Intelligent rotor cooperative communication system of multi-rotor unmanned aerial vehicle

Abstract

The invention discloses an intelligent rotor cooperative communication system of a multi-rotor unmanned aerial vehicle, which belongs to the technical field of communication and comprises an electric aero-engine and rotor integrated model, a distributed rotor cooperative communication module, a fault monitoring and prediction module and an interference observation and prediction module. The electric aero-engine and rotor wing integrated model is used for integrating each rotor wing and the corresponding electric aero-engine and describing the influence of communication faults and communication interference on the propeller wings; a communication mode between rotor control systems is converted into a self-adaptive neural network model through a distributed rotor cooperative communication module, and the communication mode between rotors is dynamically changed by adjusting the connection weight of the self-adaptive neural network model so as to adapt to different flight mission requirements; the fault monitoring and predicting module is used for monitoring and predicting a communication fault of the rotor wing cooperative communication system; and the interference observation and prediction module is used for observing and predicting communication interference on the rotor wing cooperative communication system.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to an intelligent rotor cooperative communication system for a multi-rotor unmanned aerial vehicle. Background Art

[0002] With the introduction of the national concept of "low-altitude economy", drones are increasingly used in various fields. Scenarios such as logistics and transportation, emergency rescue, and manned flights are gradually closely linked to people's daily lives, which has raised the reliability and safety of drones to an unprecedented level. In recent years, drones have been widely used in many industries, and inevitably frequently face complex environments such as "dangerous", "extreme", "special", and "harsh", and the probability of encountering such environments has increased exponentially. This situation has brought huge challenges to the safety and reliability of drones.

[0003] At present, the rotors of existing drones are mainly controlled by their corresponding electric aviation engines, which has obvious limitations. Specifically, there is a lack of mutual communication mechanism between the rotors, which makes it difficult for them to work together in "dangerous", "extreme", "special" and "harsh" environments, and it is impossible to achieve precise control of the drone. The current rotor UAV design has not yet taken the communication function between the rotors into consideration. However, this function plays a vital role in the safe operation of future drones and is very likely to become a key factor in improving the safety performance of drones.

[0004] Therefore, in order to solve the problem of lack of mutual communication mechanism between the rotors of the above-mentioned UAVs, it is urgent to develop an intelligent rotor cooperative communication system for multi-rotor UAVs, which is of great significance for improving the safety, reliability and precise control capabilities of UAVs in complex environments. Summary of the invention

[0005] The present invention aims to solve the above-mentioned problems, remedy the deficiencies of the prior art, and provide an intelligent rotor cooperative communication system for a multi-rotor UAV; the present invention can complete the cooperative communication of multiple rotors, and then realize the cooperative operation of multiple rotors, which can effectively improve the safety and reliability of the operation of multi-rotor UAVs.

[0006] To achieve the above purpose, the present invention adopts the following technical solution.

[0007] The present invention provides an intelligent rotor cooperative communication system for a multi-rotor UAV, comprising an electric aviation engine and rotor integrated model, a distributed rotor cooperative communication module, a fault monitoring and prediction module, an interference observation and prediction module, and a control module; Integrate each rotor with its corresponding electric aircraft engine through the electric aircraft engine and rotor integrated model, and describe the impact of communication failure and communication interference on the rotor; The distributed rotor cooperative communication module converts the communication mode between the rotor control systems into an adaptive neural network model, and dynamically changes the communication mode between the rotors by adjusting the connection weights of the adaptive neural network model to adapt to different flight mission requirements; The fault monitoring and prediction module monitors the occurrence of communication faults with the help of a fault monitoring model, classifies the communication faults, and thus characterizes and quantifies the communication faults; the fault monitoring and prediction module predicts the possibility of the communication fault with the help of a fault prediction model, and once it occurs, activates the fault monitoring model to characterize and quantify the communication fault; The interference observation and prediction module observes the generation of communication interference with the help of the interference observer, classifies the communication interference, and thus characterizes and quantifies the communication interference; the interference observation and prediction module predicts the possible occurrence of communication interference with the help of the interference predictor, and once it occurs, activates the interference observer to characterize and quantify the communication interference; The above communication failure and communication interference information is transmitted to the control module, and the control module processes the communication failure and communication interference through the control end.

[0008] As a preferred solution of the present invention, the electric aircraft engine and rotor integrated model is established by the following mathematical formula, which is expressed as follows: , ; in, represents the torque on each rotor, which is the initial control input of the electric aircraft engine and rotor integration model; is the friction coefficient; is the electrical angular velocity; is the moment of inertia; is the disturbance torque caused by communication failure; is the disturbance torque caused by communication interference.

