If using two cornerwise intersection points of ground base station as the origin of coordinates, processing system will be with
Machine selects the TOF that three receivers are obtained as one group, calculates the coordinate residing for unmanned plane, and carry out by other combinations
Redundant computation, reduces error.
Because the precision of positioning is directly decided by range accuracy, therefore patent of the present invention is to location algorithm and ultrasound
Particular design has been carried out away from system.
Described high-precision ultrasonic location algorithm is a kind of bifrequency telemetry based on deep learning, can be accurate
Measuring Propagation Time of Ultrasonic Wave TOF.Described bifrequency range measurement principle is as follows：Received as shown in figure 3, range-measurement system is utilized
Different frequency signals related zero crossing between relative time it is poorCome true
Determine ultrasonic propagation timeF in formula1It is the frequency for sending signal for the first time, f2It is second
The frequency of signal is sent,WithIt is that frequency is f in i-th group of related zero crossing respectively1And f2Acoustic signals it is corresponding when
Between, Δ f is both difference on the frequencies, and Δ t is the time interval of twice emitting, and i is the group number of related zero crossing.Described zero passage
The extracting method of point time data is Schmidt's shaping method, can avoid the interference of noise, but can make it that the zero crossing extracted is
Approximate zero crossing is so as to bring measurement error.The error school that the measurement error can be extracted by reverse comparator shaping circuit
Quasi- time data is calibrated, as shown in Figure 4.Described schmidt shaping circuit and reverse comparator integer circuit configuration
Threshold size it is equal.At the time of described time data is the rising edge correspondence of shaping circuit output waveform.Described is near
Then need to be judged using the confidence level size that deep learning algorithm is exported like the correlation of zero crossing.Described deep learning
Algorithm can determine one group of data of confidence level highest, thus obtain accurate ultrasonic propagation time.Described deep learning
The confidence level of algorithm output has merged the detection error of Schmidt's shaping method, so as to strengthen the accuracy of ranging, reliability, Shandong
Rod.
The described bifrequency telemetry based on deep learning is comprised the following steps that：
Step 1：It is different that unmanned plane sends two band frequencies by spherical omnidirectional transducer in the same time interval, earthward
Ultrasonic signal and notify ground base station ultrasonic positioning system to start timing.Two kinds of described frequencies are in transducer band
In wide, such as 50KHz and 51.3KHz；
Step 2：Ground base station ultrasonic ranging system receives priority two sections of acoustic signals described in step 1, passes through
At the time of the approximate zero crossing of this two segment signal of Schmidt's integer circuit extraction, corresponding two groups of time data P are obtained1=
{t11, t12, t13..., t1m... } and P2={ t21, t22, t23..., t2n... }；Carried by reverse comparator integer circuit
Take the time data Q that calibrates for error of this two segment signal1={ t '11, t '12, t '13..., t '1m... } and Q2={ t '21, t '22,
t′23..., t '2n... }.In the footmark of the time data, first is acoustic signals label, and second is data sequence
Label；
Step 3：Due to the P described in step 21And P2In the corresponding approximate zero crossing of j-th of data differ on waveform
Surely it is corresponding, i.e. t1jAnd t2jIt is not necessarily related.In addition, the zero crossing that Schmidt's shaping method is obtained is approximate zero crossing.Cause
This range-measurement system will utilize time data P1、P2、Q1And Q2, and one group of number of confidence level highest is obtained by deep learning algorithm
According to
Step 4：Pass through the data described in step 3Calculate ultrasonic propagation time.Calculation formula is WithIt is that frequency is f in the related approximate zero crossing of γ groups respectively1With
f2The acoustic signals corresponding time,ForIt is corresponding to calibrate for error the time,ForCorresponding error time, Δ t is two
The time interval of secondary transmitting,It is that relative time between the related approximate zero crossing of γ groups is poor, γ is related zero crossing
Group number.
Wherein, described deep learning algorithm, mainly using BP neural network as deep learning network model, its
Middle training sample comes from experimental data.
