CN111132258B - Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method - Google Patents
Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method Download PDFInfo
- Publication number
- CN111132258B CN111132258B CN201911423735.4A CN201911423735A CN111132258B CN 111132258 B CN111132258 B CN 111132258B CN 201911423735 A CN201911423735 A CN 201911423735A CN 111132258 B CN111132258 B CN 111132258B
- Authority
- CN
- China
- Prior art keywords
- node
- candidate relay
- unmanned aerial
- aerial vehicle
- nodes
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
- H04W40/22—Communication route or path selection, e.g. power-based or shortest path routing using selective relaying for reaching a BTS [Base Transceiver Station] or an access point
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
- H04W40/04—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/02—Hierarchically pre-organised networks, e.g. paging networks, cellular networks, WLAN [Wireless Local Area Network] or WLL [Wireless Local Loop]
- H04W84/04—Large scale networks; Deep hierarchical networks
- H04W84/08—Trunked mobile radio systems
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention discloses an unmanned aerial vehicle cluster collaborative opportunity routing method based on a virtual potential field method. The method introduces the concept of virtual force in a virtual potential field method, the virtual force is divided into topological force, navigation force and barrier force again, the three force vectors are added to obtain resultant force, and the motion state of the unmanned plane node is depicted through the resultant force. The method is used for introducing the node advancing distance, the virtual force direction and the unmanned aerial vehicle node residual energy into the selection of a next hop forwarding node as key parameters. The method comprises three key steps of candidate relay set selection, optimal candidate relay node selection, data forwarding and confirmation. Simulation experiments in a simulation environment based on MATLAB prove the superiority of the opportunistic routing method in an unmanned aerial vehicle cluster network environment.
Description
Technical Field
The invention belongs to the field of wireless networks, and particularly relates to an unmanned aerial vehicle cluster collaborative opportunity routing method based on a virtual potential field method.
Background
The high mobility of the nodes in the unmanned aerial vehicle cluster brings the results of rapid change of network topology, short service life of links and the like, so that the routing information in the unmanned aerial vehicle ad hoc network is frequently changed. Meanwhile, network nodes in the unmanned aerial vehicle ad hoc network are easy to destroy, and challenges are presented to the design of a routing protocol. The existing traditional routing protocols applied to mobile ad hoc networks (MANET) and vehicle ad hoc networks (VANET) cannot be well adapted to unmanned aerial vehicle ad hoc networks.
Virtual potential field methods, also known as artificial potential field methods, were originally applied to robot navigation-related studies. The basic idea of the virtual potential field method is to abstract the environment where the agent is located into a virtual potential field, and the agent moves under the action of the force in the virtual potential field. The forces to which an intelligent individual is subjected can be divided into two categories: attraction and repulsion. The attractive force and the repulsive force are overlapped to form resultant force, and the individual is directed towards the eyes under the action of the resultant forcePunctuation movement. Attraction force F applied to current position of virtual potential field by intelligent individual A And repulsive force F R Respectively by gravitation potential field function U A And repulsive potential field function U R And carrying out negative gradient solution. However, in a large-scale unmanned aerial vehicle cluster scene, as the unmanned aerial vehicle has high movement speed, the number of unmanned aerial vehicles in the cluster is large, and the change of the cluster topology structure rapidly leads to complex and changeable network scenes. Based on the original gravitation and repulsion, the types of forces possibly received by a single unmanned aerial vehicle in the unmanned aerial vehicle cluster are divided again, virtual forces received by the single unmanned aerial vehicle are divided into three parts of navigation force, topology force and barrier force, the three forces are overlapped to form resultant forces finally received by the unmanned aerial vehicle, and the resultant force direction is the movement direction of the unmanned aerial vehicle. And after normalization calculation is carried out on the received resultant force, the movement speed and the movement direction of the unmanned aerial vehicle are obtained. The method comprises the following specific steps:
(1) Topology force
The unmanned aerial vehicle cluster is composed of a large number of unmanned aerial vehicle nodes, and two unmanned aerial vehicles within a certain range can generate interaction force. The interaction force between unmanned aerial vehicles can be called as 'topological force', the topological force plays a key role in maintaining the topological structure of the clusters, and collision accidents caused by too close distance between unmanned aerial vehicles can be effectively avoided. Fig. 1 shows a schematic diagram of the topology forces to which the nodes in the unmanned aerial vehicle cluster network are subjected. The attractive force and repulsive force between unmanned aerial vehicles can be expressed as:
wherein k is A And k R The gravity gain coefficient and the repulsion gain coefficient are respectively shown, and D (u, v) is the Euclidean distance between the unmanned aerial vehicles u and v. At this time, for the unmanned plane u, the topology force it receives can be expressed as:
wherein the gravitational weight alpha A Repulsive force weight alpha R The method meets the following conditions:
by adjusting the gravitational weight alpha A Repulsive force weight alpha R The size relation between the two can obtain the best topological force effect under different cluster environments.
(2) Navigation force
When the unmanned aerial vehicle is in the motion process, a point where the unmanned aerial vehicle wants to move arrives exists, and the point can be a temporary point in the path planning process or an end point of the whole cluster motion. The navigation force in the virtual force is a force for driving the unmanned aerial vehicle to reach a specified place, and has the function of navigation. In fig. 2, a motion path has been planned for a certain unmanned plane u, which is currently at point P, and its coordinates are P (x p ，y p ，z p ) The drone knows that it should be in P' at the next moment. The navigation force at this time can be expressed as:
f in the formula G For navigational force, the direction of force is the vector direction, k, of P to P G Is the navigation force gain coefficient. As can be seen from the above formula, the navigation force and the navigation force gain factor are related to the euclidean distance of the starting position. The navigation force is greatest at the beginning of the movement and becomes smaller as it approaches the intermediate temporary point P' in order to form a well-stabilized network topology before the next movement is performed. If the unmanned aerial vehicle does not conduct path planning in advance, the unmanned aerial vehicle can directly move to the terminal under the action of navigation force from the terminal.
