Cellular network coverage hole detection and repair method based on unmanned aerial vehicle

Abstract

The invention provides a cellular network coverage hole detection and repair method based on an unmanned aerial vehicle, and belongs to the technical field of coverage optimization of wireless communication networks. The method comprises the following steps: constructing one or more detection tracks traversing all detection points; each patrol unmanned aerial vehicle executes a cruising task along one detection track, hovers at the detection point, receives a signal sent by the ground base station and judges whether a hole exists or not; the patrol unmanned aerial vehicle sends mark information and geographical location information of the detection points to a satellite or an adjacent ground base station; the satellite or the adjacent ground base station determines an unmanned aerial vehicle base station scheduling scheme of the first two detection points based on the mark information of every three continuous detection points; and the dispatched unmanned aerial vehicle base station flies to a specified position, so that the network coverage hole is repaired. The method is suitable for detecting and repairing cellular network coverage holes in various scenes, especially cellular network coverage recovery capability in emergency scenes such as natural disasters and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of coverage optimization of wireless communication networks, and in particular to a method for detecting and repairing coverage holes in cellular networks based on unmanned aerial vehicles. Background Art

[0002] In order to support various use cases in future wireless communication scenarios, 6G wireless communication networks must provide wider high-capacity coverage with low latency. In order to provide optimal communication services to ubiquitous user devices, telecom operators must ensure that there are no uncovered or poorly served areas within the cellular network coverage area. In actual engineering applications, coverage holes refer to areas where the signal strength received by users from the serving base station or the adjacent cooperative base station is lower than the level required to maintain the minimum service quality. Therefore, when the user terminal enters a coverage hole, problems such as dropped calls, poor communication services, and even radio link failures may occur, resulting in a poor user experience. There are many reasons for this phenomenon in actual engineering applications, such as attenuation caused by physical obstacles, inappropriate antenna parameters, hardware failures, improper spectrum planning, and natural disasters. In addition, as one of the key technologies of 6G wireless networks, millimeter wave communication may increase the number of coverage holes due to its severe attenuation characteristics.

[0003] Traditional methods for detecting cellular network coverage holes rely on a combination of strategies such as drive tests, customer complaints, and software / hardware alerts, which are time-consuming, costly, and have low reliability. To overcome these challenges, the 3rd Generation Partnership Project standardized the minimization of drive tests (MDT) technology, which allows telecom operators to automatically store user measurements and signaling messages. However, due to its reliance on terminals, MDT reports have some limitations. For example, the terminal cannot stably provide instant reports, and the terminal may not provide accurate geographic location due to factors such as privacy exposure. Therefore, the above methods are only applicable to limited situations such as cities and suburbs, and cannot be used for rural areas and emergencies caused by natural disasters such as earthquakes and hurricanes, where the network infrastructure is destroyed or even the communication status of ground base stations is unknown.

[0004] Currently, the most advanced solution to the above problems is to utilize mobile robotic platforms such as unmanned aerial vehicles. In particular, the assistance of satellites and aerial navigation systems for traditional wireless communications can provide full-time and three-dimensional broadband access for ubiquitous coverage on the earth. Compared with satellites, flexible drones can provide low-cost temporary wireless service delivery solutions for difficult-to-access areas such as caves and tunnels. At the same time, the strong line-of-sight links of drones can quickly restore communication services in emergency situations caused by natural disasters. Furthermore, drone swarms consisting of multiple drones can perform various complex tasks more efficiently because they can jointly collect more spatiotemporal data than a single drone. In this case, coordinated multipoint transmission technology can be applied to enhance drone-supported wireless communications. In order to effectively develop drone swarms in wireless networks, the mathematically tractable Delaunay triangulation, a typical geometric structure in random geometry theory, can be used to establish a multipoint cooperative transmission set.

[0005] In order to fully utilize the mobility of drones for wireless transmission, motion control is necessary. More specifically, motion control aims to design appropriate strategies to drive drones to form the desired formation structure and guide their overall maneuvers. However, there is no such solution in the prior art. Summary of the invention

[0006] In view of this, the present invention provides a cellular network coverage hole detection and repair method based on drone, aiming to improve coverage quality, reduce communication interruptions, and ensure stable and efficient communication coverage.

