CN113126647A - Collaborative guidance method based on leader and follower principle - Google Patents
Collaborative guidance method based on leader and follower principle Download PDFInfo
- Publication number
- CN113126647A CN113126647A CN201911421873.9A CN201911421873A CN113126647A CN 113126647 A CN113126647 A CN 113126647A CN 201911421873 A CN201911421873 A CN 201911421873A CN 113126647 A CN113126647 A CN 113126647A
- Authority
- CN
- China
- Prior art keywords
- aircraft
- following
- pilot
- leader
- layer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
Abstract
The invention discloses a collaborative guidance method and a collaborative guidance system based on a leader and follower principle. The collaborative guidance method and the collaborative guidance system based on the leader and follower principle have the advantages that the control process of a plurality of aircrafts during guidance is simplified, the flight path in the guidance process is stable and efficient, the system stability is good, and the like.
Description
Technical Field
The invention relates to a guidance method, in particular to a cooperative guidance method based on a leader and follower principle, and belongs to the field of unmanned aerial vehicle control.
Background
Nowadays, unmanned aerial vehicles are already used in great quantities in engineering practice. During the process of cluster flight of the aircraft, the aircraft needs to be cooperatively controlled.
In the process of forming a flight set by a plurality of aircrafts to carry out long-distance flight (middle-distance guidance), the calculation amount is obviously increased due to the need of simultaneously carrying out guidance, attitude and aircraft cooperative control, so that a relatively simple and efficient cooperative guidance method is needed to carry out long-distance formation flight guidance on the aircrafts.
In addition, the target to which the flight set needs to approach may be in a moving state, and the flight set may not be able to reach the target position after the movement using a predetermined guidance law.
Therefore, it is necessary to develop a method for saving the amount of computation and cooperatively controlling accurate guidance.
Disclosure of Invention
In order to overcome the above problems, the present inventors have conducted intensive studies to design a cooperative guidance method and system based on the principle of leader and follower using BP neural network learning, thereby completing the present invention.
In one aspect, the present invention provides a collaborative guidance system based on a leader-follower principle, the system comprising a plurality of aircraft,
the plurality of aircraft has a pilot aircraft and a plurality of following aircraft.
On the other hand, the invention provides a collaborative guidance method based on the principle that a leader follows a follower, wherein a pilot aircraft and a following aircraft are set, the pilot aircraft conducts guidance, and the following aircraft follows the pilot aircraft to achieve collaborative guidance.
One or more optional aircrafts in the plurality of aircrafts are set as a pilot aircraft, the other aircrafts are set as following aircrafts, after the pilot aircraft is set, the following aircrafts following the pilot aircraft are in communication connection with the pilot aircraft, and the following aircrafts following the same pilot aircraft are in communication with each other.
The guidance of the piloting aircraft is finished by constructing a neural network before takeoff and acquiring guidance parameters by using a neural network learning method in the process of long-distance flight.
The neural network construction comprises the following sub-steps:
s1, making a training sample;
s2, setting a BP neural network model;
and S3, substituting the training samples into the BP neural network model for training to obtain model parameters.
The training sample is provided with a flight path, and the flight path comprises a plurality of groups of flight data, and the flight data comprises: the distance between the aircraft and the target, the sight angle between the aircraft and the target, the moving speed of the target and the proportional guidance law coefficient of the aircraft.
In step S2, the number of input layer nodes of the neural network is 3, which are the distance between the pilot vehicle and the target, the line-of-sight angle between the pilot vehicle and the target, and the current moving speed of the target;
and 1 output layer node is a proportionality coefficient used by a piloting aircraft.
The input level node n is 3, the hidden level node l is 4,
different input layer node pair outputs K of different hidden layer nodesijComprises the following steps:
Kij＝ωijxi
output value L of the hidden layerjComprises the following steps:
the output M of the output layer is:
the neural network is more novel in the learning process as follows:
where i denotes different input level nodes, xiInputs for different input level nodes, where j denotes different hidden level nodes, ωijIs the weight, ω, of the output layer to the hidden layerjIs the weight from hidden layer to output layer, aijIs the biasing of the input layer to the hidden layer, bjIs the bias from the hidden layer to the output layer, δ is the learning rate, g (x) is the excitation function, where ω isi′jFor updated weights, ω, of the output layer to the hidden layerj' updated weight of hidden layer to output layer, ai′jFor updated biasing of the input layer to the hidden layer, bj' is the updated hidden layer to output layer bias.
