CN111506104A - Method and device for planning position of unmanned aerial vehicle - Google Patents
Method and device for planning position of unmanned aerial vehicle Download PDFInfo
- Publication number
- CN111506104A CN111506104A CN202010258490.0A CN202010258490A CN111506104A CN 111506104 A CN111506104 A CN 111506104A CN 202010258490 A CN202010258490 A CN 202010258490A CN 111506104 A CN111506104 A CN 111506104A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- moving
- moving direction
- neural network
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
Abstract
The embodiment of the invention provides a method and a device for planning the position of an unmanned aerial vehicle, wherein the method comprises the following steps: obtaining the current position coordinate of an unmanned aerial vehicle and the current position coordinate of a mobile terminal in a service area, calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal, and determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability. The average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the position coordinate of the next moment is the maximum, and the communication quality is higher. Therefore, the communication quality of the unmanned aerial vehicle and the mobile terminal can be improved under the condition that the communication efficiency of the unmanned aerial vehicle and the mobile terminal is not reduced.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a method and a device for planning the position of an unmanned aerial vehicle.
Background
The unmanned aerial vehicle is a key role of wireless communication, and when the mobile terminal is far away from the ground base station and the ground base station cannot provide communication for the mobile terminal, the unmanned aerial vehicle can be used as an aerial base station or a relay node to provide temporary communication for the mobile terminal. In a mobile scene, the position of the mobile terminal is changed in real time, and the communication quality of the communication between the mobile terminal and the unmanned aerial vehicle is influenced by the position of the unmanned aerial vehicle, so that the planning of the position of the unmanned aerial vehicle is very important.
The specific process of planning the position of the unmanned aerial vehicle in the prior art is as follows:
firstly, the area needing the service of the unmanned aerial vehicle, the mobile terminal in the area and the position of the unmanned aerial vehicle at the current moment are obtained, the area needing the service is divided into a plurality of cells, after the unmanned aerial vehicle selects to move, the position of the unmanned aerial vehicle after moving and the average transmission rate of the mobile terminal are used as rewards, the next cell is determined according to the direction of the incremental rewards, and then the unmanned aerial vehicle moves to the central position of the next cell.
The above-described scheme has the following problems in implementation: when the cell division is less, the central point that unmanned aerial vehicle was located has probably to be far away from mobile terminal to lead to mobile terminal and unmanned aerial vehicle to carry out the line of sight way loss of communication higher, communication quality is lower. In order to make the unmanned aerial vehicle closer to these mobile terminals, the area needing service can be divided more carefully, but such division makes the time for the unmanned aerial vehicle to select the next cell increase, and the time for the mobile terminal to wait for the unmanned aerial vehicle to perform position planning is lengthened, thereby resulting in the reduction of the communication efficiency of the unmanned aerial vehicle and the mobile terminal.
Disclosure of Invention
The embodiment of the invention aims to provide a method and a device for planning the position of an unmanned aerial vehicle, which can improve the communication quality of the unmanned aerial vehicle and a mobile terminal under the condition of not reducing the communication efficiency of the unmanned aerial vehicle and the mobile terminal. The specific technical scheme is as follows:
in a first aspect, a method for planning a position of an unmanned aerial vehicle provided in an embodiment of the present invention includes:
acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served;
calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal;
the training process of the neural network model comprises the following steps: sequentially inputting position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the iteration times are reached;
and determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
Optionally, the step of calculating, based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal, the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through the trained neural network model includes:
inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain the output result of the trained neural network model, wherein the output result comprises: the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
constructing Gaussian distribution of the moving step length based on the variance of the moving step length and the mean value of the moving step length, and constructing Gaussian distribution of the moving direction based on the variance of the moving direction and the mean value of the moving direction;
and selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions, wherein the preset searching probability is obeyed.
Optionally, after the step of determining the position coordinate of the unmanned aerial vehicle at the next time based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability, and the moving step with the highest probability, the method for planning the position of the unmanned aerial vehicle provided in the embodiment of the first aspect of the present invention further includes:
and when the position coordinate of the unmanned aerial vehicle at the next moment is outside the service area, selecting the moving step with the second highest probability value from the Gaussian distribution of the moving step as the moving step of the unmanned aerial vehicle, and selecting the moving direction with the second highest probability value from the Gaussian distribution of the moving direction as the moving direction of the unmanned aerial vehicle.
Optionally, the trained neural network model is obtained by training through the following steps:
step one, obtaining a training set, wherein the training set comprises: the position coordinates of the mobile terminal and the position coordinates of the unmanned aerial vehicle at a plurality of historical moments;
inputting the position coordinates of the mobile terminal at the designated moment and the position coordinates of the unmanned aerial vehicle at the designated moment into a preset neural network model aiming at one designated moment in a plurality of historical moments to obtain the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
thirdly, constructing Gaussian distribution of the moving step length based on the mean value of the moving step length and the variance of the moving step length, and constructing Gaussian distribution of the moving direction based on the mean value of the moving direction and the variance of the moving direction;
step four, selecting the moving step with the highest probability value from the Gaussian distribution of the moving step, and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving direction;
step five, determining the moving direction with the highest probability and the target position after the unmanned aerial vehicle moves according to the moving step with the highest probability;
calculating the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the target position;
step seven, aiming at the specified time, taking the average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical time of the specified time as a training target, and repeatedly executing the step two to the step six until the iteration times are reached;
and step eight, determining the neural network model reaching the iteration times as a trained neural network model.
Optionally, the training set is obtained by:
forming a data set by the position coordinates of the mobile terminal and the unmanned aerial vehicle at each historical moment within a preset time before the current moment;
sampling the data set with a release function to obtain a training sample;
and forming the training samples into a training set.
Optionally, after the step of determining the position coordinate of the drone at the next time based on the first coordinate position of the drone, the moving direction with the highest probability, and the moving step with the highest probability, the method for planning the position of the drone provided in the embodiment of the first aspect of the present invention further includes:
when unmanned aerial vehicle is in the position coordinate department at next moment, when the access request number of receiving reaches preset upper limit value, the access request of receiving after reaching the upper limit value with unmanned aerial vehicle forwards other unmanned aerial vehicles, and other unmanned aerial vehicles are the unmanned aerial vehicle in unmanned aerial vehicle communication range.
