An intelligent algorithm driven UAV and control vehicle communication optimization system and method

Abstract

The present invention discloses an intelligent algorithm driven unmanned aerial vehicle and control vehicle communication optimization system and method, and relates to the field of unmanned aerial vehicle communication technology. The system acquires the environmental data of the unmanned aerial vehicle operation area and the surroundings of the operation area, as well as the communication quality parameters between the unmanned aerial vehicle and the control vehicle in real time through a data acquisition module; establishes and maintains a dual data communication link between the unmanned aerial vehicle and the ground control vehicle through a data communication module, monitors the state of the communication link in real time, and dynamically adjusts the transmission strategy; uses the received environmental data and communication quality parameters, uses an intelligent algorithm to calculate the optimal communication position point of the control vehicle, and realizes the global and local optimal communication position selection by introducing the gray wolf algorithm and the reinforcement learning model in the intelligent algorithm module; issues a control command to the control vehicle through a control command module, adjusts the position of the control vehicle, and ensures the data communication quality and stability between the unmanned aerial vehicle and the ground control vehicle.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle communication technology, and in particular to an intelligent algorithm-driven unmanned aerial vehicle and control vehicle communication optimization system and method. Background Art

[0002] When the drone group is monitoring the operation area, it needs to communicate data with the ground control vehicle in real time. The data communication between the drone and the control vehicle mainly relies on fixed communication base stations or preset communication paths. However, fixed communication base stations cannot provide the best communication position anytime and anywhere, and preset communication paths are difficult to cope with dynamic changes in the environment.

[0003] First, changes in communication distance have a direct impact on the stability of data transmission. When performing a mission, the drone needs to move in a wide area, causing the distance between it and the control vehicle to change constantly. Traditional fixed communication base stations or paths cannot adjust their positions at any time, often resulting in communication interruptions or data loss, affecting the smooth progress of the mission. Secondly, electromagnetic interference is an important factor affecting the quality of data communication. In complex environments such as cities, mountains, and industrial areas, electromagnetic interference sources are widely distributed and change at any time, further increasing the uncertainty of data communication. In the existing technology, there is a lack of effective means to monitor and avoid these interference sources in real time, resulting in the inability to guarantee communication quality.

[0004] In order to overcome the above difficulties, some improvement measures have been proposed in the prior art, such as increasing the number of communication base stations, using high-frequency communication equipment, etc. However, these methods do not fundamentally solve the communication location and interference problems, and still have many limitations. Summary of the invention

[0005] In view of the impact of factors such as communication distance and electromagnetic interference on the quality of data communication between UAVs and control vehicles, the present invention proposes a UAV and control vehicle communication optimization system and method driven by an intelligent algorithm. The UAV collects environmental data of the operation area and its surroundings in real time, and uses an intelligent algorithm to calculate the optimal communication position point of the control vehicle, which helps managers to regulate the control vehicle, thereby ensuring the stability and security of data communication in complex environments.

[0006] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0007] A communication optimization system between a drone and a control vehicle driven by an intelligent algorithm, the system comprising:

[0008] A data acquisition module is used to obtain real-time environmental data of the UAV operation area and the surrounding area, as well as communication quality parameters between the UAV and the control vehicle;

[0009] The data communication module is used to establish and maintain a dual data communication link between the UAV and the ground control vehicle, monitor the status of the communication link in real time, and dynamically adjust the transmission strategy;

[0010] An intelligent algorithm module, used to calculate the optimal communication position point of the control vehicle by using the received environmental data and the communication quality parameters using an intelligent algorithm;

[0011] The control instruction module is used to send control instructions to the control vehicle and adjust the position of the control vehicle according to the optimal communication position point calculated by the intelligent algorithm module.

[0012] As a preferred solution of the present invention, the environmental data includes terrain data, meteorological data, obstacle distribution data and electromagnetic interference source data, and the communication quality parameters include signal strength, signal-to-noise ratio and data transmission rate between the UAV and the control vehicle.

[0013] As a preferred solution of the present invention, the data acquisition module includes:

[0014] A data acquisition unit, used to collect the environmental data in real time through sensors and cameras carried by the drone, and to collect the communication quality parameters in real time by monitoring the communication link through a data communication module;

[0015] The data preprocessing unit is used to clean, filter and standardize the environmental data and the communication quality parameters.