(3) Force of obstruction
When the unmanned aerial vehicle cluster moves in an open and barrier-free field, only navigation force and topology are neededThe force and the force are subjected to vector superposition, so that the stress condition of the single unmanned aerial vehicle can be better depicted. However, in a real business scenario, the unmanned aerial vehicle cluster often encounters obstacles such as peaks, buildings and the like during traveling. Meanwhile, when the unmanned aerial vehicle cluster performs military operation, an electromagnetic interference area set by enemy force needs to be effectively avoided. In the virtual potential field method, an area threatening the unmanned aerial vehicle cluster is abstracted into a geometrical obstacle, and the force obstructing the travel of the unmanned aerial vehicle cluster is defined as an obstacle force, namely, the position of the obstacle cannot be close to the unmanned aerial vehicle, and the unmanned aerial vehicle needs to avoid detouring. As shown in fig. 3, once the drone comes into the area of the obstacle, the rack is considered to have been damaged. The distance between the unmanned aerial vehicle and the two unmanned aerial vehicles in the topological force is smaller than R Opt When the repulsive force is generated, the obstacle force acts on the unmanned aerial vehicle in a repulsive mode. In the virtual potential field, a set of obstacles around the drone u is defined as O (u). Defining the safety distance of the obstacle as R Safe Namely, when the distance between the unmanned plane and the center is required to be kept to be more than or equal to R Safe . When the unmanned aerial vehicle moves to the center of the obstacle, the distance between the unmanned aerial vehicle and the center of the obstacle is smaller than R Safe When the drone is considered to have been damaged. Definition of the range of action of the obstacle R Sense I.e. when the unmanned aerial vehicle moves to a distance from the centre of the obstacle less than R Sense When the unmanned aerial vehicle is in use, the unmanned aerial vehicle can move away from the obstacle due to the repulsive force from the obstacle. When the distance between the unmanned aerial vehicle and the center of the obstacle is larger than R Sense In this case, the unmanned aerial vehicle is considered to be not subjected to repulsive force from the obstacle. For a single frame unmanned aerial vehicle u, it is subjected to an obstacle force F from an obstacle ob Ob It can be expressed as the following, the direction of the obstacle force is u in the opposite direction to the center of the obstacle. The obstacle forces experienced by the unmanned cluster nodes may be expressed as:
in formula (6), k O And D (u, ob) is the distance between the current unmanned aerial vehicle and the center of the obstacle, and is the obstacle force gain coefficient. All obstacles to the unmanned aerial vehicle uThe sum of the forces is:
(4) Resultant force
Vector addition of the three forces can calculate the final virtual force resultant force F of the unmanned aerial vehicle in the virtual potential field RS The method comprises the following steps:
wherein beta is G 、β T And beta O Navigation force weight coefficient, topology force weight coefficient and obstacle force weight coefficient respectively. The three coefficients all have values between 0 and 1, and their sum is 1, i.e.:
in different task scenarios, β can be varied G 、β T And beta O And the ratio relation among the three is used for planning an optimal action path for the unmanned aerial vehicle cluster. Virtual force is a virtual concept, and the magnitude of the force cannot be intuitively perceived. From the above definition, the unmanned aerial vehicle in the virtual potential field is finally subjected to virtual force resultant force F RS Ranging from 0 to plus infinity. The movement speed of the unmanned aerial vehicle is 0, V max ]Within the interval of V max Is the maximum value of the movement speed of the unmanned aerial vehicle. We need to map the virtual force magnitude experienced by the drone to the corresponding speed magnitude. To accomplish this mapping, the magnitude of the virtual force needs to be normalized first. Here normalization is performed using an arctan () function. The arctan () is one of the inverse trigonometric functions, whose domain is the whole real number R and whose value is (-pi/2, pi/2). For the unmanned plane u, the mapping relationship between the virtual force and the speed is as follows:
disclosure of Invention
Aiming at the characteristic that the unmanned aerial vehicle cluster network nodes move at a high speed and the cluster topology changes rapidly, the unmanned aerial vehicle collaborative opportunity routing method based on the virtual potential field method is provided. The method introduces a virtual potential field method into the design consideration of a routing protocol, takes factors such as virtual force, advancing distance, unmanned plane node energy and the like as key parameters of the next hop selection of the routing, and adopts the following steps:
step 1: when the period starts, the destination node can broadcast its own state information in the whole network, and the neighbor nodes can exchange their position information with each other, so that other nodes can make routing decisions. In the candidate relay set selection stage, the nodes carrying data need to perform route discovery. This node will send a Route Request (RREQ) to other nodes within its communication range. If no other node in the communication range of the node can forward data, the node enters a storage-carrying mode, continues to move, and tries to perform route discovery again when the next movement period starts. If there are other nodes in the communication range of the node, the set formed by the nodes is the candidate relay set of the node. After receiving the route request, the nodes in the candidate relay set can perform priority measurement calculation according to the node states of the nodes. When the timer of the candidate relay node expires, the node broadcasts a Route Reply (RREP). When the forward distance Mar () of the candidate relay node is negative, it indicates that the direction of the candidate relay node is opposite to the direction of the destination node, and the node will not perform route reply in order to avoid data transmission in the opposite direction. Similarly, when the remaining energy of a candidate relay node is lower than a preset minimum threshold E min The node will also discard the route reply. The node carrying the data decides the priority order according to the received route response, and determines the best candidate relay node.