And the following aircraft keeps the distance from the pilot aircraft and the same flight direction as the pilot aircraft in the process of following the pilot aircraft.
After the pilot vehicle is intercepted, one of the following vehicles can be promoted to the pilot vehicle.
The collaborative guidance method and the collaborative guidance system based on the leader and follower principle have the advantages that:
(1) a BP neural network learning method is adopted, and a guidance law is designed through training and learning of the optimal flight path, so that the flight path of the aircraft in the guidance process is more stable and efficient.
(2) The aircraft in the guidance process is cooperatively controlled by a leader-follower method, so that the control process of a plurality of aircraft in the guidance process is simplified.
(3) The navigation aircraft is abnormal, the following aircraft can be promoted into the navigation aircraft, and the system stability is good.
Drawings
FIG. 1 shows a BP neural network learning architecture diagram of a preferred embodiment;
FIG. 2 is a schematic diagram illustrating a relative spatial relationship between a following aircraft and a pilot aircraft according to a preferred embodiment;
FIG. 3 shows a structural diagram of a leader follower method-based aircraft formation spatial relationship for a preferred embodiment;
fig. 4 shows a structural diagram of an overlay attack trajectory of an aircraft on a target in an embodiment.
Detailed Description
The invention is explained in further detail below with reference to the drawing. The features and advantages of the present invention will become more apparent from the description.
The invention provides a collaborative guidance system based on the principle that a leader follows a follower, which comprises a plurality of aircrafts.
According to the invention, one of the plurality of aircraft is set up as a pilot aircraft and the other aircraft is set up as a following aircraft.
In a preferred embodiment, when the number of clusters of aircraft is high, a plurality of lead aircraft may be set up, each having a plurality of following aircraft to follow,
more preferably, each pilot aircraft has no more than 9 following aircrafts to follow, so that the situation that the number of following aircrafts following the same pilot aircraft is too large and collision occurs between the following aircrafts is avoided, and meanwhile, under the application occasion needing to be concealed, the situation that the number of following aircrafts is limited and the situation that the target of the pilot aircraft is too large can be avoided.
Furthermore, the hardware structures of the piloting aircraft and the following aircraft are the same, so that the following aircraft can be changed into the piloting aircraft after the piloting aircraft is abnormal, and the guidance process is continuously completed.
According to the invention, the aircraft is provided with a distance measuring module, the aircraft obtains the distance between the aircraft and the target through the distance measuring module, the aircraft is also provided with a target measuring module, the target measuring module can measure the movement direction and the movement speed of the target so as to adjust the running state of the aircraft,
in a preferred embodiment, the distance measurement module and the target measurement module are GPS navigation modules, the GPS navigation modules can receive satellite signals, and the distance relationship between the aircraft and the target can be accurately obtained through satellite guidance in a long-distance flight task;
the aircraft is also provided with a line-of-sight angle measuring module, the line-of-sight angle between the aircraft and the target is obtained through the line-of-sight angle measuring module, preferably, the line-of-sight angle measuring module is an observer, and the observation angle obtained by the observer is divided equally to obtain the line-of-sight angle between the aircraft and the target.
In a more preferable embodiment, when the aircraft is far away from the target, the line-of-sight angle between the aircraft and the target is obtained through the GPS navigation module so as to save energy in the cruising stage, and when the aircraft approaches the target, the line-of-sight angle measurement module is started so as to more accurately obtain the real-time line-of-sight angle,
in the invention, the approaching target is 2-5 km, preferably 3km, of the distance between the aircraft and the target.
According to the invention, the communication module is also arranged on the aircraft, so that the following aircraft can communicate with each other and communicate with the leader aircraft, when the leader is abnormal, the following aircraft can determine the leader abnormality in time and generate a new leader aircraft from the following aircraft,
further, the communication module can also enable the aircraft to be in real-time communication with the ground control station, and the aircraft obtains the position relation between the aircraft and other aircraft through communication.
On the other hand, the invention provides a collaborative guidance method based on the principle that a leader follows a follower.
According to the invention, the aircraft is guided and controlled by a proportional guidance method, wherein the proportional guidance law is
Wherein N is a proportionality coefficient, VmIn order to pilot the speed of the aircraft,
is the bullet eye line-of-sight angular rate of the piloted aircraft.
When the proportionality coefficient is smaller, the motion trail of the aircraft is closer to the tracking method, and the target can be tracked with larger overload; when the proportionality coefficient is larger, the aircraft is closer to a parallel approach method, is suitable for coping with a target with more determined motion information,
in a preferred embodiment, the value of the proportionality coefficient N is 2-6, and the inventor finds that in most aircraft adopting proportionality guidance, when the proportionality guidance coefficient is less than 2, the acceleration of the aircraft diverges, which leads to the aircraft being out of control; when the proportionality coefficient is larger than 6, the available overload of the aircraft can be exceeded, the control effect of the aircraft is influenced, the control stability of the aircraft is ensured, the method provided by the invention can be applied to most of aircraft, and the application range is expanded.
In the invention, the guidance parameters are acquired by constructing the neural network before take-off and utilizing the neural network learning method in the remote flight process, thereby completing the guidance of the piloting aircraft.
Preferably, the neural network learning method is a BP neural network learning method.
A guidance law is designed by learning the optimal flight path by using the BP neural network, so that the flight path of the aircraft in the guidance process is more stable and efficient.
Specifically, the building of the neural network includes the following sub-steps:
and S1, making a training sample.
And designing an optimal flight path of the aircraft according to the characteristics and the application scene of the aircraft, taking the flight path as a training sample, wherein the optimal flight path is considered as the optimal flight path of the aircraft by a designer.
In a preferred embodiment, the optimal flight path is according to the classic case in the field of practice.
Further, a plurality of sets of flight data are included in the flight path, the flight data including: the distance between the aircraft and the target, the sight angle between the aircraft and the target, the moving speed of the target and the proportional guidance law coefficient of the aircraft.
And S2, setting a BP neural network model.
The BP neural network is a multi-layer feedforward network trained according to an error inverse propagation algorithm, and the BP network can learn and store a large number of input-output mode mapping relations without disclosing a mathematical equation describing the mapping relations in advance. The learning rule is that the steepest descent method is used, and the weight and the threshold value of the network are continuously adjusted through back propagation, so that the error square sum of the network is minimum. The BP neural network model topological structure comprises an input layer, a hidden layer and an output layer.
According to the invention, the number of input layer nodes of the neural network is 3, which are respectively the distance between the aircraft and the target, the sight angle between the aircraft and the target and the current moving speed of the target; and 1 output layer node is a proportionality coefficient used by the aircraft.
In the BP neural network, the number of nodes of the hidden layer is crucial and directly influences the performance of the neural network, the inventor finds that when the nodes of the hidden layer are too small, the training requirement cannot be met, and when the nodes of the hidden layer are too much, the operation speed is too slow, so that the guidance effect is influenced,
in the invention, the number of the hidden layer nodes is 3-6, preferably 4, so that the required precision requirement can be trained, the calculated amount can be reduced, and the time cost can be saved.
In the invention, as the number of input layer nodes n is 3, the number of hidden layer nodes l is 4, i represents different input layer nodes, and j represents different hidden layer nodes, i is 1,2, and 3; j is 1,2,3,4, as shown in fig. 1, wherein xiIs an input layer node, omegaijIs the weight, ω, of the output layer to the hidden layerjIs the weight from hidden layer to output layer, aijIs the biasing of the input layer to the hidden layer, bjIs the bias from the hidden layer to the output layer, δ is the learning rate, and g (x) is the excitation function.
According to the invention, different input level nodes xiFor different implicationsOutput K of layer nodeijComprises the following steps:
Kij＝ωijxi
in a preferred embodiment, the excitation function g (x) is a Sigmoid function, which has more precise function and better fault tolerance, and the output value L of the hidden layerjComprises the following steps:
further, the output M of the output layer is:
and S3, substituting the training samples into the BP neural network model for training to obtain model parameters.
Calculating the BP neural network model by using the training sample, specifically, taking the distance between the aircraft and the target, the sight line angle between the aircraft and the target and the moving speed of the target in the flight data as an input layer xiTaking the aircraft proportion guidance law coefficient in the flight data as the expected output Y,
in the training process, the weight omega from the output layer to the hidden layerijWeight ω from hidden layer to output layerjBias of input layer to hidden layer aijBias b from hidden layer to output layerjThe updating is carried out continuously, and the updating is carried out continuously,
further, the continuously updating is performed by the following equation:
wherein ω isi′jFor updated weights, ω, of the output layer to the hidden layerj' updated weight of hidden layer to output layer, ai′jFor updated biasing of the input layer to the hidden layer, bjIs' updatedThe hidden layer to output layer bias, e-Y-M.
