CN110751266A - Unmanned aerial vehicle trajectory prediction module and prediction method thereof - Google Patents
Unmanned aerial vehicle trajectory prediction module and prediction method thereof Download PDFInfo
- Publication number
- CN110751266A CN110751266A CN201910920954.7A CN201910920954A CN110751266A CN 110751266 A CN110751266 A CN 110751266A CN 201910920954 A CN201910920954 A CN 201910920954A CN 110751266 A CN110751266 A CN 110751266A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- position information
- neural network
- prediction
- 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.)
- Pending
Links
Images
Classifications
-
- 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
- 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 invention discloses an unmanned aerial vehicle trajectory prediction module and a prediction method thereof, wherein the unmanned aerial vehicle trajectory prediction module comprises the following steps: 1. use low latitude radar and photoelectric detection equipment in the time of t, N records unmanned aerial vehicle's position and the gesture that corresponds, and the position includes: flight height and horizontal coordinates; 2. inputting the position information and the attitude information of the unmanned aerial vehicle into an anti-neural network algorithm for training; 3. after the training is successful, if coordinate point, height and attitude information of the unmanned aerial vehicle are input according to the regulations, the whole flight track of the unmanned aerial vehicle can be predicted. The invention has the advantages that: 1. further playing the roles of low-altitude radar and photoelectric detection, and increasing the function of track prediction on the basis of real-time monitoring; 2. by combining the flight attitude of the unmanned aerial vehicle, the accuracy of prediction is improved; 3. the method is more effective in predicting the flight path of the unmanned aerial vehicle with complex motion state.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle trajectory prediction module based on a deconvolution neural network and a prediction method thereof.
Background
With the development of society and the continuous progress of science and technology, the unmanned aerial vehicle technology has also been widely applied in various fields, and the irreplaceable effect of the unmanned aerial vehicle is reflected in different aspects.
For more effective control of the drone, and to understand the behavior of the drone. We need to predict the flight trajectory of an unmanned aerial vehicle for more effective, real-time knowledge of the flight status of the unmanned aerial vehicle to more effectively control the unmanned aerial vehicle. In order to avoid some unexpected problems.
A method for predicting the flight path of an unmanned aerial vehicle comprises ① a method without parameter estimation based on Kalman filtering or neural network, such as convolution and deconvolution neural network prediction, ② a method based on aircraft dynamics or kinematics model, such as 4D flight path prediction and metabolism GM (1,1) model prediction, and many prediction modules established based on various mathematical prediction models, such as gray model prediction, particle filtering, time series prediction, Markov prediction, and the like.
Prior art 1
The 4D track is in a space and time form, and accurate description on longitude, latitude, height and time of each point spatial position in a certain aircraft track can improve the utilization rate and safety of an airspace. The operation based on the 4D flight path is an effective means for managing the airspace under the conditions of large flow, high density and small spacing.
The specific implementation method in the module is that firstly, supermap software is used for creating an electronic airway network, and an electronic airway network diagram comprises all elements of the airway: course, distance, etc.; according to the performance parameters of the aircraft, the flying distance of the aircraft at a certain time point can be estimated, so that the expected arrival position of the aircraft on the airway can be found in the constructed airway network, and the 4D flying track of the aircraft in the airway can be predicted.
The unmanned aerial vehicle flight path prediction module based on the 4D flight path has the following defects:
1. the module algorithm is too complex, the requirements on space complexity and time complexity are high, and the module algorithm is not suitable for calculating a large amount of data;
2. the calculation needs more parameters, which has many limitations on the trajectory prediction of the unmanned aerial vehicle and is not suitable for the unmanned aerial vehicle to predict under certain specific conditions.
Prior art 2
The unmanned aerial vehicle flight path prediction module based on the metabolism GM (1,1) model is used for continuously establishing the GM (1,1) model to realize the point-to-line process. And calling an internal flight track prediction module when needed, so as to realize prediction of the next flight track of the unmanned aerial vehicle.
The specific implementation method inside the module is that the initial position of the unmanned aerial vehicle is used as the initial position at the beginning, then a GM (1,1) model is established according to various external factors such as the speed of the unmanned aerial vehicle, weather, temperature, wind direction and the like of the unmanned aerial vehicle, the position where the unmanned aerial vehicle is likely to appear next is calculated, then the new position calculated by the unmanned aerial vehicle is used as the initial position, the GM (1,1) model is established again to calculate the position where the unmanned aerial vehicle is likely to appear next, the new position is used as the initial position continuously, and the GM (1,1) model is established continuously to calculate the position where the unmanned aerial vehicle is likely to appear next. This process is called the metabolic GM (1,1) model. After the processes are carried out, the model continuously supplements new position measurement information, abandons older position measurement information, and accordingly obtains higher prediction accuracy, and then forms a track, namely the flight track of the unmanned aerial vehicle predicted by the metabolism GM (1,1) model.
Unmanned aerial vehicle flight path prediction module based on metabolism GM (1,1) model has the following defects:
1. the method is only suitable for medium and short term prediction, and cannot predict long-term behaviors of the unmanned aerial vehicle;
2. the method is only suitable for prediction approximate to exponential growth, and accurate prediction cannot be carried out under some complex flight environments of the unmanned aerial vehicle.
Abbreviations and Key term definitions
Convolution (convolution): convolution is widely applied in the field of image processing, such as filtering, edge detection, image sharpening and the like, and is realized by different convolution kernels. Features in the picture can be extracted through convolution operation in the convolutional neural network, some edges, lines, angles and other features of the picture can be extracted by the lower-layer convolutional layer, and more complex features can be learned from the lower-layer convolutional layer by the higher-layer convolution, so that the classification and identification of the picture are realized.
Deconvolution: deconvolution, which is in fact the inverse of convolution, is also called transposed convolution. There may be a wrong recognition of the deconvolution, so that the image before the convolution can be obtained by deconvolution, and actually, the image before the convolution cannot be restored by the deconvolution operation, and only the size of the image before the convolution can be restored. The process of convolution can be visualized by deconvolution, which has numerous applications in the fields of GAN and the like.
A neural network: neural networks are an important machine learning technique. It is the foundation of deep learning, which is the most popular research direction at present. The neural network is a machine learning technology which simulates the neural network of the human brain so as to realize artificial intelligence. Neural networks in the human brain are a very complex organization. The adult brain is estimated to have as many as 1000 million neurons. Machine learning is developed by simulating the neural network structure of the human brain.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an unmanned aerial vehicle trajectory prediction module and a prediction method thereof, and solves the defects in the prior art.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
an unmanned aerial vehicle trajectory prediction module comprising: the system comprises a low-altitude radar, a photoelectric detection device, a position information processing unit, a video information processing unit, a deconvolution neural network algorithm unit and a data integration unit;
the low-altitude radar is used for acquiring position information on the horizontal plane of the unmanned aerial vehicle and transmitting the position information to the position information processing unit;
the photoelectric detection device is used for acquiring attitude information of the unmanned aerial vehicle and position information of a vertical plane and transmitting the attitude information and the position information to the position information processing unit;
the position information processing unit is used for processing unmanned aerial vehicle position information acquired by photoelectric and radar and converting the unmanned aerial vehicle position information into data which can be trained by a deconvolution neural network algorithm
The video information processing unit is used for processing attitude information of the photoelectric transmission unmanned aerial vehicle and converting the attitude information into data which can be trained by a deconvolution neural network algorithm.
The deconvolution neural network algorithm unit is internally provided with a deconvolution neural network algorithm training method for training and processing the position information and the attitude information.
The data integration unit is used for processing the output result to form a prediction track and possibility probability which can be directly observed.
The invention also discloses a prediction method of the unmanned aerial vehicle trajectory prediction module, which comprises the following steps:
1. use low latitude radar and photoelectric detection equipment in the time of t, N records unmanned aerial vehicle's position and the gesture that corresponds, and the position includes: flight height and horizontal coordinates;
2. inputting the position information and the attitude information of the unmanned aerial vehicle into an anti-neural network algorithm for training;
3. after the training is successful, if coordinate point, height and attitude information of the unmanned aerial vehicle are input according to the regulations, the whole flight track of the unmanned aerial vehicle can be predicted.
Compared with the prior art, the invention has the advantages that:
1. further playing the roles of low-altitude radar and photoelectric detection, and increasing the function of track prediction on the basis of real-time monitoring;
2. by combining the flight attitude of the unmanned aerial vehicle, the accuracy of prediction is improved;
3. the method is more effective in predicting the flight path of the unmanned aerial vehicle with complex motion state.
Drawings
Fig. 1 is a schematic structural diagram of an unmanned aerial vehicle trajectory prediction module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings by way of examples.
As shown in fig. 1, an unmanned aerial vehicle trajectory prediction module includes: the system comprises a low-altitude radar, a photoelectric detection device, a position information processing unit, a video information processing unit, a deconvolution neural network algorithm unit and a data integration unit;
the low-altitude radar is used for acquiring position information on the horizontal plane of the unmanned aerial vehicle and transmitting the position information to the position information processing unit;
the photoelectric detection device is used for acquiring attitude information of the unmanned aerial vehicle and position information of a vertical plane and transmitting the attitude information and the position information to the position information processing unit;
the position information processing unit is used for processing unmanned aerial vehicle position information acquired by photoelectric and radar and converting the unmanned aerial vehicle position information into data which can be trained by a deconvolution neural network algorithm
The video information processing unit is used for processing attitude information of the photoelectric transmission unmanned aerial vehicle and converting the attitude information into data which can be trained by a deconvolution neural network algorithm.
The deconvolution neural network algorithm unit is internally provided with a deconvolution neural network algorithm training method for training and processing the position information and the attitude information.
The data integration unit is used for processing the output result to form a prediction track and possibility probability which can be directly observed.
A prediction method of an unmanned aerial vehicle trajectory prediction module comprises the following steps:
1. the low-altitude radar detects the unmanned aerial vehicle and records the position information (coordinate points) of the unmanned aerial vehicle.
2. Photoelectric detection equipment acquires unmanned aerial vehicle flight attitude information and vertical plane's positional information.
3. And transmitting the acquired unmanned aerial vehicle position information into a deconvolution neural network module digital processing system.
4. And the photoelectric detection equipment transmits the collected flight attitude information of the unmanned aerial vehicle into a deconvolution neural network module video processing system for decoding.
5. And transmitting the position information of the unmanned aerial vehicle and the decoded video information into an algorithm system of a deconvolution neural network module, predicting the position of the next stage of the unmanned aerial vehicle according to the information, and gradually realizing the prediction of the flight trajectory of the unmanned aerial vehicle in a three-dimensional space.
It will be appreciated by those of ordinary skill in the art that the examples described herein are intended to assist the reader in understanding the manner in which the invention is practiced, and it is to be understood that the scope of the invention is not limited to such specifically recited statements and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (2)
1. An unmanned aerial vehicle trajectory prediction module, comprising: the system comprises a low-altitude radar, a photoelectric detection device, a position information processing unit, a video information processing unit, a deconvolution neural network algorithm unit and a data integration unit;
the low-altitude radar is used for acquiring position information on the horizontal plane of the unmanned aerial vehicle and transmitting the position information to the position information processing unit;
the photoelectric detection device is used for acquiring attitude information of the unmanned aerial vehicle and position information of a vertical plane and transmitting the attitude information and the position information to the position information processing unit;
the position information processing unit is used for processing unmanned aerial vehicle position information acquired by photoelectric and radar and converting the unmanned aerial vehicle position information into data which can be trained by a deconvolution neural network algorithm;
the video information processing unit is used for processing attitude information of the photoelectric transmission unmanned aerial vehicle and converting the attitude information into data which can be trained by a deconvolution neural network algorithm;
the deconvolution neural network algorithm unit is internally provided with a deconvolution neural network algorithm training method for training and processing position information and attitude information;
the data integration unit is used for processing the output result to form a prediction track and possibility probability which can be directly observed.