[0016] As a preferred solution of the present invention, the intelligent algorithm module includes:

[0017] A map search unit, configured to construct an environmental map of the UAV operation area according to the environmental data, search on the environmental map using an efficient local map search algorithm, and determine a preliminary candidate position list of the best communication position point according to the communication quality parameter;

[0018] A global optimization unit, used for performing global optimization on the basis of the preliminary candidate location list by using a grey wolf algorithm to generate an optimized communication location list;

[0019] The fine optimization unit is used to use the reinforcement learning algorithm to perform model training on the optimized communication location list, and generate the final optimal communication location point in combination with real-time environmental data.

[0020] As a preferred solution of the present invention, in the map search unit, the process of constructing the environment map includes:

[0021] Construct a terrain elevation model based on terrain data, and use grids to represent the altitude of each grid point;

[0022] Convert meteorological data into impact values ​​on communication signals and construct a meteorological impact matrix;

[0023] Mark the distribution and height of obstacles in three-dimensional space according to the obstacle distribution data to form an obstacle influence map;

[0024] Generate an electromagnetic interference distribution map according to the location and intensity of the electromagnetic interference source in the electromagnetic interference source data;

[0025] Integrate terrain elevation model, meteorological influence matrix, obstacle influence map and electromagnetic interference distribution map into a unified spatial model to form a comprehensive environmental map;

[0026] The efficient local map search algorithm includes:

[0027] Set the initial search range and resolution, and define communication quality evaluation indicators, including the comprehensive signal strength index CSI and the signal-to-noise ratio weighted indicator SNR;

[0028] The environment map is divided into several 10m×10m grids, each grid is regarded as a local area, and the A* algorithm is used to search for the candidate location point with the best communication quality in each local area. The heuristic function in the A* algorithm is the communication quality score f(x,y), and the calculation formula is:

[0029] f(x,y)=g(x,y)+h(x,y);

[0030] Where g(x,y) is the actual loss from the starting point to the current position (x,y), and h(x,y) is the estimated loss from the current position (x,y) to the target point;

[0031] Calculate the communication quality evaluation index of the candidate location point with the best communication quality in each local area, and then calculate the comprehensive communication quality score. The calculation formula is as follows:

[0032] CSI i =P t -L i -F i ;

[0033]

[0034] Q i =ω1·CSI i +ω2·SNR i ;

[0035] Where, CSI i is the comprehensive signal strength index of the i-th candidate location point, P t is the UAV transmission signal power, L i is the path loss, F iis the environmental interference loss; SNR i is the signal-to-noise ratio weighted index of the i-th candidate location point, N i is the noise power; Q i is the communication quality score of the i-th candidate location point; ω1 and ω2 are the weight coefficients of the comprehensive signal strength index and the signal-to-noise ratio weighted index respectively;

[0036] The first m candidate locations with the highest comprehensive communication quality scores are added to the preliminary candidate location list.

[0037] As a preferred solution of the present invention, the grey wolf algorithm specifically includes:

[0038] Set the number of gray wolves N and the maximum number of iterations and initialize them;

[0039] Randomly generate the position X of each gray wolf individual j (j=1, 2, ..., N) are distributed in the search space of the preliminary candidate position list;

[0040] For each individual gray wolf, j , calculate the fitness value G(X j ), the calculation formula is:

[0041] G(X j )=q1signal_strength(X j )+q2B(X j )+q3data_rate(X j )-q4interference(X j );

[0042] Where q1, q2, q3, q4 are weighted coefficients; signal_strength(X j )、B(X j )、data_rate(X j )、interference(X j ) are respectively the positions X j Signal strength, signal-to-noise ratio, data transmission rate and electromagnetic interference intensity at the location;

[0043] Sort the fitness values ​​and select the three positions with the highest fitness values ​​as the optimal positions X α , suboptimal position X β and the third best position X δ ;

[0044] Calculate the position X of each individual gray wolf j Update value X j+1 :

[0045]

[0046]

[0047] Where D α , D β , D δ are the positions X of the gray wolf individuals j With the optimal position X α , suboptimal position X β and the third best position X δ distance; C1, C2, C3 represent adjustment coefficients, which are random numbers; A1, A2, A3 are dynamic adjustment parameters, which decrease in sequence;

[0048] Repeat the above steps and iteratively update the gray wolf position until the maximum number of iterations is reached or the convergence condition is met, and add all the gray wolf positions after iterative optimization to the optimized communication position list.

[0049] As a preferred solution of the present invention, the reinforcement learning model is specifically: the state space includes environmental data and communication quality parameters, the action space includes the moving direction and step length of the vehicle, and the reward function is designed to evaluate the quality of each state-action pair. The calculation formula of the reward r is:

[0050] r=w1·signal_strength(s′)+w2·B(s′)+w3·data_rate(s′)-w4·Cost(a);

[0051] Where s′ is the next state after executing action a; signal_strength(s′), B(s′), data_rate(s′) are the signal strength, signal-to-noise ratio, and data transmission rate in the next state s′ after executing action a; w1, w2, w3, and w4 are weight coefficients; Cost(a) is the movement cost of executing action a;

[0052] The training steps of the reinforcement learning model include:

[0053] Initialize the Q table Q(s,a), which represents the expected benefit of taking action a in the current state s;

[0054] Initialize model parameters, including learning rate Discount factor γ and exploration rate ε;

[0055] The optimized communication location list and environmental data are combined to generate a training data set. Each training sample (s, a, r, s′) includes: the current state s constructed according to the environmental data and communication quality parameters, the action a selected by the ε-greedy strategy, the reward r obtained by executing the selected action a, and the next state s′;

[0056] For each training sample (s, a, r, s′), the Q-learning algorithm is used to update the Q value. The Q value update formula is:

[0057]

[0058] The Q value is updated iteratively until the Q table converges or the preset number of training rounds is reached.

[0059] As a preferred solution of the present invention, the control instruction module includes a control instruction generation unit and a position adjustment unit; the control instruction generation unit is used to generate control instructions based on the calculated optimal communication position point, including the moving direction and step length of the control vehicle; the position adjustment unit is used to adjust the position of the control vehicle according to the control instruction.

[0060] A method for optimizing communication between a drone and a control vehicle driven by an intelligent algorithm, the method comprising:

[0061] Establish and maintain dual data communication links between the UAV and the ground control vehicle, monitor the status of the communication links in real time, and dynamically adjust the transmission strategy;

[0062] Acquire environmental data of the UAV operation area and the surrounding area in real time, as well as communication quality parameters between the UAV and the control vehicle, and clean, filter and standardize the environmental data and the communication quality parameters;

[0063] Utilizing the environmental data and the communication quality parameters, an intelligent algorithm is used to calculate the optimal communication position point for controlling the vehicle;

[0064] According to the optimal communication position point calculated by the intelligent algorithm module, a control instruction is sent to the control vehicle to adjust the position of the control vehicle.

[0065] As a preferred solution of the present invention, the method of using an intelligent algorithm to calculate the optimal communication position point of the control vehicle includes:

[0066] constructing an environmental map of the drone operation area based on the environmental data, searching on the environmental map using an efficient local map search algorithm, and determining a preliminary candidate location list of optimal communication location points based on communication quality parameters;

[0067] Perform global optimization based on the preliminary candidate location list through the grey wolf algorithm to generate an optimized communication location list;

[0068] The reinforcement learning algorithm is used to train the model of the optimized communication location list, and the final optimal communication location point is generated by combining the real-time environmental data.

[0069] The beneficial effects of the present invention are as follows: comprehensive acquisition and processing of environmental data and communication quality parameters ensure stable communication in complex environments; the intelligent algorithm module realizes global and local optimal communication position selection by introducing the gray wolf algorithm and reinforcement learning model, thereby improving the intelligence level of the system; dynamically adjusts the position of the control vehicle according to real-time environmental data to adapt to environmental changes and ensure that the communication link is always in the best state; optimizes the position of the control vehicle through the intelligent algorithm to ensure the data communication quality and stability between the UAV and the ground control vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0071] in:

[0072] Figure 1 It is a system modular structure diagram of the present invention;

[0073] Figure 2 4 is a flow chart of a method in an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0075] like Figure 1 As shown, it is an embodiment of the present invention, which provides a communication optimization system between a UAV and a control vehicle driven by an intelligent algorithm, mainly including a data acquisition module, a data communication module, an intelligent algorithm module and a control instruction module.

[0076] (1) A data acquisition module, which is used to obtain real-time environmental data of the UAV operation area and the surrounding area, as well as communication quality parameters between the UAV and the control vehicle;

[0077] In one embodiment, the data acquisition module includes:

[0078] The data acquisition unit is used to collect environmental data in real time through the sensors and cameras carried by the drone, and to collect communication quality parameters in real time through the monitoring of the communication link by the data communication module;

[0079] Environmental data include terrain data (terrain height, slope, ground cover type, etc.), meteorological data (temperature, humidity, wind speed, etc.), obstacle distribution data (location and height of obstacles such as buildings, trees, and mountains), electromagnetic interference source data (location and interference intensity of substations, radar stations, etc.), etc. Communication quality parameters include signal strength, signal-to-noise ratio, data transmission rate, etc. of communication between the drone and the control vehicle;

[0080] A data preprocessing unit, used for cleaning, filtering and standardizing environmental data and communication quality parameters;

[0081] Remove noise and outliers through data cleaning, fill in the actual data, use filtering techniques (such as Kalman filtering) to smooth the data, and standardize data from different sources and scales to ensure consistency in subsequent algorithm processing. Store the preprocessed data in a database for subsequent use.

[0082] (2) Data communication module, used to establish and maintain dual data communication links between the UAV and the ground control vehicle, monitor the status of the communication link in real time, and dynamically adjust the transmission strategy;

[0083] Initialize the communication link between the UAV and the control vehicle to ensure the establishment of dual links (such as the main link and the backup link); monitor the status of the communication link in real time, including signal strength, signal-to-noise ratio, data transmission rate, etc.; dynamically adjust the transmission strategy (such as switching to the backup link, adjusting the transmission power) according to the status of the communication link.

[0084] (3) an intelligent algorithm module, which is used to calculate the optimal communication position point of the control vehicle using the received environmental data and communication quality parameters using an intelligent algorithm;

[0085] In one embodiment, the intelligent algorithm module includes:

[0086] A map search unit, used to construct an environmental map of the UAV operation area based on environmental data, search on the environmental map using an efficient local map search algorithm, and determine a preliminary candidate location list of the best communication location point based on communication quality parameters;

[0087] Furthermore, the process of constructing the environment map includes:

[0088] Construct a terrain elevation model (Digital Elevation Model, DEM) based on terrain data, using grids to represent the altitude of each grid point;

[0089] Convert meteorological data into impact values ​​on communication signals and construct a meteorological impact matrix;

[0090] Mark the distribution and height of obstacles in three-dimensional space according to the obstacle distribution data to form an obstacle influence map;

[0091] Generate an electromagnetic interference distribution map according to the location and intensity of the electromagnetic interference source in the electromagnetic interference source data;

[0092] The terrain elevation model, meteorological impact matrix, obstacle impact map and electromagnetic interference distribution map are integrated into a unified spatial model to form a comprehensive environmental map.

[0093] Efficient local map search algorithms include:

[0094] Set the initial search range and resolution, and define communication quality evaluation indicators, including the comprehensive signal strength index CSI and the signal-to-noise ratio weighted indicator SNR;

[0095] The environment map is divided into several 10m×10m grids, each grid is regarded as a local area, and the A* algorithm is used to search for the candidate location point with the best communication quality in each local area. The heuristic function in the A* algorithm is the communication quality score f(x,y), and the calculation formula is:

[0096] f(x,y)=g(x,y)+h(x,y);

[0097] Where g(x,y) is the actual loss from the starting point to the current position (x,y) (such as path loss, signal strength loss), and h(x,y) is the estimated loss from the current position (x,y) to the target point (such as the signal loss estimate of the remaining distance).

[0098] Calculate the communication quality evaluation index of the candidate location point with the best communication quality in each local area, and then calculate the comprehensive communication quality score. The calculation formula is as follows:

[0099] CSI i =P t -L i -F i ;

[0100]

[0101] Q i =ω1·CSI i +ω2·SNR i ;

[0102] Where, CSI i is the comprehensive signal strength index of the i-th candidate location point, P t is the UAV transmission signal power, L i is the path loss, F i is the environmental interference loss; SNR i is the signal-to-noise ratio weighted index of the i-th candidate location point, N i is the noise power; Q i is the communication quality score of the i-th candidate location point; ω1 and ω2 are the weight coefficients of the comprehensive signal strength index and the signal-to-noise ratio weighted index respectively;

[0103] Path loss L i Calculated by the path loss model, such as: L i =20log 10 (d i )+20log 10 (f)+32.45, d i is the propagation distance, f signal is the frequency;

[0104] Environmental interference loss F i Calculated from meteorological and electromagnetic interference data, such as F i = meteorological influence (x, y) + electromagnetic interference (x, y);

[0105] The first m candidate locations with the highest comprehensive communication quality scores are added to the preliminary candidate location list. The candidate locations in the preliminary candidate location list need to meet basic conditions, such as accessibility of the communication link, terrain suitability, etc.

[0106] A global optimization unit, used for performing global optimization on the basis of the preliminary candidate location list by using a grey wolf algorithm to generate an optimized communication location list;

[0107] The Grey Wolf Algorithm specifically includes:

[0108] Set the number of gray wolves N and the maximum number of iterations M and initialize them;

[0109] Randomly generate the position X of each gray wolf individual j (j=1,2,...,N) are distributed in the search space of the preliminary candidate position list, and each random position represents a potential communication position point;

[0110] For each individual gray wolf, j , calculate the fitness value G(X j ), the calculation formula is:

[0111] G(X j ) = q1signal_strength(Xj )+q2B(X j )+q3data_rate(X j )-q4interference(X j );

[0112] Where q1, q2, q3, q4 are weighted coefficients; signal_strength(X j )、B(X j )、data_rate(X j )、interference(X j ) are respectively the positions X j Signal strength, signal-to-noise ratio, data transmission rate and electromagnetic interference intensity at the location;

[0113] Sort the fitness values ​​and select the three positions with the highest fitness values ​​as the optimal positions X α , suboptimal position X β and the third best position X δ ;

[0114] Calculate the position X of each individual gray wolf j Update value X j+1 :

[0115]

[0116]

[0117] Where D α , D β , D δ are the positions X of the gray wolf individuals j With the optimal position X α , suboptimal position X β and the third best position X δ distance; C1, C2, C3 represent adjustment coefficients, which are random numbers; A1, A2, A3 are dynamic adjustment parameters, which decrease in sequence;

[0118] Repeat the above steps and iteratively update the gray wolf position until the maximum number of iterations is reached or the convergence condition is met, and add all the gray wolf positions after iterative optimization to the optimized communication position list.

[0119] A fine optimization unit is used to train the optimized communication location list using a reinforcement learning model and generate the final optimal communication location point in combination with real-time environmental data;

[0120] In a preferred embodiment, the reinforcement learning model is specifically as follows: the state space includes environmental data and communication quality parameters, the action space includes a set of actions that can be taken by the control vehicle (movement direction and step length), and a reward function is designed to evaluate the quality of each state-action pair. The calculation formula of the reward r is:

[0121] r=w1·signal_strength(s′)+w2·B(s′)+w3·data_rate(s′)-w4·Cost(a);

[0122] Where s′ is the next state after executing action a; signal_strength(s′), B(s′), data_rate(s′) are the signal strength, signal-to-noise ratio, and data transmission rate in the next state s′ after executing action a; w1, w2, w3, and w4 are weight coefficients; Cost(a) is the movement cost of executing action a;

[0123] The training steps of the reinforcement learning model include:

[0124] Initialize the Q table Q(s,a), which represents the expected benefit of taking action a in the current state s;

[0125] Initialize model parameters, including learning rate Discount factor γ and exploration rate ε;

[0126] The optimized communication location list and environmental data are combined to generate a training data set. Each training sample (s, a, r, s′) includes: the current state s constructed according to the environmental data and communication quality parameters, the action a selected by the ε-greedy strategy, the reward r obtained by executing the selected action a, and the next state s′;

[0127] For each training sample (s, a, r, s′), the Q-learning algorithm is used to update the Q value. The Q value update formula is:

[0128]

[0129] The Q value is updated iteratively until the Q table converges or the preset number of training rounds is reached.

[0130] (4) A control instruction module, which is used to send control instructions to the control vehicle and adjust the position of the control vehicle according to the optimal communication position point calculated by the intelligent algorithm module.

[0131] In one embodiment, the control instruction module includes a control instruction generation unit and a position adjustment unit; the control instruction generation unit is used to generate a control instruction based on the calculated optimal communication position point, including the moving direction and step length of the control vehicle; the position adjustment unit is used to adjust the position of the control vehicle according to the control instruction.

[0132] like Figure 2 FIG. 1 is another embodiment of the present invention, which provides a method for optimizing communication between a drone and a control vehicle driven by an intelligent algorithm, comprising the following steps:

[0133] S1: Establish and maintain dual data communication links between the UAV and the ground control vehicle, monitor the status of the communication links in real time, and dynamically adjust the transmission strategy;

[0134] S2: Real-time acquisition of environmental data of the UAV operation area and its surroundings, as well as communication quality parameters between the UAV and the control vehicle, and cleaning, filtering and standardization of the environmental data and communication quality parameters;

[0135] S3: Using environmental data and communication quality parameters, an intelligent algorithm is used to calculate the optimal communication position point for controlling the vehicle, including:

[0136] S31: construct an environmental map of the UAV operation area based on the environmental data, search on the environmental map using an efficient local map search algorithm, and determine a preliminary candidate location list of the best communication location point based on the communication quality parameters;

[0137] S32: performing global optimization based on the preliminary candidate location list by using the grey wolf algorithm to generate an optimized communication location list;

[0138] S33: Use reinforcement learning algorithm to train the model of the optimized communication location list, and generate the final optimal communication location point in combination with real-time environmental data.

[0139] S4: Based on the optimal communication position point calculated by the intelligent algorithm module, a control instruction is issued to the control vehicle to adjust the position of the control vehicle.

[0140] In summary, the present invention improves the comprehensiveness and accuracy of environmental data by introducing terrain data, meteorological data, obstacle distribution data and electromagnetic interference source data. The cleaning, filtering and standardization of data remove noise and outliers, improve the accuracy and consistency of data, and provide a reliable data basis for subsequent intelligent algorithms. The intelligent algorithm module can efficiently determine the best communication location point through the cooperation of three units: map search, global optimization and fine optimization, realize the global and local optimal location selection, and significantly improve the intelligence level of the system. By constructing a terrain elevation model, a meteorological influence matrix, an obstacle influence map and an electromagnetic interference distribution map, the accurate construction of the environmental map is realized. The use of the A* algorithm enables the candidate location point with the best communication quality to be quickly found in each local area. The iterative update mechanism of the gray wolf algorithm ensures that the optimal location is continuously found in a dynamic environment. The introduction of the reinforcement learning model realizes self-learning and optimization of the communication location selection process. The reasonable design of the state space and action space, combined with the evaluation mechanism of the reward function, enables the system to adaptively adjust the communication location in a constantly changing environment, improves the flexibility of the system, enables the drone to maintain stable communication in a complex environment, and further improves the robustness and reliability of the system.

[0141] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or other arbitrary combinations. When implemented using software, it can be implemented in whole or in part in the form of a computer program product, which includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0142] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.

