CN116520890A - Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection - Google Patents
Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection Download PDFInfo
- Publication number
- CN116520890A CN116520890A CN202310812730.0A CN202310812730A CN116520890A CN 116520890 A CN116520890 A CN 116520890A CN 202310812730 A CN202310812730 A CN 202310812730A CN 116520890 A CN116520890 A CN 116520890A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- stage
- aerial vehicle
- unmanned aerial
- moment
- 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
- 238000007689 inspection Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 60
- 230000008569 process Effects 0.000 claims abstract description 28
- 230000005284 excitation Effects 0.000 claims abstract description 21
- 239000013598 vector Substances 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 7
- 230000033001 locomotion Effects 0.000 description 13
- 238000001914 filtration Methods 0.000 description 7
- 239000004973 liquid crystal related substance Substances 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 210000003746 feather Anatomy 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000005381 potential energy Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention relates to the technical field of unmanned aerial vehicle intelligent obstacle avoidance, in particular to an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection. The platform comprises: and a data acquisition module: the method comprises the steps of acquiring actual speed and predicted speed of an obstacle at each moment in a first stage in the unmanned aerial vehicle inspection process; and a danger judging module: the method comprises the steps of obtaining excitation noise according to the actual speed and the predicted speed of the obstacle at each moment in a first stage, and further determining a predicted active area of the obstacle at each moment in a second stage; obtaining threat coefficients according to the predicted active area, and further judging whether to re-plan the tour path; and a path planning module: and generating an optimal patrol path according to the distance between the unmanned aerial vehicle and the target point, the distance between the unmanned aerial vehicle and the obstacle, the predicted activity area and the threat coefficient when the patrol path is re-planned. The obstacle avoidance path planned by the invention is more reasonable, and reduces the crash risk of the unmanned aerial vehicle during obstacle avoidance.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle intelligent obstacle avoidance, in particular to an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection.
Background
The intelligent unmanned aerial vehicle inspection system has the advantages that the investment of personnel is greatly reduced, the operation efficiency is improved, powerful guarantee is provided for large-area popularization of the multi-rotor unmanned aerial vehicle, and a more convenient and intelligent solution is provided for application environments such as expressways, urban roads, security warehouses, petroleum pipelines, coast inspection and the like. After the airway planning is completed, the unmanned aerial vehicle receives the patrol task, and can automatically carry out daily patrol and hidden trouble investigation according to a preset route. The line operation and maintenance personnel can realize the functions of real-time inspection, photo returning, automatic return and the like by only carrying out one-key operation, thereby realizing intelligent inspection. Although a preset channel line exists, the unmanned aerial vehicle often needs to avoid high-altitude obstacles in the actual inspection process, so that an obstacle avoidance system needs to be carried, the unmanned aerial vehicle is realized by combining a laser radar ranging system with a vision system, the randomness of the motion trail of birds is strong, the obstacle is assumed to be a uniform linear motion model by the existing method, the uncertainty of the obstacle is introduced by adding a Gaussian model on the basis, so that the obstacle avoidance trail which is far away from the obstacle as far as possible is obtained, the bird flight trail is predicted only through an ideal uniform linear motion model, the accuracy can only be guaranteed to be certain in a short time, and the obstacle avoidance reaction is only carried out when the unmanned aerial vehicle approaches to the obstacle during obstacle avoidance, so that the crash risk of the unmanned aerial vehicle can be increased.
Disclosure of Invention
In order to solve the problem that the existing method is inaccurate in prediction of the flight path of an obstacle and unreasonable in obstacle avoidance path planning, the invention aims to provide an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection, and the adopted technical scheme is as follows:
the invention provides an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection, which comprises:
and a data acquisition module: the method comprises the steps of acquiring actual speed and predicted speed of an obstacle at each moment in a first stage in the unmanned aerial vehicle inspection process; the unmanned aerial vehicle inspection process comprises a first stage and a second stage, wherein the distance between the unmanned aerial vehicle and the obstacle at each moment in the first stage is larger than a preset safety distance, and the distance between the unmanned aerial vehicle and the obstacle at the initial moment in the second stage is smaller than or equal to the preset safety distance;
and a danger judging module: the method comprises the steps of obtaining excitation noise corresponding to a second stage according to the actual speed and the predicted speed of the obstacle at each moment in the first stage; determining a predicted active area of the obstacle at each moment in the second stage based on the actual position of the obstacle at each moment in the first stage, the actual speed of the obstacle at the last moment in the first stage and the excitation noise; obtaining threat coefficients of the obstacle to the preset channel according to the predicted positions of the unmanned aerial vehicle on the preset channel and the predicted active areas at each moment in the second stage; judging whether to re-plan the patrol path of the unmanned aerial vehicle based on the threat coefficient;
and a path planning module: when the patrol path of the unmanned aerial vehicle is re-planned, an artificial potential field is constructed according to the distance between the unmanned aerial vehicle and a target point, the distance between the unmanned aerial vehicle and an obstacle, a predicted active area and the threat coefficient, wherein the target point is the shortest distance point for the unmanned aerial vehicle to return to a preset channel after obstacle avoidance; and generating an optimal patrol path based on the artificial potential field.
Preferably, the method for obtaining excitation noise corresponding to the second stage according to the actual speed and the predicted speed of the obstacle at each moment in the first stage includes:
determining the difference between the actual speed of the obstacle at each moment in the first stage and the predicted speed at the corresponding moment as an error vector at the corresponding moment, wherein the speed comprises a speed size and a speed direction; and obtaining a covariance matrix corresponding to the error vector, and taking the covariance matrix as excitation noise corresponding to the second stage.