In a second aspect, an apparatus for planning a position of an unmanned aerial vehicle provided in an embodiment of the present invention includes:
the acquisition module is used for acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served;
the calculation module is used for calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal;
the training process of the neural network model comprises the following steps: sequentially inputting position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the iteration times are reached;
and the determining module is used for determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
Optionally, the calculation module is specifically configured to:
inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain the output result of the trained neural network model, wherein the output result comprises: the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
constructing Gaussian distribution of the moving step length based on the variance of the moving step length and the mean value of the moving step length, and constructing Gaussian distribution of the moving direction based on the variance of the moving direction and the mean value of the moving direction;
and selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions, wherein the preset searching probability is obeyed.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a selection module to:
and when the position coordinate of the unmanned aerial vehicle at the next moment is outside the service area, selecting the moving step with the second highest probability value from the Gaussian distribution of the moving step as the moving step of the unmanned aerial vehicle, and selecting the moving direction with the second highest probability value from the Gaussian distribution of the moving direction as the moving direction of the unmanned aerial vehicle.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a training module to:
step one, obtaining a training set, wherein the training set comprises: the position coordinates of the mobile terminal and the position coordinates of the unmanned aerial vehicle at a plurality of historical moments;
inputting the position coordinates of the mobile terminal at the designated moment and the position coordinates of the unmanned aerial vehicle at the designated moment into a preset neural network model aiming at one designated moment in a plurality of historical moments to obtain the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
and thirdly, constructing Gaussian distribution of the moving step length based on the mean value of the moving step length and the variance of the moving step length, and constructing Gaussian distribution of the moving direction based on the mean value of the moving direction and the variance of the moving direction.
And step four, selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps, and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions.
Step five, determining the moving direction with the highest probability and the target position after the unmanned aerial vehicle moves according to the moving step with the highest probability;
calculating the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the target position;
step seven, aiming at the specified time, taking the average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical time of the specified time as a training target, and repeatedly executing the step two to the step six until the iteration times are reached;
and step eight, determining the neural network model reaching the iteration times as a trained neural network model.
Optionally, the training module is specifically configured to:
forming a data set by the position coordinates of the mobile terminal and the unmanned aerial vehicle at each historical moment within a preset time before the current moment;
sampling the data set with a release function to obtain a training sample;
and forming the training samples into a training set.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a forwarding module to:
when unmanned aerial vehicle is in the position coordinate department at next moment, when the access request number of receiving reaches preset upper limit value, the access request of receiving after reaching the upper limit value with unmanned aerial vehicle forwards other unmanned aerial vehicles, and other unmanned aerial vehicles are the unmanned aerial vehicle in unmanned aerial vehicle communication range.
The embodiment of the invention uses the trained neural network model to obtain the moving step length and the moving direction of the unmanned aerial vehicle with the highest probability, further determines that the position coordinate of the unmanned aerial vehicle at the next moment is irrelevant to the form of division of the area to be served and is not limited by the central position of the cell, the trained neural network model is obtained by taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target and taking the probability of each moving direction and the probability of each moving step length as the output of the neural network model for iterative training, wherein the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the position coordinate of the next moment is the maximum, therefore, the communication quality of the unmanned aerial vehicle and the mobile terminal can be improved under the condition that the communication efficiency of the unmanned aerial vehicle and the mobile terminal is not reduced.
In a third aspect, an embodiment of the present invention provides an unmanned aerial vehicle, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete mutual communication through the communication bus; the machine-readable storage medium stores machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: the method steps for planning the position of the unmanned aerial vehicle provided by the first aspect of the embodiment of the invention are realized.
In a fourth aspect, an embodiment of the present invention provides a controller, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete mutual communication through the communication bus; the machine-readable storage medium stores machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: the method steps for planning the position of the unmanned aerial vehicle provided by the first aspect of the embodiment of the invention are realized.
In a fifth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, where the computer program is executed by a processor to perform the method steps for planning the position of a drone according to the first aspect of the present invention.
In a sixth aspect, embodiments of the present invention further provide a computer program product containing instructions, which when run on a computer, cause the computer to perform the method steps for planning the position of a drone according to the first aspect of embodiments of the present invention.
The method and the device for planning the position of the unmanned aerial vehicle, provided by the embodiment of the invention, are used for acquiring the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal in a region to be served, calculating the probability of each moving direction and the probability of each moving step length of the unmanned aerial vehicle through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal, and determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability. Compared with the prior art, the embodiment of the invention uses the trained neural network model to obtain the moving step length and the moving direction with the highest probability of the unmanned aerial vehicle, further determines that the position coordinate of the unmanned aerial vehicle at the next moment is irrelevant to the form of division of the area to be served and is not limited by the center position of the cell, the trained neural network model is obtained by taking the maximum average transmission rate of the unmanned aerial vehicle at the position after the unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target and taking the probability of each moving direction and the probability of each moving step length as the output iteration training of the neural network model, and the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the position coordinate of the next moment is the maximum, so the embodiment of the invention can obtain the unmanned aerial vehicle and the mobile terminal with the highest probability without reducing the communication efficiency between the unmanned, improve unmanned aerial vehicle and mobile terminal's communication quality. Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a method for planning a position of an unmanned aerial vehicle according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an internal structure of a predetermined neural network model according to an embodiment of the present invention;
FIG. 3 is a flowchart of a process for implementing step S102 according to an embodiment of the present invention;
FIG. 4 is a flow chart of a process for obtaining a training set according to an embodiment of the present invention;
fig. 5 is a schematic diagram of positions of a plurality of drones and a plurality of mobile terminals according to an embodiment of the present invention;
fig. 6 is a structural diagram of an apparatus for planning the position of an unmanned aerial vehicle according to an embodiment of the present invention;
fig. 7 is a structural diagram of an unmanned aerial vehicle according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
First, a method for planning the position of an unmanned aerial vehicle according to an embodiment of the present invention will be described below.
The method for planning the position of the unmanned aerial vehicle provided by the embodiment of the invention can be applied to the unmanned aerial vehicle and can also be applied to a controller, and the controller can be communicated with the unmanned aerial vehicle and is used for controlling the unmanned aerial vehicle.
As shown in fig. 1, a method for planning a position of an unmanned aerial vehicle provided in an embodiment of the present invention includes:
s101, acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served.