Described training sample acquisition methods are as follows：By in advance by testing one group of obtained P1And P2Carry out group two-by-two
Close, will { t11, t12, t13..., t1m... } and { t11, t12, t13..., t2n... } and it is converted into P={ t11t21；t11t22；
t11t23；...；t1mt2n... }；Then, P, Q are utilized1And Q2Constitute matrixT in formula12-mn
=1/ (1+e-α(k-β))、α is that empirical parameter, β are to receive the corresponding periodicity of signal amplitude maximum
(generally steady state value), t1-m=| t '1m-t1m-0.5/f1|/2, t2-n=| t '2n-t2n-0.5/f2|/2；It can thus be concluded that X's is every
One is classified as an input feature value, and its first row is that the characteristic value of data, the second row and the third line are Schmidt's shaping respectively
On two acoustic signals each zero crossing time-bands come error influence (numerical value is bigger, and error is bigger)；Finally, each is special
Levy the corresponding 1 dimension output (confidence level) of vector to set by artificial experience, (the bigger expression of numerical value is more between 0 and 1 for its value
It is likely to be used for calculating TOF).
The realization of described BP neural network model is as follows：Firstly, since the numerical values recited of input sample differs, gap
It is larger, it is necessary to carry out input data normalized；Secondly, build network and carry out the initial work of network, input is set
Layer neuron number is that 3, hidden layer is that 10, output layer is 1, and to weight w, threshold value b, learning rate, training objective minimum by mistake
Poor and maximum allowable train epochs are initialized；Then, the training of deep learning network is started, by substantial amounts of input and output
In training sample input network, the weights in network and the training of threshold value are carried out, finally when error is in tolerance interval or super
When going out frequency of training limitation, the deep learning network model with each new weights and threshold value is obtained；Finally, network survey is carried out
Examination, test result is contrasted with desired value, to assess the training result of the network model, and network parameter is carried out
Amendment.It can thus be concluded that the deep learning detection model of reference point.
Described deep learning algorithm steps are as follows：
Step 1：Set each parameter of neutral net, including the hidden layer number of plies, the number of each layer neuron and its weight w and
Threshold value b, builds basic neural network model；
Step 2：First five set of two groups of time datas is selected, feature is extracted according to the method for described input training sample
Vector, is used as the input data of detection；
Step 3：The corresponding output matrix of characteristic vector is calculated using neural network model；
Step 4：Output matrix described in traversal step 3, maximizing simultaneously records its position in array；
Step 5：One group of time data of confidence level highest is obtained by step 4It can thus be concluded that the propagation time；
The core of described ultrasonic ranging algorithm is deep learning algorithm.When carrying out reference point detection, depth is used
Learning network model can describe the non-linear relation of complexity, and merge the detection error of Schmidt's shaping method, so as to have
Beneficial to the accuracy and robustness for improving reference point detection.As shown in figure 5, using BP neural network model prediction output with
Desired output is sufficiently close to.
Described ultrasonic ranging system is constituted as shown in fig. 6, being made up of signal processing module and data processing module.
Described signal processing module is by prime process circuit, filter circuit, programmable amplifying circuit, Schmidt's shaping electricity
Road and reverse comparator shaping circuit composition.The signal that ultrasonic receiver is captured is that comparison is faint, generally only several
Millivolt is to tens millivolts, and the interference signal for the surrounding environment that can adulterate.Therefore, signal will be received and is sent to Schmidt's shaping
Circuit carries out relevant treatment with being needed before reverse comparator circuit.First, acoustic signals are believed by prime process circuit
Number amplitude limit and primary enhanced processing.Then, signal will be sent to filter circuit, and the quality factor of the circuit is very high, trap
Depth is deep, can be substantially filtered out interference signal.Finally, filtered signal will be sent to programmable amplifying circuit.Complete above-mentioned pre- place
After reason, described schmidt shaping circuit and anti-phase comparison circuit will be while receive the output signal of programmable amplifying circuit.In letter
In number processing module, described programmable amplifying circuit can adjust multiplication factor as the case may be to improve system ranging
Robustness；It is to filter out, prevent by interference signal in advance that described programmable amplifying circuit, which is designed after filter circuit,
It is further magnified, and influences range performance；Described schmidt shaping circuit and the threshold of reverse comparator integer circuit configuration
Being worth size must be equal.