[0007] The technical solution adopted by the present invention is:

[0008] A method for detecting and repairing cellular network coverage holes based on drones uses a patrol drone to collect network coverage data and uses a drone base station to repair network coverage holes, including the following steps:

[0009] Step 1: Set detection points and construct one or more detection trajectories that traverse all detection points;

[0010] Step 2: Each patrol drone performs a cruise mission along a detection trajectory, hovers at the detection point and receives the signal sent by the ground base station, calculates the SINR of the received signal and compares it with the preset threshold γ th For comparison, if SINR is lower than the threshold γ th , then it is marked that there is a hole at the detection point, otherwise it is marked that there is no hole at the detection point;

[0011] Step 3, the patrol drone sends the marking information and geographic location information of the detection point to the satellite or the nearby ground base station;

[0012] Step 4: The satellite or the adjacent ground base station determines the UAV base station scheduling scheme for the first two detection points based on the marking information of every three consecutive detection points;

[0013] Step 5: The dispatched drone base station flies to the designated location to repair the network coverage hole.

[0014] Furthermore, the specific method of step 4 is: if there are at least two consecutive detection points marked as having holes among three consecutive detection points, the drone base station group is scheduled for repair, and the two-dimensional projection of the centroid position of the drone base station group is consistent with the position of any one of the first two detection points with holes; otherwise, if there is a detection point marked as having a hole among the first two detection points, a single drone base station is scheduled for repair, and the two-dimensional projection of the position of the single drone base station is consistent with the position of the detection point with holes among the first two detection points; if there is no detection point with holes among the first two detection points, the drone base station is not scheduled for repair.

[0015] Furthermore, the drone base station group is composed of four drone base stations forming a regular tetrahedron formation, and uses multi-point cooperative transmission technology to collaboratively serve ground users.

[0016] Furthermore, the specific method of step 5 is:

[0017] Step 501, dynamic modeling of the drone base station:

[0018]

[0019] Among them, x i (t),v i (t) respectively represent the position and speed of the i-th UAV at time t; f R (·) represents the inherent dynamics of the i-th UAV; u i (t) represents the control item, which is used to implement the anti-collision motion control strategy of UAV flight;

[0020] f R (·) Includes the air resistance, gravity and a portion of the lift force used to balance the gravity, expressed as:

[0021]

[0022] Among them, M i represents the mass of the i-th UAV, and the units of positive parameters k1 and k2 are kilograms per second and kilograms per meter respectively;

[0023] Step 502, for the i The collision area of ​​the i-th drone at The communication area is p i Indicates the position, r c ≤r d , the effective anti-collision area of ​​the i-th UAV is Ψ i ∩Ω i ; Assuming i≠j, if the jth UAV enters the effective anti-collision area of ​​the ith UAV, the anti-collision mechanism will be activated;

[0024] Step 503, given vector represents the destination position of the i-th UAV, and defines the distance between the i-th UAV in the system and its destination as The corresponding speed is Define the potential function V for collision avoidance based on local perception information (i,k) (x i ,x k )for:

[0025]

[0026] Among them, r c and r d They represent the sensing distance and anti-collision distance of the UAV respectively, and ||·|| represents the l2 norm;

[0027] Rewrite the dynamic equation of the drone base station as:

[0028]

[0029] in, represents the set of all drones that perform a single drone base station repair mission, Represents the set of all drones that perform drone base station group repair tasks, represents the set of drones in the nth drone base station group; constant parameter b i ≥0 represents the local feedback gain obtained by the i-th UAV; a ij is a weighted Boolean value, indicating whether the jth UAV in the UAV base station group can obtain information from the i-th UAV; the vector represents the target position of the i-th UAV, vector The lth component in is expressed as:

[0030]

[0031] Here, l∈{1,2,3}; sgn(·) is the sign function; c i and ε i,l are all positive control parameters, is a vector The lth component of ;

[0032] Step 504, control the drone base station to fly according to the dynamic equation to achieve collision avoidance. The drone base station starts working after flying to the designated position to repair the network coverage hole.

[0033] The beneficial effects of the present invention are:

[0034] 1. The present invention detects coverage holes in cellular networks by using patrol drones, and dispatches drone base stations based on the detection results of the patrol drones, thereby being able to efficiently detect and restore coverage holes in cellular networks.

[0035] 2. The present invention adopts an anti-collision motion control strategy for the two types of UAV base station formation structures in network coverage hole detection and repair, which can ensure the flight safety of UAV base stations with different formation configurations during movement, so as to ensure the robustness of the UAV-based cellular network coverage hole detection and repair solution.