[0143] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent algorithm driven UAV and control vehicle communication optimization system, characterized in that: The system comprises: A data acquisition module is used to obtain real-time environmental data of the UAV operation area and the surrounding area, as well as communication quality parameters between the UAV and the control vehicle; The data communication module is used to establish and maintain a dual data communication link between the UAV and the ground control vehicle, monitor the status of the communication link in real time, and dynamically adjust the transmission strategy; An intelligent algorithm module is used to use the received environmental data and the communication quality parameters to calculate the optimal communication position point of the control vehicle using an intelligent algorithm; the intelligent algorithm module includes: A map search unit, configured to construct an environmental map of the UAV operation area according to the environmental data, search on the environmental map using an efficient local map search algorithm, and determine a preliminary candidate position list of the best communication position point according to the communication quality parameter; A global optimization unit, used for performing global optimization on the basis of the preliminary candidate location list by using a grey wolf algorithm to generate an optimized communication location list; A fine optimization unit is used to use a reinforcement learning algorithm to perform model training on the optimized communication location list and generate the final optimal communication location point in combination with real-time environmental data; The control instruction module is used to send control instructions to the control vehicle and adjust the position of the control vehicle according to the optimal communication position point calculated by the intelligent algorithm module.

2. According to the intelligent algorithm driven UAV and control vehicle communication optimization system of claim 1, it is characterized in that: The environmental data includes terrain data, meteorological data, obstacle distribution data and electromagnetic interference source data, and the communication quality parameters include signal strength, signal-to-noise ratio and data transmission rate of communication between the UAV and the control vehicle.

3. According to the intelligent algorithm driven UAV and control vehicle communication optimization system of claim 1, it is characterized in that: The data acquisition module comprises: A data acquisition unit, used to collect the environmental data in real time through sensors and cameras carried by the drone, and to collect the communication quality parameters in real time by monitoring the communication link through a data communication module; The data preprocessing unit is used to clean, filter and standardize the environmental data and the communication quality parameters.

4. According to the intelligent algorithm driven UAV and control vehicle communication optimization system of claim 1, it is characterized in that: In the map search unit, the process of constructing the environment map includes: Construct a terrain elevation model based on terrain data, and use grids to represent the altitude of each grid point; Convert meteorological data into impact values ​​on communication signals and construct a meteorological impact matrix; Mark the distribution and height of obstacles in three-dimensional space according to the obstacle distribution data to form an obstacle influence map; Generate an electromagnetic interference distribution map according to the location and intensity of the electromagnetic interference source in the electromagnetic interference source data; Integrate terrain elevation model, meteorological influence matrix, obstacle influence map and electromagnetic interference distribution map into a unified spatial model to form a comprehensive environmental map; The efficient local map search algorithm includes: Set the initial search range and resolution, and define communication quality evaluation indicators, including the comprehensive signal strength index CSI and the signal-to-noise ratio weighted indicator SNR; The environment map is divided into several 10m×10m grids, each grid is regarded as a local area, and the A* algorithm is used to search for the candidate location point with the best communication quality in each local area. The heuristic function in the A* algorithm is the communication quality score f(x,y), and the calculation formula is: f(x,y)=g(x,y))h(x,y); Where g(x,y) is the actual loss from the starting point to the current position (x,y), and h(x,y) is the estimated loss from the current position (x,y) to the target point; Calculate the communication quality evaluation index of the candidate location point with the best communication quality in each local area, and then calculate the comprehensive communication quality score. The calculation formula is as follows: CSI i =P t -L i -F i ; Q i =ω1×CSI i )ω2×SNR i ; Where, CSI i is the comprehensive signal strength index of the i-th candidate location point, P t is the UAV transmission signal power, L i is the path loss, F i is the environmental interference loss; SNR i is the signal-to-noise ratio weighted index of the i-th candidate location point, N i is the noise power; Q i is the communication quality score of the i-th candidate location point; ω1 and ω2 are the weight coefficients of the comprehensive signal strength index and the signal-to-noise ratio weighted index respectively; The first m candidate locations with the highest comprehensive communication quality scores are added to the preliminary candidate location list.