Described data processing module is made up of TDC-GP21, MCU, outside sound velocity calibration module and main control computer.Institute
When the TDC-GP21 stated is used for the rising edge of the output signal of extracted with high accuracy schmidt shaping circuit and reverse comparator circuit
Between, and it is transferred to MCU.Described outside sound velocity calibration module is used for the real time calibration velocity of sound, utilizes ultrasonic propagation known distance
Time data is transferred to MCU by the time used to be calibrated.Described MCU will be from TDC-GP21 and the outside velocity of sound
The data of calibration module are pre-processed, and are transferred to main control computer.Described main control computer is to the data from MCU
Advanced treating is carried out, and the data after having handled are transferred to unmanned plane.
Based on above-mentioned location algorithm and range-measurement system, what described ultrasonic wave was positioned comprises the following steps that：
Step 1：It is different that unmanned plane sends two band frequencies by spherical omnidirectional transducer in the same time interval, earthward
Ultrasonic signal and notify ground base station ultrasonic positioning system to start timing.
Step 2：After a period of time, ultrasonic receiver array receives priority the acoustic signals of two kinds of frequencies.
Step 3：Acoustic signals described in step 2 will be transmitted to the respective signal transacting mould of ultrasonic receiver array
Block, respectively obtains corresponding two sections of square-wave signals.
Step 4：Two sections of square-wave signals described in step 3 will be transmitted to ultrasound data processing module.The module according to
Bifrequency telemetry based on deep learning obtains the distance of unmanned plane and ultrasonic receiver array, thus obtains unmanned plane
Position coordinates, distance, height and attitude angle relative to level point, and be transferred to unmanned plane.
Step 5：Unmanned machine automatic drive system is according to the control instruction of input, the parameter that collection sensor is provided, according to
The control method and logic of setting produce control instruction, and realize relevant control by executing agency.
Step 6：Circulation step 1~5, until position error is in allowed band.
Landing step based on image recognition and graviational interaction
The main purpose of the precision approach based on image recognition and graviational interaction is so that 4 of unmanned plane are universal
Precision approach is taken turns in defined 4 circular arc concave bottoms.Unmanned plane is positioned at after the centre of ground base station overhead, is started slow
Decline.Positioning accurate in the fine setting for being positioned for unmanned plane air position based on image recognition, enhancing descent
Degree.It is described that gravitation is located through so that 4 universal wheels of unmanned plane are recessed along 4 smooth circular arcs of correspondence based on graviational interaction
Face falls, and is finally parked in accurate 4 circular arc concave bottoms, as shown in Figure 8.The fixed point orientation of 4 universal wheels passes through electricity
Sub- compass is adjusted so that body front is towards due north when unmanned plane lands, so that the orientation of 4 universal wheels is respectively east
South, northeast, southwest, northwest.This can prevent the level point of 4 universal wheels to be located on the joint face of 4 circular arc concave surfaces, it is impossible under
It is sliding.
It is fixed that described image identification specifically includes image acquisition, perspective transform, dynamic threshold binaryzation, circle detection, the center of circle
Position.Wherein, described image is obtained by cradle head camera immediately below body；Described cradle head camera will adjust shooting in real time
The position of head, keeps camera vertically downward, so as to stabilize image；Described perspective transform is used to correct lopsided image；Institute
The dynamic threshold binaryzation stated is used to extract the profile of 4 large circle icons and center small circle ring icon in image, by background with
Prospect is separated, and facilitates subsequent detection to operate；Described circle detection is used to recognize the annulus icon profile in figure；Institute
The center of circle positioning stated specifically includes coarse positioning and fine positioning.