Wherein the area to be served is an area in which a communication service needs to be provided. The service area can be a circular area formed by taking the initial position of the unmanned aerial vehicle as the circle center and taking the maximum distance in which the unmanned aerial vehicle can travel as the radius. When the area needing to provide the communication service is large, one unmanned aerial vehicle cannot provide the communication service; or when there are many mobile terminals in an area that needs to provide communication service and one unmanned aerial vehicle cannot provide communication service for all mobile terminals, the embodiment of the invention can use an unmanned aerial vehicle group. Each unmanned aerial vehicle in the unmanned aerial vehicle group is responsible for a part of the area needing to provide communication service or a part of the mobile terminals in the area to be served.
The current position coordinate of the unmanned aerial vehicle is a three-dimensional coordinate, the height of the unmanned aerial vehicle can be a preset fixed value, and the current position coordinate of the mobile terminal is a two-dimensional coordinate.
For example, if the unmanned aerial vehicle is required to provide communication service for tourists trapped in a scenic spot, an area where the tourists may appear can be used as a to-be-served area, if the unmanned aerial vehicle is required to provide communication service for a city where an earthquake occurs, a cell of the city where the earthquake occurs can be used as the to-be-served area, and the to-be-served area can be set in advance according to a specific scene.
It can be understood that the unmanned aerial vehicle model is different, and the maximum distance that unmanned aerial vehicle can travel is different, and the maximum distance that can travel is when unmanned aerial vehicle stops moving from beginning to move to the battery exhaustion, the distance of traveling.
And S102, calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal.
The training process of the neural network model comprises the following steps: the method comprises the steps of sequentially inputting position coordinates of a mobile terminal and an unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking an average transmission rate of a position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until iteration times are reached.
The current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal are input into the trained neural network model, and the output result of the trained neural network model comprises the probability of each moving direction and the probability of each moving step.
The current position coordinate is the position coordinate of the current moment, the current moment is the current time of the unmanned aerial vehicle system, and the difference time between the two moments is the time required by one iteration when a preset neural network model is trained. The iteration times are preset integer values, the range of the moving direction of the unmanned aerial vehicle is 0-360 degrees, and the moving step length refers to the distance moving within the preset time.
As shown in fig. 2, the preset neural network model may include: strategy sub-neural network and evaluation sub-neural network, StIndicates the state at time t, AtRepresenting the estimated values of the moving step length and the moving direction profit with the maximum probability of the sub-neural network for evaluating the sub-neural network to output the strategy executed by the unmanned aerial vehicle at the time t, atRepresenting the moving step length with the maximum probability and the moving direction r in the output result of the strategy sub-neural network at the time ttAnd the moving step with the maximum selection probability at the time t and the profit value of the moving direction are shown. The environment includes unmanned aerial vehicle position coordinates and mobile terminal's position coordinates. The strategy sub-neural network is used for outputting the probability of the moving direction of the unmanned aerial vehicle and the probability of the moving step length. Evaluating a sub-neural network for highest probability of outputThe moving step length and the moving direction are evaluated, namely the position of the unmanned aerial vehicle is determined according to the moving step length with the highest probability and the moving direction. The number of hidden layers in the strategy sub-neural network and the number of each hidden layer neuron are related to the size of the area to be served, and the larger the area to be served is, the larger the number of the hidden layers in the strategy sub-neural network and the number of each hidden layer neuron are.
It can be understood that the higher the average transmission rate of the unmanned aerial vehicle and the mobile terminal is, the higher the evaluation value representing the moving step length with the highest probability and the moving direction is, in the iterative training process, the evaluation sub-neural network feeds back the evaluation result to the strategy sub-neural network, and the strategy sub-neural network adjusts the internal parameters according to the increasing direction of the evaluation value until the iteration number is reached.
It can be understood that when there are a plurality of mobile terminals, the closer the distance between the unmanned aerial vehicle and the mobile terminal is, the larger the average transmission rate between the unmanned aerial vehicle and the mobile terminal is, and the better the communication quality between the unmanned aerial vehicle and the mobile terminal is. When each mobile terminal moves in real time, the moved mobile terminals are not gathered together, and a certain distance exists between part of the mobile terminals. The number of mobile terminals accessed by the unmanned aerial vehicle and the total transmission rate can be acquired, the average transmission rate is obtained through calculation, the positions of the unmanned aerial vehicle at the next historical moment and the maximization of the average transmission rate of the mobile terminals are used as the training targets of the preset neural network model, and the position coordinates of the unmanned aerial vehicle and the position coordinates of the mobile terminals are used as the input of the preset neural network model at the historical moment. During initial training, the maximum speed of the unmanned aerial vehicle, the range of the moving direction of the unmanned aerial vehicle and the time required by iteratively training a preset neural network model are preset. The time of iteratively training the neural network model once is used as the difference time between two continuous moments, and the maximum speed of the unmanned aerial vehicle is multiplied by the time required by one iteration, so that the upper limit value of the moving step length can be obtained. The trained neural network models have different inputs, and the moving step length and the moving direction obtained by the trained neural network models are different.
S103, determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
It can be understood that the unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability, and the moving time is the time of one iteration during the training of the neural network model, namely the difference time between the current time and the next time, so that the position of the unmanned aerial vehicle at the next time can be predicted.
The embodiment of the invention uses a trained neural network model to obtain the moving step length of the unmanned aerial vehicle with the highest probability and the moving direction with the highest probability, further determines that the position coordinate of the unmanned aerial vehicle at the next moment is irrelevant to the form of division of the area to be served and is not limited by the center position of the cell, and the trained neural network model takes the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target and takes the probability of each moving direction and the probability of each moving step length as the output of the neural network model for iterative training, so that the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the position coordinate at the next moment is the maximum, and the communication quality is higher. The embodiment of the invention can improve the communication quality of the unmanned aerial vehicle and the mobile terminal under the condition of not reducing the communication efficiency of the unmanned aerial vehicle and the mobile terminal.
As an implementation manner provided by the embodiment of the present invention, as shown in fig. 3, the step of S102 may be implemented by the following steps:
and S1021, inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain an output result of the trained neural network model.
Wherein, the output result includes: the mean value of the unmanned aerial vehicle moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length.
S1022, a gaussian distribution of the moving step is constructed based on the mean of the moving step and the variance of the moving step, and a gaussian distribution of the moving direction is constructed based on the mean of the moving direction and the variance of the moving direction.