[0036] 3. The present invention integrates air and ground resources to enhance network communication capabilities and is suitable for rapid network reconstruction in emergency situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the principle of cellular network coverage hole detection and repair based on drones.

[0038] Figure 2 This is a schematic diagram of the inspection route of a single patrol drone.

[0039] Figure 3 It is a schematic diagram of the judgment result of the drone base station cluster configuration.

[0040] Figure 4 This is a heat map of the signal-to-noise ratio in an emergency scenario, where the left picture shows the detection path and results, and the right picture shows the recovery results.

[0041] Figure 5 It is a schematic diagram of the two-dimensional trajectory of the drone base station used to restore coverage holes in emergency scenarios.

[0042] Figure 6 It is a graph of the minimum distance variation between drone base stations used to restore coverage holes in emergency scenarios. DETAILED DESCRIPTION

[0043] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0044] A method for detecting and repairing cellular network coverage holes based on drones comprises the following steps:

[0045] Step 1: Set detection points and construct one or more detection trajectories that traverse all detection points;

[0046] Step 2: Each patrol drone performs a cruise mission along a detection trajectory, hovers at the detection point and receives the signal sent by the ground base station, calculates the SINR of the received signal and compares it with the preset threshold γ th For comparison, if SINR is lower than the threshold γ th , then mark the detection point as having a hole, otherwise mark the detection point as not having a hole. The inspection drone detects the cellular coverage area whose location information and communication status are unknown, and feeds back the real-time detection results to the adjacent base station or satellite equipment to determine the scale of the coverage hole; based on the judgment of the scale of the coverage hole, the drone base station is reasonably scheduled to ensure the effective coverage rate of the target area.

[0047] Step 3, the patrol drone sends the marking information and geographic location information of the detection point to the satellite or the nearby ground base station;

[0048] Step 4: The satellite or the adjacent ground base station determines the scheduling scheme of the drone base station for the first two detection points based on the marking information of every three consecutive detection points. This method discretizes the collected coverage data to reduce the system's decision and scheduling delays, and is suitable for large-scale networks. There are two scheduling schemes: one for covering small-scale holes, and one for covering large-scale holes, which consists of four drone base stations, and uses coordinated multi-point technology to enhance communication capabilities.

[0049] Step 5: The dispatched drone base station flies to the designated location to repair the network coverage hole. The motion control of the drone adopts an anti-collision method based on potential function to ensure the flight safety of drone base stations in different formations; this strategy has the characteristics of decentralization by sensing local information.

[0050] After detecting a coverage hole, the UAV base station can execute different cluster configurations according to the scheduling information and quickly deploy to the designated area to provide wireless services to restore communication under the control of the anti-collision motion control strategy.

[0051] Consider the wireless communication scenario of downlink transmission of a terrestrial cellular network in a given area. In the actual wireless transmission process, in order to effectively cover a given area, the wireless communication network system uses a variety of network deployment optimization technologies to optimize the base station deployment. However, due to new obstacles or natural disasters in the transmission environment, there may be coverage holes in the ideal coverage area of ​​the terrestrial cellular network. Assume that the height is H BS The location distribution of ground base stations follows the density λ B The homogeneous Poisson point process (PPP) of B For the position (x,y)∈Φ BThe coverage hole A in the coverage area of ​​a base station is defined as the received signal to interference and noise ratio (SINR) lower than the predetermined threshold γ th The mathematical expression for the geographical area is:

[0052]

[0053] According to the SINR threshold γ th With different selections of , the coverage holes (i.e., areas where users cannot register on the network) and the weak coverage areas (i.e., areas with low connection rate, high drop rate and poor user perception) involved in actual projects can be regarded as the coverage holes mentioned in this method.

[0054] Patrol drones perform regular patrols according to pre-set flight paths to detect potential coverage holes in ground cellular networks. During this process, patrol drones send detection results to remote satellites or nearby ground base stations; the latter determine the size of the coverage hole based on the detection results and then decide the size of the drone base station to be dispatched to restore the ground coverage hole at the target location.