[0009] As another preferred embodiment of the present invention, the adaptive neural network model is specifically expressed as the following formula: ; in, Output of the adaptive neural network model; , is the input of the adaptive neural network model and represents each rotor control system, ; is the connection weight; n is the number of rotor control systems; BP algorithm or RLS algorithm is used as the learning algorithm of the adaptive neural network model to optimize the connection weight. , there is no communication between any two rotor control systems.

[0010] As another preferred solution of the present invention, the fault monitoring and prediction module uses the Lagrange network as a fault monitoring model, marks no fault, network packet loss and data disorder fault as 0, -1, and 1, and converts it into a three-classification problem. The monitoring data is used as the input of the fault monitoring model, and the network packet loss and data disorder fault are used as the output of the fault monitoring model. The network is trained to complete the characterization and quantification of communication faults.

[0011] As another preferred solution of the present invention, the fault monitoring and prediction module also uses the Lagrange network as a fault prediction model. The fault prediction model uses historical data of two types of faults, network packet loss and data disorder, to predict faults, explore data change patterns, and intervene in impending faults in advance.

[0012] As another preferred solution of the present invention, the interference observation and prediction module uses the Lagrange network as an interference observer, marks no interference, electromagnetic interference and transmission interference as 0, -1, and 1, converts it into a three-classification problem, and completes the characterization and quantification of interference by training the observed data.

[0013] As another preferred solution of the present invention, the interference observation and prediction module also uses the Lagrange network as an interference predictor. The interference predictor uses historical data of electromagnetic interference and transmission interference to predict interference, explore data change rules, and intervene in impending interference in advance.

[0014] As another preferred embodiment of the present invention, the Lagrange network is used as an approximation model and a classification model, and its expression is as follows: ; in, Output for Lagrange network; is the Lagrange network weight; Input for Lagrange network; for The corresponding output: is the coefficient.

[0015] As another preferred solution of the present invention, the Lagrange network optimizes its parameters using any one of the learning algorithms selected from the group consisting of a recursive least squares algorithm, a back propagation algorithm, and a wake-sleep algorithm.

[0016] As another preferred embodiment of the present invention, the control module uses the control input of the updated electric aircraft engine and rotor integration model to process the communication failure and communication interference between the rotors. The control input of the updated electric aircraft engine and rotor integration model is expressed as: ; in, is the initial control input for the electric aero engine and rotor integrated model, The disturbance torque caused by communication failure Estimated value, The disturbance torque caused by communication interference Estimated value.

[0017] Beneficial effects of the present invention: The intelligent rotor cooperative communication system of a multi-rotor UAV provided by the present invention can complete the cooperative communication between the multi-rotors of the UAV, thereby realizing the cooperative operation of the multi-rotors, effectively improving the safety and reliability of UAV applications, monitoring and predicting the communication failures of the rotor cooperative communication system itself, and observing and predicting the communication interference of the rotor cooperative communication system itself.

[0018] To sum up, the intelligent rotor cooperative communication system of the multi-rotor UAV of the present invention has the capabilities of high reliability, strong security, strong autonomy and high intelligence, which improves the intelligence, safety and reliability of the multi-rotor UAV. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a communication mode diagram between the rotor control systems of the intelligent rotor cooperative communication system of a quad-rotor drone taking a quad-rotor as an example of the present invention.

[0020] Figure 2 The present invention is a schematic structural diagram of an adaptive neural network model of an intelligent rotor cooperative communication system of a multi-rotor UAV.

[0021] Markings in the figure: 1 is rotor control system No. 1, 2 is rotor control system No. 2, 3 is rotor control system No. 3, and 4 is rotor control system No. 4. DETAILED DESCRIPTION

[0022] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific implementation methods. It should be understood that the specific implementation methods described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] Combination Figure 1 and Figure 2 As shown, an embodiment of the present invention provides an intelligent rotor cooperative communication system for a multi-rotor UAV, including an electric aircraft engine and rotor integrated model, a distributed rotor cooperative communication module, a fault monitoring and prediction module, an interference observation and prediction module and a control module; each rotor is integrated with its corresponding electric aircraft engine through the electric aircraft engine and rotor integrated model, and the influence of communication faults and communication interference on the rotor is described; the communication mode between each rotor control system is converted into an adaptive neural network model through the distributed rotor cooperative communication module, and the communication mode between each rotor is dynamically changed by adjusting the connection weights of the adaptive neural network model to adapt to different flight mission requirements; the fault monitoring and prediction module is used to The fault monitoring model is used to monitor the occurrence of communication faults, classify the communication faults, and thus characterize and quantify the communication faults; the fault monitoring and prediction module predicts the possible occurrence of communication faults with the help of the fault prediction model, and once it occurs, the fault monitoring model is enabled to characterize and quantify the communication faults; the interference observation and prediction module observes the generation of communication interference with the help of the interference observer, classifies the communication interference, and thus characterizes and quantifies the communication interference; the interference observation and prediction module predicts the possible occurrence of communication interference with the help of the interference predictor, and once it occurs, the interference observer is enabled to characterize and quantify the communication interference; the above-mentioned communication fault and communication interference information is transmitted to the control module, and the control module processes the communication fault and communication interference through the control end.

[0024] in, Figure 1 The communication mode diagram between the rotor control systems of the intelligent rotor cooperative communication system of a quad-rotor UAV taking a quad-rotor as an example is shown. Figure 1 The communication connection mode of any two rotor control systems in the form of a combination between rotor control system No. 1, rotor control system No. 2, rotor control system No. 3, and rotor control system No. 4 is shown. Figure 1 It can be seen that there are 6 communication connection modes, namely the communication connection modes between rotor control system 1 No. 1 and rotor control system 2 No. 2 w 1 , Communication connection mode between rotor control system No. 2 and rotor control system No. 4 w 2 , Communication connection mode between rotor control system 3 No. 3 and rotor control system 4 No. 4 w 3 , Communication connection mode between rotor control system 1 No. 1 and rotor control system 3 No. 3 w 4 , Communication connection mode between rotor control system 2 No. 2 and rotor control system 3 No. 3 w 5, Communication connection mode between rotor control system 1 No. 1 and rotor control system 4 No. 4 w 6 .

[0025] Specifically, the electric aircraft engine and rotor integrated model is used to integrate each rotor with its corresponding electric aircraft engine, and describe the impact of communication failures and communication interference on the rotor; the distributed rotor cooperative communication module includes multiple rotor control systems, and each rotor control system is interconnected to form an adaptive neural network model; wherein, whether any two rotor control systems communicate is determined by the connection weight of the adaptive neural network model; the fault monitoring and prediction module is used to monitor and predict communication failures of the rotor cooperative communication system, and the communication failures include network packet loss and data disorder and other phenomena; the interference observation and prediction module is used to observe and predict communication interference to the rotor cooperative communication system, and the communication interference includes electromagnetic interference and transmission interference; the control module processes communication failures and communication interference through the control end.

[0026] Specifically, the electric aircraft engine and rotor integration model is established by the following mathematical formula, which is expressed as follows: , ; in, represents the torque on each rotor, which is the initial control input of the electric aircraft engine and rotor integration model; is the friction coefficient; is the electrical angular velocity; is the moment of inertia; is the disturbance torque caused by communication failure; is the disturbance torque caused by communication interference.

[0027] like Figure 2 The structural diagram of the adaptive neural network model shown in FIG. 1 is a schematic diagram of the structure of the adaptive neural network model, wherein the adaptive neural network model is specifically expressed as follows: ; in, Output of the adaptive neural network model; , is the input of the adaptive neural network model and represents each rotor control system, ; is the connection weight; n is the number of rotor control systems; BP algorithm or RLS algorithm is used as the learning algorithm of the adaptive neural network model to optimize the connection weight. , there is no communication between any two rotor control systems. Specifically, taking a quadrotor as an example, the adaptive neural network model can be specifically expressed as follows: .

[0028] Specifically, the fault monitoring and prediction module uses the Lagrange network as a fault monitoring model, marks no fault, network packet loss and data disorder fault as 0, -1, and 1, and converts it into a three-classification problem. The monitoring data is used as the input of the fault monitoring model, and the network packet loss and data disorder faults are used as the output of the fault monitoring model. The network is trained to complete the characterization and quantification of communication faults. Among them, the establishment of the fault monitoring model is as follows: using the Lagrange network as a fault monitoring model, the input layer of the fault monitoring model is the network transmission data between each rotor control system, and the output layer of the fault monitoring model is the no-fault data and the faulty data, respectively. The no-fault data is marked as "0", the network packet loss fault is marked as "-1", and the data disorder fault is marked as "1", and the network is trained.

[0029] Specifically, the fault monitoring and prediction module also uses the Lagrange network as a fault prediction model. The fault prediction model uses historical data of two types of faults, network packet loss and data disorder, to predict faults, explore the law of data changes, and intervene in impending faults in advance. Among them, the establishment of the fault prediction model is as follows: using the Lagrange network as a fault prediction model, the input layer of the fault prediction model uses historical data of network packet loss and data disorder faults, and the output layer of the fault prediction model is two types of fault prediction data to train the network.

[0030] Specifically, the interference observation and prediction module uses the Lagrange network as an interference observer, marks no interference, electromagnetic interference and transmission interference as 0, -1, and 1, converts it into a three-classification problem, and completes the characterization and quantification of interference by training the observation data; wherein, the establishment of the interference observer is as follows: using the Lagrange network as an interference observer, the input layer of the interference observer is the network transmission data between each rotor control system, the output layer of the interference observer is the no-interference data and the faulty data, the no-interference data is marked as "0", the electromagnetic interference is marked as "-1", the transmission interference is marked as "1", and the network is trained.

[0031] Specifically, the interference observation and prediction module also uses the Lagrange network as an interference predictor. The interference predictor uses historical data of electromagnetic interference and transmission interference to predict interference, explore the law of data changes, and intervene in the impending interference in advance. Among them, the establishment of the interference predictor is as follows: using the Lagrange network as an interference predictor, the input layer of the interference predictor uses historical data of electromagnetic interference and transmission interference, and the output layer of the interference predictor is two types of interference prediction data to train the network.

[0032] Specifically, the Lagrange network is used as an approximation model and a classification model. The Lagrange network has strong nonlinear approximation, fast convergence, strong generalization ability, "self-learning ability", robustness and anti-interference performance. Its expression is as follows: ; in, Output for Lagrange network; is the Lagrange network weight; Input for Lagrange network; for The corresponding output: is the coefficient.

[0033] Specifically, the Lagrange network optimizes its parameters using any one of the learning algorithms selected from the recursive least squares algorithm, the back propagation algorithm, and the wake-sleep algorithm; the learning algorithm for optimizing the parameters of the Lagrange network may also use other algorithms that can meet the optimization function, and is not limited to the above learning algorithms.

[0034] Specifically, the control module uses the control input of the updated electric aircraft engine and rotor integration model to process the communication failure and communication interference between the rotors. The control input of the updated electric aircraft engine and rotor integration model is expressed as: ; in, is the initial control input for the electric aero engine and rotor integrated model, The disturbance torque caused by communication failure Estimated value, The disturbance torque caused by communication interference Estimated value.

[0035] Based on the technical solutions described above, the following advantages are summarized: (1) The intelligent rotor cooperative communication system of the multi-rotor UAV of the present invention is constituted by the combination of the electric aviation engine and rotor integrated model, the distributed rotor cooperative communication module, the fault monitoring and prediction module, the interference observation and prediction module and the control module. The system has high reliability, strong security, strong autonomy and high intelligence, thereby improving the intelligence, safety and reliability of the multi-rotor UAV.

[0036] (2) The present invention adopts a distributed rotor cooperative communication module to realize the mutual communication connection between the rotor control systems, and transforms the multi-rotor communication problem into an adaptive neural network model. Whether any two rotor control systems communicate is determined by the connection weight; in this way, the problem of the rotor control system being difficult to control in "extremely dangerous" situations can be solved collaboratively.

[0037] (3) The fault monitoring and prediction module of the present invention can be used to monitor and predict communication faults of the rotor cooperative communication system; the fault monitoring and prediction module uses the Lagrange network as a fault monitoring model, takes the monitoring data as the fault monitoring model input, takes the network packet loss and data disorder faults as the fault monitoring model output, trains the network, and completes the characterization and quantification of communication faults; the fault monitoring and prediction module uses the Lagrange network as a fault prediction model, and the fault prediction model uses the historical data of two types of faults, network packet loss and data disorder, to predict faults, explore the law of data changes, and intervene in impending faults in advance.

[0038] (4) The interference observation and prediction module of the present invention can be used to observe and predict communication interference to the rotor cooperative communication system, wherein the interference includes electromagnetic interference and transmission interference; the interference observation and prediction module uses a Lagrange network as an interference observer, the input layer of the interference observer is the network transmission data between the rotor control systems, and the output layer of the interference observer is interference-free data and faulty data, respectively. By training the observed data, the interference is characterized and quantified; the interference observation and prediction module uses a Lagrange network as an interference predictor, and the interference predictor uses historical data of electromagnetic interference and transmission interference to predict interference, explore the law of data change, and intervene in the impending interference in advance.

[0039] (5) The Lagrange network used in the present invention has strong nonlinear approximation, fast convergence, strong generalization ability, "self-learning ability", robustness and anti-interference performance, and can be well applied as an approximation model and classification model.

[0040] It can be understood that the above specific description of the present invention is only used to illustrate the present invention and is not limited to the technical solutions described in the embodiments of the present invention. Those skilled in the art should understand that the present invention can still be modified or replaced by equivalents to achieve the same technical effects; as long as the use requirements are met, they are within the protection scope of the present invention.

Claims

1. An intelligent rotor cooperative communication system for a multi-rotor UAV, characterized in that: It includes an electric aviation engine and rotor integrated model, a distributed rotor collaborative communication module, a fault monitoring and prediction module, an interference observation and prediction module, and a control module; Integrate each rotor with its corresponding electric aircraft engine through the electric aircraft engine and rotor integrated model, and describe the impact of communication failure and communication interference on the rotor; The distributed rotor cooperative communication module converts the communication mode between the rotor control systems into an adaptive neural network model, and dynamically changes the communication mode between the rotors by adjusting the connection weights of the adaptive neural network model to adapt to different flight mission requirements; The fault monitoring and prediction module monitors the occurrence of communication faults with the help of a fault monitoring model, classifies the communication faults, and thus characterizes and quantifies the communication faults; the fault monitoring and prediction module predicts the possibility of the communication fault with the help of a fault prediction model, and once it occurs, activates the fault monitoring model to characterize and quantify the communication fault; The interference observation and prediction module observes the generation of communication interference with the help of the interference observer, classifies the communication interference, and thus characterizes and quantifies the communication interference; the interference observation and prediction module predicts the possible occurrence of communication interference with the help of the interference predictor, and once it occurs, activates the interference observer to characterize and quantify the communication interference; The above communication failure and communication interference information is transmitted to the control module, and the control module processes the communication failure and communication interference through the control end.

2. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 1, characterized in that: The electric aircraft engine and rotor integration model is established by the following mathematical formula, which is expressed as follows: ， ； in, represents the torque on each rotor, which is the initial control input of the electric aircraft engine and rotor integration model; is the friction coefficient; is the electrical angular velocity; is the moment of inertia; is the disturbance torque caused by communication failure; is the disturbance torque caused by communication interference.

3. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 1, characterized in that: The adaptive neural network model is specifically expressed as follows: ； in, Output of the adaptive neural network model; , is the input of the adaptive neural network model and represents each rotor control system, ; is the connection weight; n is the number of rotor control systems; BP algorithm or RLS algorithm is used as the learning algorithm of the adaptive neural network model to optimize the connection weight. , there is no communication between any two rotor control systems.

4. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 1, characterized in that: The fault monitoring and prediction module uses the Lagrange network as a fault monitoring model, marks no fault, network packet loss and data disorder fault as 0, -1, and 1, and converts it into a three-classification problem. The monitoring data is used as the input of the fault monitoring model, and the network packet loss and data disorder fault are used as the output of the fault monitoring model. The network is trained to complete the characterization and quantification of communication faults.

5. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 4, characterized in that: The fault monitoring and prediction module also uses the Lagrange network as a fault prediction model. The fault prediction model uses historical data of two types of faults, network packet loss and data disorder, to predict faults, explore data change patterns, and intervene in impending faults in advance.

6. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 1, characterized in that: The interference observation and prediction module uses the Lagrange network as an interference observer, marks no interference, electromagnetic interference and transmission interference as 0, -1, 1, converts it into a three-classification problem, and completes the characterization and quantification of interference by training the observed data.

7. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 6, characterized in that: The interference observation and prediction module also uses the Lagrange network as an interference predictor. The interference predictor uses historical data of electromagnetic interference and transmission interference to predict interference, explore data change rules, and intervene in the impending interference in advance.

8. An intelligent rotor cooperative communication system for a multi-rotor UAV according to any one of claims 4 to 7, characterized in that: The Lagrange network is used as an approximation model and a classification model, and its expression is as follows: ； in, Output for Lagrange network; is the Lagrange network weight; Input for Lagrange network; for The corresponding output: is the coefficient.

9. The intelligent rotor cooperative communication system for a multi-rotor UAV according to claim 8, characterized in that: The Lagrange network optimizes its parameters using any one of the learning algorithms of the recursive least squares algorithm, the back propagation algorithm, and the wake-sleep algorithm.

10. According to the intelligent rotor cooperative communication system of a multi-rotor UAV in claim 2, the control module uses the control input of the updated electric aircraft engine and rotor integration model to process the communication failure and communication interference between the rotors, and the control input of the updated electric aircraft engine and rotor integration model is expressed as: ； in, is the initial control input for the electric aero engine and rotor integrated model, The disturbance torque caused by communication failure Estimated value, The disturbance torque caused by communication interference Estimated value.

Images