CN116522802B - Intelligent flight planning method for unmanned airship based on meteorological data - Google Patents
Intelligent flight planning method for unmanned airship based on meteorological data Download PDFInfo
- Publication number
- CN116522802B CN116522802B CN202310780343.3A CN202310780343A CN116522802B CN 116522802 B CN116522802 B CN 116522802B CN 202310780343 A CN202310780343 A CN 202310780343A CN 116522802 B CN116522802 B CN 116522802B
- Authority
- CN
- China
- Prior art keywords
- unmanned airship
- intelligent
- decision
- information
- wind field
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/092—Reinforcement learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/094—Adversarial learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an unmanned airship intelligent flight planning method based on meteorological data, which designs a deep neural network model according to an intelligent flight planning application data structure; the intelligent decision model is built by combining the perceived information with a deep countermeasure circulation Q neural network after passing through a long-term memory artificial network layer; an intelligent decision system framework is designed, and state information and wind field information of the airship are fused; establishing an unmanned airship flight simulation environment, and setting a reward mechanism to enable an intelligent body to obtain real-time feedback in the interaction with the environment; model training is carried out by utilizing reinforcement learning ideas, and a performance evaluation of intelligent decision is carried out on whether targets are completed or not by maximizing a cost function to enable interaction of the airship and the environment to learn Xi Zuiyou strategy. By adopting the intelligent flight planning method of the unmanned airship based on meteorological data, the energy consumption of the airship is saved, the unmanned airship can quickly reach a task target area and efficiently stay, and the development of a task is supported.
Description
Technical Field
The invention relates to the technical field of high-end equipment manufacturing, in particular to an unmanned airship intelligent flight planning method based on meteorological data.
Background
The unmanned airship has the advantages of long residence time, heavy task load, capability of residence at fixed points and the like, and is an ideal flight platform for application development such as communication, remote sensing, meteorological detection and the like. In recent years, with the gradual development of unmanned airship platforms and task load technologies, related application requirements of unmanned airships are more and more urgent, and the industry is urgent to have unmanned flight applications stably land. The unmanned airship application requirements put forth higher and higher requirements on the unmanned airship application capability, the unmanned airship application requirements are required to stably and reliably execute tasks and promote autonomous flight capability under complex environmental conditions, and intelligent flight planning becomes a key link for promoting the unmanned airship application capability.
Under urgent application requirements, unmanned airship technology is continuously developed, but the unmanned airship intelligent flight planning method still needs to break through. The unmanned airship is used as an aircraft which flies by buoyancy lighter than air, the flight of the unmanned airship is obviously influenced by wind fields in the environment, and the unmanned airship has certain wind resistance capability by providing power through a propeller, but has limited wind resistance capability due to huge volume.
Disclosure of Invention
The invention aims to provide an intelligent flight planning method for unmanned airship based on meteorological data, which can quickly reach a task target area and realize efficient residence while saving energy consumption of the unmanned airship and supporting task development.
In order to achieve the above purpose, the invention provides an unmanned airship intelligent flight planning method based on meteorological data, which comprises the following steps:
s1, designing a deep neural network model for sensing wind field information by utilizing a convolutional neural network according to an intelligent flight planning application data structure;
the perceived wind field information is integrated with the state information and then passes through a long-short-term memory neural network layer and then is combined with a deep countermeasure cycle (Q) neural network, and an intelligent decision model for outputting action decisions is established, so that an intelligent agent has global memory, and the intelligent decision model can obtain strategies in time sequence decisions;
s2, designing an intelligent decision system framework by utilizing the deep neural network selected in the step S1, and regarding the complete intelligent flight planning process as a Markov decision process;
s3, establishing an unmanned airship flight simulation environment, including a wind field generation model and an unmanned airship state change model, and setting a corresponding rewarding mechanism to enable an intelligent body to obtain real-time feedback in interaction with the environment;
s4, training a deep neural network model for sensing wind field information and an intelligent decision model for action decision in the simulation environment obtained in the step S3 by utilizing the reinforcement learning idea and the intelligent decision system frame in the step S2, and enabling the unmanned airship to learn strategies in interaction with the environment by maximizing a cost function, wherein a calculation formula of the maximizing the cost function is as follows:
；
wherein ,for attenuation factor->Rewards for the decision at the present moment +.>For the input of the current moment of the intelligent decision model, +.>Outputting the action decision of the unmanned airship at the current moment, < > for the unmanned airship at the current moment>For the next moment of input of the intelligent decision model, < >>Outputting the decision for the action of the unmanned airship at the next moment>Representing +.>Is not limited to the desired one;
and then testing whether the target is completed or not by using the actual wind field data, and performing intelligent decision performance evaluation on the tested result.