The trained neural network model output layer is a classifier and used for outputting the variance of the moving step length, the mean of the moving direction and the variance of the moving direction, the variation trend of the mean and the variance of the moving step length meets Gaussian distribution, the mean and the variance of the moving direction also meet the Gaussian distribution, and the variance is used for measuring the dispersion degree of the Gaussian distribution.
And S1023, selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions according to the preset searching probability.
It can be understood that, in the gaussian distribution, the horizontal axis is the mean value, the vertical axis is the probability, the probability of the moving step corresponds to the mean value of the moving step at the maximum, the mean value is the moving step size of the unmanned aerial vehicle, the probability of the moving direction corresponds to the mean value of the moving direction at the maximum, and the mean value is the moving direction size of the unmanned aerial vehicle.
It can be understood that, in the embodiment, a greedy algorithm and a randomization algorithm are combined, a search probability is preset, a random number is generated, and when the random number is smaller than the preset search probability, according to the gaussian distribution of the moving step length and the gaussian distribution of the moving direction, the moving step length with the maximum probability and the moving direction with the maximum probability are selected as the action to be executed by the unmanned aerial vehicle; and if the generated random number is larger than the preset search probability, randomly selecting a moving step length and a moving direction as the action to be executed by the unmanned aerial vehicle. The method and the device improve the possibility of executing the action with low probability to a certain extent, and can avoid trapping into local optimum when the trained neural network model is used to obtain the moving direction and the moving step length with the maximum probability.
As an alternative embodiment of the present invention, the moving step with the highest probability value may be directly selected from the gaussian distribution of the moving steps, and the moving direction with the highest probability value may be selected from the gaussian distribution of the moving directions.
As an implementation manner provided by the embodiment of the present invention, before step S103, the method for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes:
and when the position coordinate of the unmanned aerial vehicle at the next moment is outside the service area, selecting the moving step with the second highest probability value from the Gaussian distribution of the moving step as the moving step of the unmanned aerial vehicle, and selecting the moving direction with the second highest probability value from the Gaussian distribution of the moving direction as the moving direction of the unmanned aerial vehicle.
It can be understood that the unmanned aerial vehicle can predict the position reached by the unmanned aerial vehicle after moving according to the maximum moving direction and the maximum moving step length of the probability value according to the current coordinate position, the maximum moving direction and the maximum moving step length of the probability value, and if the position is outside the region to be served, the unmanned aerial vehicle cannot reach the position. In order to save resources, the unmanned aerial vehicle gives up the moving direction and the moving step length, and then selects the moving step length and the moving direction with the second highest probability value, so that even if the position with the highest average transmission rate cannot be reached, the unmanned aerial vehicle is ensured to reach the position with the second highest average transmission rate, and the communication between the unmanned aerial vehicle and the mobile terminal can be continuously maintained.
As an implementation manner provided by the embodiment of the present invention, the neural network model trained in the step S102 is obtained by training through the following steps:
step one, a training set is obtained.
Wherein, the training set includes: the position coordinates of the mobile terminal at a plurality of historical moments and the position coordinates of the unmanned aerial vehicle.
And step two, inputting the position coordinates of the mobile terminal at the designated moment and the position coordinates of the unmanned aerial vehicle at the designated moment into a preset neural network model aiming at one designated moment in the plurality of historical moments to obtain the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle.
The designated time is any one of a plurality of historical times.
And thirdly, constructing Gaussian distribution of the moving step length based on the mean value of the moving step length and the variance of the moving step length, and constructing Gaussian distribution of the moving direction based on the mean value of the moving direction and the variance of the moving direction.
And step four, selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps, and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions.
And step five, determining the moving direction with the highest probability and the target position after the unmanned aerial vehicle moves according to the moving step with the highest probability.
And the target position is the position of the unmanned aerial vehicle after moving according to the moving direction with the highest probability and the moving step with the highest probability.
And step six, calculating the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the target position.
It can be understood that the unmanned aerial vehicle obtains the total number of the mobile terminals and the transmission rate of communication with each mobile terminal at the target position, and calculates the total transmission rate, and the ratio of the total transmission rate to the total number of the mobile terminals is the average transmission rate.
And step seven, aiming at the specified time, taking the maximum average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical time of the specified time as a training target, and repeatedly executing the step two to the step six until the iteration times are reached.
And step eight, determining the neural network model reaching the iteration times as a trained neural network model.
According to the embodiment, the average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical moment of the specified moment is taken as a training target, the position coordinate of the mobile terminal at the specified moment and the position coordinate of the unmanned aerial vehicle at the specified moment are taken as the input of a preset neural network model, the mean value of the moving direction of the unmanned aerial vehicle, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length are taken as the output of the neural network model, Gaussian distribution is constructed, the moving step length and the moving direction with the maximum probability are selected from the Gaussian distribution, the average transmission rate of communication between the unmanned aerial vehicle and the mobile terminal at the target position is calculated, and the accuracy of the trained neural network model obtained through iterative training can be improved.
As an implementation manner provided by the embodiment of the present invention, as shown in fig. 4, the training set may be obtained by training through the following steps:
s401, forming a data set by the position coordinates of the mobile terminal and the unmanned aerial vehicle at each historical moment in the preset time before the current moment.
S402, performing playback sampling on the data set to obtain a training sample.
The existing sample putting-back method comprises the steps that data set is subjected to existing sample putting-back sampling, the position coordinates of the unmanned aerial vehicle and the mobile terminal at a historical moment are randomly selected, the position coordinates serve as a training sample, then the training sample is put back into the data set, and the existing sample putting-back sampling is executed again until the number of the training samples reaches the preset number of samples.
And S403, forming the training samples into a training set.
According to the embodiment of the invention, the data set is subjected to the sample with the release, so that the balance degree of training samples in a training set can be improved, and a neural network model does not depend on a single training sample during training; meanwhile, when the training set is used for training the neural network model for the first time, parameters of the neural network model need to be initialized, and the initialized result of the parameters of the neural network model is related to the characteristics of the training samples, so that the efficiency of training the neural network model can be improved.
As an implementation manner provided by the embodiment of the present invention, after step S103, the method for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes:
when the number of the received access requests reaches a preset upper limit value at the position coordinate of the unmanned aerial vehicle at the next moment, the access requests received after the unmanned aerial vehicle reaches the upper limit value are forwarded to other unmanned aerial vehicles.
Wherein, other unmanned aerial vehicles are unmanned aerial vehicles within unmanned aerial vehicle communication range.
It can be understood that after the unmanned aerial vehicle reaches the position of the next moment after the current moment, the wireless network of the unmanned aerial vehicle can be opened, and the unmanned aerial vehicle can be accessed to the mobile terminal. The number of the mobile terminals accessed by the unmanned aerial vehicle is limited, when the access request received by the unmanned aerial vehicle reaches the upper limit, the unmanned aerial vehicle can not access the mobile terminals any more, and the access request received after reaching the upper limit can be forwarded to other unmanned aerial vehicles. Other unmanned aerial vehicles decide whether to access the mobile terminals or continue to forward the access requests to other unmanned aerial vehicles according to the distance between the mobile terminals sending the access requests and the unmanned aerial vehicles and the number of the access requests accessed by the unmanned aerial vehicles. If the mobile terminal sending the access request is in the network coverage range of the unmanned aerial vehicle, and the number of the unmanned aerial vehicle accessing the mobile terminal does not reach the preset upper limit value, the unmanned aerial vehicle accesses the mobile terminal sending the access request.
For example, as shown in fig. 5, a solid circle represents an unmanned aerial vehicle, a hollow diamond represents a mobile terminal, the circular area is an area to be served, the upper limit of the number of mobile terminals accessed by the unmanned aerial vehicle is 4, the number of mobile terminals connected by the unmanned aerial vehicle a is 4, the unmanned aerial vehicle a forwards an access request of the mobile terminal a to the unmanned aerial vehicle B, the number of mobile terminals accessed by the unmanned aerial vehicle B is 3, and the mobile terminal a is within the network coverage range of the unmanned aerial vehicle B, so that the unmanned aerial vehicle B accesses the mobile terminal.
As an optional implementation manner of the present invention, a connection line of the position coordinates of the unmanned aerial vehicle at each time may be used as the trajectory of the unmanned aerial vehicle, so as to implement trajectory planning of the unmanned aerial vehicle, and provide a cushion for planning the position of the unmanned aerial vehicle in the area to be served next time.
As shown in fig. 6, an apparatus for planning a position of an unmanned aerial vehicle according to an embodiment of the present invention includes:
an obtaining module 601, configured to obtain a current position coordinate of the unmanned aerial vehicle and a current position coordinate of the mobile terminal in the area to be served;
a calculating module 602, configured to calculate, based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal, probabilities of the unmanned aerial vehicle in each moving direction and the probabilities of each moving step through a trained neural network model;
the training process of the neural network model comprises the following steps: sequentially inputting position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the iteration times are reached;
the determining module 603 is configured to determine a position coordinate of the drone at the next time based on the current coordinate position of the drone, the moving direction with the highest probability, and the moving step with the highest probability.
Optionally, the calculation module is specifically configured to:
inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain the output result of the trained neural network model, wherein the output result comprises: the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
constructing Gaussian distribution of the moving step length based on the variance of the moving step length and the mean value of the moving step length, and constructing Gaussian distribution of the moving direction based on the variance of the moving direction and the mean value of the moving direction;
and selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions, wherein the preset searching probability is obeyed.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a selection module to:
and when the position coordinate of the unmanned aerial vehicle at the next moment is outside the service area, selecting the moving step with the second highest probability value from the Gaussian distribution of the moving step as the moving step of the unmanned aerial vehicle, and selecting the moving direction with the second highest probability value from the Gaussian distribution of the moving direction as the moving direction of the unmanned aerial vehicle.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a training module to:
step one, obtaining a training set, wherein the training set comprises: the position coordinates of the mobile terminal and the position coordinates of the unmanned aerial vehicle at a plurality of historical moments;
inputting the position coordinates of the mobile terminal at the designated moment and the position coordinates of the unmanned aerial vehicle at the designated moment into a preset neural network model aiming at one designated moment in a plurality of historical moments to obtain the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
and thirdly, constructing Gaussian distribution of the moving step length based on the mean value of the moving step length and the variance of the moving step length, and constructing Gaussian distribution of the moving direction based on the mean value of the moving direction and the variance of the moving direction.
And step four, selecting the moving step with the highest probability value from the Gaussian distribution of the moving steps, and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving directions.
Step five, determining the moving direction with the highest probability and the target position after the unmanned aerial vehicle moves according to the moving step with the highest probability;
calculating the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the target position;
step seven, aiming at the specified time, taking the average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical time of the specified time as a training target, and repeatedly executing the step two to the step six until the iteration times are reached;
and step eight, determining the neural network model reaching the iteration times as a trained neural network model.
Optionally, the training module is specifically configured to:
forming a data set by the position coordinates of the mobile terminal and the unmanned aerial vehicle at each historical moment within a preset time before the current moment;
sampling the data set with a release function to obtain a training sample;
and forming the training samples into a training set.
Optionally, the apparatus for planning the position of the unmanned aerial vehicle provided by the embodiment of the present invention further includes: a forwarding module to:
when unmanned aerial vehicle is in the position coordinate department at next moment, when the access request number of receiving reaches preset upper limit value, the access request of receiving after reaching the upper limit value with unmanned aerial vehicle forwards other unmanned aerial vehicles, and other unmanned aerial vehicles are the unmanned aerial vehicle in unmanned aerial vehicle communication range.
The embodiment of the invention uses the trained neural network model to obtain the moving step length and the moving direction of the unmanned aerial vehicle with the highest probability, further determines that the position coordinate of the unmanned aerial vehicle at the next moment is irrelevant to the form of division of the area to be served and is not limited by the central position of the cell, the trained neural network model is obtained by taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target and taking the probability of each moving direction and the probability of each moving step length as the output of the neural network model for iterative training, wherein the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the position coordinate of the next moment is the maximum, therefore, the communication quality of the unmanned aerial vehicle and the mobile terminal can be improved under the condition that the communication efficiency of the unmanned aerial vehicle and the mobile terminal is not reduced.
An embodiment of the present invention further provides an unmanned aerial vehicle, as shown in fig. 7, including a processor 701, a communication interface 702, a memory 703 and a communication bus 704, where the processor 701, the communication interface 702, and the memory 703 complete mutual communication through the communication bus 704,
a memory 703 for storing a computer program;
the processor 701 is configured to implement the following steps when executing the program stored in the memory 703:
acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served;
calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal;
the training process of the neural network model comprises the following steps: sequentially inputting position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate of the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the iteration times are reached;
and determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
The communication bus mentioned in the above-mentioned unmanned aerial vehicle may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
Communication interface is used for the communication between above-mentioned unmanned aerial vehicle and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
In a further embodiment provided by the present invention, there is also provided a computer readable storage medium having stored therein a computer program which, when executed by a processor, performs the steps of any one of the above methods of planning the position of a drone.