Preferably, the determining the predicted active area of the obstacle at each moment in the second stage based on the actual position of the obstacle at each moment in the first stage, the actual speed of the obstacle at the last moment in the first stage and the excitation noise includes:
constructing a flight track of an obstacle based on the actual positions of the obstacle at all moments of the first stage, and determining the minimum three-dimensional ellipsoid of the flight track as an active area corresponding to the obstacle at the last moment of the first stage;
taking the excitation noise corresponding to the second stage as the excitation noise of the Kalman filter, taking the active region corresponding to the obstacle at the last moment in the first stage and the actual speed of the obstacle at the corresponding moment as the observation data of the Kalman filter, and obtaining the predicted active regions of the obstacles at all moments in the second stage; wherein the predicted active area at the later time is greater than or equal to the predicted active area at the previous time.
Preferably, according to the predicted position of the unmanned aerial vehicle on the preset channel and the predicted active area in each moment in the second stage, the threat coefficient of the obstacle to the preset channel is obtained, including:
acquiring the center of a predicted active area of the obstacle at each moment in the second stage; calculating the distance between the predicted position of the unmanned aerial vehicle on the preset channel at each moment in the second stage and the center of the corresponding moment, fitting all the distances in the second stage to obtain a first function, and deriving the first function to obtain a first derivative corresponding to each moment in the second stage;
obtaining the length of a long half shaft corresponding to a predicted active area of the obstacle at each moment in the second stage, fitting all the lengths in the second stage to obtain a second function, and deriving the second function to obtain a second derivative corresponding to each moment in the second stage;
and obtaining threat coefficients of the obstacle to a preset channel according to all the first derivatives and all the second derivatives.
Preferably, obtaining threat coefficients of the obstacle to the preset channel according to all the first derivatives and all the second derivatives includes:
for any time in the second stage, obtaining the opposite number of the first derivative corresponding to the time, and recording the hyperbolic tangent value of the opposite number as a first index; the hyperbolic tangent value of the second derivative corresponding to the moment is recorded as a second index;
and obtaining threat coefficients of the obstacle to the preset channel according to the first index and the second index corresponding to each moment in the second stage, wherein the first index and the second index are in positive correlation with the threat coefficients.
Preferably, constructing the artificial potential field according to the distance between the unmanned aerial vehicle and the target point, the distance between the unmanned aerial vehicle and the obstacle, the predicted active area and the threat coefficient includes:
constructing a gravitational field according to the distance between the unmanned plane and the target point;
constructing a repulsive field according to the distance between the unmanned aerial vehicle and the obstacle, the length of a long axis of a predicted active area and the threat coefficient;
an artificial potential field is obtained based on the gravitational field and the repulsive field.
Preferably, the determining whether to re-plan the tour path of the unmanned aerial vehicle based on the threat coefficient includes:
if the threat coefficient is smaller than or equal to the threat coefficient threshold, determining that the patrol path of the unmanned aerial vehicle is not planned again; and if the threat coefficient is greater than the threat coefficient threshold, determining to re-plan the patrol path of the unmanned aerial vehicle.
The invention has at least the following beneficial effects:
aiming at the problems that the effectiveness of a prediction result is short and the danger coefficient of emergency obstacle avoidance is extremely high in the existing mode of predicting the moving track of the obstacle by assuming uniform linear motion of the dynamic obstacle, the method and the device analyze the moving track in stages, wherein the distance between the unmanned aerial vehicle and the obstacle in the first stage is larger than the preset safety distance, namely the unmanned aerial vehicle is in a safety area, the obstacle in the first stage cannot interfere the flying of the unmanned aerial vehicle, the method and the device predict the speed of the obstacle in each moment in the first stage, the distance between the unmanned aerial vehicle and the obstacle in the second stage is smaller than or equal to the preset safety distance, namely the obstacle possibly interferes the unmanned aerial vehicle to fly in the preset navigation channel, the method and the device predict the moving area of the obstacle in the second stage based on the flying condition of the first stage, and the method and the device can obtain the predicted noise coefficient of the obstacle in the second stage based on the predicted speed of each obstacle in the first stage, and the device can estimate the predicted noise of the obstacle in the second stage based on the predicted speed of the real obstacle in the first stage; when the patrol path of the unmanned aerial vehicle is re-planned, the threat coefficient is used as an adjusting coefficient corresponding to the obstacle in the subsequent artificial potential field algorithm, when the threat is larger, the repulsive force is larger, the safety of the obtained obstacle avoidance track is higher, the planned obstacle avoidance path is more reasonable, and the crash risk when the unmanned aerial vehicle is in obstacle avoidance is greatly reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a system block diagram of an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection provided by an embodiment of the invention;
fig. 2 is a schematic view of a drone and a target point in an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of a three-dimensional holographic inspection unmanned aerial vehicle control platform according to the invention with reference to the attached drawings and the preferred embodiment.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection, which is specifically described below with reference to the accompanying drawings.
An unmanned aerial vehicle control platform embodiment capable of three-dimensional holographic inspection:
the specific scene aimed at by this embodiment is: although the unmanned aerial vehicle has a preset channel line in the inspection process, high-altitude obstacles possibly appear in the inspection process, birds are the most of the high-altitude obstacles, in order to avoid the obstacles, the unmanned aerial vehicle is usually required to carry an obstacle avoidance system, commonly used laser radar ranging and a vision system are combined to realize unmanned aerial vehicle flight obstacle avoidance, but the motion trail randomness of the birds is strong, the existing method assumes the obstacles as a uniform linear motion model, the uncertainty of the obstacles is introduced by adding a Gaussian model on the basis, so that the obstacle avoidance trail which is far away from the obstacles as much as possible is obtained, but the bird flight trail is predicted only through the ideal uniform linear motion model, the obstacle avoidance reaction can be carried out only in a short time, and the risk of the unmanned aerial vehicle is increased if the unmanned aerial vehicle waits for approaching the obstacles in the obstacle avoidance process.