A final BP neural network model can be obtained through training of a plurality of groups of flight data,
in a preferred embodiment, the trained BP neural network model is determined to meet the requirement by observing a calculation error E, which is 0.5E2，
Further, when E is less than 0.1Y, the BP neural network model is considered to meet the training requirement, and the BP neural network model can be used by the aircraft in the process of guidance and is solidified in the aircraft. In a preferred embodiment, the BP neural network model meeting the training requirements is consolidated into all aircraft, so that any one aircraft can be set up as a pilot aircraft to perform a pilot guidance task.
In another preferred embodiment, the BP neural network model meeting the training requirements is consolidated into a predetermined number of aircraft, so that only a part of the aircraft can be set up as a pilot aircraft and the rest of the aircraft can be used as follow-up aircraft, thereby reducing the configuration requirements of the follow-up aircraft and further reducing the manufacturing cost.
According to the cooperative guidance method based on the leader and follower principle, a pilot aircraft is set before the aircraft takes off.
One or more of the plurality of aircraft is optionally configured as a pilot aircraft, and the remaining aircraft are configured as follower aircraft.
In the invention, different numbers of pilot aircrafts can be set according to different aircraft cluster sizes so as to deal with the burst problem.
In a preferred embodiment, one pilot aircraft is set up for every 4-10 aircrafts, and each pilot aircraft has 3-9 following aircrafts to follow.
After the pilot aircraft is set up, the following aircraft following the pilot aircraft establishes communication connection with the pilot aircraft, and the following aircraft following the same pilot aircraft establishes communication with each other.
According to the invention, the piloting vehicle acquires guidance parameters by using a neural network model. Specifically, the takeoff position of the pilot aircraft and the position of the target are set.
Due to the possible reasons of target movement or detection precision and the like, the position of the target is a predicted value, and the error between the predicted position of the target and the actual position of the target is not more than 100 m.
In a preferred embodiment, the takeoff position and the target position of the pilot aircraft are provided by a GPS navigation module of the pilot aircraft.
And in the flight process of the piloting aircraft, outputting a proportionality coefficient of a proportional guidance law of the piloting aircraft by using a neural network model, and controlling the piloting aircraft to fly.
Further, in the input of the neural network model, the distance between the piloting vehicle and the target is obtained from the distance measurement module, the line-of-sight angle between the piloting vehicle and the target is obtained from the GPS navigation module and/or the line-of-sight angle measurement module, and the moving speed of the target is obtained from the target measurement module. Because the BP neural network is used for continuously correcting the actual position of the target, the pilot aircraft can obtain the optimal flight path and accurately reach the actual position of the target.
According to the invention, the following aircraft can fly along with the piloting aircraft by constructing the following model.
The following model is constructed, namely a proportion guidance law of the following aircraft is obtained by constructing a dynamic model, so that the following aircraft follows around the piloting aircraft, as shown in fig. 2.
In the invention, because the following aircraft and the pilot aircraft have the same structure and fly along with the pilot aircraft, the proportion guidance laws of the following aircraft and the pilot aircraft are the same, and the overload control formula is as follows:
wherein V is the following aircraft speed,
the bullet eye line-of-sight angular rate.
Before the aircraft takes off, setting the preset positions of different following aircrafts,
specifically, the position vector between the preset position of the following aircraft and the piloting aircraft is L, the included angle between the speed of the piloting aircraft and the position vector L is eta, and the preset position of the following aircraft can be determined by setting L and eta in advance.
Further, the proportionality coefficient N of the following aircraft during flight may be the same as the lead aircraft, or may be a fixed value. The specific selection is determined according to the position relation between the following aircraft and the pilot aircraft, the distance between the following aircraft and the pilot aircraft is far, the smaller the proportionality coefficient is, and the preferable proportionality coefficient is 3-5.
In a preferred embodiment, the proportionality coefficient N of the following aircraft is a fixed value in the formation process, and the fixed proportionality coefficient can avoid selecting an unreasonable proportionality coefficient on the premise of known aircraft performance, so as to affect the following effect.
In the invention, the following aircraft needs to keep the distance between the following aircraft and the pilot aircraft and the same flight direction as the pilot aircraft in the process of following the pilot aircraft.
Specifically, the following aircraft tracks its predetermined position through the guidance law, reaches the predetermined position, and flies with the pilot aircraft. In the following process, the following aircraft uses the same overload instruction as the pilot aircraft to achieve the synchronization purpose.
During the flight of the following aircraft, the following aircraft targets its predetermined position at a line-of-sight angle e relative to the predetermined position.
In a preferred embodiment, the sight angle epsilon is obtained by resolving through a ground control system, and since the preset position information of the following aircraft can be obtained through position calculation of the pilot aircraft, the position and the speed of the following aircraft can also be obtained in real time in the ground control system. The sight angle epsilon is calculated by the ground control system, so that the use frequency of the sight angle measuring module of the following aircraft is reduced, the control precision of each unit in the formation process is improved, and the formation flight cost and energy consumption are reduced.
According to the invention, the following aircraft needs to be corrected when it is not in the predetermined position. Specifically, when the following aircraft is monitored by the ground control system to be beyond the preset range thereof, the following aircraft does not keep the same overload instruction as the pilot aircraft, and independent guidance control is carried out by taking the preset position as a target again.
The preset range is the allowable deviation between the positions of the following aircrafts and the position of the pilot aircraft, and can be flexibly adjusted according to the number of the following aircrafts, and preferably, the preset range is L +/-10 m; eta + -5 deg.
Further, when a plurality of following aircrafts follow one pilot aircraft, the distances L between different following aircrafts and the pilot aircraftiDifferent, different following aircraft epsiloniAlso different, where i represents a different following aircraft.
The inventor finds that too close distance between the following aircraft and the pilot aircraft easily causes the following aircraft to collide with the leading aircraft, and too far distance causes small-angle attitude change of the pilot aircraft to have too large influence on the preset position of the following aircraft, so that the following aircraft needs to switch the following mode frequently, and in a preferred embodiment, the set range of the L is 90 m-110 m to ensure the following effect of the following aircraft.
In a preferred embodiment, the distance L between the following aircraft and the pilot aircraft is determined according to the working range of the aircraft, and L is 70% -90% of the working range of the aircraft, so that the cooperative guidance system consisting of the pilot aircraft and the following aircraft can realize full coverage on the target and a certain area around the target.
In a preferred embodiment, the following aircraft is also able to measure the distance S to other following aircraft and process it as a monitored parameter to avoid collisions between the following aircraft. And S is obtained through a ground control system, when the distance S between the two following aircrafts is too small, at least one aircraft is far away from the set position, the positions of the two aircrafts are monitored, the following aircrafts far away from the set position are controlled to not keep the same overload instruction as the pilot aircraft, and independent guidance control is carried out by taking the preset position as a target again.
In a preferred embodiment, the piloting aircraft is located at the front end of the following aircraft, so that the following aircraft does not block the measurement of the distance measurement module and the line-of-sight angle measurement module of the piloting aircraft, and the target is prevented from being lost to influence the guidance precision of the piloting aircraft.
In a further preferred embodiment, the following vehicles can be distributed uniformly around the pilot vehicle, more preferably, in a spherical area centered on the pilot vehicle and having a radius L.
In the invention, if the pilot aircraft is intercepted, any following aircraft with the BP neural network model can be promoted to be the pilot aircraft, guidance is carried out according to the guidance mode of the pilot aircraft, and communication is established with other following aircraft, so that other following aircraft can fly along with the following aircraft.
In a preferred embodiment, when the pilot vehicle is intercepted, the following vehicle with the smallest distance from the flight path of the following vehicle is promoted to be the new pilot vehicle, so that the magnitude of the flight path adjustment of the new pilot vehicle is the smallest, and the system can be rapidly stabilized, as shown in fig. 3, when the pilot vehicle is intercepted, the following vehicle 2 is promoted to be the new pilot vehicle.
Examples
Example 1
The simulation experiment was set as follows: and 3 aircrafts form a cooperative guidance system, one of the aircrafts is a pilot aircraft, the other aircraft is a following aircraft, the aircrafts need to perform close-range tracking investigation on targets beyond 4000m, the targets fly at a constant speed of 50m/s and horizontally fly along the negative direction of the x axis, the initial speed of the pilot aircraft is 200m/s, and the initial line-of-sight angle is 45 degrees. The initial position of the pilot aircraft is 0, the value of a position vector L between the preset position of the following aircraft and the pilot aircraft is 100m,
in the flight process, the distance between the following aircraft and the pilot aircraft is always 100 +/-10 m, and the position relation between the aircraft of the guidance terminal and the target is as follows:
| location information | | Following aircraft | 1 | Piloting | Following aircraft | 2 |
| Horizontal position/m | 1966.43 | 2063.52 | 1964.56 | 1866.28 | ||
| Vertical position/m | 2828.43 | 2825.27 | 2826.42 | 2824.42 |
As shown in fig. 4, the pilot vehicle can track the target, plan an optimal flight path to reach the moving target position, and the following vehicle can follow the pilot vehicle and maintain the distance from the pilot vehicle.
In the description of the present invention, it should be noted that the terms "upper", "lower", "inner" and "outer" indicate the orientation or positional relationship based on the operation state of the present invention, and are only for convenience of description and simplification of description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and be operated, and thus should not be construed as limiting the present invention.
The present invention has been described above in connection with preferred embodiments, but these embodiments are merely exemplary and merely illustrative. On the basis of the above, the invention can be subjected to various substitutions and modifications, and the substitutions and the modifications are all within the protection scope of the invention.
Claims (10)
1. A cooperative guidance system based on leader-follower principle, the system comprising a plurality of aircraft,
the plurality of aircraft has a pilot aircraft and a plurality of following aircraft.
2. A collaborative guidance method based on a leader and follower principle is characterized in that a pilot aircraft and a following aircraft are set, the pilot aircraft conducts guidance, and the following aircraft follows the pilot aircraft to achieve collaborative guidance.
3. The collaborative guidance method based on leader-follower principle according to claim 2,
one or more optional aircrafts in the plurality of aircrafts are set as a pilot aircraft, the other aircrafts are set as following aircrafts, after the pilot aircraft is set, the following aircrafts following the pilot aircraft are in communication connection with the pilot aircraft, and the following aircrafts following the same pilot aircraft are in communication with each other.
4. The collaborative guidance method based on leader-follower principle according to claim 2,
the guidance of the piloting aircraft is finished by constructing a neural network before takeoff and acquiring guidance parameters by using a neural network learning method in the process of long-distance flight.
5. The collaborative guidance method based on leader-follower principle according to claim 4,
the neural network construction comprises the following sub-steps:
s1, making a training sample;
s2, setting a BP neural network model;
and S3, substituting the training samples into the BP neural network model for training to obtain model parameters.
6. The collaborative guidance method based on leader-follower principle according to claim 5,
the training sample is provided with a flight path, and the flight path comprises a plurality of groups of flight data, and the flight data comprises: the distance between the aircraft and the target, the sight angle between the aircraft and the target, the moving speed of the target and the proportional guidance law coefficient of the aircraft.
7. The collaborative guidance method based on leader-follower principle according to claim 5,
in step S2, the number of input layer nodes of the neural network is 3, which are the distance between the pilot vehicle and the target, the line-of-sight angle between the pilot vehicle and the target, and the current moving speed of the target;
and 1 output layer node is a proportionality coefficient used by a piloting aircraft.
8. The collaborative guidance method based on leader-follower principle according to claim 5,
the input level node n is 3, the hidden level node l is 4,
different input layer node pair outputs K of different hidden layer nodesijComprises the following steps:
Kij＝ωijxi
output value L of the hidden layerjComprises the following steps:
the output M of the output layer is:
the neural network is more novel in the learning process as follows:
where i denotes different input level nodes, xiInputs for different input level nodes, where j denotes different hidden level nodes, ωijIs the weight, ω, of the output layer to the hidden layerjIs the weight from hidden layer to output layer, aijIs the biasing of the input layer to the hidden layer, bjIs the bias from hidden layer to output layer, δ is the learning rate, g (x) is the excitation function, ω'ijIs the updated weight, ω ', of the output layer to the hidden layer'jIs the updated weight, a ', from hidden layer to output layer'ijFor updated bias of input layer to implicit layer, b'jIs the updated hidden layer to output layer bias.
9. The collaborative guidance method based on leader-follower principle according to claim 2,
and the following aircraft keeps the distance from the pilot aircraft and the same flight direction as the pilot aircraft in the process of following the pilot aircraft.