Preferably, in step S1, the deep neural network model designed to sense wind field information includes: extracting wind field information characteristics at the current moment by using a convolutional neural network, and integrating wind field and state information by using a full-connection layer neural network;
designing an intelligent decision model includes: based on deep antagonism circulation Q neural network construction, fused wind field and state heterogeneous information are taken as input, a long-short-term memory neural network is utilized to process sequence decision problems, the change trend of the fused information is sensed, the intelligent decision model has a memory function, and the change trend of the fused data is prejudged through the sequence data.
Preferably, in step S2, designing an intelligent decision system framework specifically includes: the task decision process is a Markov decision process, the unmanned airship action fusion state is decided based on the unmanned airship state, task and environment composite information, the unmanned airship action fusion state is updated according to the unmanned airship state change model after the unmanned airship takes action decision, the aircraft state at the next moment is obtained, and the task planning result of the unmanned airship is obtained through loop iteration.
Preferably, the action decision comprises an increment of speed and heading angle.
Preferably, in step S2, the process of framework loop iteration of the intelligent decision system includes:
s21, extracting wind field characteristics at the current moment in a two-dimensional time sequence wind field generated by a wind field generation model by using a convolutional neural network;
s22, splicing the generated one-dimensional vector with the current state information of the unmanned airship, and taking the spliced one-dimensional vector and the current state information of the unmanned airship as input of the intelligent decision model at the current moment after passing through the long-short-period memory neural network layerOutputting the unmanned airship action decision +.>；
S23, acquiring state information of the unmanned airship at the next moment by using a state change model of the unmanned airship, and acquiring new state information and a wind field at the next momentInformation fusion into；
S24, obtaining the track planning point through the post-loop decision of the long-short-term memory neural network layer.
Preferably, in step S3, the reward mechanism is an evaluation mechanism for adopting an action decision for the current unmanned airship, and after the unmanned airship adopts the action decision in the fusion information at the current moment, the unmanned airship will transfer to the fusion information at the next moment, and the reward mechanism makes an evaluation by comparing the fusion information at the current moment with the change of the fusion information at the next moment.
Preferably, in step S4, the performance evaluation of the intelligent decision includes:
s41, capturing predicted data of an actual wind field, and taking the initial state, expected position coordinates and a sequence of the actual wind field of the unmanned airship as input of an intelligent decision system framework at the same time to realize future task decisions;
s42, evaluating the target state of the unmanned airship through an intelligent decision result generated by the actual wind field data.
Preferably, in step S42, the method for evaluating the target state of the unmanned airship is: and judging through the distance between the final position of the unmanned airship and the target point.
Therefore, by adopting the intelligent flight planning method of the unmanned airship based on the meteorological data, the unmanned airship can quickly reach a task target area and realize efficient residence while saving the energy consumption of the unmanned airship, and supports the development of tasks.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a design diagram of an intelligent decision model of an unmanned airship intelligent flight planning method based on meteorological data;
FIG. 2 is a schematic diagram of an intelligent decision system framework of an unmanned airship intelligent flight planning method based on meteorological data according to the invention;
FIG. 3 is a flow chart of intelligent decision performance evaluation of an unmanned airship intelligent flight planning method based on meteorological data according to the invention;
FIG. 4 is a learning process diagram of an intelligent planning method of the unmanned airship intelligent flight planning method based on meteorological data;
fig. 5 is a test process diagram of an intelligent planning method of the unmanned airship intelligent flight planning method based on meteorological data.
Detailed Description
The technical scheme of the invention is further described below through the attached drawings and the embodiments.
Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs.
Example 1
An unmanned airship intelligent flight planning method based on meteorological data comprises the following steps:
s1, designing a deep neural network model for sensing wind field information by utilizing a convolutional neural network according to an intelligent flight planning application data structure; extracting wind field information characteristics at the current moment by using a convolutional neural network, and integrating wind field and state information by using a full-connection layer neural network;
the intelligent decision model for outputting action decisions is established by integrating the perceived information with the state information, then combining the perceived information with a depth countermeasure Q network after passing through a long-short-term memory artificial network layer, so that the intelligent agent has global memory, and the model can obtain an optimal strategy in time sequence decisions; based on deep antagonism circulation Q neural network construction, fused wind field and state heterogeneous information are taken as input, a long-short-term memory neural network is utilized to process a sequence decision problem, the change trend of the fused information is sensed, the model has a memory function, and the change trend of fused data is prejudged through sequence data. The intelligent decision model design is shown in figure 1.