Step 2: for a node to forward data, the node may prioritize according to the received routing replies. Since the timer value of the candidate relay node with the largest priority metric result is the smallest, the candidate relay node with the first route reply is generally taken as the best candidate relay node. However, due to the high-speed mobility and the rapid topology change of the nodes of the unmanned aerial vehicle cluster network, a routing loop phenomenon can occur, and data packets can be repeatedly transmitted among a plurality of nodes, so that the waste of network resources is caused. In order to avoid the phenomenon of routing loops, nodes in the candidate relay set need to be filtered. If the nodes in the candidate relay set have participated in data forwarding in the previous path, the nodes which have participated in data forwarding are screened out of the candidate relay set. And the node to be forwarded with the highest priority is selected from the filtered candidate relay set to forward the data. However, in the course of the movement of the unmanned aerial vehicle cluster, the obstacle in the scene may cause the cluster to have a short split, and the candidate relay node may not exist in the candidate relay set after filtering. When the node carrying the data has no candidate relay set point, the node enters a storage-carrying mode, continues to move, and tries route discovery again when waiting for the beginning of the next movement period.
Step 3: when the node carrying the data finds the best candidate relay node, the data packet will be forwarded to this relay node. The relay node receiving the data packet writes the own node serial number into the header information of the data packet, and records the forwarding path undergone by the data packet. For subsequent routing loop verification. When the relay node receives the data packet from the previous hop node, the node will repeat the opportunistic routing process to find out the best candidate relay node in the candidate relay set of the node for data forwarding. Until the destination node is within the transmission range of the carrying data node, the data will be directly sent to the destination node.
The performance of the unmanned aerial vehicle cluster collaborative opportunity routing method based on the virtual potential field method provided by the invention is verified in simulation software. In the simulation experiment, 30 nodes in the unmanned aerial vehicle cluster are assumed. One node is a source node and the other node is a destination node. And when each movement period starts, the source node sends a data packet to the destination node, and the data packet is forwarded by other nodes in the unmanned aerial vehicle cluster network and finally accepted by the destination node. The simulation duration is set to 265 motion periods, the simulation scene is a square area with a side length of 5000m, and two barriers and a destination end point of an upper right corner are also arranged in the scene. The simulation scenario is shown in fig. 4.
Table 1 shows two typical virtual potential field method based unmanned cluster co-ordinated opportunity routing (VPFC-OR) scenarios with two different weight coefficient assignments. Fig. 8, 9 and 10 are graphs of two VPFC-OR opportunistic routing algorithms versus hop-based opportunistic routing algorithm (ExOR) and topology link state based opportunistic routing (TLG-OR). As can be seen from fig. 8, the opportunistic routing provided by the method has a small improvement in the average propagation delay performance of a single data packet, and as can be seen from fig. 9, the opportunistic routing provided by the method can effectively reduce the average forwarding times of the single data packet, and meanwhile, as can be seen from fig. 10, the opportunistic routing provided by the method has lower node normalized residual energy variance. Simulation results show that compared with EoOR and TLG-OR, the unmanned aerial vehicle cluster collaborative opportunity route based on the virtual potential field method provided by the invention has better network performance.
TABLE 1 VPFC-OR routing protocol parameter settings
Drawings
FIG. 1 is a schematic diagram of a topology force of a cluster node of an unmanned aerial vehicle
FIG. 2 is a schematic diagram of unmanned aerial vehicle cluster node navigation forces
FIG. 3 is a schematic diagram of a force applied to a cluster node of an unmanned aerial vehicle
Fig. 4 is a simulation scenario of the movement of an unmanned aerial vehicle cluster, the topology state of the cluster at the initial time
FIG. 5 is a basic flow of VPFC-OR opportunistic routing
FIG. 6 is a schematic view of node travel distance
FIG. 7 is a schematic diagram of virtual force angles between candidate relay nodes and destination nodes
FIG. 8 is a graph of simulated contrast of average propagation delay of a single packet
FIG. 9 is a graph of simulated comparison of average times a packet is forwarded
FIG. 10 is a comparison graph of unmanned aerial vehicle node normalized residual energy variance simulations
Detailed Description
The invention is described in further detail below with reference to the drawings and examples.
The unmanned aerial vehicle collaborative opportunity network routing method based on the virtual potential field method is realized in a simulation environment built by MATLAB, and the effectiveness of the method is proved through a simulation result. The unmanned aerial vehicle collaborative opportunity routing method based on the Virtual Potential Field method is simply referred to as VPFC-OR (Virtual-Potential-Field-Based Cooperative Opportunistic Routing) in the specification. Fig. 5 is a basic flow of VPFC-OR opportunistic routing. The specific implementation steps for VPFC-OR are given below:
step 1: candidate relay set selection
The VPFC-OR routing protocol combines the specific characteristics of the unmanned aerial vehicle cluster network, and is mainly considered from three aspects of node advancing distance, node virtual force direction and node residual energy when the priority of the candidate relay set is ordered.
(1) Node travel distance
The ultimate goal of opportunistic routing is to forward packets faster and better to the destination node. When the current node selects the best candidate relay node, the relative positions of the candidate relay node and the destination node should be considered, so that each transmission can advance towards the destination node as much as possible, the hop count of network transmission is reduced, and the overall performance of the network is improved. Fig. 6 is a schematic view of the node advancing distance.