The precision approach based on image recognition and graviational interaction is comprised the following steps that：
Step 1：Unmanned plane carries out bearing calibration by electronic compass；
Step 2：Unmanned plane camera obtains ground base station image；
Step 3：When unmanned plane and ground base station farther out when, processor will carry out image recognition to the image in step 1,
The position in 4 great circle centers of circle is obtained, so that the intersection point of the intersection line in 4 centers of circle is ground base station center；When nobody
When machine and nearer ground base station, processor will carry out image recognition to the image in step 1, obtain the position in the middle roundlet center of circle
Put, i.e. ground base station center；
Step 4：Processor obtains nobody according to the relative position of ground base station center in the picture in step 3
The control parameter of machine so that moved to the centre of image ground base station center.
Step 5：Repeat step 1~4, until unmanned plane drops to ground base station, and then execution step 6；
Step 6：Unmanned plane realizes automatic downslide by graviational interaction by 4 smooth circular arc concave surfaces, stops at 4 and determines
On site, i.e., 4 universal wheels have been stuck in 4 smooth circular arc concave bottoms.
The detecting step positioned based on optoelectronic switch
The detection and localization step, main purpose is whether detection unmanned plane has accurately been parked in assigned position.Described inspection
It is the optoelectronic switch positioned at locating calibration device lower half to survey sensor, as illustrated in figs. 7 and 8.Whether positioning successfully detects
According to for 4 pairs of optoelectronic switches of detection whether Close All.Illustrate that universal wheel is all blocked if optoelectronic switch Close All, enter
And position successfully；Failure is positioned if the non-Close All of optoelectronic switch, unmanned plane will receive the signal of positioning failure, after taking off
It will relocate.
Claims (9)
1. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition, unmanned plane is used for the ten thousand of landing including N number of
To wheel, N >=3, it is characterised in that 4 angles of the ground base station are provided with ultrasonic receiver, ultrasonic receiver and ultrasound
Ripple range-measurement system is connected, and two cornerwise intersection points of ground base station are base station center, and the heart has 1 image recognition pair in a base station
As small icon, N number of image recognition object large icons surrounds image recognition object small icon, each image recognition object large icons
There is 1 locating calibration device in region, the orientation phase of the direction of N number of locating calibration device respectively with N number of universal wheel of unmanned plane
Matching；
Be also equipped with unmanned plane electronic compass, the spherical all-around ultrasonic wave transducer for omnidirectional emission ultrasonic wave and
Cradle head camera, cradle head camera vertically ground；
The localization method comprises the following steps：
The first step, unmanned plane are calculated using airborne navigation positioning module and obtain present position, and earthward base station sends nothing
Line signal and data, ground base station are received after signal, it is allowed to which unmanned plane lands, and the azimuth information of base station location is sent into nothing
Man-machine, unmanned aerial vehicle (UAV) control module flies to the approximate location of base station according to azimuth information, control unmanned plane；
By spherical omnidirectional transducer in the same time interval, earthward base station sends two band frequencies difference for second step, unmanned plane
For f1、f2Ultrasonic signal, ultrasonic ranging system determines the super of 4 ultrasonic receivers respectively using bifrequency telemetry
Acoustic transit time TOF, randomly chooses the ultrasonic propagation time TOF of 3 ultrasonic receivers as one group, calculates nobody
Machine coordinate, using base station center as the origin of coordinates, the unmanned plane coordinate obtained according to calculating, control unmanned plane is flown in base station
The heart, and hovered in setting height, in this step：
Determine the ultrasonic propagation time TOF of any one ultrasonic receiver specific steps respectively using bifrequency telemetry
For：The frequency respectively f that ultrasonic ranging system is received using current ultrasonic receiver1、f2Ultrasonic signal between phase
The relative time difference of zero crossing is closed to determine ultrasonic propagation time, wherein, frequency is respectively f1、f2Ultrasonic signal between
The related zero crossing of one group of confidence level highest is judged by deep learning algorithm；
After 3rd step, unmanned plane are hovered over above base station center, start slow decline, during decline, pass through electronic compass
Adjust the orientation of unmanned plane universal wheel so that direction setting orientation in body front during unmanned plane land, meanwhile, by cradle head camera
Image recognition object small icon and image recognition object large icons are obtained, so that base station center is obtained, according to base station center in cloud
Relative position in the image that platform camera is obtained in real time obtains the control parameter of unmanned plane, and control unmanned plane causes base station center
The centre movement of the image obtained in real time to cradle head camera, final unmanned plane dropped on ground base station, unmanned plane it is N number of
Universal wheel is fallen under graviational interaction in N number of locating calibration device.
2. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 1, its feature exists
In described image identification object small icon is small circle ring icon；Described image identification object large icons is large circle icon.
3. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 1, its feature exists
In the top half of the locating calibration device is the circular arc concave surface that a Ge Xiang centers are sunk, and is matched with the universal wheel of unmanned plane
Close, and be accurately positioned by graviational interaction realization；The latter half of the locating calibration device is shaped as cylinder, and universal
The size matching of wheel, for blocking universal wheel.
4. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 3, its feature exists
In the upper optoelectronic switch equipped with docking form of facing the wall and meditating of the latter half of the locating calibration device terminates laggard for positioning
Row detection and localization, in the 3rd step, after unmanned plane drops to ground base station, if optoelectronic switch Close All, is positioned
Success, otherwise, positioning failure.
5. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 1, its feature exists
In in the second step, determining the ultrasonic propagation time TOF of any one ultrasonic receiver method includes following step
Suddenly：
The frequency respectively f that step 2.1, ultrasonic ranging system are received using current ultrasonic receiver1、f2Ultrasonic wave
After signal, at the time of extracting the approximate zero crossing after two sections of ultrasonic signals, corresponding two groups of time data P are obtained1、P2, P1=
{t11, t12, t13..., t1m... }, P2={ t21, t22, t23..., t2n... }, in formula, t1mFor frequency f1Ultrasonic signal
In approximate zero crossing moment for being extracted to for m-th, itself and frequency f1Ultrasonic signal in m-th zero crossing correspondence moment
Between relation it is unknown；t2nFor frequency f2Ultrasonic signal in approximate zero crossing moment for being extracted to for n-th, itself and frequency
f2Ultrasonic signal in n-th zero crossing correspondence momentBetween relation it is unknown, then, extract P1、P2When calibrating for error
Between data, obtain Q1、Q2, Q1={ t '11, t '12, t '13..., t '1m... }, Q2={ t '21, t '22, t '23..., t
′2n... }, in formula, t '1m、t′2nRespectively t1m、t2nCalibrate for error the time；
Step 2.2, by P1With P2In time data match to form multigroup time data group two-by-two, with reference to Q1、Q2, by depth
Practise algorithm and obtain one group of time data group of confidence level highest in time data group.If one group of time data group of confidence level highest
Corresponding to the γ groups in all related zero crossing groups, then approximately the zero crossing correspondence moment is expressed as correlationWith
Step 2.3, calculating ultrasonic propagation time TOF：
In formula,WithIt is the related approximate zero crossing intermediate frequency of γ groups respectively
Rate is f1And f2The acoustic signals corresponding time,ForIt is corresponding to calibrate for error the time,ForCorresponding error time,
Δ t is the time interval of twice emitting,It is that relative time between the related approximate zero crossing of γ groups is poor, γ is correlation
The group number of zero crossing.
6. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 5, its feature exists
In deep learning algorithm described in step 2.2 is used as deep learning network model using BP neural network.
7. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 6, its feature exists
In in the step 2.1, at the time of by approximate zero crossing after two sections of ultrasonic signals of Schmidt's integer circuit extraction；It is logical
Cross reverse comparator integer circuit extraction P1、P2The time data that calibrates for error.
8. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 7, its feature exists
In the training sample acquisition methods of the deep learning network model are：
By testing two groups of time data P are obtained using with step 2.1 identical method1、P2, P1={ t11, t12, t13...,
t1m... }, P2={ t21, t22, t23..., t2n... }, by two groups of time data P1And P2Combination of two is carried out, P={ t are obtained11t21；
t11t22；t11t23；...；t1mt2n... }；Then, P, Q are utilized1And Q2Constitute matrixIn formula
t12-mn=1/ (1+e-α(k-β))、α is that empirical parameter, β are to receive signal amplitude maximum corresponding week
Issue, t1-m=| t '1m-t1m-0.5/f1|/2, t2-n=| t '2n-t2n-0.5/f2|/2；It can thus be concluded that X it is each be classified as one it is defeated
Enter characteristic vector, its first row is that the characteristic value of data, the second row and the third line are that Schmidt's shaping is believed two sound waves respectively
Number each zero crossing time-bands come error influence；Finally, the corresponding 1 dimension output of each characteristic vector passes through artificial experience
Setting, its value is between 0 and 1.