S2, designing an intelligent decision system framework by utilizing the deep neural network selected in the step S1, and regarding the complete intelligent flight planning process as a Markov decision process; the design flow of the intelligent decision system framework is shown in fig. 2.
The design of the intelligent decision system frame specifically comprises the following steps: regarding the task decision process as a Markov decision process, and deciding the action fusion state of the aircraft as based on the composite information of the state, the task and the environment of the aircraftAnd updating the unmanned airship state change model to obtain the state of the aircraft at the next moment after the unmanned airship takes an action decision.
The action decision includes an increment of speed and heading angle. And (3) obtaining a mission planning result of the whole mission period of the aircraft through cyclic iteration, and supporting the decision.
The framework loop iteration process of the intelligent decision system comprises the following steps:
s21, extracting wind field characteristics at the current moment in a two-dimensional time sequence wind field generated by a wind field generation model by using a convolutional neural network;
s22, splicing the generated one-dimensional vector with the current state information of the unmanned airship, and inputting the one-dimensional vector as an intelligent decision model after passing through the long-short-term memory neural network layerOutputting the unmanned airship action decision +.>；
S23, acquiring state information of the unmanned airship at the next moment by using a state change model of the unmanned airship, and fusing the acquired new state information with wind field information at the next moment again to form the unmanned airship；
S24, obtaining the track planning point through the loop decision after the long-short-term memory layer.
S3, establishing an unmanned airship flight simulation environment, including a wind field generation model and an unmanned airship state change model, and setting a corresponding rewarding mechanism to enable the model to obtain real-time feedback in interaction with the environment;
s4, in the simulation environment obtained in the step S3, model training is carried out by utilizing a reinforcement learning idea, and the unmanned airship learns an optimal strategy in interaction with the environment by maximizing a cost function, wherein a calculation formula of the maximized cost function is as follows:
；
wherein ,for attenuation factor->Rewards for the decision at the present moment +.>For the input of the current moment of the intelligent decision model, +.>Outputting the action decision of the unmanned airship at the current moment, < > for the unmanned airship at the current moment>For the next moment of input of the intelligent decision model, < >>Outputting the decision for the action of the unmanned airship at the next moment>Representing +.>Is not limited to the desired one;
and then testing whether the target is completed or not by using the actual wind field data, and performing intelligent decision performance evaluation on the tested result. The intelligent decision performance evaluation flow is shown in fig. 3.
The rewarding mechanism is an evaluation mechanism for adopting action decisions for the current unmanned airship, the unmanned airship can transfer to fusion information at the next moment after adopting the action decisions in the fusion information at the current moment, and the rewarding mechanism evaluates by comparing the changes of the unmanned airship and the fusion information.
In step S4, the performance evaluation of the intelligent decision specifically includes:
s41, taking the initial state, expected position coordinates and the sequence of the actual wind field of the unmanned airship as the input of an intelligent decision system framework simultaneously through grabbing the predicted data of the actual wind field, so as to realize the decision of at least 24h task in the future and update the unmanned airship quickly along with the data update; the initial state of the unmanned airship comprises information such as initial speed, heading, coordinates, current time, residual energy ratio and the like of the unmanned airship.
S42, evaluating whether the unmanned airship can complete the target state through an intelligent decision result generated by the actual wind field data.
In step S42, the speed and heading of the unmanned airship are updated according to the incremental decisions of the speed and heading angle generated at each moment, and the expected route is drawn according to the state change model, so that the path of the unmanned airship is drawn, and the method for evaluating whether the unmanned airship can complete the target state is to judge through the distance between the final position of the unmanned airship and the target point.
The following simulation tests were performed:
the training process is shown in FIG. 4, the wind field used for training is a simulated wind field, and the data format isThe format tensor, wherein each unit represents the change of 0.25 longitude and latitude, the state information of the unmanned airship comprises initial coordinates (2.5, 5), the remaining energy ratio of 55% at the current time of 0, the current speed of 5m/s, the heading of 60 degrees and task target point coordinates (17.5,9), wherein the wind field is changed continuously along with the time. Unmanned airships are continually learned in this environment until path planning can be completed with any initial data.