2. The prediction method of the unmanned aerial vehicle trajectory prediction module according to claim 1, wherein:
1) use low latitude radar and photoelectric detection equipment in the time of t, N records unmanned aerial vehicle's position and the gesture that corresponds, and the position includes: flight height and horizontal coordinates;
2) inputting the position information and the attitude information of the unmanned aerial vehicle into an anti-neural network algorithm for training;
3) after the training is successful, if coordinate point, height and attitude information of the unmanned aerial vehicle are input according to the regulations, the whole flight track of the unmanned aerial vehicle can be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910920954.7A CN110751266A (en) | 2019-09-26 | 2019-09-26 | Unmanned aerial vehicle trajectory prediction module and prediction method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910920954.7A CN110751266A (en) | 2019-09-26 | 2019-09-26 | Unmanned aerial vehicle trajectory prediction module and prediction method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110751266A true CN110751266A (en) | 2020-02-04 |
Family
ID=69277169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910920954.7A Pending CN110751266A (en) | 2019-09-26 | 2019-09-26 | Unmanned aerial vehicle trajectory prediction module and prediction method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751266A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461292A (en) * | 2020-03-17 | 2020-07-28 | 南京航空航天大学 | Real-time trajectory prediction method for unmanned aerial vehicle |
| CN111461437A (en) * | 2020-04-01 | 2020-07-28 | 北京工业大学 | Data-driven crowd movement simulation method based on generation of confrontation network |
| CN112069889A (en) * | 2020-07-31 | 2020-12-11 | 北京信息科技大学 | Civil aircraft trajectory prediction method, electronic device and storage medium |
| CN112799031A (en) * | 2021-03-31 | 2021-05-14 | 长沙莫之比智能科技有限公司 | Clutter suppression method for millimeter wave ground-like radar data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902282A (en) * | 2012-09-25 | 2013-01-30 | 中国兵器工业第二0五研究所 | Optic axis and inertia axis superposition-based geographic tracking method |
| CN106873628A (en) * | 2017-04-12 | 2017-06-20 | 北京理工大学 | A kind of multiple no-manned plane tracks the collaboration paths planning method of many maneuvering targets |
| CN107846258A (en) * | 2017-09-07 | 2018-03-27 | 新疆美特智能安全工程股份有限公司 | A kind of unmanned plane system of defense |
| CN109635793A (en) * | 2019-01-31 | 2019-04-16 | 南京邮电大学 | A kind of unmanned pedestrian track prediction technique based on convolutional neural networks |
| CN109947126A (en) * | 2019-03-07 | 2019-06-28 | 中国科学院深圳先进技术研究院 | Control method, device, equipment and the readable medium of quadrotor drone |
-
2019
- 2019-09-26 CN CN201910920954.7A patent/CN110751266A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902282A (en) * | 2012-09-25 | 2013-01-30 | 中国兵器工业第二0五研究所 | Optic axis and inertia axis superposition-based geographic tracking method |
| CN106873628A (en) * | 2017-04-12 | 2017-06-20 | 北京理工大学 | A kind of multiple no-manned plane tracks the collaboration paths planning method of many maneuvering targets |
| CN107846258A (en) * | 2017-09-07 | 2018-03-27 | 新疆美特智能安全工程股份有限公司 | A kind of unmanned plane system of defense |
| CN109635793A (en) * | 2019-01-31 | 2019-04-16 | 南京邮电大学 | A kind of unmanned pedestrian track prediction technique based on convolutional neural networks |
| CN109947126A (en) * | 2019-03-07 | 2019-06-28 | 中国科学院深圳先进技术研究院 | Control method, device, equipment and the readable medium of quadrotor drone |
Non-Patent Citations (2)
| Title |
|---|
| 万刚等: "《无人机测绘技术及应用》", 31 December 2015 * |
| 徐贵力等: "《光电检测技术与系统设计》", 31 August 2013 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461292A (en) * | 2020-03-17 | 2020-07-28 | 南京航空航天大学 | Real-time trajectory prediction method for unmanned aerial vehicle |
| CN111461437A (en) * | 2020-04-01 | 2020-07-28 | 北京工业大学 | Data-driven crowd movement simulation method based on generation of confrontation network |
| CN111461437B (en) * | 2020-04-01 | 2023-11-07 | 北京工业大学 | Data-driven crowd motion simulation method based on generation of countermeasure network |
| CN112069889A (en) * | 2020-07-31 | 2020-12-11 | 北京信息科技大学 | Civil aircraft trajectory prediction method, electronic device and storage medium |
| CN112069889B (en) * | 2020-07-31 | 2021-08-03 | 北京信息科技大学 | Civil aircraft trajectory prediction method, electronic device and storage medium |
| CN112799031A (en) * | 2021-03-31 | 2021-05-14 | 长沙莫之比智能科技有限公司 | Clutter suppression method for millimeter wave ground-like radar data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110751266A (en) | Unmanned aerial vehicle trajectory prediction module and prediction method thereof | |
| CN109348707A (en) | For the method and apparatus of the Q study trimming experience memory based on deep neural network | |
| CN109255286B (en) | Unmanned aerial vehicle optical rapid detection and identification method based on deep learning network framework | |
| CN111176309B (en) | Multi-unmanned aerial vehicle self-group mutual inductance understanding method based on spherical imaging | |
| US11789466B2 (en) | Event camera based navigation control | |
| Hentati et al. | Mobile target tracking mechanisms using unmanned aerial vehicle: Investigations and future directions | |
| CN113326826A (en) | Network model training method and device, electronic equipment and storage medium | |
| CN113359843A (en) | Unmanned aerial vehicle autonomous landing method and device, electronic equipment and storage medium | |
| Vemprala et al. | Representation learning for event-based visuomotor policies | |
| Silva et al. | Landing area recognition by image applied to an autonomous control landing of VTOL aircraft | |
| CN112651374B (en) | Future trajectory prediction method based on social information and automatic driving system | |
| CN112419345A (en) | Patrol car high-precision tracking method based on echo state network | |
| CN111798518A (en) | Mechanical arm posture detection method, device and equipment and computer storage medium | |
| CN115937801A (en) | Vehicle track prediction method and device based on graph convolution | |
| US20230137541A1 (en) | Switching recurrent kalman network | |
| Rahmania et al. | Exploration of the impact of kernel size for yolov5-based object detection on quadcopter | |
| Gunawan et al. | Geometric deep particle filter for motorcycle tracking: Development of intelligent traffic system in Jakarta | |
| Yoon et al. | Learning when to use adaptive adversarial image perturbations against autonomous vehicles | |
| CN109669180B (en) | Continuous wave radar unmanned aerial vehicle detection method | |
| Zhou et al. | A vision-based autonomous detection scheme for obstacles on the runway | |
| Bie et al. | UAV recognition and tracking method based on YOLOv5 | |
| Senanayake et al. | 3D Radar Velocity Maps for Uncertain Dynamic Environments | |
| Liu et al. | Generative AI for Unmanned Vehicle Swarms: Challenges, Applications and Opportunities | |
| Kainth et al. | Chasing the Intruder: A Reinforcement Learning Approach for Tracking Unidentified Drones | |
| Kabore et al. | Deep Learning Based Formation Control of Drones |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200204 |
|
| RJ01 | Rejection of invention patent application after publication |