9. a kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition as claimed in claim 8, its feature exists
In the realization of the deep learning network model is as follows：First, input data is normalized；Secondly, network is built
And the initial work of network is carried out, it is that 3, hidden layer is that 10, output layer is 1 to set input layer number, and to weights
W, threshold value b, learning rate, training objective minimal error and maximum allowable train epochs are initialized；Then, depth is started
The training of network is practised, substantial amounts of input and output training sample is put into network, the weights in network and the training of threshold value are carried out,
It is final when error limit in tolerance interval or beyond frequency of training, obtain carrying the depth of each new weights and threshold value
Practise network model；Finally, network test is carried out, test result is contrasted with desired value, to assess the network model
Training result, and network parameter is modified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710109786.4A CN107018522B (en) | 2017-02-27 | 2017-02-27 | Positioning method of unmanned aerial vehicle ground base station based on multi-information fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710109786.4A CN107018522B (en) | 2017-02-27 | 2017-02-27 | Positioning method of unmanned aerial vehicle ground base station based on multi-information fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107018522A true CN107018522A (en) | 2017-08-04 |
| CN107018522B CN107018522B (en) | 2020-05-26 |
Family
ID=59440597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710109786.4A Expired - Fee Related CN107018522B (en) | 2017-02-27 | 2017-02-27 | Positioning method of unmanned aerial vehicle ground base station based on multi-information fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107018522B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107902049A (en) * | 2017-10-30 | 2018-04-13 | 上海大学 | The autonomous fuel loading system of unmanned boat based on image and laser sensor |
| CN109125004A (en) * | 2018-09-26 | 2019-01-04 | 张子脉 | A kind of supersonic array obstacle avoidance apparatus, method and its intelligent blind crutch |
| CN110018239A (en) * | 2019-04-04 | 2019-07-16 | 珠海市一微半导体有限公司 | A kind of carpet detection method |
| CN110155350A (en) * | 2019-04-23 | 2019-08-23 | 西北大学 | A kind of unmanned plane landing-gear and its control method |
| CN110287271A (en) * | 2019-06-14 | 2019-09-27 | 南京拾柴信息科技有限公司 | A kind of method for building up of wireless base station and domain type geography atural object incidence matrix |
| CN110758136A (en) * | 2019-09-23 | 2020-02-07 | 广西诚新慧创科技有限公司 | Charging parking apron and unmanned aerial vehicle charging system |
| CN112068160A (en) * | 2020-04-30 | 2020-12-11 | 东华大学 | Unmanned aerial vehicle signal interference method based on navigation positioning system |
| CN112558626A (en) * | 2020-11-11 | 2021-03-26 | 安徽翼讯飞行安全技术有限公司 | Air control system for small civil unmanned aerial vehicle |
| CN112649570A (en) * | 2020-12-11 | 2021-04-13 | 河海大学 | Tail gas detection device and method based on infrared thermal imaging double vision and ultrasonic positioning |
| CN112731442A (en) * | 2021-01-12 | 2021-04-30 | 桂林航天工业学院 | Surveying instrument with adjustable unmanned aerial vehicle survey and drawing is used |
| US20210149046A1 (en) * | 2017-06-30 | 2021-05-20 | Gopro, Inc. | Ultrasonic Ranging State Management for Unmanned Aerial Vehicles |
| CN112835021A (en) * | 2020-12-31 | 2021-05-25 | 杭州海康机器人技术有限公司 | Positioning method, device, system and computer readable storage medium |
| CN116700354A (en) * | 2023-08-01 | 2023-09-05 | 众芯汉创(江苏)科技有限公司 | Spatial position checking and judging method based on visible light data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102883430A (en) * | 2012-09-12 | 2013-01-16 | 南京邮电大学 | Range-based wireless sensing network node positioning method |
| CN105182994A (en) * | 2015-08-10 | 2015-12-23 | 普宙飞行器科技（深圳）有限公司 | Unmanned-aerial-vehicle fixed-point landing method |
| CN106168808A (en) * | 2016-08-25 | 2016-11-30 | 南京邮电大学 | A kind of rotor wing unmanned aerial vehicle automatic cruising method based on degree of depth study and system thereof |
| CN106200677A (en) * | 2016-08-31 | 2016-12-07 | 中南大学 | A kind of express delivery delivery system based on unmanned plane and method |
| CN106184786A (en) * | 2016-08-31 | 2016-12-07 | 马彦亭 | A kind of automatic landing system of unmanned plane and method |
| US20170043871A1 (en) * | 2014-07-08 | 2017-02-16 | X Development Llc | Interaction During Delivery from Aerial Vehicle |
| CN106444824A (en) * | 2016-05-23 | 2017-02-22 | 重庆零度智控智能科技有限公司 | UAV (unmanned aerial vehicle), and UAV landing control method and device |
-
2017
- 2017-02-27 CN CN201710109786.4A patent/CN107018522B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102883430A (en) * | 2012-09-12 | 2013-01-16 | 南京邮电大学 | Range-based wireless sensing network node positioning method |
| US20170043871A1 (en) * | 2014-07-08 | 2017-02-16 | X Development Llc | Interaction During Delivery from Aerial Vehicle |
| CN105182994A (en) * | 2015-08-10 | 2015-12-23 | 普宙飞行器科技（深圳）有限公司 | Unmanned-aerial-vehicle fixed-point landing method |
| CN106444824A (en) * | 2016-05-23 | 2017-02-22 | 重庆零度智控智能科技有限公司 | UAV (unmanned aerial vehicle), and UAV landing control method and device |
| CN106168808A (en) * | 2016-08-25 | 2016-11-30 | 南京邮电大学 | A kind of rotor wing unmanned aerial vehicle automatic cruising method based on degree of depth study and system thereof |
| CN106200677A (en) * | 2016-08-31 | 2016-12-07 | 中南大学 | A kind of express delivery delivery system based on unmanned plane and method |
| CN106184786A (en) * | 2016-08-31 | 2016-12-07 | 马彦亭 | A kind of automatic landing system of unmanned plane and method |
Non-Patent Citations (1)
| Title |
|---|
| 姚金杰: "基于地面基站的目标定位技术研究", 《中国博士学位论文全文数据库》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210149046A1 (en) * | 2017-06-30 | 2021-05-20 | Gopro, Inc. | Ultrasonic Ranging State Management for Unmanned Aerial Vehicles |
| US11982739B2 (en) * | 2017-06-30 | 2024-05-14 | Gopro, Inc. | Ultrasonic ranging state management for unmanned aerial vehicles |
| CN107902049A (en) * | 2017-10-30 | 2018-04-13 | 上海大学 | The autonomous fuel loading system of unmanned boat based on image and laser sensor |
| CN109125004A (en) * | 2018-09-26 | 2019-01-04 | 张子脉 | A kind of supersonic array obstacle avoidance apparatus, method and its intelligent blind crutch |
| CN110018239A (en) * | 2019-04-04 | 2019-07-16 | 珠海市一微半导体有限公司 | A kind of carpet detection method |
| CN110155350A (en) * | 2019-04-23 | 2019-08-23 | 西北大学 | A kind of unmanned plane landing-gear and its control method |
| CN110155350B (en) * | 2019-04-23 | 2022-07-26 | 西北大学 | Control method of unmanned aerial vehicle landing device |
| CN110287271A (en) * | 2019-06-14 | 2019-09-27 | 南京拾柴信息科技有限公司 | A kind of method for building up of wireless base station and domain type geography atural object incidence matrix |
| CN110758136A (en) * | 2019-09-23 | 2020-02-07 | 广西诚新慧创科技有限公司 | Charging parking apron and unmanned aerial vehicle charging system |
| CN112068160A (en) * | 2020-04-30 | 2020-12-11 | 东华大学 | Unmanned aerial vehicle signal interference method based on navigation positioning system |
| CN112068160B (en) * | 2020-04-30 | 2024-03-29 | 东华大学 | Unmanned aerial vehicle signal interference method based on navigation positioning system |
| CN112558626A (en) * | 2020-11-11 | 2021-03-26 | 安徽翼讯飞行安全技术有限公司 | Air control system for small civil unmanned aerial vehicle |