5. The intelligent algorithm driven UAV and control vehicle communication optimization system according to claim 1, characterized in that: The grey wolf algorithm specifically includes: Set the number of gray wolves N and the maximum number of iterations and initialize them; Randomly generate the position X of each gray wolf individual j (j=1, 2, ..., N) are distributed in the search space of the preliminary candidate position list; For each individual gray wolf, position X j , calculate the fitness value G(X j ), the calculation formula is: G(X j )=q1signal_strength(X j )+q2B(X j )+q3data_rate(X j )-q4interference(X j ); Where q1, q2, q3, q4 are weighted coefficients; signal_strength(X j )、B(X j )、data_rate(X j )、interference(X j ) are respectively the positions X j Signal strength, signal-to-noise ratio, data transmission rate and electromagnetic interference intensity at the location; Sort the fitness values ​​and select the three positions with the highest fitness values ​​as the optimal positions X α , suboptimal position X β and the third best position X δ ; Calculate the position X of each individual gray wolf j Update value X j+1 : Where D α , D β , D δ are the positions X of the gray wolf individuals j With the optimal position X α , suboptimal position X β and the third best position X δ distance; C1, C2, C3 represent adjustment coefficients, which are random numbers; A1, A2, A3 are dynamic adjustment parameters, which decrease in sequence; repeat the above steps, iteratively update the gray wolf position until the maximum number of iterations is reached or the convergence condition is met, and add all the gray wolf positions after iterative optimization to the optimized communication position list.

6. The intelligent algorithm driven UAV and control vehicle communication optimization system according to claim 1, characterized in that: The reinforcement learning algorithm is specifically as follows: the state space includes environmental data and communication quality parameters, the action space includes the moving direction and step length of the vehicle, and a reward function is designed to evaluate the quality of each state-action pair. The calculation formula of the reward r is: r=w1·signal_strength(s′)+w2·B(s′)+w3·data_rate(s′)-w4·Cost(a); Where s′ is the next state after executing action a; signal_strength(s′), B(s′), data_rate(s′) are the signal strength, signal-to-noise ratio and data transmission rate in the next state s′ after executing action a; w1, w2, w3, w4 are weight coefficients; Cost(a) is the movement cost of executing action a; The training steps of a reinforcement learning algorithm include: Initialize the Q table Q(s,a), which represents the expected benefit of taking action a in the current state s; Initialize model parameters, including learning rate Discount factor γ and exploration rate ε; The optimized communication location list and environmental data are combined to generate a training data set. Each training sample (s, a, r, s′) includes: the current state s constructed according to the environmental data and communication quality parameters, the action a selected by the ε-greedy strategy, the reward r obtained by executing the selected action a, and the next state s′; For each training sample (s, a, r, s′), the Q-learning algorithm is used to update the Q value. The Q value update formula is: The Q value is updated iteratively until the Q table converges or the preset number of training rounds is reached.

7. The intelligent algorithm driven UAV and control vehicle communication optimization system according to claim 1, characterized in that: The control instruction module includes a control instruction generation unit and a position adjustment unit; the control instruction generation unit is used to generate a control instruction according to the calculated optimal communication position point, including the moving direction and step length of the control vehicle; the position adjustment unit is used to adjust the position of the control vehicle according to the control instruction.

8. The communication optimization method of the UAV and control vehicle communication optimization system driven by an intelligent algorithm according to any one of claims 1 to 7, characterized in that: The method comprises: Establish and maintain dual data communication links between the UAV and the ground control vehicle, monitor the status of the communication links in real time, and dynamically adjust the transmission strategy; Acquire environmental data of the UAV operation area and the surrounding area in real time, as well as communication quality parameters between the UAV and the control vehicle, and clean, filter and standardize the environmental data and the communication quality parameters; Utilizing the environmental data and the communication quality parameters, an intelligent algorithm is used to calculate the optimal communication position point for controlling the vehicle; According to the optimal communication position point calculated by the intelligent algorithm module, a control instruction is sent to the control vehicle to adjust the position of the control vehicle.

9. The communication optimization method of the intelligent algorithm driven UAV and control vehicle communication optimization system according to claim 8, characterized in that: The method of using an intelligent algorithm to calculate the optimal communication position point for controlling the vehicle includes: constructing an environmental map of the drone operation area based on the environmental data, searching on the environmental map using an efficient local map search algorithm, and determining a preliminary candidate location list of optimal communication location points based on communication quality parameters; Perform global optimization based on the preliminary candidate location list through the grey wolf algorithm to generate an optimized communication location list; The reinforcement learning algorithm is used to train the model of the optimized communication location list, and the final optimal communication location point is generated by combining the real-time environmental data.

Images