[0055] Figure 1 The working principle of the method is demonstrated, mainly from the perspectives of communication link and control link. Specifically, from the communication perspective, the patrol drone receives the signal sent by the ground base station at the detection point, and compares the calculated received SINR with the preset threshold γ th Comparison is performed to detect whether the detection point is located in a ground coverage hole. In engineering implementation, the patrol drone performs a cruise mission along a preset trajectory and hovers at a predetermined detection point to complete the task of detecting potential ground coverage holes. If a ground coverage hole is found, the patrol drone sends the geographic location information of the corresponding detection point to the satellite or the adjacent ground base station to determine the scheduling instructions for the drone base station. From a control perspective, the patrol drone, the satellite or the adjacent ground base station forms a control loop with the drone base station. When the satellite or the adjacent ground base station receives the detection information of the patrol drone, it makes relevant decisions on scheduling the drone base station based on the detection information of the connected detection points around the coverage hole. Then, according to the size of the detected ground coverage hole, the control decision is sent to a single drone base station or a group of drone base stations. Finally, the dispatched single drone base station or group of drone base stations flies to the target location to restore the coverage hole.

[0056] The specific process of this method is as follows:

[0057] (1) The path of the patrol drone is related to the randomly generated detection points. Assume that the initial detection points obey the density λ CHDPPP is used to ensure that the initial detection points are evenly distributed in the area. Next, any detection point whose distance to its nearest neighbor is less than a given positive constant D is excluded to form the final detection point set, where the given constant D is positively correlated with the coverage radius of the cellular base station. The final detection point set formed after refining the initial detection point PPP can effectively reduce the number of detection points, thereby reducing the system cost of the inspection process. Considering the consumption of UAV maneuvering energy and detection efficiency, patrol UAVs should not traverse all detection points according to the shortest distance criterion repeatedly. Therefore, for the path planning of patrol UAVs, many existing path planning algorithms, such as the classic Dijkstra algorithm, can be applied to meet the above requirements.

[0058] For ease of explanation, Figure 2 The figure shows a schematic diagram of the path used by patrol drones to detect ground coverage holes. The three circular areas in the figure represent the coverage of three adjacent ground base stations, the squares represent the preset detection points, the red / blue squares represent the final detection points after the second selection, and the directed line segments with arrows represent the shortest path that traverses all these detection points according to the Dijkstra algorithm.

[0059] (2) The patrol drone flies along the path and calculates the received SINR at each detection point. Based on the calculation result, combined with the preset threshold γ th Add a red or blue label to each detection point. Specifically, the detection point outside the coverage area is marked as red (i.e., the SINR value received by the patrol drone is lower than the threshold γ th ), while the detection points located in the coverage area are marked in blue (i.e., the SINR value received by the patrol drone is equal to or greater than the threshold γ th ).

[0060] (3) The patrol drone sends the marking information and geographic location information of the detection point to the satellite or nearby ground base station.

[0061] (4) The satellite or the nearby ground base station determines the UAV base station scheduling scheme for the first two detection points based on the marking information of every three consecutive detection points. In order to ensure the effective utilization of the UAV base station, this method selects UAV base stations with different cluster configurations according to the scale of the detected coverage holes. Specifically, this patent considers two cluster configurations of UAV base stations, namely configuration 1 (a single UAV base station) and configuration 2 (a cluster consisting of four UAV base stations). Configuration 1 is used to recover "small" scale coverage holes; configuration 2 is used to recover "large" scale coverage holes. In particular, the four UAVs in configuration 2 can be formed into a regular tetrahedron and use multi-point cooperative transmission technology to collaboratively serve ground users.

[0062] In a large-scale wireless communication network, it takes a long time for patrol drones to traverse the preset flight path. Figure 2 The detection process of the patrol drone shown can be discretely represented as a sequence of three consecutive detection points, and the scheduling strategy of the first two drones at each consecutive detection point is determined by their detection results. Figure 3 All situations of the scheduling strategy determined by the detection results of three consecutive detection points are exhaustively enumerated. Among them, the scheduling decision of the drone base station for the first two detection points is based on the information of the ground coverage holes obtained at every three consecutive detection points. For the simplicity of description, "R" is used to represent red (i.e., the detection point located in the coverage hole) and "B" is used to represent blue (i.e., the detection point located outside the coverage hole). Specifically, when the color labels of the three detection points belong to the set {RBR, RBB, BRB}, it is judged that there are small-scale coverage holes near the first two monitoring points, and the cluster scheduling scheme of the drone base station is configuration 1; when the color labels of the three detection points belong to the set {RRR, RRB, BRR}, the cluster scheduling scheme of the drone base station is configuration 2. Since the current drone base station scheduling scheme is to improve the conditions of the first two detection points, no action will be taken if the label belongs to the set {BBR, BBB}.