In fig. 6, it is assumed that the one-hop transmission distance at the source node i is R tx Then there are three candidate relay nodes N within one hop range of the source node i i (1)，N i (2)，N i (3) Dist (i, d) is defined as the Euclidean distance from the source node to the destination node. Candidate relay node N for source node i i (j) In the sense that the number of the cells,its distance to the source node i is Dist (i, N i (j) Distance to destination node is Dist (N) i (j) D), at this time, the candidate relay node N can be solved by a trigonometric function i (j) And the included angle theta of the destination node d relative to the source node i j The cosine values of (2) are:
θ 1 then it is node N i (1) And an included angle formed by the source node i and the destination node d. N in the figure i ' (j) is candidate relay node N i (j) Projection points on the connecting line of the source node i and the destination node d define candidate relay nodes N i (j) Is advanced by Dist (i, N) i ' (j)). Candidate relay node N i (j) The advance distance of (a) uses March (N) i (j) A) represents that the expression is:
March(N i (j))＝Dist(i，N i (j))·cosθ j (12)
available candidate relay node N i (j) The expression of the advancing distance of (c) is:
as can be seen from the equation (13), when the included angle between the candidate relay node and the source node or the destination node is greater than 90 DEG, the forward distance is negative, such as the candidate relay node N i (3) As shown. If N is to be i (3) As a relay node for the next hop, the packet will be farther from the destination node. When the candidate relay set is actually selected, the candidate relay node with the negative advancing distance is screened out of the candidate relay set first, so that data is prevented from being transmitted to the destination node in the opposite direction. To facilitate the subsequent calculation, we normalize the forward distance. Normalized candidate relay node N i (j) Is a distance of travel Mar (N) i (j) Is) is:
at this time, for node i, its candidate relay set N i (j) After filtering, the normalized advancing distance value range of the nodes in the candidate relay set is [0,1 ]]。
(2) Node virtual force
In the virtual potential field method, the unmanned plane node is affected by three forces, i.e., navigation force from a destination point, topology force of surrounding neighbor nodes, and obstacle force of an obstacle. The three forces are multiplied by the corresponding weight coefficients respectively, so that the virtual force finally born by the unmanned plane node can be obtained
Is of the formula wherein->
And->
The unmanned plane nodes are respectively subjected to navigation force from a destination point, topology force of surrounding neighbor nodes and barrier force of barriers, and beta G 、β T And beta O Navigation force weight coefficient, topology force weight coefficient and obstacle force weight coefficient respectively.
When the best candidate relay selection of the node is carried out, not only the current position condition of the candidate relay node is considered, but also the motion condition of the candidate relay node is considered. As shown in fig. 7, the drone cluster moves toward the destination. Active node i and relay node N in unmanned aerial vehicle cluster i (1) And N i (2) The data needs to be finally forwarded to the destination node d. The virtual forces experienced by some nodes in the cluster are as shown. The candidate relay set of the source node i has N i (1) And N i (2) Two candidate relay nodes for the source nodei, node N i (1) And node N i (2) The forward distances of (a) are not quite different, and it is not appropriate to select the best candidate relay node from the viewpoint of the forward distance only. At this time, the influence of the virtual force needs to be considered when the priority ordering of the candidate relay nodes is performed. The virtual force directly affects the motion direction of the node, and for a dynamic topology, if the node carrying the data can move towards the destination node, the network hop count can be reduced to a certain extent, and the performance of the network can be effectively improved. If the destination node is also in a moving state at this time, analysis of the relative speed direction between the candidate relay node and the destination node is required. The candidate relay node N in the unmanned aerial vehicle cluster is shown in fig. 7 i (1)、N i (2) And the magnitude and direction of the virtual force experienced by the destination node d,
for the virtual forces to which the candidate relay node j is ultimately subjected, +.>
For the virtual force finally received by the destination node d, the included angle omega between the virtual force direction of the candidate relay node j and the virtual force direction of the destination node d j The expression of (2) is:
from equation (16), ω of the source node i candidate relay set node j The value range of (2) is [0, pi ]]. When omega j The smaller the time, the more consistent the virtual force direction between the candidate relay node and the target node is shown; when omega j The more pi is approached, the opposite direction of motion of the candidate relay node and the destination node is indicated. For the convenience of calculation, normalizing the virtual force included angle between the candidate relay node and the target node:
normalized virtualVir (N) i (j) A) represents a value range of [0,1 ]]. Vir (N) i (j) Monotonicity and ω) j In contrast, ω j The smaller Vir (N) i (j) The higher the priority of the candidate relay node, the more likely the candidate relay node will be the best candidate for data forwarding.
(3) Node residual energy
The unmanned aerial vehicle cluster network is one of wireless sensor networks, and the unmanned aerial vehicle cluster network is composed of a plurality of unmanned aerial vehicle nodes carrying transceivers. The unmanned aerial vehicle is typically powered by its onboard battery, which is affected by the unmanned aerial vehicle design. The energy possessed by a single frame unmanned aerial vehicle is limited. When the cluster motion is carried out, the unmanned plane needs to spend a great deal of energy for moving and hovering, and meanwhile, the unmanned plane also needs to consume energy for performing tasks such as routing decision and forwarding data packets. If a certain unmanned aerial vehicle node is always used for data forwarding, the unmanned aerial vehicle node can consume energy rapidly and cannot support the unmanned aerial vehicle node to continue to fly to finish subsequent tasks, so that unmanned aerial vehicle damage is caused, and overall network performance is affected. Therefore, when the priority ranking of the candidate relay nodes is carried out, the residual energy of the unmanned aerial vehicle needs to be brought into an evaluation standard, and the forwarding task is distributed to the unmanned aerial vehicle nodes with more residual energy as much as possible, so that the duration of the cluster network can be effectively prolonged, the overall load balance of the unmanned aerial vehicle cluster is realized, and the overall network performance is improved.
Defining the energy at the initial moment of the single-frame unmanned aerial vehicle as E 0 E is consumed by the single unmanned aerial vehicle after the unit distance of movement m Is consumed by E for each data forwarding f Is a function of the energy of the (c). When the unmanned aerial vehicle cluster moves to the current movement period, defining the energy consumed by the unmanned aerial vehicle cluster due to movement as E M The energy consumed by forwarding the data packet is E F Then the unmanned node remaining energy can be expressed as:
E r ＝E 0 -E M -E F (18)
in order to ensure that the unmanned plane node reserves enough energy to fly, E is set min Data transfer for unmanned aerial vehicleThe minimum threshold for transmission, i.e. when the energy remaining in the unmanned aerial vehicle nodes in the candidate node set exceeds E min When the node performs priority ranking of the candidate relays; if the remaining energy of the unmanned plane node is less than or equal to E min The node does not participate in the candidate relay node prioritization. For the convenience of calculation, the remaining energy of the unmanned plane node is normalized, and the remaining energy of the normalized node j is calculated by Ene (N) i (j) A) represents that the expression is:
normalized node j residual energy Ene (N i (j) A value range of [0,1 ]]The more node residual energy, the more Ene (N i (j) The closer the value is to 1; the node residual energy is less than or equal to E min Ene (N) i (j) A value of 0) when the node does not participate in the prioritization of candidate relay nodes.
Step 2: best candidate relay node selection
Mar(N i (j) For normalized candidate relay node N i (j) Is a forward travel distance of Vir (N) i (j) For the virtual force included angle between the candidate relay node and the target node after normalization, ene (N) i (j) For the candidate relay node j of the source node i, the priority metric equation is:
Val(N i (j))＝γ M Mar(N i (j))+γ V Vir(N i (j))+γ E Ene(N i (j)) (20)
wherein gamma is M 、γ V 、γ E Respectively a node forward distance weight coefficient, a node virtual force weight coefficient and a node residual energy weight coefficient, which satisfy the following relation:
γ M +γ V +γ E ＝1 (21)
the candidate relay node calculates its own priority metric value by using the above-mentioned timer-based or contention-based coordination methodAnd the source node sorts the priority of the nodes in the candidate relay set according to the route response information, and selects the best candidate relay node for data forwarding. In different combat missions, the weight coefficients of the three indexes can be adjusted according to specific combat environments. If the hop count of data transmission is required to be reduced, the node forward distance weight coefficient gamma can be increased M Reducing the values of the other two weight coefficients; if the randomness of the network topology change is large, the node virtual force weight coefficient gamma can be improved V The proportion of the materials is as follows; if the energy consumption of the unmanned aerial vehicle cluster network nodes is kept as uniform as possible, the residual energy weight coefficient gamma can be properly increased E Is a value of (2). Under different network environments, an optimal coefficient allocation scheme needs to be found, so that the overall performance of the network can be effectively improved.
Step 3: data forwarding and acknowledgement
For a node to forward data, the node may prioritize according to the received routing replies. Since the timer value of the candidate relay node with the largest priority metric result is the smallest, the candidate relay node with the first route reply is generally taken as the best candidate relay node. However, due to the high-speed mobility and the rapid topology change of the nodes of the unmanned aerial vehicle cluster network, a routing loop phenomenon can occur, and data packets can be repeatedly transmitted among a plurality of nodes, so that the waste of network resources is caused. In order to avoid the phenomenon of routing loops, nodes in the candidate relay set need to be filtered. If the nodes in the candidate relay set have participated in data forwarding in the previous path, the nodes which have participated in data forwarding are screened out of the candidate relay set. The verification method can greatly reduce the generation of routing loop phenomenon. And the node to be forwarded with the highest priority is selected from the filtered candidate relay set to forward the data. However, in the course of the movement of the unmanned aerial vehicle cluster, the obstacle in the scene may cause the cluster to have a short split, and the candidate relay node may not exist in the candidate relay set after filtering. When the node carrying the data has no candidate relay set point, the node enters a storage-carrying mode, continues to move, and tries route discovery again when waiting for the beginning of the next movement period.
When the node carrying the data finds the best candidate relay node, the data packet will be forwarded to this relay node. The relay node receiving the data packet writes the own node serial number into the header information of the data packet, and records the forwarding path undergone by the data packet. For subsequent routing loop verification. When the relay node receives the data packet from the previous hop node, the node will repeat the opportunistic routing process to find out the best candidate relay node in the candidate relay set of the node for data forwarding. Until the destination node is within the transmission range of the carrying data node, the data will be directly sent to the destination node.
What is not described in detail in the present application belongs to the prior art known to those skilled in the art.