For the trained intelligent decision model, real meteorological wind field data are adopted for testing, the testing process is as shown in fig. 5, and the wind field data are in the same formatThe format tensor, wherein each unit represents the change of 0.25 longitude and latitude, the state information of the unmanned airship comprises initial coordinates (2.5, 4), the remaining energy ratio is 65% at the current time of 8, the current speed is 7m/s, the heading is 15 degrees, and task target point coordinates (17.7,9.7), wherein the decision interval is 450s, the displayed test process interval is 1h for convenience in display, and 1000 groups of wind fields are used for testing, so that the probability of reaching the target point is 99.8%.
Therefore, by adopting the intelligent flight planning method of the unmanned airship based on the meteorological data, the unmanned airship can quickly reach a task target area and realize efficient residence while saving the energy consumption of the unmanned airship, and supports the development of tasks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention and not for limiting it, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that: the technical scheme of the invention can be modified or replaced by the same, and the modified technical scheme cannot deviate from the spirit and scope of the technical scheme of the invention.
Claims (7)
1. An unmanned airship intelligent flight planning method based on meteorological data is characterized by comprising the following steps of: the method comprises the following steps:
s1, designing a deep neural network model for sensing wind field information according to an intelligent flight planning application data structure;
the perceived wind field information passes through a long-term and short-term memory neural network layer and then is combined with a deep countermeasure cycle (Q) neural network, and an intelligent decision model for outputting action decisions is established, so that an intelligent agent has global memory, and the intelligent decision model can obtain strategies in time sequence decisions;
s2, designing an intelligent decision system framework, and fusing state information and wind field information of the unmanned airship;
the fusion of the state information and the wind field information of the unmanned airship comprises the following steps:
s21, extracting wind field characteristics at the current moment in a two-dimensional time sequence wind field by using a convolutional neural network;
s22, splicing the generated one-dimensional vector with the current state information of the unmanned airship, and taking the spliced one-dimensional vector and the current state information of the unmanned airship as input of the intelligent decision model at the current moment after passing through the long-short-period memory neural network layerOutputting the unmanned airship action decision +.>；
S23, acquiring state information of the unmanned airship at the next moment by using a state change model of the unmanned airship, and fusing the acquired new state information with wind field information at the next moment to form the unmanned airship；
S24, obtaining a track planning point through a long-short-term memory neural network layer post-loop decision;
s3, establishing an unmanned airship flight simulation environment, including a wind field generation model and an unmanned airship state change model, and setting a corresponding rewarding mechanism to enable an intelligent body to obtain real-time feedback in interaction with the environment;
s4, in the simulation environment obtained in the step S3, model training is carried out by utilizing a reinforcement learning idea, and the unmanned airship learns strategies in interaction with the environment by maximizing a cost function, wherein a calculation formula of the maximizing the cost function is as follows:
；
wherein ,for attenuation factor->For deciding at the current momentRewarding (I)>For the input of the current moment of the intelligent decision model, +.>Outputting the action decision of the unmanned airship at the current moment, < > for the unmanned airship at the current moment>For the input of the next moment of the intelligent decision model,outputting the decision for the action of the unmanned airship at the next moment>Representing +.>Is not limited to the desired one;
and then testing whether the target is completed or not by using the actual wind field data, and performing intelligent decision performance evaluation on the tested result.
2. The unmanned airship intelligent flight planning method according to claim 1, wherein in step S1, the deep neural network model designed to sense wind field information comprises: extracting wind field information characteristics at the current moment by using a convolutional neural network, and integrating wind field and state information by using a full-connection layer neural network;
designing an intelligent decision model includes: based on deep antagonism circulation Q neural network construction, fused wind field and state heterogeneous information are taken as input, a long-short-term memory neural network is utilized to process sequence decision problems, the change trend of the fused information is sensed, the intelligent decision model has a memory function, and the change trend of the fused data is prejudged through the sequence data.
3. The unmanned airship intelligent flight planning method based on meteorological data according to claim 1, wherein in step S2, the design of the intelligent decision system framework specifically includes: the task decision process is a Markov decision process, the unmanned airship action fusion state is decided based on the unmanned airship state, task and environment composite information, the unmanned airship action fusion state is updated according to the unmanned airship state change model after the unmanned airship takes action decision, the aircraft state at the next moment is obtained, and the task planning result of the unmanned airship is obtained through loop iteration.
4. A method of intelligent flight planning for unmanned airship based on meteorological data according to claim 3, wherein the action decision comprises an increment of speed and heading angle.