10. The collaborative guidance method based on leader-follower principle according to claim 2,
after the pilot vehicle is intercepted, one of the following vehicles can be promoted to the pilot vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911421873.9A CN113126647B (en) | 2019-12-31 | 2019-12-31 | Collaborative guidance method based on leader and follower principle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911421873.9A CN113126647B (en) | 2019-12-31 | 2019-12-31 | Collaborative guidance method based on leader and follower principle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113126647A true CN113126647A (en) | 2021-07-16 |
| CN113126647B CN113126647B (en) | 2022-07-19 |
Family
ID=76770694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911421873.9A Active CN113126647B (en) | 2019-12-31 | 2019-12-31 | Collaborative guidance method based on leader and follower principle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113126647B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113126647B (en) * | 2019-12-31 | 2022-07-19 | 北京理工大学 | Collaborative guidance method based on leader and follower principle |
| CN115047877A (en) * | 2022-06-08 | 2022-09-13 | 中国船舶集团有限公司系统工程研究院 | Unmanned ship target tracking method and system based on proportional guidance |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5521817A (en) * | 1994-08-08 | 1996-05-28 | Honeywell Inc. | Airborne drone formation control system |
| CN104238552A (en) * | 2014-09-19 | 2014-12-24 | 南京理工大学 | Redundancy multi-robot forming system |
| CN105527960A (en) * | 2015-12-18 | 2016-04-27 | 燕山大学 | Mobile robot formation control method based on leader-follow |
| US20180074520A1 (en) * | 2016-09-13 | 2018-03-15 | Arrowonics Technologies Ltd. | Formation flight path coordination of unmanned aerial vehicles |
| CN108827312A (en) * | 2018-08-08 | 2018-11-16 | 清华大学 | A kind of coordinating game model paths planning method based on neural network and Artificial Potential Field |
| CN109144076A (en) * | 2018-10-31 | 2019-01-04 | 吉林大学 | A kind of more vehicle transverse and longitudinals coupling cooperative control system and control method |
| EP3438950A1 (en) * | 2017-08-03 | 2019-02-06 | Airbus Operations S.A.S. | Method for anticipating the movement of a wake vortex in a formation flight of two aircraft |
| CN109712424A (en) * | 2019-02-26 | 2019-05-03 | 辽宁工业大学 | A kind of automobile navigation method based on Internet of Things |
| CN110398975A (en) * | 2019-09-04 | 2019-11-01 | 西北工业大学 | A kind of navigator's follower type multiple aircraft formation fault tolerant control method based on broadcast operation framework |
| CN113568425A (en) * | 2020-04-28 | 2021-10-29 | 北京理工大学 | Cluster cooperative guidance method based on neural network learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113126647B (en) * | 2019-12-31 | 2022-07-19 | 北京理工大学 | Collaborative guidance method based on leader and follower principle |
-
2019
- 2019-12-31 CN CN201911421873.9A patent/CN113126647B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5521817A (en) * | 1994-08-08 | 1996-05-28 | Honeywell Inc. | Airborne drone formation control system |
| CN104238552A (en) * | 2014-09-19 | 2014-12-24 | 南京理工大学 | Redundancy multi-robot forming system |
| CN105527960A (en) * | 2015-12-18 | 2016-04-27 | 燕山大学 | Mobile robot formation control method based on leader-follow |
| US20180074520A1 (en) * | 2016-09-13 | 2018-03-15 | Arrowonics Technologies Ltd. | Formation flight path coordination of unmanned aerial vehicles |
| EP3438950A1 (en) * | 2017-08-03 | 2019-02-06 | Airbus Operations S.A.S. | Method for anticipating the movement of a wake vortex in a formation flight of two aircraft |
| CN108827312A (en) * | 2018-08-08 | 2018-11-16 | 清华大学 | A kind of coordinating game model paths planning method based on neural network and Artificial Potential Field |
| CN109144076A (en) * | 2018-10-31 | 2019-01-04 | 吉林大学 | A kind of more vehicle transverse and longitudinals coupling cooperative control system and control method |
| CN109712424A (en) * | 2019-02-26 | 2019-05-03 | 辽宁工业大学 | A kind of automobile navigation method based on Internet of Things |
| CN110398975A (en) * | 2019-09-04 | 2019-11-01 | 西北工业大学 | A kind of navigator's follower type multiple aircraft formation fault tolerant control method based on broadcast operation framework |
| CN113568425A (en) * | 2020-04-28 | 2021-10-29 | 北京理工大学 | Cluster cooperative guidance method based on neural network learning |
Non-Patent Citations (9)
| Title |
|---|
| HAIZHAOLIANG 等: "《Guidance strategies for interceptor against active defense spacecraft in two-on-two engagement》", 《AEROSPACESCIENCEANDTECHNOLOGY》 * |
| RUNLEDU 等: "《Design of three-dimensional nonlinear guidance law with bounded acceleration command》", 《AEROSPACE SCIENCE AND TECHNOLOGY》 * |
| SHICHUN YANG 等: "《Distributed formation control of nonholonomic autonomous vehicle via RBF neural network》", 《MECHANICAL SYSTEMS AND SIGNAL PROCESSING》 * |
| XIA CHEN 等: "《Tracking Control of UAV in Formation Based on FHM》", 《PROCEEDINGS OF THE 37TH CHINESE CONTROL CONFERENCE》 * |
| YAN YU 等: "《Design and Optimization of Press Bend Forming Path for Producing Aircraft Integral Panels with Compound Curvatures》", 《CHINESE JOURNAL OF AERONAUTICS》 * |
| 任章 等: "《飞行器集群协同制导控制方法及应用研究》", 《导航定位与授时》 * |
| 冯运铎 等: "《一种分布式多无人机协同定距盘旋跟踪制导律》", 《兵工学报》 * |
| 方科 等: "《高超声速飞行器时间协同再入制导》", 《航空学报》 * |
| 陈昶荣 等: "《基于神经网络的飞行器协同编队控制研究》", 《控制与信息技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113126647B (en) * | 2019-12-31 | 2022-07-19 | 北京理工大学 | Collaborative guidance method based on leader and follower principle |
| CN115047877A (en) * | 2022-06-08 | 2022-09-13 | 中国船舶集团有限公司系统工程研究院 | Unmanned ship target tracking method and system based on proportional guidance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113126647B (en) | 2022-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108549407B (en) | Control algorithm for multi-unmanned aerial vehicle cooperative formation obstacle avoidance | |
| Oh et al. | Decentralised standoff tracking of moving targets using adaptive sliding mode control for UAVs | |
| Chen et al. | Path planning for multi-UAV formation | |
| CN105022401A (en) | SLAM method through cooperation of multiple quadrotor unmanned planes based on vision | |
| CN102591358B (en) | Multi-UAV (unmanned aerial vehicle) dynamic formation control method | |
| Tisdale et al. | Autonomous UAV path planning and estimation | |
| Tandale et al. | Trajectory tracking controller for vision-based probe and drogue autonomous aerial refueling | |
| CN112684807A (en) | Unmanned aerial vehicle cluster three-dimensional formation method | |
| US20100121503A1 (en) | Collision avoidance system and a method for determining an escape manoeuvre trajectory for collision avoidance | |
| Lai et al. | On-board trajectory generation for collision avoidance in unmanned aerial vehicles | |
| CN111897316A (en) | Multi-aircraft autonomous decision-making method under scene fast-changing condition | |
| CN113126647B (en) | Collaborative guidance method based on leader and follower principle | |
| Oh et al. | Coordinated standoff tracking of groups of moving targets using multiple UAVs | |
| Frew et al. | Obstacle avoidance with sensor uncertainty for small unmanned aircraft | |
| CN111811513B (en) | Flight path planning method for cooperative coverage and obstacle avoidance of multiple unmanned aerial vehicles | |
| CN107957686B (en) | Unmanned helicopter auto landing on deck control system based on prediction control | |
| CN109901387A (en) | A kind of automatic near-earth anti-collision system Self-adaptive flight trajectory predictions method of aircraft | |
| CN110793522B (en) | Flight path planning method based on ant colony algorithm | |
| Bryson et al. | Active airborne localisation and exploration in unknown environments using inertial SLAM | |
| Bodi et al. | Reinforcement learning based UAV formation control in GPS-denied environment | |
| Yang et al. | A decentralised control strategy for formation flight of unmanned aerial vehicles | |
| Zhang et al. | UAV formation flight cooperative tracking controller design | |
| CN109857146B (en) | Layered unmanned aerial vehicle tracking control method based on feedforward and weight distribution | |
| CN110162084B (en) | Formation control method of flying missile cluster system based on consistency theory | |
| CN116719239A (en) | Trace underactuated satellite incomplete information tracking game control method |
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 |