The embodiment provides an unmanned aerial vehicle control platform capable of three-dimensional holographic inspection, as shown in fig. 1, the unmanned aerial vehicle control platform capable of three-dimensional holographic inspection of the embodiment comprises a data acquisition module, a danger judgment module and a path planning module, and the modules are respectively introduced.
I, data acquisition module
The data acquisition module in the embodiment is used for acquiring the actual speed and the predicted speed of the obstacle at each moment in the first stage in the unmanned aerial vehicle inspection process; the unmanned aerial vehicle inspection process comprises a first stage and a second stage, wherein the distance between the unmanned aerial vehicle and the obstacle at each moment in the first stage is larger than a preset safety distance, and the distance between the unmanned aerial vehicle and the obstacle at the initial moment in the second stage is smaller than or equal to the preset safety distance.
The unmanned aerial vehicle has the flight channel of predetermineeing in the inspection process, intelligent flight platform outputs a predetermined flight channel to the three-dimensional modeling of inspection region, the unmanned aerial vehicle needs report the emergency to the platform in the real-time flight process, adopt 5G communication between the two, have high data transmission, processing, feedback efficiency, unmanned aerial vehicle utilizes camera and laser radar to gather image data and point cloud data in the inspection flight process, wherein, image data includes the aerial image in the camera field of vision, point cloud data includes the positional information of barrier, image data and the point cloud data that gathers all send to flight platform, carry out analysis by flight platform, processing and obtain emergency scheme feedback to unmanned aerial vehicle execution rapidly, it is to be noted that, emergency scheme in this embodiment is unmanned aerial vehicle obstacle avoidance path's planning. The unmanned aerial vehicle obstacle avoidance path planning not only needs to identify the obstacle, but also needs to predict the flight path of the obstacle, evaluate the threat of the obstacle to the unmanned aerial vehicle flight path, and then obtain the accurate obstacle avoidance path.
In the inspection flight process of the unmanned aerial vehicle, the pose of the unmanned aerial vehicle is always parallel to a preset channel line, the carried camera and the laser radar system continuously collect image data and point cloud data in front of the unmanned aerial vehicle, the collected data are transmitted back to the flight platform, the flight platform is carried with a YOLOv5 flight biological recognition model trained by a large number of obstacle atlas, the obstacles appearing in the image are recognized, and the distance between the unmanned aerial vehicle and the obstacles appearing in the picture at the current moment is acquired by combining the transmitted laser radar point cloud data.
According to the method, the unmanned aerial vehicle inspection process is divided into two stages, the distance between the unmanned aerial vehicle and the obstacle at each moment in the first stage is larger than the preset safety distance, the distance between the unmanned aerial vehicle and the obstacle at the initial moment in the second stage is smaller than or equal to the preset safety distance, namely, the moment when the obstacle is smaller than or equal to the preset safety distance is taken as a dividing point, the historical moment before the dividing point in the unmanned aerial vehicle inspection process is taken as the first stage, and the dividing point in the unmanned aerial vehicle inspection process and the moment after the dividing point are taken as the second stage. The preset safety distance in this embodiment is 100 meters, and in a specific application, an implementer can set according to a specific situation. The first stage is following prediction, namely, assuming that each movement of the obstacle is uniform linear movement, and predicting the flight state of the next moment only according to the gesture and the speed of the obstacle at the current moment; however, when the distance between the obstacle and the unmanned aerial vehicle is smaller than or equal to the preset safety distance, namely, the unmanned aerial vehicle enters the early warning range, the following prediction of the first stage is stopped, the unmanned aerial vehicle enters the second stage, the noise item of the Kalman filtering prediction of the second stage is determined according to the monitoring and following prediction results of the first stage, the speed and the activity range at the last moment before entering the early warning range are used as initial input values, the subsequent prediction results are obtained through continuous iteration, and whether the obstacle activity track threatens the flight of the unmanned aerial vehicle is judged.
The principle of bird cruise flapping wings proves that the steering is performed in the flying process of birds, and the steering has a direct relation with the tail feather and double-wing gestures: 1. when birds turn left, the left wing is required to hang down and the right wing is required to lift up, and then tail feathers are tilted up; 2. when birds turn right, the left wing is required to raise the right wing to droop, and then the tail feather is tilted upwards; 3. when the double wings swing in the oblique rear direction, climbing upwards is carried out at the current flying height; 4. when the double wings are backward inclined, namely the front edge is high and the rear edge is low, the double wings keep the current flying height or downward gliding action; 5. when the double wings swing perpendicular to the flight direction, the double wings hover in the air. The bird flight attitude recognition model is trained through the classified neural network, a large number of bird flight attitudes are obtained through the large data set, the corner direction corresponding to the corresponding attitude is input into the neural network for training, the graphic data collected by the unmanned aerial vehicle can be input into the trained neural network after training is completed, and the flight direction of the next moment is predicted. The training process of the neural network is the prior art, and will not be described in detail here. And acquiring the distance between the unmanned aerial vehicle and the obstacle, the actual speed of the obstacle at each moment in the first stage and the position of the obstacle based on the point cloud data, wherein the speed comprises the speed and the speed direction. According to the method, the actual speed of each obstacle at each moment in the first stage of the unmanned aerial vehicle inspection process is collected, the speed of each obstacle at each moment in the first stage is predicted by adopting the existing prediction model, the predicted speed of each obstacle at each moment in the first stage of the unmanned aerial vehicle inspection process is obtained, and the method for predicting data based on the existing model is the existing method and is not repeated here.
So far, the actual speed and the predicted speed of the obstacle at each moment in the first stage of the unmanned aerial vehicle inspection process are obtained.
II, dangerous judgement module
The risk judging module of the embodiment is used for obtaining excitation noise corresponding to the second stage according to the actual speed and the predicted speed of the obstacle at each moment in the first stage; determining a predicted active area of the obstacle at each moment in the second stage based on the actual position of the obstacle at each moment in the first stage, the actual speed of the obstacle at the last moment in the first stage and the excitation noise; obtaining threat coefficients of the obstacle to the preset channel according to the predicted positions of the unmanned aerial vehicle on the preset channel and the predicted active areas at each moment in the second stage; and judging whether to re-plan the patrol path of the unmanned aerial vehicle based on the threat coefficient.
The data acquisition module acquires the actual speed and the predicted speed of the obstacle at each moment in the first stage of the unmanned aerial vehicle inspection process, and for any moment in the first stage: determining the difference between the actual speed of the obstacle at the moment and the predicted speed at the moment as an error vector at the moment, wherein the error vector at the moment isWherein->For the error vector at this instant, +.>For the actual speed of the obstacle at that moment, +.>A predicted speed for the obstacle at that time; by adopting the method, the error vector of each moment in the first stage is obtained, namely an error vector set is obtained; the error vector is a noise item of the actual flying speed deviating from the predicted flying speed in the obstacle flying process, and the obstacleWhen the flight direction changes each time, the traditional prediction is performed by taking uniform linear motion as an assumption, but the actual flight of birds at the current moment is not performed by uniform linear motion, and certain errors or other uncertain factors possibly exist between the actual flight angle and the predicted flight angle of the obstacle to interfere the preliminary prediction, so that the prediction result is inaccurate, and the obtained error vector also influences the prediction process of the subsequent Kalman filtering. If the unmanned aerial vehicle makes obstacle avoidance reaction near the obstacle, the risk is extremely high, so that the unmanned aerial vehicle immediately stops following prediction once the distance between the unmanned aerial vehicle and the obstacle enters the early warning range, the second-stage prediction is carried out, the obstacle avoidance track is determined according to the prediction result, and the safety can be guaranteed to the greatest extent only by making reaction in advance. In this embodiment, covariance matrixes corresponding to all error vectors are obtained, the covariance matrixes are used as excitation noise corresponding to the second stage, random variables at the last moment before entering an early warning range are used as initial iteration inputs, namely random variables at the last moment in the first stage are used as initial iteration inputs, and all prediction results in the subsequent fault-tolerant time are iteratively output.
Kalman filtering is an algorithm for optimally estimating the state of a system by using a linear system state equation and through system input and output observation data. Since the observed data includes the effects of noise and interference in the system, the optimal estimate can also be considered as a filtering process, i.e. the predicted value obtained by the linear mathematical model+the sensed measurement value=a more accurate measurement value. The parameters of the kalman filtering can be obtained through iteration of input items, wherein the parameters are adjustable parameters, the most important parameters are observation errors and excitation errors, the upper limit of accuracy of a prediction result is determined by two noise items, and the upper limit is generally set manually, but an unmanned plane observation object is the flight state of a remote obstacle, so that the measurement accuracy is not high, in order to ensure the accuracy of the prediction result, two variables are set for each obstacle, namely the speed of each obstacle at each moment and a corresponding active area, and the active area is obtained based on a flight track. According to the actual positions of the obstacles at all moments in the first stage, a continuous-change point cloud three-dimensional reconstruction technology is utilized to construct a complete flight path of the obstacles in the first stage, the minimum three-dimensional ellipsoid of the flight path is determined as an active area corresponding to the obstacle at the last moment in the first stage, and the active area is determined based on the actual flight path of the obstacle, so that the active area is an actual active range.
Considering that the speed and the moving range are not directly related, but the faster the speed is, the more likely the moving area of the obstacle changes, the more likely the moving area threatens the preset channel of the unmanned aerial vehicle, and because the flying direction of the obstacle cannot be determined in a longer prediction process, the embodiment takes the moving area as a prediction variable, takes an error vector following the prediction in a first stage as an uncertain factor of flying direction change and non-uniform flying, namely an excitation noise item, inputs the flying speed of each moment of the obstacle and the moving area corresponding to the moment into a classical Kalman filter as observation data, carries out classical Kalman filtering on the basis of the observation data to obtain a gain value and a state transition matrix, and the measurement noise covariance of the Kalman filter is self-set by a flying platform of the unmanned aerial vehicle according to the data acquisition precision of the unmanned aerial vehicle, the precision of a bird recognition module, the precision of a bird gesture acquisition module and the like; and predicting the active area of the obstacle at the next moment by taking the covariance matrix as excitation noise of interference of the directional randomness on the active area and taking the observed data corresponding to the last moment of the first stage as an initial iteration value, and then carrying out the next iteration by utilizing the covariance of the optimal estimated value at the next moment to obtain the active area of the obstacle which is not monitored in a follow-up manner in the fault-tolerant time, namely, each moment of the second stage.
Thus, the predicted active area of the obstacle at each moment in the second stage is obtained.
Considering that as the distance between the obstacle and the unmanned aerial vehicle gets closer and the moving area of the obstacle gets larger and faster, the obstacle is more likely to threaten the normal tour of the unmanned aerial vehicle, the more the navigation path of the unmanned aerial vehicle needs to be readjusted in order to avoid collision of the unmanned aerial vehicle with the obstacle. Based on the method, firstly, the predicted positions of the unmanned aerial vehicle at each moment in the second stage on a preset channel are obtained, then the center of the predicted active area of the obstacle at each moment in the second stage is obtained, the Euclidean distance between the predicted positions of the unmanned aerial vehicle at each moment in the second stage on the preset channel and the center at the corresponding moment is calculated, all Euclidean distances in the second stage are fitted to obtain a first function, and the first function is derived to obtain a first derivative corresponding to each moment in the second stage; obtaining the length of a long half shaft corresponding to a predicted active area of the obstacle at each moment in the second stage, fitting all the lengths in the second stage to obtain a second function, and deriving the second function to obtain a second derivative corresponding to each moment in the second stage; for any time in the second stage, obtaining the opposite number of the first derivative corresponding to the time, and recording the hyperbolic tangent value of the opposite number as a first index; the hyperbolic tangent value of the second derivative corresponding to the moment is recorded as a second index; and obtaining threat coefficients of the obstacle to the preset channel according to the first index and the second index corresponding to each moment in the second stage, wherein the first index and the second index are in positive correlation with the threat coefficients. The positive correlation relationship indicates that the dependent variable increases with the increase of the independent variable, and the dependent variable decreases with the decrease of the independent variable, and the specific relationship may be a multiplication relationship, an addition relationship, an idempotent of an exponential function, and is determined by practical application. As a specific embodiment, a specific calculation formula of threat coefficients of the obstacle to the preset channel is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,threat coefficient for obstacle to preset channel, < ->For the number of moments in the second phase, +.>For the predicted position of the unmanned aerial vehicle on the preset channel and the predicted activity area of the obstacle at the ith moment of the second stageEuclidean distance between centers of>For the first derivative corresponding to the i-th moment of the second phase,/and>length of long half axis corresponding to predicted active area of obstacle at i-th moment of second stage, +.>For the second derivative corresponding to the i-th moment of the second phase,/and>as a hyperbolic tangent function.
Representing a first derivative corresponding to the ith moment of the second stage, and representing the change condition of the distance between the center of the obstacle active area and the unmanned aerial vehicle at the ith moment of the second stage, wherein the distance can be reduced or increased; when the first derivative corresponding to the ith moment in the second stage is a negative number, the distance between the predicted position of the unmanned aerial vehicle on the preset channel and the center of the predicted active area of the obstacle is reduced; when the first derivative corresponding to the ith moment in the second stage is a positive number, the distance between the predicted position of the unmanned aerial vehicle on the preset channel and the center of the predicted active area of the obstacle is increased. />Introducing a hyperbolic tangent function to logically correct the derivative value for the first index, namely calculating the hyperbolic tangent value of the opposite number of the derivative, namely when the derivative value is negative and smaller, the threat degree of the predicted active area of the obstacle to the channel is larger and larger, and the first index is gradually increased between 0 and 1; when the derivative value is positive and is larger, the threat degree of the predicted active area of the obstacle to the channel is smaller and smaller, and the first index is gradually reduced from-1 to 0. />For the second index, ++>The greater the value of (2) is greater than or equal to 0, which means that the obstacle activity area is gradually increasing, the threat to the channel is increasing, and the greater the derivative value is by introducing a hyperbolic tangent function, the greater the value of the second index is between 0 and 1. And T represents the number of moments in the second stage, namely the maximum fault-tolerant time for which the unmanned aerial vehicle approaches at the current speed under the assumption that the position of the obstacle is fixed just after entering the early warning range. In the embodiment, the threat value of the obstacle to the unmanned aerial vehicle flying on the preset channel at the ith moment in the second stage is represented by the sum of the first index and the second index, and the threat coefficient of the obstacle to the preset channel is obtained.
Setting a threat coefficient threshold, wherein the threat coefficient is 0 in the embodiment, and in a specific application, an implementer can set according to specific conditions; when the threat coefficient of the obstacle to the preset channel is smaller than or equal to 0, judging that the obstacle has no threat to the unmanned aerial vehicle flying on the preset channel; when the threat coefficient of the obstacle to the preset channel is greater than 0, the threat of the obstacle to the unmanned aerial vehicle flying on the preset channel is judged, and at the moment, obstacle avoidance measures are needed to be taken, namely, the patrol path of the unmanned aerial vehicle is needed to be planned again.
So far, the danger judging module of the embodiment judges whether to re-plan the patrol path of the unmanned aerial vehicle based on the threat coefficient.
III path planning module
The path planning module of the embodiment is used for constructing an artificial potential field according to the distance between the unmanned aerial vehicle and the target point, the distance between the unmanned aerial vehicle and the obstacle, the predicted active area and the threat coefficient when the tour path of the unmanned aerial vehicle is planned again, wherein the target point is the shortest distance point for the unmanned aerial vehicle to return to the preset channel after obstacle avoidance; and generating an optimal patrol path based on the artificial potential field.
When the path of the unmanned aerial vehicle is re-planned, the embodiment completes the planning of the patrol path of the unmanned aerial vehicle by combining the artificial potential field, takes the obstacle as a potential energy high point, calculates a potential field diagram of the three-dimensional space obtained by the whole known point cloud, and automatically avoids the obstacle to roll to the target point like a rolling ball under ideal conditions. Wherein the artificial potential field comprises a gravitational field and a repulsive field, and the gravitational field is specifically:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the attractive force of the target point on the unmanned aerial vehicle, < ->Is Unmanned Aerial Vehicle (UAV)>For the corresponding target point when avoiding obstacles, < > for>For the distance between the unmanned aerial vehicle and the target point, +.>The gain coefficient is proportional to the attraction force.
The target point is the shortest distance point of the unmanned aerial vehicle returning to the preset channel after obstacle avoidance, according to the detour radian, the maximum obstacle avoidance diameter is adopted for limiting under the condition that the obstacle avoidance track is unknown, and the length of the long axis of the active area of the obstacle in the Kalman filtering prediction and the sum value of the distances between the unmanned aerial vehicle and the obstacle at the corresponding moment are recorded as a first length, namelyWherein->The length of the long half shaft corresponding to the predicted active area of the obstacle at the i-th moment of the second stage,the distance between the unmanned aerial vehicle and the obstacle at the ith moment of the second stage. Based on the first length and the position of the unmanned aerial vehicle, the target point on the preset channel can be determined, as shown in fig. 2, which is a schematic diagram of the unmanned aerial vehicle and the target point. The present embodiment sets->The value of 1, in a particular application, the practitioner can set according to the particular circumstances. The distance in this embodiment is a euclidean distance.
The repulsive force is specifically as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,the repulsive force of the target point to the unmanned aerial vehicle is +.>Is Unmanned Aerial Vehicle (UAV)>In the form of an obstacle to the formation of a solid,for the distance between the unmanned aerial vehicle and the obstacle, < >>To predict the length of the long axis of the active area, +.>The threat coefficient of the obstacle to the preset channel is given, and k is the repulsive force proportional gain coefficient.
Representing the maximum range of influence of obstacles on the unmanned aerial vehicle, in the present invention +.>Is set to the length of the long axis of the active area of the obstacle in the kalman filter prediction. />As a repulsive force formula of the artificial potential field, when the threat coefficient of the obstacle to the preset channel is larger, the repulsive force between the unmanned aerial vehicle and the obstacle needs to be larger to avoid the obstacle, so that ∈>Multiplied by->Representing an increase in repulsive force with an increase in obstacle threat coefficient. In this embodiment, k is set to 1, and in a specific application, an implementer may set according to a specific situation.
By adopting the method, the repulsive force of the unmanned aerial vehicle on the obstacle and the attractive force of the target point on the unmanned aerial vehicle can be obtained, and the controlled object, namely the unmanned aerial vehicle, is subjected to the repulsive force action and the attractive force action in the composite field formed by the two potential fields, so that the resultant force of the repulsive force and the attractive force guides the movement of the unmanned aerial vehicle, and the optimal inspection path of the unmanned aerial vehicle is generated. The manual potential field regulation path is the prior art and will not be repeated here.
So far, by adopting the platform provided by the embodiment, the re-planning of the inspection path of the unmanned aerial vehicle is completed, and the optimal inspection path of the unmanned aerial vehicle is obtained.
Aiming at the problems that the existing assumed dynamic obstacle moves at a uniform speed to predict the moving track of the obstacle and avoid the obstacle, the method has the defects that the effectiveness of a prediction result is short and the danger coefficient of emergency avoidance is extremely high, the method analyzes the problems in stages, the distance between the unmanned aerial vehicle and the obstacle in the first stage is larger than the preset safety distance, namely the unmanned aerial vehicle is in a safety area, the obstacle in the first stage cannot interfere the flying of the unmanned aerial vehicle, the method predicts the speed of the obstacle in each moment in the first stage, the distance between the unmanned aerial vehicle and the obstacle in the second stage is smaller than or equal to the preset safety distance, namely the obstacle possibly interferes the unmanned aerial vehicle to fly in the preset channel, the method predicts the moving area of the obstacle in the second stage based on the flying condition of the obstacle in the first stage, and the method predicts the moving area of the obstacle in the second stage based on the assumption that the actual flying angle of the obstacle is not uniform speed linear motion, and a certain error is possibly present between the actual flying angle of the obstacle in the first stage and the predicted flying angle, the method predicts the moving noise in the second stage based on the actual speed of the obstacle in the second stage, the method is estimated by the method, the method obtains the estimated moving coefficient of the obstacle in the second stage of the flight of the obstacle in the first stage, the method is estimated on the estimated moving area of the actual threat of the obstacle in the second stage, the estimated moving area of the obstacle is estimated based on the actual speed of the estimated moving area of the obstacle in the first stage, and the estimated moving coefficient of the obstacle in the second stage, and the estimated moving area is estimated on the actual speed of the actual estimated real speed of the obstacle in the second stage, the threat coefficient is used as an adjusting coefficient corresponding to the obstacle in the subsequent artificial potential field algorithm, when the threat is larger, the repulsive force is larger, the safety of the obtained obstacle avoidance track is higher, and the crash risk when the unmanned aerial vehicle is in obstacle avoidance is greatly reduced.
Claims (7)
1. Unmanned aerial vehicle control platform that can three-dimensional holographic inspection, its characterized in that, this platform includes:
and a data acquisition module: the method comprises the steps of acquiring actual speed and predicted speed of an obstacle at each moment in a first stage in the unmanned aerial vehicle inspection process; the unmanned aerial vehicle inspection process comprises a first stage and a second stage, wherein the distance between the unmanned aerial vehicle and the obstacle at each moment in the first stage is larger than a preset safety distance, and the distance between the unmanned aerial vehicle and the obstacle at the initial moment in the second stage is smaller than or equal to the preset safety distance;
and a danger judging module: the method comprises the steps of obtaining excitation noise corresponding to a second stage according to the actual speed and the predicted speed of the obstacle at each moment in the first stage; determining a predicted active area of the obstacle at each moment in the second stage based on the actual position of the obstacle at each moment in the first stage, the actual speed of the obstacle at the last moment in the first stage and the excitation noise; obtaining threat coefficients of the obstacle to the preset channel according to the predicted positions of the unmanned aerial vehicle on the preset channel and the predicted active areas at each moment in the second stage; judging whether to re-plan the patrol path of the unmanned aerial vehicle based on the threat coefficient;
and a path planning module: when the patrol path of the unmanned aerial vehicle is re-planned, an artificial potential field is constructed according to the distance between the unmanned aerial vehicle and a target point, the distance between the unmanned aerial vehicle and an obstacle, a predicted active area and the threat coefficient, wherein the target point is the shortest distance point for the unmanned aerial vehicle to return to a preset channel after obstacle avoidance; and generating an optimal patrol path based on the artificial potential field.
2. The unmanned aerial vehicle control platform for three-dimensional holographic inspection according to claim 1, wherein the method for obtaining excitation noise corresponding to the second stage according to the actual speed and the predicted speed of the obstacle at each moment in the first stage comprises:
determining the difference between the actual speed of the obstacle at each moment in the first stage and the predicted speed at the corresponding moment as an error vector at the corresponding moment, wherein the speed comprises a speed size and a speed direction; and obtaining a covariance matrix corresponding to the error vector, and taking the covariance matrix as excitation noise corresponding to the second stage.
3. The unmanned aerial vehicle control platform of claim 1, wherein the determining the predicted active area of the obstacle at each time of the second stage based on the actual position of the obstacle at each time of the first stage, the actual speed of the obstacle at the last time of the first stage, and the excitation noise comprises:
constructing a flight track of an obstacle based on the actual positions of the obstacle at all moments of the first stage, and determining the minimum three-dimensional ellipsoid of the flight track as an active area corresponding to the obstacle at the last moment of the first stage;
taking the excitation noise corresponding to the second stage as the excitation noise of the Kalman filter, taking the active region corresponding to the obstacle at the last moment in the first stage and the actual speed of the obstacle at the corresponding moment as the observation data of the Kalman filter, and obtaining the predicted active regions of the obstacles at all moments in the second stage; wherein the predicted active area at the later time is greater than or equal to the predicted active area at the previous time.
4. The unmanned aerial vehicle control platform capable of three-dimensional holographic inspection according to claim 3, wherein obtaining threat coefficients of the obstacle to the preset channel according to the predicted position of the unmanned aerial vehicle on the preset channel and the predicted active area at each moment in the second stage comprises:
acquiring the center of a predicted active area of the obstacle at each moment in the second stage; calculating the distance between the predicted position of the unmanned aerial vehicle on the preset channel at each moment in the second stage and the center of the corresponding moment, fitting all the distances in the second stage to obtain a first function, and deriving the first function to obtain a first derivative corresponding to each moment in the second stage;
obtaining the length of a long half shaft corresponding to a predicted active area of the obstacle at each moment in the second stage, fitting all the lengths in the second stage to obtain a second function, and deriving the second function to obtain a second derivative corresponding to each moment in the second stage;
and obtaining threat coefficients of the obstacle to a preset channel according to all the first derivatives and all the second derivatives.
5. The unmanned aerial vehicle control platform of claim 4, wherein obtaining threat coefficients of the obstacle to the preset channel based on all the first derivatives and all the second derivatives comprises:
for any time in the second stage, obtaining the opposite number of the first derivative corresponding to the time, and recording the hyperbolic tangent value of the opposite number as a first index; the hyperbolic tangent value of the second derivative corresponding to the moment is recorded as a second index;
and obtaining threat coefficients of the obstacle to the preset channel according to the first index and the second index corresponding to each moment in the second stage, wherein the first index and the second index are in positive correlation with the threat coefficients.
6. A three-dimensional holographic inspection unmanned aerial vehicle control platform according to claim 3, wherein constructing an artificial potential field from the distance between the unmanned aerial vehicle and the target point, the distance between the unmanned aerial vehicle and the obstacle, the predicted active area and the threat coefficients comprises:
constructing a gravitational field according to the distance between the unmanned plane and the target point;
constructing a repulsive field according to the distance between the unmanned aerial vehicle and the obstacle, the length of a long axis of a predicted active area and the threat coefficient;
an artificial potential field is obtained based on the gravitational field and the repulsive field.