5. The intelligent flight planning method of the unmanned airship based on meteorological data according to claim 1, wherein in the step S3, the rewarding mechanism is an evaluation mechanism for adopting action decisions for the current unmanned airship, the unmanned airship can transfer to fusion information of the next moment after adopting the action decisions in the fusion information of the current moment, and the rewarding mechanism evaluates by comparing the fusion information of the current moment and the fusion information of the next moment.
6. The unmanned airship intelligent flight planning method according to claim 1, wherein in step S4, the performance evaluation of the intelligent decision comprises:
s41, capturing predicted data of an actual wind field, and taking the initial state, expected position coordinates and a sequence of the actual wind field of the unmanned airship as input of an intelligent decision system framework at the same time to realize future task decisions;
s42, evaluating the target state of the unmanned airship through an intelligent decision result generated by the actual wind field data.
7. The intelligent flight planning method for unmanned airship based on meteorological data according to claim 1, wherein in step S42, the method for evaluating the target state of unmanned airship completion is as follows: and judging through the distance between the final position of the unmanned airship and the target point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310780343.3A CN116522802B (en) | 2023-06-29 | 2023-06-29 | Intelligent flight planning method for unmanned airship based on meteorological data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310780343.3A CN116522802B (en) | 2023-06-29 | 2023-06-29 | Intelligent flight planning method for unmanned airship based on meteorological data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116522802A CN116522802A (en) | 2023-08-01 |
| CN116522802B true CN116522802B (en) | 2023-09-19 |
Family
ID=87390541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310780343.3A Active CN116522802B (en) | 2023-06-29 | 2023-06-29 | Intelligent flight planning method for unmanned airship based on meteorological data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116522802B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119004979A (en) * | 2024-08-09 | 2024-11-22 | 中国人民解放军32180部队 | Unmanned aerial vehicle target task autonomous region searching and distributing method based on field data |
| CN119126889B (en) * | 2024-08-22 | 2025-06-27 | 北京市农林科学院智能装备技术研究中心 | Multi-objective solar greenhouse ventilation decision-making method, device, electronic device and storage medium |
| CN119129641B (en) * | 2024-09-10 | 2025-06-17 | 海南大学 | Multi-agent cooperative control method for traffic scenarios |
| CN119717642A (en) * | 2024-12-23 | 2025-03-28 | 河南通航无人机科技有限公司 | Automatic take-off and landing method and system based on unmanned airport |
| CN120176684B (en) * | 2025-05-21 | 2025-08-26 | 中国科学院国家空间科学中心 | Airship path planning method and system based on deep reinforcement learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111538241A (en) * | 2020-04-30 | 2020-08-14 | 中国人民解放军国防科技大学 | Intelligent control method for horizontal track of stratospheric airship |
| WO2021086532A1 (en) * | 2019-10-29 | 2021-05-06 | Loon Llc | Navigating aerial vehicles using deep reinforcement learning |
| CN112766499A (en) * | 2021-02-02 | 2021-05-07 | 电子科技大学 | Method for realizing autonomous flight of unmanned aerial vehicle through reinforcement learning technology |
| CN113848974A (en) * | 2021-09-28 | 2021-12-28 | 西北工业大学 | Aircraft trajectory planning method and system based on deep reinforcement learning |
| KR20220037027A (en) * | 2020-09-16 | 2022-03-24 | 이민형 | System and method for monitoring the ground using hybrid unmanned airship |
| CN115453914A (en) * | 2022-10-19 | 2022-12-09 | 哈尔滨理工大学 | Unmanned ship recovery distributed decision simulation system considering sea wave interference |
| CN116339388A (en) * | 2023-05-29 | 2023-06-27 | 北京航空航天大学 | Method and system for controlling coverage of cluster area of stratospheric airship in uncertain wind field |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111694365B (en) * | 2020-07-01 | 2021-04-20 | 武汉理工大学 | A Deep Reinforcement Learning Based Path Tracking Method for Unmanned Vessel Formation |
-
2023
- 2023-06-29 CN CN202310780343.3A patent/CN116522802B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021086532A1 (en) * | 2019-10-29 | 2021-05-06 | Loon Llc | Navigating aerial vehicles using deep reinforcement learning |
| CN111538241A (en) * | 2020-04-30 | 2020-08-14 | 中国人民解放军国防科技大学 | Intelligent control method for horizontal track of stratospheric airship |
| KR20220037027A (en) * | 2020-09-16 | 2022-03-24 | 이민형 | System and method for monitoring the ground using hybrid unmanned airship |
| CN112766499A (en) * | 2021-02-02 | 2021-05-07 | 电子科技大学 | Method for realizing autonomous flight of unmanned aerial vehicle through reinforcement learning technology |
| CN113848974A (en) * | 2021-09-28 | 2021-12-28 | 西北工业大学 | Aircraft trajectory planning method and system based on deep reinforcement learning |
| CN115453914A (en) * | 2022-10-19 | 2022-12-09 | 哈尔滨理工大学 | Unmanned ship recovery distributed decision simulation system considering sea wave interference |
| CN116339388A (en) * | 2023-05-29 | 2023-06-27 | 北京航空航天大学 | Method and system for controlling coverage of cluster area of stratospheric airship in uncertain wind field |
Non-Patent Citations (2)
| Title |
|---|
| Guest Editorial Introduction to the Special Issue on Deep learning Models for Safe and Secure Intelligent Transportation Systerms;Alireza Jolfaei;IEEE;第22卷(第7期);4224-4229 * |
| 平流层飞艇抗风场干扰约束Unscented卡尔曼滤波算法设计与应用;赵岩;高社生;丁晓;韦道知;;国防科技大学学报;第39卷(第2期);64-70 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116522802A (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116522802B (en) | Intelligent flight planning method for unmanned airship based on meteorological data | |
| Bellemare et al. | Autonomous navigation of stratospheric balloons using reinforcement learning | |
| Baboo et al. | An efficient weather forecasting system using artificial neural network | |
| CN112947592B (en) | A Re-Entry Vehicle Trajectory Planning Method Based on Reinforcement Learning | |
| CN108594858B (en) | Unmanned aerial vehicle searching method and device for Markov moving target | |
| CN112698646B (en) | Aircraft path planning method based on reinforcement learning | |
| CN115391924B (en) | Deep reinforcement learning design method for intelligent satellite in digital twin environment | |
| CN113110550B (en) | Unmanned aerial vehicle flight control method based on reinforcement learning and network model distillation | |
| CN114169066A (en) | Space target characteristic measuring and reconnaissance method based on micro-nano constellation approaching reconnaissance | |
| CN112141369B (en) | Decision and control method for autonomous rendezvous and docking of translational closing sections of spacecraft | |
| Conlon et al. | Generalizing competency self-assessment for autonomous vehicles using deep reinforcement learning | |
| CN120129016B (en) | Intelligent trajectory planning method for relay unmanned aerial vehicle of near space communication network | |
| CN116167254A (en) | Multidimensional city simulation deduction method and system based on city big data | |
| CN119668277A (en) | Flight control network construction method, control method and system | |
| CN116817909A (en) | Unmanned aerial vehicle relay type navigation method based on deep reinforcement learning | |
| Lv et al. | Hierarchical reinforcement learning method for long-horizon path planning of stratospheric airship | |
| Sidenko et al. | Machine learning for unmanned aerial vehicle routing on rough terrain | |
| Zhao et al. | Clustering-based hyper-heuristic algorithm for multi-region coverage path planning of heterogeneous UAVs | |
| CN110826810A (en) | A regional rainfall prediction method combining spatial reasoning and machine learning | |
| Hou et al. | Trajectory Planning Based On Continuous Decision Deep Reinforcement Learning for Stratospheric Airship | |
| CN119623789A (en) | Route path planning scheme generation method and system based on deep reinforcement learning | |
| CN119272938A (en) | A method for energy consumption prediction and management based on large model intelligent agent | |
| Liu et al. | Comprehensive tradeoff and utilization of airborne renewable energy and uncertain stratospheric wind potential based on reinforcement learning | |
| CN119004979A (en) | Unmanned aerial vehicle target task autonomous region searching and distributing method based on field data | |
| Meng et al. | Optimization of Stratospheric Airship Station-Keeping Strategy Based on LSTM-DQN Algorithm |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230926 Address after: 3-2-90 (self declaration), No. 5 Xingbin Road, Sino Italian Ningbo Ecological Park, Yuyao, Ningbo, Zhejiang 315000 Patentee after: Linzhou (Ningbo) Technology Co.,Ltd. Address before: 100000 1720, 17/F, Shining Building, No. 35, Xueyuan Road, Haidian District, Beijing Patentee before: BEIJING TIANHANG HUACHUANG TECHNOLOGY Co.,Ltd. Patentee before: Linzhou (Ningbo) Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |