CN114374951B - Dynamic pre-deployment method for multiple unmanned aerial vehicles - Google Patents
Dynamic pre-deployment method for multiple unmanned aerial vehicles Download PDFInfo
- Publication number
- CN114374951B CN114374951B CN202210046566.2A CN202210046566A CN114374951B CN 114374951 B CN114374951 B CN 114374951B CN 202210046566 A CN202210046566 A CN 202210046566A CN 114374951 B CN114374951 B CN 114374951B
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- vehicle
- sequence
- vehicles
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000004891 communication Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000003062 neural network model Methods 0.000 claims abstract description 14
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 13
- 230000000694 effects Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000006855 networking Effects 0.000 abstract description 4
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/18—Network planning tools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention relates to a dynamic pre-deployment method for multiple unmanned aerial vehicles, which belongs to the field of unmanned aerial vehicle auxiliary communication, and the method adopts a distributed networking mode to establish communication between unmanned aerial vehicles and ground vehicles; training a sequence-to-sequence gating cyclic neural network model by adopting a distributed federal learning training method; the method comprises the steps of adopting a time slot dividing mode, inputting the position of each time slot of a vehicle into a sequence-to-sequence gating cyclic neural network model to predict the future movement trend of the vehicle; and a virtual force deployment algorithm is adopted to realize the determination of the deployment position of the unmanned aerial vehicle and the switching of vehicle signals. The method can fully consider the movement characteristics of the vehicle, predict the future information of the vehicle, and deploy the unmanned aerial vehicle at a reasonable position so as to ensure that the vehicle is accessed to the unmanned aerial vehicle base station.
Description
Technical Field
The invention belongs to the field of unmanned aerial vehicle auxiliary communication, and relates to a multi-unmanned aerial vehicle dynamic pre-deployment method.
Background
During the past 20 years, the unmanned aerial vehicle industry is developing with the high speed of wave tide of science and technology. In the development of the fifth generation of mobile communication, a plurality of application scenes are not satisfied with the modes of a fixed base station and a fixed access network, such as disaster quick rescue, and urban vehicles are flexibly and quickly accessed. The unmanned aerial vehicle is used as an air base station to provide a scheme of a wireless access network for ground mobile users and vehicles, and is gradually popularized by various communication manufacturers.
The unmanned aerial vehicle is deployed in the high altitude, so that the unmanned aerial vehicle is not influenced by geographical environment factors, and more flexible wireless base station coverage requirements can be provided for ground users. However, the ability of single drones to provide base station services is limited, so multi-drone distributed networking has become a mainstream to provide communication services.
The existing unmanned aerial vehicle provides base station services only to be fixed at a certain position in the air or provided according to the planned flight track in advance. The position of the ground vehicle may change over time and this trend is random and difficult to characterize. When the position of the ground vehicle changes, the fixed access networking mode or the service provision according to the track planned in advance is unable to meet the randomness of the movement. Therefore, there is a need for an unmanned aerial vehicle deployment networking scheme that can sense the movement trend of ground vehicles in real time and provide flexibility as required.
Disclosure of Invention
In view of the above, the present invention aims to provide a dynamic pre-deployment method for multiple unmanned aerial vehicles considering movement trend, which adopts a neural network training method to learn movement characteristics of each vehicle and acquire a neural network model capable of predicting future position information, and deploys the model on the centroid position of each responsible vehicle by predicting the future position information of each vehicle, so that each vehicle can evaluate the signal quality of the multiple unmanned aerial vehicles and select to access to the unmanned aerial vehicle with the strongest signal quality.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a dynamic pre-deployment method of a multi-unmanned aerial vehicle comprises the following steps:
s1: the vehicle is connected with each unmanned aerial vehicle in the area through a communication transmitting device to establish a communication link;
S2: the vehicle uploads the positioning information of the vehicle to each unmanned aerial vehicle through an uplink;
s3: constructing a distributed communication network among unmanned aerial vehicles through communication links, and distributing a sequence-to-sequence gating cyclic neural network model and training rounds;
S4: each unmanned aerial vehicle trains a local sequence-to-sequence gating circulating neural network model through an edge computing server by utilizing a plurality of pieces of vehicle positioning information collected in the area;
s5: each unmanned aerial vehicle performs election according to the data volume participating in training to obtain a leader and a common group;
s6: the common group unmanned aerial vehicle uploads the locally trained sequence-to-sequence gating cyclic neural network model parameters to an unmanned aerial vehicle leader;
s7: the unmanned aerial vehicle leader updates model parameters uploaded by unmanned aerial vehicles of all ordinary groups on a local edge calculation server by using a federal average method, updates model parameters of a sequence-to-sequence gated cyclic neural network and training rounds and transmits the model parameters and training rounds to the unmanned aerial vehicles of all ordinary groups;
S8: each unmanned aerial vehicle judges whether the current training round reaches the target round, if so, the step S9 is executed, otherwise, the steps S4-S7 are executed;
S9: each unmanned aerial vehicle is provided with time slots with equal intervals, and vehicles in the unmanned aerial vehicle area transmit own positioning information to the unmanned aerial vehicle through a communication link when each time slot starts;
s10: each unmanned aerial vehicle inputs the positioning information of the vehicle in the step S9 into a sequence-to-sequence cyclic neural network model, and predicts the future position of the vehicle in the next time slot;
S11: each unmanned aerial vehicle calculates the position with zero total force by using a virtual force guide deployment algorithm and flies to the position with zero total force;
s12: the vehicle sorts the signal quality intensity of the locally received unmanned aerial vehicle, if the maximum signal quality is smaller than the set signal quality threshold value, the vehicle selects to give up to be connected to the unmanned aerial vehicle base station, otherwise, the vehicle selects to be connected to the unmanned aerial vehicle base station with the maximum signal quality.
Further, the federal averaging method in step S7 specifically includes:
Where N represents the total number of unmanned aerial vehicles, w i (k) represents the model parameters of the ith unmanned aerial vehicle, D i represents the data size of the ith unmanned aerial vehicle, D represents the total data size, and w (k) is the federally averaged model.
Further, the virtual force guidance algorithm described in step S11 includes the steps of:
s111: each unmanned aerial vehicle utilizes the predicted position information of S10 to construct the position distribution information of vehicles in the whole unmanned aerial vehicle group;
s112: each unmanned aerial vehicle calculates the mass center position of the vehicle position distribution according to the vehicle predicted position in the area;
S113: the centroid position generates a virtual gravitation effect on the unmanned aerial vehicle, and the unmanned aerial vehicle is towed to the position;
S114: each unmanned aerial vehicle updates its own energy information; if the energy is smaller than or equal to the energy threshold of the unmanned aerial vehicle, the unmanned aerial vehicle flies back to the charging center, the vehicle connected to the unmanned aerial vehicle is disconnected with the unmanned aerial vehicle, the signal quality intensity of the nearby unmanned aerial vehicle base station is searched, and the unmanned aerial vehicle base station with the largest signal quality intensity is selected to be connected.
Further, the signal quality in step S12 is the signal-to-interference-and-noise ratio of the communication, which is defined as follows:
Where p m,k is the transmit power of drone m to vehicle k, g m,k is the power gain between drone m and ground vehicle k, And N p is the Gaussian white noise power of the environment, which represents the sum of the products of the power and the power gain of N-1 unmanned aerial vehicles except the unmanned aerial vehicle m.
The invention has the beneficial effects that: according to the invention, by introducing a distributed federal learning training method, training of the sequence-to-sequence gated cyclic neural network model of each unmanned aerial vehicle is accelerated, the position privacy of a user is protected, and meanwhile, an unmanned aerial vehicle leader is elected to perform a model parameter aggregation task according to the data size, so that model training termination caused by attack of a single aggregation node can be effectively prevented. According to the scheme, the moving characteristic of the vehicle can be fully considered, the future information of the vehicle can be predicted, and the unmanned aerial vehicle is deployed at a reasonable position, so that the vehicle is ensured to be connected to the unmanned aerial vehicle base station.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and other advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the specification.
Drawings
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in the following preferred detail with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart that considers dynamic pre-deployment of a drone;
FIG. 2 is a flow chart of virtual force deployment for a drone;
FIG. 3 is a diagram of a communication link established between a drone and a vehicle;
Fig. 4 is a diagram of the centroid position of the drone in the vehicle distribution for the virtual force deployment algorithm.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration, and the following embodiments and features in the embodiments may be combined with each other without conflict.
Wherein the drawings are for illustrative purposes only and are shown in schematic, non-physical, and not intended to limit the invention; for the purpose of better illustrating embodiments of the invention, certain elements of the drawings may be omitted, enlarged or reduced and do not represent the size of the actual product; it will be appreciated by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numbers in the drawings of embodiments of the invention correspond to the same or similar components; in the description of the present invention, it should be understood that, if there are terms such as "upper", "lower", "left", "right", "front", "rear", etc., that indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, it is only for convenience of describing the present invention and simplifying the description, but not for indicating or suggesting that the referred device or element must have a specific azimuth, be constructed and operated in a specific azimuth, so that the terms describing the positional relationship in the drawings are merely for exemplary illustration and should not be construed as limiting the present invention, and that the specific meaning of the above terms may be understood by those of ordinary skill in the art according to the specific circumstances.
As shown in fig. 1, the unmanned aerial vehicle dynamic pre-deployment includes three part processes. The first part is S1-S3 of the procedure of establishing a communication link, which corresponds to the access scheme, and is specifically shown in fig. 3: the unmanned aerial vehicle comprises N unmanned aerial vehicles, and the bottom layer of each unmanned aerial vehicle is connected with different vehicles.
The second part is a sequence-to-sequence gated cyclic neural network model training process, adopts a distributed federal learning method, and corresponds to S4-S8 of an access scheme. The method comprises the following specific steps:
1) Vehicle access unmanned aerial vehicle positioning information Tk,n＝[(x1,y1),(x2,y2),…(xi,yi),…(xn,yn)],
Wherein (x i,yi) represents longitude and latitude data of the ith moment of the vehicle k, T k,n represents longitude and latitude data sequences of the k n moments of the vehicle,
Uploading the data to an accessed unmanned aerial vehicle through an uplink communication link;
2) The unmanned aerial vehicle locally updates the local model parameters by using the data sequence in the step 1) and using the following gradient descent algorithm:
wi(k)＝wi(k-1)-ηΔwi(k-1)
Wherein w i (k-1) is a model parameter of the ith unmanned aerial vehicle of the previous round (k-1), w i (k) is a model parameter of the ith unmanned aerial vehicle of the current round (k), and eta is a learning rate adopted by the unmanned aerial vehicle for local training;
3) The unmanned aerial vehicle group performs election according to the data size participating in training, the unmanned aerial vehicle with the largest local training data size of the unmanned aerial vehicle in the round (k) is selected as a leader of the unmanned aerial vehicle group, and the rest unmanned aerial vehicles are common groups;
4) The unmanned aerial vehicle of the common group transmits the local model parameter w i (k) to the unmanned aerial vehicle leader through a communication link;
5) The unmanned aerial vehicle leader updates the global model according to the federal average method, and specifically comprises the following steps:
The model parameters of each unmanned aerial vehicle are w i (k), the data size of each unmanned aerial vehicle is D i, the total data size is D, and w (k) is a federally averaged model.
6) The unmanned aerial vehicle leader judges whether the current round (k) is the training round originally set: if yes, the updated model parameters are issued to unmanned aerial vehicles of all the ordinary groups, training is stopped, otherwise, the updated model parameters are issued to unmanned aerial vehicles of all the ordinary groups, and the steps 1) to 6) are continuously repeated.
The third part is pre-deployment of the unmanned aerial vehicle, and corresponds to S9-S12 of the access scheme. The method comprises the following specific steps:
1) Each unmanned aerial vehicle is provided with time slots with equal intervals, and a vehicle in the unmanned aerial vehicle area uploads own positioning information (x t,yt) to the unmanned aerial vehicle through a communication link when each time slot starts, wherein t is the current time slot;
2) Each unmanned aerial vehicle inputs the positioning information (x t,yt) of the vehicle in S9 into a cyclic neural network model listed in a sequence, predicts the future position (x t+1,yt+1) of the vehicle in the next time slot, wherein t+1 is the next time slot;
3) Each unmanned aerial vehicle calculates the position with zero total force by using a virtual force guide deployment algorithm and flies to the position with zero total force;
4) The vehicle sorts the signal quality intensity of the locally received unmanned aerial vehicle, if the maximum signal quality is smaller than the set signal quality threshold value, the vehicle selects to give up to access to the unmanned aerial vehicle base station, otherwise, the vehicle selects to access to the unmanned aerial vehicle base station with the maximum signal quality.
Step 3) the virtual force guide deployment algorithm is shown in fig. 2, and specifically comprises the following steps:
3-1): each unmanned aerial vehicle utilizes the predicted position information of S10 to construct the position distribution information of vehicles in the whole unmanned aerial vehicle group as shown in figure 4;
3-2): each unmanned aerial vehicle calculates the mass center position of the vehicle position distribution according to the vehicle predicted position in the area, and the mass center position is shown in figure 4;
3-3): the centroid position has a virtual gravitation effect on the unmanned aerial vehicle, and the unmanned aerial vehicle is pulled to the position;
3-4): the unmanned aerial vehicle updates own energy information; if the energy is smaller than or equal to the energy threshold of the unmanned aerial vehicle, the unmanned aerial vehicle flies back to the charging center, the vehicle connected to the unmanned aerial vehicle is disconnected with the unmanned aerial vehicle, the signal quality intensity of the nearby unmanned aerial vehicle base station is searched, and the unmanned aerial vehicle base station with the maximum signal quality intensity is selected to be connected;
wherein the signal quality strength in step 3-4) is defined as follows:
the unmanned aerial vehicle comprises N unmanned aerial vehicles, wherein the transmitting power of the unmanned aerial vehicle m to a vehicle k is p m,k, and the unmanned aerial vehicle m is defined as follows:
Where g m,k is the power gain between the drone m and the ground vehicle k, The sum of the products of power gain and power gain of other N-1 unmanned aerial vehicles except unmanned aerial vehicle m is shown, and N p is the Gaussian white noise power of the environment.
Finally, it is noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made thereto without departing from the spirit and scope of the present invention, which is intended to be covered by the claims of the present invention.
Claims (4)
1. A dynamic pre-deployment method of a multi-unmanned aerial vehicle is characterized in that: the method comprises the following steps:
s1: the vehicle is connected with each unmanned aerial vehicle in the area through a communication transmitting device to establish a communication link;
S2: the vehicle uploads the positioning information of the vehicle to each unmanned aerial vehicle through an uplink;
s3: constructing a distributed communication network among unmanned aerial vehicles through communication links, and distributing a sequence-to-sequence gating cyclic neural network model and training rounds;
S4: each unmanned aerial vehicle trains a local sequence-to-sequence gating circulating neural network model through an edge computing server by utilizing a plurality of pieces of vehicle positioning information collected in the area;
s5: each unmanned aerial vehicle performs election according to the data volume participating in training to obtain a leader and a common group;
s6: the common group unmanned aerial vehicle uploads the locally trained sequence-to-sequence gating cyclic neural network model parameters to an unmanned aerial vehicle leader;
s7: the unmanned aerial vehicle leader updates model parameters uploaded by unmanned aerial vehicles of all ordinary groups on a local edge calculation server by using a federal average method, updates model parameters of a sequence-to-sequence gated cyclic neural network and training rounds and transmits the model parameters and training rounds to the unmanned aerial vehicles of all ordinary groups;
S8: each unmanned aerial vehicle judges whether the current training round reaches the target round, if so, the step S9 is executed, otherwise, the steps S4-S7 are executed;
S9: each unmanned aerial vehicle is provided with time slots with equal intervals, and vehicles in the unmanned aerial vehicle area transmit own positioning information to the unmanned aerial vehicle through a communication link when each time slot starts;
s10: each unmanned aerial vehicle inputs the positioning information of the vehicle in the step S9 into a sequence-to-sequence cyclic neural network model, and predicts the future position of the vehicle in the next time slot;
S11: each unmanned aerial vehicle calculates the position with zero total force by using a virtual force guide deployment algorithm and flies to the position with zero total force;
s12: the vehicle sorts the signal quality intensity of the locally received unmanned aerial vehicle, if the maximum signal quality is smaller than the set signal quality threshold value, the vehicle selects to give up to be connected to the unmanned aerial vehicle base station, otherwise, the vehicle selects to be connected to the unmanned aerial vehicle base station with the maximum signal quality.
2. The multi-unmanned aerial vehicle dynamic pre-deployment method of claim 1, wherein: the federal averaging method in step S7 specifically includes:
Where N represents the total number of unmanned aerial vehicles, w i (k) represents the model parameters of the ith unmanned aerial vehicle, D i represents the data size of the ith unmanned aerial vehicle, D represents the total data size, and w (k) is the federally averaged model.
3. The multi-unmanned aerial vehicle dynamic pre-deployment method of claim 1, wherein: the virtual force guidance algorithm described in step S11 includes the steps of:
s111: each unmanned aerial vehicle utilizes the predicted position information of S10 to construct the position distribution information of vehicles in the whole unmanned aerial vehicle group;
s112: each unmanned aerial vehicle calculates the mass center position of the vehicle position distribution according to the vehicle predicted position in the area;
S113: the centroid position generates a virtual gravitation effect on the unmanned aerial vehicle, and the unmanned aerial vehicle is towed to the position;
S114: each unmanned aerial vehicle updates its own energy information; if the energy is smaller than or equal to the energy threshold of the unmanned aerial vehicle, the unmanned aerial vehicle flies back to the charging center, the vehicle connected to the unmanned aerial vehicle is disconnected with the unmanned aerial vehicle, the signal quality intensity of the nearby unmanned aerial vehicle base station is searched, and the unmanned aerial vehicle base station with the largest signal quality intensity is selected to be connected.
4. The multi-unmanned aerial vehicle dynamic pre-deployment method of claim 1, wherein: the signal quality in step S12 is the signal-to-interference-and-noise ratio of the communication, and is defined as follows:
Where p m,k is the transmit power of drone m to vehicle k, g m,k is the power gain between drone m and ground vehicle k, And N p is the Gaussian white noise power of the environment, which represents the sum of the products of the power and the power gain of N-1 unmanned aerial vehicles except the unmanned aerial vehicle m.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210046566.2A CN114374951B (en) | 2022-01-12 | 2022-01-12 | Dynamic pre-deployment method for multiple unmanned aerial vehicles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210046566.2A CN114374951B (en) | 2022-01-12 | 2022-01-12 | Dynamic pre-deployment method for multiple unmanned aerial vehicles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114374951A CN114374951A (en) | 2022-04-19 |
| CN114374951B true CN114374951B (en) | 2024-04-30 |
Family
ID=81143373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210046566.2A Active CN114374951B (en) | 2022-01-12 | 2022-01-12 | Dynamic pre-deployment method for multiple unmanned aerial vehicles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114374951B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114641007B (en) * | 2022-05-18 | 2022-08-26 | 鹏城实验室 | Aerial communication platform deployment method and device, intelligent terminal and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018033926A1 (en) * | 2016-08-18 | 2018-02-22 | Tevel Advanced Technologies Ltd. | System and method for plantation agriculture tasks management and data collection |
| CN110488867A (en) * | 2019-08-28 | 2019-11-22 | 中国人民解放军国防科技大学 | A kind of unmanned plane cluster convoy behavior generation method based on the virtual field of force of improvement |
| WO2020097562A1 (en) * | 2018-11-09 | 2020-05-14 | Iocurrents, Inc. | Machine learning-based prediction, planning, and optimization of trip time, trip cost, and/or pollutant emission during navigation |
| CN111786713A (en) * | 2020-06-04 | 2020-10-16 | 大连理工大学 | Unmanned aerial vehicle network hovering position optimization method based on multi-agent deep reinforcement learning |
| CN112291795A (en) * | 2020-10-23 | 2021-01-29 | 广州大学 | Distributed networking system and method based on backbone network of unmanned aerial vehicle |
| CN112423270A (en) * | 2020-10-12 | 2021-02-26 | 南京航空航天大学 | Unmanned aerial vehicle cluster low interception deployment method based on virtual force and beam parameter optimization |
| CN112580537A (en) * | 2020-12-23 | 2021-03-30 | 中国人民解放军国防科技大学 | Deep reinforcement learning method for multi-unmanned aerial vehicle system to continuously cover specific area |
| CN112929866A (en) * | 2021-01-20 | 2021-06-08 | 河北工程大学 | Unmanned aerial vehicle deployment method for self-adaptively optimizing urban disaster area network coverage |
| CN113177645A (en) * | 2021-06-29 | 2021-07-27 | 腾讯科技（深圳）有限公司 | Federal learning method and device, computing equipment and storage medium |
| CN113286314A (en) * | 2021-05-25 | 2021-08-20 | 重庆邮电大学 | Unmanned aerial vehicle base station deployment and user association method based on Q learning algorithm |
| CN113780462A (en) * | 2021-09-24 | 2021-12-10 | 华中科技大学 | Vehicle detection network establishment method based on unmanned aerial vehicle aerial image and application thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9849807B2 (en) * | 2012-12-28 | 2017-12-26 | General Electric Company | System and method for determining operational group assignments of vehicles in a vehicle system |
-
2022
- 2022-01-12 CN CN202210046566.2A patent/CN114374951B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018033926A1 (en) * | 2016-08-18 | 2018-02-22 | Tevel Advanced Technologies Ltd. | System and method for plantation agriculture tasks management and data collection |
| WO2020097562A1 (en) * | 2018-11-09 | 2020-05-14 | Iocurrents, Inc. | Machine learning-based prediction, planning, and optimization of trip time, trip cost, and/or pollutant emission during navigation |
| CN110488867A (en) * | 2019-08-28 | 2019-11-22 | 中国人民解放军国防科技大学 | A kind of unmanned plane cluster convoy behavior generation method based on the virtual field of force of improvement |
| CN111786713A (en) * | 2020-06-04 | 2020-10-16 | 大连理工大学 | Unmanned aerial vehicle network hovering position optimization method based on multi-agent deep reinforcement learning |
| CN112423270A (en) * | 2020-10-12 | 2021-02-26 | 南京航空航天大学 | Unmanned aerial vehicle cluster low interception deployment method based on virtual force and beam parameter optimization |
| CN112291795A (en) * | 2020-10-23 | 2021-01-29 | 广州大学 | Distributed networking system and method based on backbone network of unmanned aerial vehicle |
| CN112580537A (en) * | 2020-12-23 | 2021-03-30 | 中国人民解放军国防科技大学 | Deep reinforcement learning method for multi-unmanned aerial vehicle system to continuously cover specific area |
| CN112929866A (en) * | 2021-01-20 | 2021-06-08 | 河北工程大学 | Unmanned aerial vehicle deployment method for self-adaptively optimizing urban disaster area network coverage |
| CN113286314A (en) * | 2021-05-25 | 2021-08-20 | 重庆邮电大学 | Unmanned aerial vehicle base station deployment and user association method based on Q learning algorithm |
| CN113177645A (en) * | 2021-06-29 | 2021-07-27 | 腾讯科技（深圳）有限公司 | Federal learning method and device, computing equipment and storage medium |
| CN113780462A (en) * | 2021-09-24 | 2021-12-10 | 华中科技大学 | Vehicle detection network establishment method based on unmanned aerial vehicle aerial image and application thereof |
Non-Patent Citations (3)
| Title |
|---|
| Artificial Intelligence for UAV-Enabled Wireless Networks: A Survey;Mohamed-Amine Lahmeri;《IEEE Open Journal of the Communications Society》;20210423;全文 * |
| 基于5G和无人机智能组网的应急通信技术;钟剑峰;《电讯技术》;20201130;全文 * |
| 植保无人机飞控系统与航线规划研究进展分析;曹光乔;李亦白;南风;刘东;陈聪;张进龙;;农业机械学报;20200819(第08期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114374951A (en) | 2022-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Proactive drone-cell deployment: Overload relief for a cellular network under flash crowd traffic | |
| CN111786713B (en) | Unmanned aerial vehicle network hovering position optimization method based on multi-agent deep reinforcement learning | |
| Challita et al. | Deep reinforcement learning for interference-aware path planning of cellular-connected UAVs | |
| CN110364031A (en) | The path planning and wireless communications method of unmanned plane cluster in ground sensors network | |
| Chakraborty et al. | SkyRAN: a self-organizing LTE RAN in the sky | |
| Athanasiadou et al. | LTE ground-to-air field measurements in the context of flying relays | |
| Peng et al. | A predictive on-demand placement of UAV base stations using echo state network | |
| CN114374951B (en) | Dynamic pre-deployment method for multiple unmanned aerial vehicles | |
| CN115499921A (en) | Three-dimensional trajectory design and resource scheduling optimization method for complex unmanned aerial vehicle network | |
| Liu et al. | Multiobjective optimization for improving throughput and energy efficiency in UAV-enabled IoT | |
| CN112672361A (en) | Large-scale MIMO capacity increasing method based on unmanned aerial vehicle cluster deployment | |
| Luo et al. | A two-step environment-learning-based method for optimal UAV deployment | |
| CN114980169A (en) | Unmanned aerial vehicle auxiliary ground communication method based on combined optimization of track and phase | |
| CN113919483A (en) | Method and system for constructing and positioning radio map in wireless communication network | |
| Guo et al. | 3D aerial vehicle base station (UAV-BS) position planning based on deep Q-learning for capacity enhancement of users with different QoS requirements | |
| Singh et al. | Placement of mmWave base stations for serving urban drone corridors | |
| Liu et al. | A novel hybrid split and federated learning architecture in wireless UAV networks | |
| Chen et al. | Decentralized trajectory and power control based on multi-agent deep reinforcement learning in UAV networks | |
| Huang et al. | System-level metrics for non-terrestrial networks under stochastic geometry framework | |
| Akin et al. | Multiagent Q-learning based UAV trajectory planning for effective situationalawareness | |
| Arribas et al. | Optimizing fairness in cellular networks with mobile drone relays | |
| CN117793723A (en) | Air base station deployment method for emergency scene | |
| Ryu et al. | Energy efficient deployment of aerial base stations for mobile users in multi-hop UAV networks | |
| Wang et al. | Communications and networking technologies for intelligent drone cruisers | |
| Charlesworth | Using non-cooperative games to coordinate communications UAVs |
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 |