Claims (5)
1. The unmanned aerial vehicle cluster collaborative opportunity routing method based on the virtual potential field method comprises the following steps:
step 1: selecting a candidate relay set; when each cluster motion period starts, firstly judging whether all data packets in a network are forwarded to a destination node, if all the data packets are received by the destination node, the motion period is not operated, new data packets exist in the cluster network or the data packets which are terminated to be forwarded due to the disconnection of a link in the previous period wait for the next forwarding, meanwhile, when the period starts, the destination node broadcasts the state information of the destination node in the whole network, and the neighbor nodes exchange the position information of the destination node so that other nodes can perform routing decision; in the candidate relay set selection stage, a node carrying data needs to perform route discovery, the node sends a route request RREQ to other nodes in the communication range of the node, if no other nodes in the communication range of the node can forward the data, the node enters a storage-carrying mode, continues to move, and tries to perform route discovery again when waiting for the beginning of the next movement period, if other nodes exist in the communication range of the node, the set formed by the nodes is the candidate of the nodeThe relay set, after the node in the candidate relay set receives the route request, the node will calculate the priority metric according to the node state of the node, when the timer of the candidate relay node is overtime, the node will broadcast the route response RREP, when the advancing distance of the candidate relay node is negative, it indicates that the direction of the candidate relay node is opposite to the direction of the destination node, in order to avoid the data to be transmitted to the opposite direction, the node will not reply the route, and similarly, when the residual energy of a certain candidate relay node is lower than the preset minimum threshold E min The node also gives up the route reply, the node carrying the data decides the priority order according to the received route reply, and the best candidate relay node is determined;
step 2: selecting the best candidate relay node; for the nodes of the data to be forwarded, the nodes are subjected to priority ranking according to the received route responses, the node which is the candidate relay node with the largest priority measurement result is used as the best candidate relay node because the timer value of the candidate relay node with the largest priority measurement result is the smallest, however, due to the high-speed mobility and the rapid topology change of the nodes of the unmanned aerial vehicle cluster network, the phenomenon of a route loop can occur, the data packet can be repeatedly transmitted among a plurality of nodes, the waste of network resources is caused, the phenomenon of the route loop is avoided, the nodes in the candidate relay set need to be filtered, if the nodes in the candidate relay set participate in the data forwarding in the previous path, the node which is the data to be forwarded can be screened out of the candidate relay set, the node with the highest priority is selected in the filtered candidate relay set, but the unmanned aerial vehicle cluster is in the process of moving, the barrier in the scene can cause the cluster to be split, the filtered candidate relay set possibly has no candidate relay node, the node is carried in the candidate relay point, and the node is carried in the candidate relay set, and the node is in a temporary mode, and the next motion is continued until the candidate relay set is found in a short period;
step 3: data forwarding and confirmation; when the node carrying the data finds the best candidate relay node, the data packet is forwarded to the relay node, the relay node receiving the data packet writes the own node serial number into the data packet header information, records the forwarding path undergone by the data packet so as to carry out the route loop test subsequently, and when the relay node receives the data packet from the previous hop node, the node repeats the opportunistic route flow to find the best candidate relay node in the candidate relay set of the node for data forwarding until the destination node is in the transmission range carrying the data node, and the data is directly sent to the destination node.
2. The unmanned aerial vehicle cluster cooperation opportunity routing method based on the virtual potential field method according to claim 1, wherein the specific determination method of the forward distance of the candidate relay node selection node is as follows:
the final purpose of opportunistic routing is to forward data packets faster and better to the destination node; when the current node selects the best candidate relay node, the relative positions of the candidate relay node and the target node should be considered, so that each transmission can advance towards the target node as far as possible, the hop count of network transmission is reduced, and the overall performance of the network is improved; assume that the one-hop transmission distance of the source node i is R tx Then there are three candidate relay nodes N within one hop range of the source node i i (1)，N i (2)，N i (3) Defining Dist (i, d) as Euclidean distance from a source node to a destination node; candidate relay node N for source node i i (j) For example, the distance from the source node i is Dist (i, N i (j) Distance to destination node is Dist (N) i (j) D), at this time, the candidate relay node N can be solved by a trigonometric function i (j) And the included angle theta of the destination node d relative to the source node i j The cosine values of (2) are:
candidate relay node N i (j) The advance distance of (a) uses March (N) i (j) A) represents that the expression is:
March(N i (j))＝Dist(i，N i (j))·cosθ j (2)
available candidate relay node N i (j) The expression of the advancing distance of (c) is:
as shown in the formula (3), when the included angle formed by the candidate relay node, the source node and the destination node is larger than 90 degrees, the advancing distance is a negative number; if N is to be i (3) As a relay node for the next hop, the packet will be farther from the destination node; when the candidate relay set is actually selected, the candidate relay node with the negative advancing distance is screened out of the candidate relay set at first, so that data is prevented from being transmitted to the destination node in the opposite direction; to facilitate the subsequent calculation, we normalize the advancing distance; normalized candidate relay node N i (j) Is a distance of travel Mar (N) i (j) Is) is:
at this time, for node i, its candidate relay set N i (j) After filtering, the normalized advancing distance value range of the nodes in the candidate relay set is [0,1 ]]。
3. The unmanned aerial vehicle cluster collaborative opportunity routing method based on the virtual potential field method according to claim 1, wherein the specific determination method of the candidate relay node selection node virtual force is as follows:
in the virtual potential field method, unmanned plane nodes are influenced by navigation force from target nodes, topology force of surrounding neighbor nodes and barrier force of barriers; the three forces are multiplied by the corresponding weight coefficients respectively, so that the virtual force finally born by the unmanned plane node can be obtained
Wherein the method comprises the steps of
And->
The unmanned plane nodes are respectively subjected to navigation force from the target node, topology force of surrounding neighbor nodes and barrier force of barrier, and beta G 、β T And beta O Navigation force weight coefficient, topology force weight coefficient and obstacle force weight coefficient respectively;
when the optimal candidate relay node is selected, not only the current position condition of the candidate relay node is considered, but also the motion condition of the candidate relay node is considered; defining an active node i and a relay node N in the unmanned aerial vehicle cluster i (1) And N i (2) The destination node d to which the data needs to be finally forwarded; the candidate relay set of the source node i has N i (1) And N i (2) Two candidate relay nodes, node N for source node i i (1) And node N i (2) The forward distances of the relay node are not different, if the best candidate relay node is selected from the angle of the forward distance only, the relay node is not suitable; at this time, when the priority ordering of the candidate relay nodes is performed, the influence of the virtual force needs to be considered; the virtual force directly affects the motion direction of the node, and for a dynamic topology, if the node carrying the data can move towards the destination node, the network hop count can be reduced; if the destination node is in a moving state at the moment, analyzing the relative speed direction between the candidate relay node and the destination node;
for the virtual forces to which the candidate relay node j is ultimately subjected, +.>
For the final virtual force of the destination node d, the included angle omega between the virtual force direction of the candidate relay node j and the virtual force direction of the destination node d j The expression of (2) is: />
From equation (6), ω of the source node i candidate relay set node j The value range of (2) is [0, pi ]]The method comprises the steps of carrying out a first treatment on the surface of the When omega j The smaller the time, the more consistent the virtual force direction between the candidate relay node and the target node is shown; when omega j The more pi is approached, the opposite direction of the candidate relay node and the destination node is indicated; for the convenience of calculation, normalizing the virtual force included angle between the candidate relay node and the target node:
vir (N) i (j) A) represents a value range of [0,1 ]]The method comprises the steps of carrying out a first treatment on the surface of the From formula (7), ω j The smaller Vir (N) i (j) The higher the priority of the candidate relay node, the more likely the candidate relay node will be the best candidate for data forwarding.