[0063] (5) The dispatched UAV base station flies to the designated location to repair the network coverage hole. In order to effectively restore the ground coverage holes detected in a given area, it is often necessary to dispatch multiple UAVs to their corresponding target locations at the same time and form a UAV base station formation that meets the configuration. In this case, in order to meet the needs of engineering practice, the motion control and anti-collision mechanism of multiple UAVs needs to be reasonably designed.

[0064] First, the dynamics of the drone base station needs to be modeled. Assume that at a given time, there are M drone base stations that meet configuration 1 and N drone base stations that meet configuration 2 in a given area to repair cellular network coverage holes. Let and Respectively represent the sets of drones with these two configurations. Obviously, there are and In other words, there are a total of M+4N drone base stations that need to be dispatched. Considering the air resistance encountered by the drone during flight, its dynamic equation can be modeled as

[0065]

[0066] Among them, x i (t),v i (t) respectively represent the position information and speed information of the i-th UAV at time t; f R (·) represents the inherent dynamics of the i-th UAV; ui (t) represents the designed control term. The inherent nonlinear dynamics of the UAV considered here include the air resistance, gravity, and a part of the lift used to balance the gravity, which can be expressed as:

[0067]

[0068] Among them, M i represents the mass of the i-th UAV, and the units of the positive parameters k1 and k2 are kilograms per second (kg / s) and kilograms per meter (kg / m), respectively.

[0069] Next, the anti-collision motion control strategy is designed. i The collision area of ​​the drone base station labeled i is defined as Define its communication area as And r c ≤r d Therefore, the effective anti-collision area of ​​the drone base station labeled i is Ψ i ∩Ω i That is, assuming i≠j, if a UAV with the label j enters the collision avoidance zone of the UAV with the label i (i.e. ), the anti-collision mechanism will be activated. Based on the above definition, the motion control strategy of the drone base station corresponding to different cluster configurations can be given, that is, u in the formula i (t). In addition, given the vector represents the destination position of the i-th UAV, and defines the distance between the i-th UAV in the system and its destination as Then, the corresponding speed is Based on the above definition, the dynamic equation of the drone base station in the drone-based cellular network coverage hole detection and repair system is:

[0070]

[0071] Among them, the constant parameter b i ≥0 represents the local feedback gain obtained by the i-th UAV; a ij Indicates that the jth UAV in the UAV swarm in configuration 2 can obtain information from the i-th UAV; the potential function V for collision avoidance based on local perception information (i,k) (x i ,x k ) can be defined as:

[0072]

[0073] vector The lth component in can be expressed as:

[0074]

[0075] Here, l∈{1,2,3}, the function sgn(·) is a sign function; c i and ε i,l are all positive control parameters, is a vector When the initial positions of the drone base stations are outside the collision area of ​​each other and there are no isolated nodes in configuration 2, the drone base stations in the system can achieve collision avoidance while reaching the target position.

[0076] The effectiveness of this method is illustrated below by taking ground coverage holes in a wireless communication network as an example.

[0077] Consider an emergency communication scenario where a large number of ground base stations cannot work properly. The task of the drone in this scenario is to repair as many ground communications as possible and reduce coverage holes. Figure 4 As shown in the left figure, the patrol drone starts its inspection from the rightmost side of the given area, and makes a judgment on the status of each detection point based on the given signal-to-noise ratio threshold. The information is then transmitted back to the main control console through the satellite communication network or the remaining cellular base stations that are still functioning normally, so as to achieve real-time scheduling of the drone base stations that perform the task, and finally the repair results. Figure 4 As shown in the figure on the right. In this embodiment, the drone group used to perform the task consists of four drone base stations. Figure 4 Taking the upper right area in the left figure as an example, the two-dimensional trajectory of the drone base station used to perform the mission is as follows Figure 5 As shown in the figure, the “UAV” in the legend refers to the drone base station used to repair cellular coverage holes; at the same time, the minimum distance between these drone base stations is as follows Figure 6 As shown, even in the most extreme conditions, the drone base stations can still be guaranteed to be outside the collision zone of each other.

[0078] The present invention mainly includes a cellular network coverage hole detection method based on patrol drones, a drone base station scheduling algorithm based on patrol results, and an anti-collision motion control strategy to ensure the flight safety of drone base stations. The present invention is suitable for the detection and repair of cellular network coverage holes in various scenarios, especially the cellular network coverage recovery capability in emergency scenarios such as natural disasters. The drone base station scheduling algorithm of the present invention can reasonably schedule drone base stations according to the scale of coverage holes, thereby making the detection and repair of cellular network coverage holes based on drones more reasonable and effective.

Claims

1. A method for detecting and repairing cellular network coverage holes based on drones, characterized in that: Using inspection drones to collect network coverage data and using drone base stations to repair network coverage holes includes the following steps: Step 1: Set detection points and construct one or more detection trajectories that traverse all detection points; Step 2: Each patrol drone performs a cruise mission along a detection trajectory, hovers at the detection point and receives the signal sent by the ground base station, calculates the SINR of the received signal and compares it with the preset threshold γ th For comparison, if SINR is lower than the threshold γ th , then it is marked that there is a hole at the detection point, otherwise it is marked that there is no hole at the detection point; Step 3, the patrol drone sends the marking information and geographic location information of the detection point to the satellite or the nearby ground base station; Step 4: The satellite or the adjacent ground base station determines the UAV base station scheduling scheme for the first two detection points based on the marking information of every three consecutive detection points; Step 5: The dispatched drone base station flies to the designated location to repair the network coverage hole.

2. The method for detecting and repairing cellular network coverage holes based on drones according to claim 1, characterized in that: The specific method of step 4 is: if there are at least two consecutive detection points marked as having holes among the three consecutive detection points, the drone base station group is dispatched for repair, and the two-dimensional projection of the centroid position of the drone base station group is consistent with the position of any of the first two detection points with holes; Otherwise, if there is a detection point marked as having a hole in the first two detection points, a single UAV base station is dispatched for repair, and the two-dimensional projection of the position of the single UAV base station is consistent with the position of the detection point with a hole in the first two detection points; If there is no empty detection point in the first two detection points, the drone base station will not be dispatched for repair.

3. The method for detecting and repairing cellular network coverage holes based on drones according to claim 2, characterized in that: The drone base station group is composed of four drone base stations forming a regular tetrahedron formation, and uses multi-point cooperative transmission technology to collaboratively serve ground users.

4. The method for detecting and repairing cellular network coverage holes based on drones according to claim 2, characterized in that: The specific method of step 5 is: Step 501, dynamic modeling of the drone base station: Among them, x i (t),v i (t) respectively represent the position and speed of the i-th UAV at time t; f R (·) represents the inherent dynamics of the i-th UAV; u i (t) represents the control item, which is used to implement the anti-collision motion control strategy of UAV flight; f R (·) Includes the air resistance, gravity and a portion of the lift force used to balance the gravity, expressed as: Among them, M i represents the mass of the i-th UAV, and the units of positive parameters k1 and k2 are kilograms per second and kilograms per meter respectively; Step 502, for the i The collision area of ​​the i-th drone at The communication area is p i Indicates the position, r c ≤r d , the effective anti-collision area of ​​the i-th UAV is Ψ i ∩Ω i ; Assuming i≠j, if the jth UAV enters the effective anti-collision area of ​​the ith UAV, the anti-collision mechanism will be activated; Step 503, given vector represents the destination position of the i-th UAV, and defines the distance between the i-th UAV in the system and its destination as The corresponding speed is Define the potential function V for collision avoidance based on local perception information (i,k) (x i ,x k )for: Among them, r c and r d They represent the sensing distance and anti-collision distance of the UAV respectively, and ||·|| represents the l2 norm; Rewrite the dynamic equation of the drone base station as: in, represents the set of all drones that perform a single drone base station repair mission, Represents the set of all drones that perform drone base station group repair tasks, represents the set of drones in the nth drone base station group; constant parameter b i ≥0 represents the local feedback gain obtained by the i-th UAV; a ij is a weighted Boolean value, indicating whether the jth UAV in the UAV base station group can obtain information from the i-th UAV; the vector represents the target position of the i-th UAV, vector The lth component in is expressed as: Here, l∈{1,2,3}; sgn(·) is the sign function; c i and ε i,l are all positive control parameters, is a vector The lth component of ; Step 504, control the drone base station to fly according to the dynamic equation to achieve collision avoidance. The drone base station starts working after flying to the designated position to repair the network coverage hole.

Images