In a further embodiment provided by the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform any one of the above-described embodiments of the method of planning the position of a drone.
The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, e.g., from one website site, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DS L)) or wireless (e.g., infrared, wireless, microwave, etc.) manner to another website site, computer, server, or data center.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method of planning a position of a drone, the method comprising:
acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served;
calculating the probability of each moving direction of the unmanned aerial vehicle and the probability of each moving step length through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal;
wherein the training process of the neural network model comprises the following steps: sequentially inputting the position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate at the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the number of iterations is reached;
and determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
2. The method of claim 1, wherein the step of calculating the probability of each moving direction and the probability of each moving step of the unmanned aerial vehicle through a trained neural network model based on the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal comprises:
inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain an output result of the trained neural network model, wherein the output result comprises: the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
constructing a Gaussian distribution of the moving step length based on the variance of the moving step length and the mean of the moving step length, and constructing a Gaussian distribution of the moving direction based on the variance of the moving direction and the mean of the moving direction;
and selecting a moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting a moving direction with the highest probability value from the Gaussian distribution of the moving directions, wherein the preset searching probability is obeyed.
3. The method of claim 2, wherein after the step of determining the location coordinates for the drone at the next time based on the current coordinate location of the drone, the most probable direction of movement, and the most probable step size of movement, the method further comprises:
and when the position coordinate of the unmanned aerial vehicle at the next moment is outside the service area, selecting the moving step with the second highest probability value from the Gaussian distribution of the moving step as the moving step of the unmanned aerial vehicle, and selecting the moving direction with the second highest probability value from the Gaussian distribution of the moving direction as the moving direction of the unmanned aerial vehicle.
4. The method according to any one of claims 1-3, wherein the trained neural network model is trained by:
step one, obtaining a training set, wherein the training set comprises: the position coordinates of the mobile terminal and the position coordinates of the unmanned aerial vehicle at a plurality of historical moments;
inputting the position coordinates of the mobile terminal at the designated moment and the position coordinates of the unmanned aerial vehicle at the designated moment into a preset neural network model aiming at one designated moment in a plurality of historical moments to obtain the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
thirdly, constructing Gaussian distribution of the moving step length based on the mean value of the moving step length and the variance of the moving step length, and constructing Gaussian distribution of the moving direction based on the mean value of the moving direction and the variance of the moving direction;
step four, selecting the moving step with the highest probability value from the Gaussian distribution of the moving step, and selecting the moving direction with the highest probability value from the Gaussian distribution of the moving direction;
step five, determining the moving direction with the highest probability and the target position after the unmanned aerial vehicle moves according to the moving step with the highest probability;
step six, calculating the average transmission rate of the communication between the unmanned aerial vehicle and the mobile terminal at the target position;
step seven, aiming at the specified time, taking the average transmission rate of the unmanned aerial vehicle and the mobile terminal at the next historical time of the specified time as a training target, and repeatedly executing the step two to the step six until the iteration times are reached;
and step eight, determining the neural network model reaching the iteration times as a trained neural network model.
5. The method of claim 4, wherein the training set is obtained by:
forming a data set by the position coordinates of the mobile terminal and the unmanned aerial vehicle at each historical moment within a preset time before the current moment;
sampling the data set with a release function to obtain a training sample;
and combining the training samples into a training set.
6. The method of any of claims 1-3, wherein after the step of determining the position coordinates of the drone at the next time based on the first coordinate position of the drone, the most probable direction of movement, and the most probable step size of movement, the method further comprises:
when the unmanned aerial vehicle is in the position coordinate department at next moment, when the number of the received access requests reaches the preset upper limit value, the access requests received after the unmanned aerial vehicle reaches the upper limit value are forwarded to other unmanned aerial vehicles, and the other unmanned aerial vehicles are unmanned aerial vehicles in the unmanned aerial vehicle communication range.
7. An apparatus for planning a position of a drone, the apparatus comprising:
the acquisition module is used for acquiring the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal in the area to be served;
the calculation module is used for calculating and obtaining the probability of each moving direction and the probability of each moving step length of the unmanned aerial vehicle through a trained neural network model based on the current position coordinate of the unmanned aerial vehicle and the current position coordinate of the mobile terminal;
wherein the training process of the neural network model comprises the following steps: sequentially inputting the position coordinates of the mobile terminal and the unmanned aerial vehicle at a plurality of historical moments into a preset neural network model, selecting a moving step length with the highest probability and a moving direction with the highest probability from output results of the preset neural network model, taking the average transmission rate at the position where the maximized unmanned aerial vehicle moves according to the moving step length with the highest probability and the moving direction with the highest probability as a training target, and training the preset neural network model until the number of iterations is reached;
and the determining module is used for determining the position coordinate of the unmanned aerial vehicle at the next moment based on the current coordinate position of the unmanned aerial vehicle, the moving direction with the highest probability and the moving step length with the highest probability.
8. The apparatus of claim 7, wherein the computing module is specifically configured to:
inputting the current position coordinates of the unmanned aerial vehicle and the current position coordinates of the mobile terminal into the trained neural network model to obtain an output result of the trained neural network model, wherein the output result comprises: the mean value of the moving direction, the variance of the moving direction, the mean value of the moving step length and the variance of the moving step length of the unmanned aerial vehicle;
constructing a Gaussian distribution of the moving step length based on the variance of the moving step length and the mean of the moving step length, and constructing a Gaussian distribution of the moving direction based on the variance of the moving direction and the mean of the moving direction;
and selecting a moving step with the highest probability value from the Gaussian distribution of the moving steps and selecting a moving direction with the highest probability value from the Gaussian distribution of the moving directions, wherein the preset searching probability is obeyed.
9. An unmanned aerial vehicle is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for finishing mutual communication through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1-6 when executing a program stored in the memory.
10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010258490.0A CN111506104B (en) | 2020-04-03 | 2020-04-03 | Method and device for planning position of unmanned aerial vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010258490.0A CN111506104B (en) | 2020-04-03 | 2020-04-03 | Method and device for planning position of unmanned aerial vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111506104A true CN111506104A (en) | 2020-08-07 |
| CN111506104B CN111506104B (en) | 2021-10-01 |
Family
ID=71874139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010258490.0A Active CN111506104B (en) | 2020-04-03 | 2020-04-03 | Method and device for planning position of unmanned aerial vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111506104B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113543066A (en) * | 2021-06-07 | 2021-10-22 | 北京邮电大学 | Sensory-guidance integrated interaction and multi-target emergency networking method and system |
| WO2022033390A1 (en) * | 2020-08-12 | 2022-02-17 | 华为技术有限公司 | Position acquisition method and apparatus thereof |
Citations (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472850A (en) * | 2013-09-29 | 2013-12-25 | 合肥工业大学 | Multi-unmanned aerial vehicle collaborative search method based on Gaussian distribution prediction |
| CN105549597A (en) * | 2016-02-04 | 2016-05-04 | 同济大学 | Unmanned vehicle dynamic path programming method based on environment uncertainty |
| US20170213070A1 (en) * | 2016-01-22 | 2017-07-27 | Qualcomm Incorporated | Object-focused active three-dimensional reconstruction |
| CN107065929A (en) * | 2017-05-05 | 2017-08-18 | 成都通甲优博科技有限责任公司 | A kind of unmanned plane is around flying method and system |
| CN107168324A (en) * | 2017-06-08 | 2017-09-15 | 中国矿业大学 | A kind of robot path planning method based on ANFIS fuzzy neural networks |
| CN107229968A (en) * | 2017-05-24 | 2017-10-03 | 北京小米移动软件有限公司 | Gradient parameter determines method, device and computer-readable recording medium |
| CN105137967B (en) * | 2015-07-16 | 2018-01-19 | 北京工业大学 | The method for planning path for mobile robot that a kind of depth autocoder is combined with Q learning algorithms |
| WO2018110963A1 (en) * | 2016-12-16 | 2018-06-21 | Samsung Electronics Co., Ltd. | Movable object and method for controlling the same |
| CN108803667A (en) * | 2018-05-30 | 2018-11-13 | 北京邮电大学 | A kind of unmanned plane synergic monitoring and tracking |
| CN108880664A (en) * | 2018-07-23 | 2018-11-23 | 北京邮电大学 | A kind of disaster assistance information processing method |
| WO2018214706A1 (en) * | 2017-05-22 | 2018-11-29 | 腾讯科技（深圳）有限公司 | Movement control method, storage medium, computer apparatus, and service robot |
| CN109032168A (en) * | 2018-05-07 | 2018-12-18 | 西安电子科技大学 | A kind of Route planner of the multiple no-manned plane Cooperative Area monitoring based on DQN |
| CN109213712A (en) * | 2018-09-06 | 2019-01-15 | 北京邮电大学 | For the service providing method of machine type communication system, device and electronic equipment |
| CN109269507A (en) * | 2018-11-26 | 2019-01-25 | 张家港江苏科技大学产业技术研究院 | Robot path planning method and device |
| US20190065933A1 (en) * | 2017-08-31 | 2019-02-28 | Ford Global Technologies, Llc | Augmenting Real Sensor Recordings With Simulated Sensor Data |
| CN109782806A (en) * | 2019-02-19 | 2019-05-21 | 北京邮电大学 | Path following method and device in a kind of unmanned plane room |
| CN109839956A (en) * | 2019-03-04 | 2019-06-04 | 北京邮电大学 | A kind of paths planning method and device of unmanned plane |
| CN109990790A (en) * | 2019-03-29 | 2019-07-09 | 北京邮电大学 | A kind of unmanned plane paths planning method and device |
| CN110083166A (en) * | 2019-05-30 | 2019-08-02 | 浙江远传信息技术股份有限公司 | For the coordinated dispatching method of multirobot, device, equipment and medium |
| CN110096051A (en) * | 2018-01-31 | 2019-08-06 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating vehicle control instruction |
| CN110389591A (en) * | 2019-08-29 | 2019-10-29 | 哈尔滨工程大学 | A kind of paths planning method based on DBQ algorithm |
| CN110428115A (en) * | 2019-08-13 | 2019-11-08 | 南京理工大学 | Maximization system benefit method under dynamic environment based on deeply study |
| CN110658852A (en) * | 2019-09-16 | 2020-01-07 | 苏州米龙信息科技有限公司 | Intelligent target searching method and system for unmanned aerial vehicle |
| CN110794842A (en) * | 2019-11-15 | 2020-02-14 | 北京邮电大学 | Reinforced learning path planning algorithm based on potential field |
| CN110809274A (en) * | 2019-10-28 | 2020-02-18 | 南京邮电大学 | Narrowband Internet of things-oriented unmanned aerial vehicle base station enhanced network optimization method |
| CN110857085A (en) * | 2018-08-17 | 2020-03-03 | 福特全球技术公司 | Vehicle path planning |
-
2020
- 2020-04-03 CN CN202010258490.0A patent/CN111506104B/en active Active
Patent Citations (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472850A (en) * | 2013-09-29 | 2013-12-25 | 合肥工业大学 | Multi-unmanned aerial vehicle collaborative search method based on Gaussian distribution prediction |
| CN105137967B (en) * | 2015-07-16 | 2018-01-19 | 北京工业大学 | The method for planning path for mobile robot that a kind of depth autocoder is combined with Q learning algorithms |
| US20170213070A1 (en) * | 2016-01-22 | 2017-07-27 | Qualcomm Incorporated | Object-focused active three-dimensional reconstruction |
| CN105549597A (en) * | 2016-02-04 | 2016-05-04 | 同济大学 | Unmanned vehicle dynamic path programming method based on environment uncertainty |
| WO2018110963A1 (en) * | 2016-12-16 | 2018-06-21 | Samsung Electronics Co., Ltd. | Movable object and method for controlling the same |
| CN107065929A (en) * | 2017-05-05 | 2017-08-18 | 成都通甲优博科技有限责任公司 | A kind of unmanned plane is around flying method and system |
| WO2018214706A1 (en) * | 2017-05-22 | 2018-11-29 | 腾讯科技（深圳）有限公司 | Movement control method, storage medium, computer apparatus, and service robot |
| CN107229968A (en) * | 2017-05-24 | 2017-10-03 | 北京小米移动软件有限公司 | Gradient parameter determines method, device and computer-readable recording medium |
| CN107168324A (en) * | 2017-06-08 | 2017-09-15 | 中国矿业大学 | A kind of robot path planning method based on ANFIS fuzzy neural networks |
| US20190065933A1 (en) * | 2017-08-31 | 2019-02-28 | Ford Global Technologies, Llc | Augmenting Real Sensor Recordings With Simulated Sensor Data |
| CN110096051A (en) * | 2018-01-31 | 2019-08-06 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating vehicle control instruction |
| CN109032168A (en) * | 2018-05-07 | 2018-12-18 | 西安电子科技大学 | A kind of Route planner of the multiple no-manned plane Cooperative Area monitoring based on DQN |
| CN108803667A (en) * | 2018-05-30 | 2018-11-13 | 北京邮电大学 | A kind of unmanned plane synergic monitoring and tracking |
| CN108880664A (en) * | 2018-07-23 | 2018-11-23 | 北京邮电大学 | A kind of disaster assistance information processing method |
| CN110857085A (en) * | 2018-08-17 | 2020-03-03 | 福特全球技术公司 | Vehicle path planning |
| CN109213712A (en) * | 2018-09-06 | 2019-01-15 | 北京邮电大学 | For the service providing method of machine type communication system, device and electronic equipment |
| CN109269507A (en) * | 2018-11-26 | 2019-01-25 | 张家港江苏科技大学产业技术研究院 | Robot path planning method and device |
| CN109782806A (en) * | 2019-02-19 | 2019-05-21 | 北京邮电大学 | Path following method and device in a kind of unmanned plane room |
| CN109839956A (en) * | 2019-03-04 | 2019-06-04 | 北京邮电大学 | A kind of paths planning method and device of unmanned plane |
| CN109990790A (en) * | 2019-03-29 | 2019-07-09 | 北京邮电大学 | A kind of unmanned plane paths planning method and device |
| CN110083166A (en) * | 2019-05-30 | 2019-08-02 | 浙江远传信息技术股份有限公司 | For the coordinated dispatching method of multirobot, device, equipment and medium |
| CN110428115A (en) * | 2019-08-13 | 2019-11-08 | 南京理工大学 | Maximization system benefit method under dynamic environment based on deeply study |
| CN110389591A (en) * | 2019-08-29 | 2019-10-29 | 哈尔滨工程大学 | A kind of paths planning method based on DBQ algorithm |
| CN110658852A (en) * | 2019-09-16 | 2020-01-07 | 苏州米龙信息科技有限公司 | Intelligent target searching method and system for unmanned aerial vehicle |
| CN110809274A (en) * | 2019-10-28 | 2020-02-18 | 南京邮电大学 | Narrowband Internet of things-oriented unmanned aerial vehicle base station enhanced network optimization method |
| CN110794842A (en) * | 2019-11-15 | 2020-02-14 | 北京邮电大学 | Reinforced learning path planning algorithm based on potential field |
Non-Patent Citations (5)
| Title |
|---|
| QIANG WANG ETAL.: "Multi-UAV Dynamic Wireless Networking With Deep Reinforcement Learning", 《IEEE COMMUNICATIONS LETTERS》 * |
| YUNPENG XU ETAL.: "A Brief Review of the Intelligent Algorithm for Traveling Salesman Problem in UAV Route Planning", 《2019 IEEE 9TH INTERNATIONAL CONFERENCE ON ELECTRONICS INFORMATION AND EMERGENCY COMMUNICATION (ICEIEC)》 * |
| 刘杰等: "基于三维空间的无人机规避攻击导引设计", 《电光与控制》 * |
| 李豆豆: "多源航空监视信息融合技术的研究与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 杨翰章: "无人机基站蜂窝网络下的部署方法及移动切换技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022033390A1 (en) * | 2020-08-12 | 2022-02-17 | 华为技术有限公司 | Position acquisition method and apparatus thereof |
| CN113543066A (en) * | 2021-06-07 | 2021-10-22 | 北京邮电大学 | Sensory-guidance integrated interaction and multi-target emergency networking method and system |
| CN113543066B (en) * | 2021-06-07 | 2023-11-03 | 北京邮电大学 | Integrated interaction and multi-target emergency networking method and system for sensing communication guide finger |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111506104B (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102030128B1 (en) | Resource allocating method for wireless backhaul network and apparatus based on machine learning | |
| CN111506104B (en) | Method and device for planning position of unmanned aerial vehicle | |
| CN111461410B (en) | Air quality prediction method and device based on transfer learning | |
| WO2016165459A1 (en) | Method and device for indoor positioning | |
| CN111352417A (en) | Rapid generation method of heterogeneous multi-unmanned aerial vehicle cooperative path | |
| CN112020103A (en) | Content cache deployment method in mobile edge cloud | |
| CN110582072B (en) | Fuzzy matching-based resource allocation method and device in cellular internet of vehicles | |
| CN112469060B (en) | Antenna parameter determining method and device | |
| CN107949025A (en) | A kind of network selecting method based on non-cooperative game | |
| CN112947591A (en) | Path planning method, device, medium and unmanned aerial vehicle based on improved ant colony algorithm | |
| CN113687875B (en) | Method and device for unloading vehicle tasks in Internet of vehicles | |
| CN111638511A (en) | Signal fusion-based multi-radar space registration cooperative detection method and device | |
| CN107786989B (en) | Lora intelligent water meter network gateway deployment method and device | |
| EP3308190A1 (en) | Determining of model parameters for positioning purposes | |
| CN112882723A (en) | Edge service deployment method facing parallel micro-service combination | |
| CN104039008A (en) | Hybrid positioning method | |
| JP2004193912A (en) | Ray receive judging method, system and electric wave propagation characteristics estimating method using the same | |
| CN115866626B (en) | Service deployment method based on NSGA-II and simulated annealing in edge environment | |
| CN116827846A (en) | Routing, routing path, multicast routing path decision method and electronic equipment | |
| CN111465057A (en) | Edge caching method and device based on reinforcement learning and electronic equipment | |
| CN110890932A (en) | ESN model-based 5G millimeter wave channel information determination method and device | |
| CN112738883B (en) | Method and device for determining position of air base station | |
| CN114783215A (en) | Unmanned aerial vehicle clustering method and device and electronic equipment | |
| CN117119585B (en) | Bluetooth positioning navigation system and method | |
| CN112512115A (en) | Method and device for determining position of air base station and electronic equipment |
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 |