7. The unmanned aerial vehicle control platform of claim 1, wherein the determining whether to re-plan the tour path of the unmanned aerial vehicle based on the threat coefficients comprises:
if the threat coefficient is smaller than or equal to the threat coefficient threshold, determining that the patrol path of the unmanned aerial vehicle is not planned again; and if the threat coefficient is greater than the threat coefficient threshold, determining to re-plan the patrol path of the unmanned aerial vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310812730.0A CN116520890B (en) | 2023-07-05 | 2023-07-05 | Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310812730.0A CN116520890B (en) | 2023-07-05 | 2023-07-05 | Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116520890A true CN116520890A (en) | 2023-08-01 |
| CN116520890B CN116520890B (en) | 2023-09-05 |
Family
ID=87406774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310812730.0A Active CN116520890B (en) | 2023-07-05 | 2023-07-05 | Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116520890B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116878471A (en) * | 2023-09-06 | 2023-10-13 | 湖南湘船重工有限公司 | Unmanned survey and drawing system on water |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105159297A (en) * | 2015-09-11 | 2015-12-16 | 南方电网科学研究院有限责任公司 | Power transmission line unmanned plane inspection obstacle avoidance system and method |
| CN105841703A (en) * | 2016-03-15 | 2016-08-10 | 电子科技大学 | Calculating method for optimal route of unmanned aerial vehicle used for positioning object in threat environment |
| WO2018058442A1 (en) * | 2016-09-29 | 2018-04-05 | 深圳市大疆创新科技有限公司 | Method and device for panning route, flight control system, omnidirectional obstacle avoidance system, and unmanned aerial vehicle |
| CN108536171A (en) * | 2018-03-21 | 2018-09-14 | 电子科技大学 | The paths planning method of multiple no-manned plane collaboration tracking under a kind of multiple constraint |
| CN112965496A (en) * | 2021-02-23 | 2021-06-15 | 武汉理工大学 | Path planning method and device based on artificial potential field algorithm and storage medium |
| US20210272466A1 (en) * | 2020-02-28 | 2021-09-02 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| CN115556769A (en) * | 2022-10-19 | 2023-01-03 | 阿波罗智联(北京)科技有限公司 | Obstacle state quantity determination method and device, electronic device and medium |
| CN115793709A (en) * | 2022-12-09 | 2023-03-14 | 大连大学 | APF unmanned aerial vehicle path planning method based on POMDP model |
| CN116107329A (en) * | 2021-11-10 | 2023-05-12 | 海鹰航空通用装备有限责任公司 | Path planning method for distributed bee colony in complex environment |
-
2023
- 2023-07-05 CN CN202310812730.0A patent/CN116520890B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105159297A (en) * | 2015-09-11 | 2015-12-16 | 南方电网科学研究院有限责任公司 | Power transmission line unmanned plane inspection obstacle avoidance system and method |
| CN105841703A (en) * | 2016-03-15 | 2016-08-10 | 电子科技大学 | Calculating method for optimal route of unmanned aerial vehicle used for positioning object in threat environment |
| WO2018058442A1 (en) * | 2016-09-29 | 2018-04-05 | 深圳市大疆创新科技有限公司 | Method and device for panning route, flight control system, omnidirectional obstacle avoidance system, and unmanned aerial vehicle |
| CN108536171A (en) * | 2018-03-21 | 2018-09-14 | 电子科技大学 | The paths planning method of multiple no-manned plane collaboration tracking under a kind of multiple constraint |
| US20210272466A1 (en) * | 2020-02-28 | 2021-09-02 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| CN112965496A (en) * | 2021-02-23 | 2021-06-15 | 武汉理工大学 | Path planning method and device based on artificial potential field algorithm and storage medium |
| CN116107329A (en) * | 2021-11-10 | 2023-05-12 | 海鹰航空通用装备有限责任公司 | Path planning method for distributed bee colony in complex environment |
| CN115556769A (en) * | 2022-10-19 | 2023-01-03 | 阿波罗智联(北京)科技有限公司 | Obstacle state quantity determination method and device, electronic device and medium |
| CN115793709A (en) * | 2022-12-09 | 2023-03-14 | 大连大学 | APF unmanned aerial vehicle path planning method based on POMDP model |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116878471A (en) * | 2023-09-06 | 2023-10-13 | 湖南湘船重工有限公司 | Unmanned survey and drawing system on water |
| CN116878471B (en) * | 2023-09-06 | 2023-11-10 | 湖南湘船重工有限公司 | Unmanned survey and drawing system on water |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116520890B (en) | 2023-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lee et al. | Deep learning-based monocular obstacle avoidance for unmanned aerial vehicle navigation in tree plantations: Faster region-based convolutional neural network approach | |
| US20190310651A1 (en) | Object Detection and Determination of Motion Information Using Curve-Fitting in Autonomous Vehicle Applications | |
| Ross et al. | Learning monocular reactive uav control in cluttered natural environments | |
| Tisdale et al. | Autonomous UAV path planning and estimation | |
| CN111932588A (en) | Tracking method of airborne unmanned aerial vehicle multi-target tracking system based on deep learning | |
| Levine et al. | Information-rich path planning with general constraints using rapidly-exploring random trees | |
| CN105759829A (en) | Laser radar-based mini-sized unmanned plane control method and system | |
| Lawrance et al. | Path planning for autonomous soaring flight in dynamic wind fields | |
| CN109885084A (en) | A kind of multi-rotor unmanned aerial vehicle Autonomous landing method based on monocular vision and fuzzy control | |
| CN109582038B (en) | Unmanned aerial vehicle path planning method | |
| CN116520890B (en) | Unmanned aerial vehicle control platform capable of three-dimensional holographic inspection | |
| Bipin et al. | Autonomous navigation of generic monocular quadcopter in natural environment | |
| CN110262555B (en) | Real-time obstacle avoidance control method for unmanned aerial vehicle in continuous obstacle environment | |
| Fu et al. | Vision-based obstacle avoidance for flapping-wing aerial vehicles | |
| CN114200471B (en) | Forest fire source detection system and method based on unmanned aerial vehicle, storage medium and equipment | |
| Chou et al. | Predictive runtime monitoring of vehicle models using Bayesian estimation and reachability analysis | |
| CN116907282B (en) | Unmanned target aircraft ultra-low altitude flight control method based on artificial intelligence algorithm | |
| WO2023109589A1 (en) | Smart car-unmanned aerial vehicle cooperative sensing system and method | |
| CN106155082A (en) | A kind of unmanned plane bionic intelligence barrier-avoiding method based on light stream | |
| CN111831010A (en) | Unmanned aerial vehicle obstacle avoidance flight method based on digital space slice | |
| CN117369479B (en) | Unmanned aerial vehicle obstacle early warning method and system based on oblique photogrammetry technology | |
| Lu et al. | Perception and avoidance of multiple small fast moving objects for quadrotors with only low-cost RGBD camera | |
| Anastasiou et al. | Hyperion: A robust drone-based target tracking system | |
| EP4002837A1 (en) | Object tracking system including stereo camera assembly and methods of use | |
| Qi et al. | Detection and tracking of a moving target for UAV based on machine vision |
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 |