4. The unmanned aerial vehicle cluster cooperation opportunity routing method based on the virtual potential field method according to claim 1, wherein the specific determination method of the residual energy of the candidate relay node selection node is as follows:
the unmanned aerial vehicle cluster network is one of wireless sensor networks, and consists of a plurality of unmanned aerial vehicle nodes carrying transceivers; the unmanned aerial vehicle is influenced by the design of the unmanned aerial vehicle, and the unmanned aerial vehicle is powered by a battery carried by the unmanned aerial vehicle; the energy possessed by a single frame unmanned aerial vehicle is limited; when the cluster motion is carried out, the unmanned plane needs to spend a large amount of energy for moving and hovering, and meanwhile, the energy is consumed for executing the tasks of routing decision and forwarding the data packet; if a certain unmanned aerial vehicle node is always used for data forwarding, the unmanned aerial vehicle node can consume energy rapidly and cannot support the unmanned aerial vehicle node to continue to fly to finish subsequent tasks, so that the unmanned aerial vehicle is damaged, and the overall network performance is affected; therefore, when the priority ranking of the candidate relay nodes is carried out, the residual energy of the unmanned aerial vehicle needs to be brought into an evaluation standard, and the forwarding task is distributed to the unmanned aerial vehicle nodes with more residual energy as much as possible, so that the duration of the cluster network can be effectively prolonged, the overall load balance of the unmanned aerial vehicle cluster is realized, and the overall network performance is improved;
defining the energy at the initial moment of the single-frame unmanned aerial vehicle as E 0 E is consumed by the single unmanned aerial vehicle after the unit distance of movement m Is consumed by E for each data forwarding f Energy of (2); when the unmanned aerial vehicle cluster moves to the current movement period, defining the energy consumed by the unmanned aerial vehicle cluster due to movement as E M The energy consumed by forwarding the data packet is E F Then the unmanned node remaining energy can be expressed as:
E r ＝E 0 -E M -E F (8)
in order to ensure that the unmanned plane node reserves enough energy to fly, E is set min Minimum threshold for data forwarding for unmanned aerial vehicle, i.e. when the energy remaining for unmanned aerial vehicle nodes in candidate node set exceeds E min When the node performs priority ranking of the candidate relays; if the remaining energy of the unmanned plane node is less than or equal to E min The node does not participate in the priority ordering of the candidate relay nodes; for the convenience of calculation, the remaining energy of the unmanned plane node is normalized, and the remaining energy of the normalized node j is calculated by Ene (N) i (j) A) represents that the expression is:
normalized node j residual energy Ene (N i (j) A value range of [0,1 ]]The more node residual energy, the more Ene (N i (j) The closer the value is to 1; the node residual energy is less than or equal to E min Ene (N) i (j) A value of 0) when the node does not participate in the prioritization of candidate relay nodes.
5. The unmanned aerial vehicle cluster collaborative opportunity routing method based on the virtual potential field method according to claim 1, wherein the specific determination method of the best candidate relay node is as follows:
Mar(N i (j) For normalized candidate relay node N i (j) Is a forward travel distance of Vir (N) i (j) For the virtual force included angle between the candidate relay node and the target node after normalization, ene (N) i (j) For the candidate relay node j of the source node i, the priority metric equation is:
Val(N i (j))＝γ M Mar(N i (j))+γ V Vir(N i (j))+γ E Ene(N i (j)) (10)
wherein gamma is M 、γ V 、γ E Respectively a node forward distance weight coefficient, a node virtual force weight coefficient and a node residual energy weight coefficient, which satisfy the following relation:
γ M +γ V +γ F ＝1 (11)
the candidate relay node calculates the priority metric value of the candidate relay node, adopts a coordination method based on a timer or a competition to carry out route reply, and the source node sorts the priority of the candidate relay concentrated node according to the route reply information, and selects the best candidate relay node for data forwarding; in different combat tasks, the weight coefficients of the three indexes can be adjusted according to specific combat environments; if the hop count of data transmission is required to be reduced, the node forward distance weight coefficient gamma can be increased M Reducing the values of the other two weight coefficients; if the randomness of the network topology change is large, the node virtual force weight coefficient gamma can be improved V The proportion of the materials is as follows; if want to keep unmanned aerial vehicle cluster networkThe energy consumption of the network nodes is as uniform as possible, so that the residual energy weight coefficient gamma can be properly increased E Is a value of (2); under different network environments, an optimal coefficient allocation scheme needs to be found, so that the overall performance of the network can be effectively improved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911423735.4A CN111132258B (en) | 2019-12-30 | 2019-12-30 | Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911423735.4A CN111132258B (en) | 2019-12-30 | 2019-12-30 | Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111132258A CN111132258A (en) | 2020-05-08 |
| CN111132258B true CN111132258B (en) | 2023-05-09 |
Family
ID=70507942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911423735.4A Active CN111132258B (en) | 2019-12-30 | 2019-12-30 | Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111132258B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111561932B (en) * | 2020-05-27 | 2023-07-18 | 武汉理工大学 | Ship navigation method based on virtual force |
| CN111935314B (en) * | 2020-09-25 | 2021-01-12 | 支付宝(杭州)信息技术有限公司 | Block chain system, message transmission method and device |
| CN111934990B (en) | 2020-09-25 | 2021-02-09 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN111935000B (en) | 2020-09-25 | 2021-01-08 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN111934999B (en) | 2020-09-25 | 2021-01-22 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN111934996B (en) | 2020-09-25 | 2021-01-12 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN111934998B (en) | 2020-09-25 | 2021-02-09 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN111934997B (en) | 2020-09-25 | 2021-01-12 | 支付宝(杭州)信息技术有限公司 | Message transmission method and device |
| CN112968967B (en) | 2020-09-25 | 2023-05-19 | 支付宝(杭州)信息技术有限公司 | Block synchronization method and device |
| CN112423270B (en) * | 2020-10-12 | 2022-03-25 | 南京航空航天大学 | Unmanned aerial vehicle cluster low interception deployment method based on virtual force and beam parameter optimization |
| CN113219857B (en) * | 2021-05-31 | 2022-07-19 | 中国人民解放军国防科技大学 | Unmanned system cluster network communication simulation method and device |
| CN115022833A (en) * | 2022-04-21 | 2022-09-06 | 哈尔滨工业大学 | Unmanned aerial vehicle swarm data transmission method based on potential energy field |
| CN114828141B (en) * | 2022-04-25 | 2024-04-19 | 广西财经学院 | UWSNs multi-hop routing method based on AUV networking |
| CN115022892B (en) * | 2022-05-31 | 2023-12-01 | 南京邮电大学 | Sensor node deployment method in chemical plant environment based on improved virtual force |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102269593B (en) * | 2010-06-01 | 2014-03-12 | 北京航空航天大学 | Fuzzy virtual force-based unmanned plane route planning method |
| CN108549407B (en) * | 2018-05-23 | 2020-11-13 | 哈尔滨工业大学（威海） | Control algorithm for multi-unmanned aerial vehicle cooperative formation obstacle avoidance |
| CN109521794A (en) * | 2018-12-07 | 2019-03-26 | 南京航空航天大学 | A kind of multiple no-manned plane routeing and dynamic obstacle avoidance method |
| CN109460064B (en) * | 2019-01-03 | 2019-11-15 | 中国人民解放军战略支援部队航天工程大学 | Unmanned plane cluster regions covering method and its device based on virtual potential field function |
-
2019
- 2019-12-30 CN CN201911423735.4A patent/CN111132258B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111132258A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111132258B (en) | Unmanned aerial vehicle cluster collaborative opportunity routing method based on virtual potential field method | |
| Zheng et al. | Adaptive communication protocols in flying ad hoc network | |
| El Houda Bahloul et al. | A flocking-based on demand routing protocol for unmanned aerial vehicles | |
| CN110475205B (en) | Relay selection method based on track relation in unmanned aerial vehicle ad hoc network and unmanned aerial vehicle | |
| CN112202848B (en) | Unmanned system network self-adaptive routing method and system based on deep reinforcement learning | |
| Toorchi et al. | Skeleton-based swarm routing (SSR): Intelligent smooth routing for dynamic UAV networks | |
| Jianmin et al. | Ardeep: Adaptive and reliable routing protocol for mobile robotic networks with deep reinforcement learning | |
| Lu et al. | Relay in the sky: A UAV-aided cooperative data dissemination scheduling strategy in VANETs | |
| CN114339936A (en) | Aircraft self-organizing network optimization link state routing mechanism based on Q learning | |
| Zhao et al. | Route discovery in flying ad-hoc network based on bee colony algorithm | |
| Albu-Salih et al. | Dynamic routing method over hybrid SDN for flying ad hoc networks | |
| Huo et al. | A UAV mobile strategy in mobile ad hoc networks | |
| Qiu et al. | QLGR: A Q-learning-based Geographic FANET Routing Algorithm Based on Multiagent Reinforcement Learning. | |
| CN114980126A (en) | Method for realizing unmanned aerial vehicle relay communication system based on depth certainty strategy gradient algorithm | |
| Gao et al. | Improvement of GPSR routing protocol for TDMA-based UAV ad-hoc networks | |
| Liu et al. | AR-GAIL: Adaptive routing protocol for FANETs using generative adversarial imitation learning | |
| Kumbhar et al. | Innovating multi-objective optimal message routing for unified high mobility networks | |
| CN112672398B (en) | 3D-GPSR routing method based on self-adaptive kalman prediction | |
| Jiang et al. | Research on OLSR adaptive routing strategy based on dynamic topology of UANET | |
| Li et al. | Ad hoc network routing protocol based on location and neighbor sensing | |
| CN110996369B (en) | Unmanned aerial vehicle network routing working method based on task driving | |
| Hao et al. | Mobility-aware trajectory design for aerial base station using deep reinforcement learning | |
| Zhao et al. | Comparison study on uav movement for adapting to multimedia burst in post-disaster networks | |
| Abdulhae et al. | Data dissemination of vehicular Ad-Hoc network in highway scenario | |
| Tang et al. | Disaster Resilient Emergency Communication With Intelligent Air-Ground Cooperation |
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 |