| CN112649570A (en) * | 2020-12-11 | 2021-04-13 | 河海大学 | Tail gas detection device and method based on infrared thermal imaging double vision and ultrasonic positioning |
| CN112835021A (en) * | 2020-12-31 | 2021-05-25 | 杭州海康机器人技术有限公司 | Positioning method, device, system and computer readable storage medium |
| CN112835021B (en) * | 2020-12-31 | 2023-11-14 | 杭州海康威视数字技术股份有限公司 | Positioning method, device, system and computer readable storage medium |
| CN112731442A (en) * | 2021-01-12 | 2021-04-30 | 桂林航天工业学院 | Surveying instrument with adjustable unmanned aerial vehicle survey and drawing is used |
| CN116700354A (en) * | 2023-08-01 | 2023-09-05 | 众芯汉创(江苏)科技有限公司 | Spatial position checking and judging method based on visible light data |
| CN116700354B (en) * | 2023-08-01 | 2023-10-17 | 众芯汉创(江苏)科技有限公司 | Spatial position checking and judging method based on visible light data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107018522B (en) | 2020-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107018522A (en) | A kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition | |
| Cesetti et al. | A vision-based guidance system for UAV navigation and safe landing using natural landmarks | |
| CN102426019B (en) | Unmanned aerial vehicle scene matching auxiliary navigation method and system | |
| CN101339244B (en) | On-board SAR image automatic target positioning method | |
| CN109856638B (en) | Method for automatically searching and positioning specific underwater target | |
| CN108613679A (en) | A kind of mobile robot Extended Kalman filter synchronous superposition method | |
| CN110880071A (en) | Operator-based passive radar combat effectiveness flexible evaluation modeling method | |
| CN102768354A (en) | Method and system for obtaining echo data of underwater target | |
| CN102944238B (en) | Method for determining relative position of planetary probe in process of approaching target | |
| US11614331B2 (en) | Position tracking inside metallic environments using magneto-electric quasistatic fields | |
| CN109738902A (en) | A kind of autonomous acoustic navigation method of underwater high-speed target with high precision based on sync beacon mode | |
| CN111090283B (en) | Unmanned ship combined positioning and orientation method and system | |
| Han et al. | Micro-Doppler-based space target recognition with a one-dimensional parallel network | |
| CN107340529A (en) | A kind of spaceborne frequency measurement localization method, device and system | |
| Loebis et al. | Review of multisensor data fusion techniques and their application to autonomous underwater vehicle navigation | |
| CN113885012A (en) | Satellite-borne photon counting laser radar denoising precision evaluation method for forest research area | |
| CN113448340B (en) | Unmanned aerial vehicle path planning method and device, unmanned aerial vehicle and storage medium | |
| CN102252674A (en) | Underwater geomagnetic positioning and navigation device | |
| CN113483885B (en) | Composite pulse vibration source positioning method based on scorpion hair seam coupling positioning mechanism | |
| Zhou et al. | Initial performance analysis on underside iceberg profiling with autonomous underwater vehicle | |
| US20210306248A1 (en) | Method for self-localizing of an ad hoc network of in-water systems | |
| Inzartsev et al. | Integrated positioning system of autonomous underwater robot and its application in high latitudes of arctic zone | |
| JP6774085B2 (en) | Active sensor signal processing system, signal processing method and signal processing program | |
| CN112379395A (en) | Positioning navigation time service system | |
| CN112394744A (en) | Integrated unmanned aerial vehicle system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200526 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |