Fixed-wing unmanned aerial vehicle trajectory tracking control system and method based on disturbance observer

Abstract

The invention discloses a fixed-wing unmanned aerial vehicle trajectory tracking control system and method based on a disturbance observer. The system acquires various state data of the unmanned aerial vehicle through the data acquisition module, the interference observer module estimates external interference in real time based on a dynamic model, the trajectory planning module generates an optimized trajectory according to tasks and environments, and the controller module realizes accurate control by adopting a model prediction control and sliding mode control composite strategy. And the execution mechanism driving module executes the control instruction. The method comprises the steps of data acquisition, interference estimation, trajectory planning, controller design and calculation, actuating mechanism driving and the like. The method can effectively improve the trajectory tracking accuracy and stability of the fixed-wing unmanned aerial vehicle in a complex environment, and has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to a fixed-wing UAV trajectory tracking control system and method based on an interference observer. Background Art

[0002] Fixed-wing drones, with their high flight efficiency and long flight times, have been widely used in fields such as surveying and mapping, inspection, logistics and distribution, agricultural and forestry plant protection, and military reconnaissance. In these applications, accurate trajectory tracking is a key factor in ensuring the safe and efficient completion of drone missions.

[0003] However, fixed-wing drones are extremely susceptible to various complex interference factors during actual flight. Externally, atmospheric turbulence can cause severe turbulence during flight, altering the drone's flight attitude and trajectory. Strong winds can not only deviate from the intended route but also pose a serious threat to flight stability. Internally, fluctuations in engine performance can cause thrust instability, further affecting the drone's flight state. Furthermore, the drone's dynamic model often has a certain degree of uncertainty, and sensor measurement inevitably introduces noise. These factors significantly hinder precise trajectory tracking control.

[0004] A review of existing fixed-wing UAV trajectory tracking control systems and methods reveals that, while certain achievements have been made, numerous shortcomings remain. Some traditional control methods have limited ability to estimate and compensate for disturbances. This makes it difficult to effectively guarantee trajectory tracking accuracy in the face of complex and changing disturbances, leading to significant deviations in actual flight. Some trajectory planning algorithms fail to comprehensively consider environmental factors and mission constraints, resulting in suboptimal trajectories that fail to fully leverage the UAV's performance advantages and may even expose the UAV to safety risks during flight. Furthermore, some actuator drive methods lack sufficient control precision and reliability, making it impossible to accurately translate control commands into actual actions, impacting the UAV's ability to track the intended trajectory.

[0005] In view of the above background, developing a fixed-wing UAV control system and method that can effectively deal with complex interference and achieve high-precision trajectory tracking has become an important issue that needs to be urgently solved in the current UAV field. The present invention came into being based on this background. Summary of the Invention

[0006] The present invention provides a fixed-wing UAV trajectory tracking control system and method based on an interference observer, aiming to solve the problems raised by the above background technology.

[0007] The present invention is implemented as follows: a fixed-wing UAV trajectory tracking control system based on an interference observer includes: a data acquisition module: including multiple sensors such as an inertial measurement unit (IMU), a global positioning system (GPS), a barometric altimeter, and an airspeed meter, for real-time acquisition of attitude information, position information, speed information, and acceleration information of the fixed-wing UAV; wherein the attitude information includes pitch angle, roll angle, and yaw angle; the position information includes longitude, latitude, and altitude; and the speed information includes airspeed and ground speed;

[0008] Disturbance Observer Module: This module is based on a dynamic model of a fixed-wing UAV that fully considers the UAV's mass, moment of inertia, aerodynamic force, and aerodynamic torque. By analyzing and processing data collected by sensors, the Disturbance Observer can estimate in real time the external disturbances acting on the UAV, including atmospheric turbulence, engine thrust fluctuations, and other unknown environmental disturbances.

[0009] Trajectory Planning Module: Receives mission target information from the ground control station or pre-set mission target information, and combines it with the UAV's current position and attitude information to plan a flight trajectory that meets mission requirements and flight constraints. The trajectory planning algorithm uses an optimization-based approach to minimize flight time, flight energy consumption, or maximize mission completion efficiency. It also considers constraints such as the UAV's dynamic limitations, flight safety area, and obstacle avoidance requirements. The planned trajectory includes a series of reference points and corresponding attitude and velocity commands, providing target input for trajectory tracking control.

[0010] Controller module: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control.

[0011] Actuator drive module: According to the control instructions output by the controller module, it drives the actuators of the fixed-wing UAV, including the servo and engine throttle. The actuator drive module has the ability of fast response and high-precision control, and can accurately convert the control instructions into the actions of the actuator, thereby achieving the adjustment of the UAV's attitude and flight status.

[0012] Preferably, the interference observer module is specifically implemented as follows:

[0013] Constructing a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, defining the state variables as the UAV's attitude, velocity, and position, and the input variables as control inputs, including rudder angle and throttle opening. The output variables are measurable sensor data.

[0014] Design disturbance observer: Based on the state-space model, use the sliding mode disturbance observer or extended state observer method to design the disturbance observer;

[0015] Adaptive parameter adjustment: The disturbance observer uses an adaptive algorithm to adjust the observation parameters. The adaptive algorithm adjusts the observer's gain parameters in real time based on the disturbance estimation error and changes in the system state variables to improve the disturbance observer's adaptability to different interference environments and flight conditions.

[0016] Preferably, the specific implementation steps of the trajectory planning module are as follows:

[0017] Mission objective analysis: Analyzes mission objective information from the ground control station or pre-set mission objective information to determine the UAV's starting position, target position, mission type, and other mission-related requirements;

[0018] Environmental modeling: Use map information, obstacle information, and weather data to model the UAV's flight environment, build an environmental model including flight no-fly zones, obstacle distribution, and weather conditions (information to provide environmental constraints for trajectory planning;

[0019] Initial trajectory generation: Using a search algorithm such as the Rapidly Exploring Random Tree (RRT) algorithm or the A* algorithm, an initial feasible trajectory is generated from the starting position to the target position, taking into account environmental constraints and the dynamic limitations of the UAV. During the initial trajectory generation process, the optimization goal is to minimize the flight distance or flight time, while ensuring that the trajectory avoids obstacles and no-fly zones.

[0020] Trajectory Optimization: The initial trajectory is optimized using an optimization-based approach. The optimization objectives include minimizing flight energy consumption, maximizing mission completion efficiency, and smoothing the trajectory to reduce the UAV's attitude changes. During the optimization process, the trajectory is adjusted and optimized to obtain the final reference trajectory, taking into account the UAV's dynamic model, flight constraints, and interference information estimated by the interference observer.

[0021] Preferably, the specific design method of the controller module is as follows:

[0022] Model predictive control part:

[0023] Prediction model establishment: Based on the dynamic model of the fixed-wing UAV, a prediction model for model predictive control is established. The prediction model predicts the state of the UAV at multiple moments in the future, including attitude, speed, and position, based on the current system state and future control inputs.

[0024] Objective function definition: Define the objective function of the model predictive control. The objective function usually includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the UAV and the reference trajectory. The control input change term is used to limit the rate of change of the control input to ensure system stability and control smoothness. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements.

[0025] Rolling optimization solution: In each control cycle, model predictive control solves the optimization problem based on the prediction model and objective function to obtain the optimal control input sequence for multiple future moments;

[0026] Sliding surface design: Based on the trajectory tracking error of the fixed-wing UAV, a sliding surface is designed. When moving on the sliding surface, it can achieve stable tracking of the reference trajectory and has a certain degree of robustness.

[0027] Sliding mode control law design: Based on the motion of the sliding surface, a sliding mode control law is designed. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface; the switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface.

[0028] Chattering suppression: The sliding mode control law is improved using methods such as the boundary layer method or integral sliding mode control. The boundary layer method sets a boundary layer near the sliding surface and performs continuous processing of the sliding mode control law within the boundary layer, reducing the frequency of switching control and thus reducing chattering. Integral sliding mode control introduces an integral term to ensure that the system is on the sliding surface at the initial moment, avoiding the chattering problem of traditional sliding mode control in the initial stage.

[0029] Composite control strategy fusion: Model predictive control and sliding mode control are combined to form a composite control strategy. In the composite control strategy, model predictive control is responsible for generating the optimal control input sequence based on the system's predicted state and reference trajectory to achieve accurate trajectory tracking; sliding mode control provides additional control for system uncertainties and interferences, ensuring that the system can still operate stably in the presence of interferences.

[0030] Preferably, a fixed-wing UAV trajectory tracking control method based on an interference observer comprises the following steps:

[0031] Data collection steps: Use sensors such as an inertial measurement unit, global positioning system, barometric altimeter, and airspeed meter to collect the fixed-wing UAV's attitude, position, velocity, and acceleration data in real time. Preprocess the collected data, including filtering and denoising operations, to improve the accuracy and reliability of the data.

[0032] Interference estimation steps: Based on the dynamic model of the fixed-wing UAV, a disturbance observer is used to estimate the external disturbance acting on the UAV in real time. By analyzing and processing sensor data, the disturbance observer separates the external disturbance from the system state variables to obtain the interference estimate. The disturbance observer uses an adaptive algorithm to automatically adjust the observation parameters according to the changes in the UAV's flight state to improve the accuracy of the interference estimation.

[0033] Trajectory planning: Receive mission objective information from the ground control station or pre-set mission objective information, combine it with the UAV's current position and attitude information, and plan a flight trajectory that meets the mission requirements and flight constraints. The trajectory planning process includes mission objective analysis, environment modeling, initial trajectory generation, and trajectory optimization. The rapid exploration random tree (RRT) algorithm or A* algorithm search algorithm is used to generate the initial feasible trajectory. The initial trajectory is then optimized through nonlinear programming (NLP) or model predictive control (MPC) optimization methods to obtain the final reference trajectory.

[0034] Controller design and calculation steps: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control. The model predictive control part uses the UAV's dynamic model to predict the system state at multiple moments in the future and calculates the optimal control input sequence by optimizing the objective function. The sliding mode control part designs the sliding surface and sliding mode control law based on the trajectory tracking error to improve the robustness of the system. In the controller design process, the UAV's dynamic constraints, control input constraints, and the disturbance information estimated by the disturbance observer are considered. By reasonably adjusting the control parameters, the controller can achieve optimal control performance under different flight conditions.

[0035] Actuator driving steps: According to the control instructions output by the controller module, the actuators of the fixed-wing UAV, including the servo and engine throttle, are driven. The actuator driving module converts the control instructions into the actions of the actuators, thereby adjusting the UAV's attitude and flight status. During the actuator driving process, the actions of the actuators are monitored and feedback is provided to ensure that the actuators can accurately execute the control instructions.

[0036] Preferably, in the interference estimation step, a sliding mode interference observer or an extended state observer is used to perform interference estimation, and the specific steps are as follows:

[0037] Construct a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, define state variables, input variables, and output variables, and treat external disturbances as uncertain inputs to the system in the state-space model.

[0038] Design a disturbance observer: Using the sliding mode disturbance observer as an example, design a sliding surface and sliding mode control law based on the state-space model. The sliding surface design estimates the system's state variables and disturbance errors, and the sliding mode control law is adjusted based on the movement of the sliding surface. By adjusting the parameters of the sliding mode control law, the observer's output can quickly track disturbance signals in the system.

[0039] Adaptive parameter adjustment: The adaptive algorithm adjusts the observer gain parameters in real time according to the disturbance estimation error and the changes in the system state variables.

[0040] Preferably, in the trajectory planning step, an optimization-based method is used to perform trajectory optimization, and the specific steps are as follows:

[0041] Define optimization objectives: Based on mission requirements and flight performance requirements, define the objective function for trajectory optimization. The objective function includes minimizing flight time, flight energy consumption, and maximizing mission completion efficiency.

[0042] Establish an optimization model: including the objective function, constraints and decision variables. The decision variables include the position, posture and speed of the trajectory points.

[0043] Solve the optimization problem: Utilize a numerical solution algorithm to adjust and optimize the trajectory, taking into account the interference information estimated by the interference observer, to obtain the optimal reference trajectory that meets the mission requirements and flight constraints.

[0044] Preferably, in the controller design and calculation steps, the specific implementation steps of the model predictive control part are as follows:

[0045] Predictive model establishment: Based on the dynamic model of the fixed-wing UAV, a predictive model for model predictive control is established. The predictive model can predict the state of the UAV at multiple moments in the future based on the current system state and future control inputs;

[0046] Objective function definition: Define the objective function of model predictive control. The objective function includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the drone and the reference trajectory. The control input change term is used to limit the rate of change of the control input. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements.

[0047] Rolling optimization solution: In each control cycle, the optimal control input sequence for multiple future moments is obtained by solving the optimization problem based on the prediction model and objective function. The optimization problem is solved numerically using the quadratic programming (QP) algorithm. During the solution process, the UAV's dynamic constraints, control input constraints, and interference information estimated by the interference observer are considered to ensure the feasibility and effectiveness of the optimization results. Only the first control variable in the optimized control input sequence is applied to the UAV system. In the next control cycle, prediction and optimization are re-performed to achieve rolling optimization control.

[0048] Preferably, in the controller design and calculation steps, the specific implementation steps of the sliding mode control part are as follows:

[0049] Sliding surface design: Based on the trajectory tracking error of the fixed-wing UAV, a sliding surface is designed. A linear sliding surface design method is used, using a linear combination of position error and attitude error as the sliding surface variable. The sliding surface should be designed so that the system can stably track the reference trajectory while moving on the sliding surface and has a certain degree of robustness.

[0050] Sliding mode control law design: Design a sliding mode control law based on the motion of the sliding surface. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface. The switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface. Based on the interference information estimated by the interference observer, the switching control gain is adjusted to ensure that the system maintains good control performance in the presence of interference.

[0051] Chattering suppression: To suppress chattering in sliding mode control, the sliding mode control law is improved using methods such as the boundary layer method or integral sliding mode control. The boundary layer method sets a boundary layer near the sliding surface, which makes the sliding mode control law continuous within the boundary layer, thereby reducing the frequency of switching control. Integral sliding mode control introduces an integral term, which enables the system to be on the sliding surface at the initial moment, thus avoiding the chattering problem of traditional sliding mode control in the initial stage.

[0052] Preferably, in the actuator driving step, the action of the actuator is monitored and feedback is provided, which is specifically implemented as follows:

[0053] Sensor feedback: Sensors, including servo angle sensors and throttle position sensors, are installed on the actuators of fixed-wing drones to monitor the actuator's motion status in real time. The data collected by the sensors is fed back to the controller module, which adjusts the control instructions based on the feedback data to ensure that the actuator can accurately execute the control instructions.

[0054] Fault diagnosis and fault-tolerant control: Real-time monitoring and fault diagnosis of the working status of the actuator are carried out. When a fault in the actuator is detected, the fault-tolerant control strategy is used to adjust the system.

[0055] Due to the adoption of the above solution, the beneficial effects of the present invention are:

[0056] Improve trajectory tracking accuracy: The interference observer estimates external interference in real time and feeds the interference information back to the controller, enabling the controller to adjust the control strategy in time according to the interference situation, effectively reducing the impact of interference on the UAV trajectory tracking and improving the trajectory tracking accuracy.

[0057] Enhanced system robustness: A composite control strategy combining model predictive control and sliding mode control is adopted. Model predictive control can generate the optimal control input sequence based on the system's predicted state and reference trajectory, while sliding mode control provides additional control for system uncertainties and disturbances. The combination of the two enables the system to operate stably in the presence of disturbances and model uncertainties, thereby enhancing the system's robustness.

[0058] Optimized trajectory planning: The trajectory planning module uses an optimization-based approach to generate the optimal trajectory based on mission requirements, flight constraints, and interference information, thereby improving the UAV's flight efficiency and mission completion capabilities.

[0059] Improve the control accuracy of the actuator: The actuator drive module ensures that the actuator can accurately execute control instructions through sensor feedback, fault diagnosis and fault-tolerant control, thereby improving the control accuracy and reliability of the actuator. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] like Figure 1 As shown: A fixed-wing UAV trajectory tracking control system based on an interference observer includes: a data acquisition module: including multiple sensors such as an inertial measurement unit (IMU), a global positioning system (GPS), a barometric altimeter, and an airspeed meter, for real-time acquisition of attitude information, position information, speed information, and acceleration information of the fixed-wing UAV; wherein the attitude information includes pitch angle, roll angle, and yaw angle; the position information includes longitude, latitude, and altitude; and the speed information includes airspeed and ground speed;

[0063] Disturbance Observer Module: This module is based on a dynamic model of a fixed-wing UAV that fully considers the UAV's mass, moment of inertia, aerodynamic force, and aerodynamic torque. By analyzing and processing data collected by sensors, the Disturbance Observer can estimate in real time the external disturbances acting on the UAV, including atmospheric turbulence, engine thrust fluctuations, and other unknown environmental disturbances.

[0064] Trajectory Planning Module: Receives mission target information from the ground control station or pre-set mission target information, and combines it with the UAV's current position and attitude information to plan a flight trajectory that meets mission requirements and flight constraints. The trajectory planning algorithm uses an optimization-based approach to minimize flight time, flight energy consumption, or maximize mission completion efficiency. It also considers constraints such as the UAV's dynamic limitations, flight safety area, and obstacle avoidance requirements. The planned trajectory includes a series of reference points and corresponding attitude and velocity commands, providing target input for trajectory tracking control.

[0065] Controller module: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control.

[0066] Actuator drive module: According to the control instructions output by the controller module, it drives the actuators of the fixed-wing UAV, including the servo and engine throttle. The actuator drive module has the ability to respond quickly and achieve high-precision control. It can accurately convert the control instructions into the actions of the actuators, thereby adjusting the UAV's attitude and flight status. The specific implementation of the interference observer module is as follows:

[0067] Constructing a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, defining the state variables as the UAV's attitude, velocity, and position, and the input variables as control inputs, including rudder angle and throttle opening. The output variables are measurable sensor data.

[0068] Design disturbance observer: Based on the state-space model, use the sliding mode disturbance observer or extended state observer method to design the disturbance observer;

[0069] Adaptive parameter adjustment: The disturbance observer uses an adaptive algorithm to adjust the observation parameters. The adaptive algorithm adjusts the observer's gain parameters in real time based on the disturbance estimation error and changes in the system state variables to improve the disturbance observer's adaptability to different interference environments and flight conditions.

[0070] The specific implementation steps of the trajectory planning module are as follows: Mission objective analysis: parsing the mission objective information from the ground control station or pre-set mission objective information to determine the UAV's starting position, target position, mission type and other mission-related requirements;

[0071] Environmental modeling: Use map information, obstacle information, and weather data to model the UAV's flight environment, build an environmental model including flight no-fly zones, obstacle distribution, and weather conditions (information to provide environmental constraints for trajectory planning;

[0072] Initial trajectory generation: Using a search algorithm such as the Rapidly Exploring Random Tree (RRT) algorithm or the A* algorithm, an initial feasible trajectory is generated from the starting position to the target position, taking into account environmental constraints and the dynamic limitations of the UAV. During the initial trajectory generation process, the optimization goal is to minimize the flight distance or flight time, while ensuring that the trajectory avoids obstacles and no-fly zones.

[0073] Trajectory Optimization: The initial trajectory is optimized using an optimization-based approach. The optimization objectives include minimizing flight energy consumption, maximizing mission completion efficiency, and smoothing the trajectory to reduce the UAV's attitude changes. During the optimization process, the trajectory is adjusted and optimized to obtain the final reference trajectory, taking into account the UAV's dynamic model, flight constraints, and interference information estimated by the interference observer.

[0074] The specific design method of the controller module is as follows:

[0075] Model Predictive Control: Prediction model establishment: Based on the dynamic model of the fixed-wing UAV, a prediction model for model predictive control is established. The prediction model predicts the state of the UAV at multiple moments in the future, including attitude, speed, and position, based on the current system state and future control inputs.

[0076] Objective function definition: Define the objective function of the model predictive control. The objective function usually includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the UAV and the reference trajectory. The control input change term is used to limit the rate of change of the control input to ensure system stability and control smoothness. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements.

[0077] Rolling optimization solution: In each control cycle, model predictive control solves the optimization problem based on the prediction model and objective function to obtain the optimal control input sequence for multiple future moments;

[0078] Sliding surface design: Based on the trajectory tracking error of the fixed-wing UAV, a sliding surface is designed. When moving on the sliding surface, it can achieve stable tracking of the reference trajectory and has a certain degree of robustness.

[0079] Sliding mode control law design: Based on the motion of the sliding surface, a sliding mode control law is designed. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface; the switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface.

[0080] Chattering suppression: The sliding mode control law is improved using methods such as the boundary layer method or integral sliding mode control. The boundary layer method sets a boundary layer near the sliding surface and performs continuous processing of the sliding mode control law within the boundary layer, reducing the frequency of switching control and thus reducing chattering. Integral sliding mode control introduces an integral term to ensure that the system is on the sliding surface at the initial moment, avoiding the chattering problem of traditional sliding mode control in the initial stage.

[0081] Composite control strategy fusion: Model predictive control and sliding mode control are combined to form a composite control strategy. In the composite control strategy, model predictive control is responsible for generating the optimal control input sequence based on the system's predicted state and reference trajectory to achieve accurate trajectory tracking; sliding mode control provides additional control for system uncertainties and interferences, ensuring that the system can still operate stably in the presence of interferences.

[0082] A fixed-wing UAV trajectory tracking control method based on interference observer includes the following steps:

[0083] Data collection steps: Use sensors such as an inertial measurement unit, global positioning system, barometric altimeter, and airspeed meter to collect the fixed-wing UAV's attitude, position, velocity, and acceleration data in real time. Preprocess the collected data, including filtering and denoising operations, to improve the accuracy and reliability of the data.

[0084] Interference estimation steps: Based on the dynamic model of the fixed-wing UAV, a disturbance observer is used to estimate the external disturbance acting on the UAV in real time. By analyzing and processing sensor data, the disturbance observer separates the external disturbance from the system state variables to obtain the interference estimate. The disturbance observer uses an adaptive algorithm to automatically adjust the observation parameters according to the changes in the UAV's flight state to improve the accuracy of the interference estimation.

[0085] Trajectory planning: Receive mission objective information from the ground control station or pre-set mission objective information, combine it with the UAV's current position and attitude information, and plan a flight trajectory that meets the mission requirements and flight constraints. The trajectory planning process includes mission objective analysis, environment modeling, initial trajectory generation, and trajectory optimization. The rapid exploration random tree (RRT) algorithm or A* algorithm search algorithm is used to generate the initial feasible trajectory. The initial trajectory is then optimized through nonlinear programming (NLP) or model predictive control (MPC) optimization methods to obtain the final reference trajectory.

[0086] Controller design and calculation steps: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control. The model predictive control part uses the UAV's dynamic model to predict the system state at multiple moments in the future and calculates the optimal control input sequence by optimizing the objective function. The sliding mode control part designs the sliding surface and sliding mode control law based on the trajectory tracking error to improve the robustness of the system. In the controller design process, the UAV's dynamic constraints, control input constraints, and the disturbance information estimated by the disturbance observer are considered. By reasonably adjusting the control parameters, the controller can achieve optimal control performance under different flight conditions.

[0087] Actuator drive step: Based on the control instructions output by the controller module, the fixed-wing UAV's actuators, including the servo and engine throttle, are driven. The actuator drive module converts the control instructions into actuator movements, thereby adjusting the UAV's attitude and flight state. During the actuator drive process, the actuator's movements are monitored and feedback is provided to ensure that the actuator can accurately execute the control instructions. In the interference estimation step, a sliding mode disturbance observer or an extended state observer is used for interference estimation. The specific steps are as follows:

[0088] Construct a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, define state variables, input variables, and output variables, and treat external disturbances as uncertain inputs to the system in the state-space model.

[0089] Design a disturbance observer: Using the sliding mode disturbance observer as an example, design a sliding surface and sliding mode control law based on the state-space model. The sliding surface design estimates the system's state variables and disturbance errors, and the sliding mode control law is adjusted based on the movement of the sliding surface. By adjusting the parameters of the sliding mode control law, the observer's output can quickly track disturbance signals in the system.

[0090] Adaptive parameter adjustment: The adaptive algorithm adjusts the observer gain parameters in real time according to the disturbance estimation error and the changes in the system state variables.

[0091] In the trajectory planning step, an optimization-based method is used to optimize the trajectory. The specific steps are as follows:

[0092] Define optimization objectives: Based on mission requirements and flight performance requirements, define the objective function for trajectory optimization. The objective function includes minimizing flight time, flight energy consumption, and maximizing mission completion efficiency.

[0093] Establish an optimization model: including the objective function, constraints and decision variables. The decision variables include the position, posture and speed of the trajectory points.

[0094] Solve the optimization problem: Utilize a numerical solution algorithm to adjust and optimize the trajectory, taking into account the interference information estimated by the interference observer, to obtain the optimal reference trajectory that meets the mission requirements and flight constraints.

[0095] In the controller design and calculation steps, the specific implementation steps of the model predictive control part are as follows:

[0096] Predictive model establishment: Based on the dynamic model of the fixed-wing UAV, a predictive model for model predictive control is established. The predictive model can predict the state of the UAV at multiple moments in the future based on the current system state and future control inputs;

[0097] Objective function definition: Define the objective function of model predictive control. The objective function includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the drone and the reference trajectory. The control input change term is used to limit the rate of change of the control input. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements.

[0098] Rolling optimization solution: In each control cycle, the optimal control input sequence for multiple future moments is obtained by solving the optimization problem based on the prediction model and objective function. The optimization problem is solved numerically using the quadratic programming (QP) algorithm. During the solution process, the UAV's dynamic constraints, control input constraints, and interference information estimated by the interference observer are considered to ensure the feasibility and effectiveness of the optimization results. Only the first control variable in the optimized control input sequence is applied to the UAV system. In the next control cycle, prediction and optimization are re-performed to achieve rolling optimization control.

[0099] In the controller design and calculation steps, the specific implementation steps of the sliding mode control part are as follows:

[0100] Sliding surface design: Based on the trajectory tracking error of the fixed-wing UAV, a sliding surface is designed. A linear sliding surface design method is used, using a linear combination of position error and attitude error as the sliding surface variable. The sliding surface should be designed so that the system can stably track the reference trajectory while moving on the sliding surface and has a certain degree of robustness.

[0101] Sliding mode control law design: Design a sliding mode control law based on the motion of the sliding surface. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface. The switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface. Based on the interference information estimated by the interference observer, the switching control gain is adjusted to ensure that the system maintains good control performance in the presence of interference.

[0102] Chattering suppression: To suppress chattering in sliding mode control, the sliding mode control law is improved using methods such as the boundary layer method or integral sliding mode control. The boundary layer method sets a boundary layer near the sliding surface, which makes the sliding mode control law continuous within the boundary layer, thereby reducing the frequency of switching control. Integral sliding mode control introduces an integral term, which enables the system to be on the sliding surface at the initial moment, thus avoiding the chattering problem of traditional sliding mode control in the initial stage.

[0103] In the actuator driving step, the action of the actuator is monitored and feedback is provided, which is specifically implemented as follows:

[0104] Sensor feedback: Sensors, including servo angle sensors and throttle position sensors, are installed on the actuators of fixed-wing drones to monitor the actuator's motion status in real time. The data collected by the sensors is fed back to the controller module, which adjusts the control instructions based on the feedback data to ensure that the actuator can accurately execute the control instructions.

[0105] Fault diagnosis and fault-tolerant control: Real-time monitoring and fault diagnosis of the working status of the actuator are carried out. When a fault is detected in the actuator, the fault-tolerant control strategy is used to adjust the system.

[0106] In this embodiment, the present invention uses an inertial measurement unit (IMU), such as the MPU-6050, which has a high sampling frequency of up to 1kHz and can accurately measure the acceleration and angular velocity of the drone, thereby obtaining attitude information such as pitch angle, roll angle and yaw angle. The global positioning system (GPS) uses u-blox NEO-M8N, which has a positioning accuracy of up to 2.5 meters and can provide the longitude and latitude information of the drone in real time. A barometric altimeter such as the MS5611 has an accuracy of up to 10 centimeters and is used to measure the altitude of the drone. The airspeed meter uses a pitot tube airspeed meter, which can accurately measure the airspeed of the drone. The data collected by these sensors is transmitted to the data processing unit via a data bus. In the data processing unit, the collected data is subjected to pre-processing operations such as Kalman filtering to remove noise interference and improve the accuracy and reliability of the data.

[0107] Constructing a state space model: The dynamic equation of the fixed-wing UAV is expressed as:

[0108] Among them, x is the state variable vector, including attitude, speed, position and other information; u is the control input vector, namely the rudder angle and throttle opening; d is the external interference vector; y is the output vector, corresponding to the sensor measurement data; f and h are the corresponding functions. Taking a typical fixed-wing UAV as an example, its state variables can be specifically defined as They represent roll angle, pitch angle, yaw angle, roll angular rate, pitch angular rate, yaw angular rate, airspeed, longitude, latitude, and altitude respectively.

[0109] Taking the sliding mode disturbance observer as an example, let the sliding surface be s=Ce, where is the state estimation error, C is the sliding surface design matrix. The sliding mode control law is designed as u s = -ρsgn(s), where ρ is the control gain and sgn(·) is the sign function. By adjusting the values ​​of C and ρ, the output of the observer can quickly track the interference signal in the system. For example, for a simple attitude control model, after multiple simulations and actual tests, when When ρ=5, the interference observer can better estimate the interference.

[0110] Adaptive law Where γ is the adaptive gain. The value of ρ is adjusted in real time based on the disturbance estimation error and changes in the system state variables. For example, when the disturbance is strong and the disturbance estimation error is large, γ is set to a larger value, such as γ = 0.1, to enable the observer to converge to an accurate disturbance estimate more quickly. When the disturbance is small, γ is set to a smaller value, such as γ = 0.01, to avoid over-adjustment of the observer parameters.

[0111] In this embodiment, the trajectory planning module:

[0112] Mission objective analysis: Assume that the mission of the drone is to survey a certain area, the starting position is (x0, y0, z0), and the center position of the target area is (x t ,y t , z t ), the mission requires that the flight altitude be maintained at z = 100 meters, the flight speed not exceed v = 20 meters per second, and the mission execution time not exceed T = 60 minutes.

[0113] Environmental modeling: Use electronic maps to obtain terrain information of the flight area, and construct an obstacle model using obstacle detection sensors or pre-acquired obstacle distribution data. For example, it is known that there are multiple peaks with a height of h = 50 meters and some buildings with a radius of r = 10 meters in the flight area. At the same time, based on meteorological data, it is determined that there are strong wind areas in the flight area with wind speeds up to v w =15 m / s, direction is northeast.

[0114] Initial trajectory generation: The Rapidly Exploring Random Tree (RRT) algorithm is used to generate an initial feasible trajectory from the starting position to the target area, taking into account environmental constraints and the UAV's dynamic limitations. For example, setting the search step size to Δs = 5 meters and the maximum number of iterations to N = 1000, ultimately generates an initial trajectory that avoids obstacles and strong winds.

[0115] Trajectory optimization: Nonlinear programming (NLP) method is used for trajectory optimization, with minimizing flight energy consumption as the objective function. Assuming that flight energy consumption is related to factors such as flight speed, acceleration, and attitude change, it can be expressed as Among them, α, β, and γ are weight coefficients. After multiple simulations and actual tests, α=0.5, β=0.3, and γ=0.2 are determined. In the optimization process, the dynamic constraints of the UAV are considered, such as the maximum acceleration a max =5 m / s 2 , maximum attitude angle limit φ max =θ max =30°, and the interference information estimated by the interference observer, the trajectory is adjusted and optimized to obtain the final reference trajectory.

[0116] In this embodiment, the model predictive control part: Prediction model establishment: Based on the dynamic model of the fixed-wing UAV, a discretized prediction model is established. For example, the Euler discretization method is used to convert the continuous-time dynamic equation into a discrete-time form: k+1 =x k +Δtf(x k ,u k ), where Δt is the control period, which is set to 0.1 seconds.

[0117] Objective function definition: The objective function is where N p For the prediction time domain, it is set to 10; Q and R are weight matrices. After multiple simulations and actual tests, it is determined that Q = diag(1,1,1,1,1,1,1,1,1), R = diag(0.1,0.1,0.1,0.1); x k | is the predicted state, is the reference trajectory state, u k | is the control input.

[0118] Rolling optimization solution: In each control cycle, the optimal control input sequence for multiple future moments is obtained by solving the quadratic programming (QP) problem. For example, the quadprog function in the MATLAB optimization toolbox is used to solve the problem. During the solution process, the dynamic constraints of the UAV, such as the control surface deflection angle limit δ max =30°, throttle opening limit u min =0, U max = 1, as well as the disturbance information estimated by the disturbance observer, to ensure the feasibility and effectiveness of the optimization results. Only the first control variable in the optimized control input sequence is applied to the UAV system. In the next control cycle, prediction and optimization are re-performed to achieve rolling optimization control.

[0119] Sliding mode control part: Sliding mode surface design: According to the trajectory tracking error of the fixed-wing UAV, the sliding mode surface is designed as where e x ,e y , e z is the position error, e x ,e y , e z is the speed error, k1, k2, k3, k4, k5, k6 are the sliding surface coefficients. After multiple simulations and actual tests, it is determined that k1=k2=k3=1, k4=k5=k6=0.5.

[0120] Objective function definition: The objective function is where N p For the prediction time domain, it is set to 10; Q and R are weight matrices. After multiple simulations and actual tests, it is determined that Q = diag(1,1,1,1,1,1,1,1,1), R = diag(0.1,0.1,0.1,0.1); x k | is the predicted state, is the reference trajectory state, u k | is the control input.

[0121] Rolling optimization solution: In each control cycle, the optimal control input sequence for multiple future moments is obtained by solving the quadratic programming (QP) problem. For example, the quadprog function in the MATLAB optimization toolbox is used to solve the problem. During the solution process, the dynamic constraints of the UAV, such as the control surface deflection angle limit δ max =30°, throttle opening limit u min =0, U max = 1, as well as the disturbance information estimated by the disturbance observer, to ensure the feasibility and effectiveness of the optimization results. Only the first control variable in the optimized control input sequence is applied to the UAV system. In the next control cycle, prediction and optimization are re-performed to achieve rolling optimization control.

[0122] Sliding mode control part: Sliding mode surface design: According to the trajectory tracking error of the fixed-wing UAV, the sliding mode surface is designed as

[0123] where e x , e y , e z is the position error, e x , e y , e z is the speed error, k1, k2, k3, k4, k5, k6 are the sliding surface coefficients. After multiple simulations and actual tests, it is determined that k1=k2=k3=1, k4=k5=k6=0.5.

[0124] Sliding mode control law design: The sliding mode control law is u s =u eq +u sw , where u eq is the equivalent control part, which is used to offset the deterministic part of the system and can be obtained by solving S=0; U SW It is the switching control part, which is used to overcome the uncertainty and interference of the system, and is set as u sw =-ρsgn(s), ρ is the switching control gain, which is adjusted according to the interference information estimated by the interference observer. When the interference is strong, ρ takes a larger value, such as ρ=3; when the interference is small, ρ takes a smaller value, such as ρ=1.

[0125] Chattering suppression: The boundary layer method is used to suppress chattering. A boundary layer Ω = {s::} is set near the sliding surface S = 0, where ε is the boundary layer thickness, set to ε = 0.1. Within the boundary layer, the sliding mode control law is continuous, that is, To reduce the frequency of switching control and reduce vibration.

[0126] Composite control strategy fusion: The composite control strategy is integrated by adjusting the weight coefficients of model predictive control and sliding mode control. Assuming the weight coefficient is λ, the final control input is u=λumpc +(1-λ)us, where u mpc is the control input of the model predictive control output, u s The control input for the sliding mode control output is λ. The value of λ is determined through simulation and actual testing based on different flight conditions, such as interference intensity and mission requirements. For example, when interference is weak and trajectory tracking accuracy is high, λ = 0.8; when interference is strong, λ = 0.5.

[0127] In this embodiment, the actuator drive module uses pulse width modulation (PWM) technology to drive the servo and engine throttle. For example, for the servo, the controller calculates the corresponding PWM signal duty cycle based on the control instructions output by the controller. By adjusting the duty cycle, the servo's rotation angle is controlled, thereby achieving control of the aileron, elevator, and rudder. For the engine throttle, a corresponding PWM signal is similarly generated based on the control instructions to control the engine throttle opening, thereby adjusting the engine thrust. Sensors are installed on the actuator, such as a potentiometer-type angle sensor for the servo with an accuracy of up to 0.1° and a Hall effect position sensor for the throttle position, to monitor the actuator's operating status in real time. The data collected by the sensors is fed back to the controller module, which adjusts the control instructions based on the feedback data to ensure that the actuator can accurately execute the control instructions. Simultaneously, the actuator's operating status is monitored in real time and fault diagnosis is performed. When an actuator fault is detected, a fault-tolerant control strategy is used to adjust the system. For example, if a servo fails, the control instructions of other servos are adjusted to ensure that the drone can still maintain a certain flight attitude and stability.

[0128] Example 1

[0129] In a real-world flight test, a fixed-wing drone with a 2-meter wingspan was equipped with the aforementioned disturbance observer-based trajectory tracking control system. The drone's mission was to take off from an airport, follow a predetermined trajectory to a target area for data collection, and then return to the airport for landing.

[0130] During flight, the data acquisition module collects real-time data on the drone's attitude, position, velocity, and other parameters. The disturbance observer module accurately estimates atmospheric turbulence and engine thrust fluctuations. The trajectory planning module plans an optimal flight trajectory based on mission objectives and environmental information. The controller module employs a combined model predictive control and sliding mode control strategy to generate precise control instructions based on the disturbance information estimated by the disturbance observer and the reference trajectory. The actuator driver module converts these control instructions into actuator actions, enabling the drone to accurately track the predetermined trajectory.

[0131] In the presence of strong atmospheric turbulence interference, the position tracking error of the traditional trajectory tracking control system can reach more than 5 meters. However, using the system of the present invention, the position tracking error can be controlled within 1 meter and the attitude tracking error can be controlled within 1°, effectively improving the accuracy and stability of the UAV trajectory tracking.

[0132] The above description of the embodiments is intended to facilitate the understanding and use of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without having to go through creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art based on the principles of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A fixed-wing UAV trajectory tracking control system based on interference observer, characterized in that: include: Data acquisition module: Contains multiple sensors such as inertial measurement unit (IMU), global positioning system (GPS), barometric altimeter, and airspeed meter, used to collect attitude information, position information, speed information, and acceleration information of fixed-wing UAV in real time; The attitude information includes pitch angle, roll angle, and yaw angle; the position information includes longitude, latitude, and altitude; and the speed information includes airspeed and ground speed; Disturbance Observer Module: This module is based on a dynamic model of a fixed-wing UAV that fully considers the UAV's mass, moment of inertia, aerodynamic force, and aerodynamic torque. By analyzing and processing data collected by sensors, the Disturbance Observer can estimate in real time the external disturbances acting on the UAV, including atmospheric turbulence, engine thrust fluctuations, and other unknown environmental disturbances. Trajectory Planning Module: Receives mission target information from the ground control station or pre-set mission target information, and combines it with the UAV's current position and attitude information to plan a flight trajectory that meets mission requirements and flight constraints. The trajectory planning algorithm uses an optimization-based approach to minimize flight time, flight energy consumption, or maximize mission completion efficiency. It also considers constraints such as the UAV's dynamic limitations, flight safety area, and obstacle avoidance requirements. The planned trajectory includes a series of reference points and corresponding attitude and velocity commands, providing target input for trajectory tracking control. Controller module: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control. Actuator drive module: According to the control instructions output by the controller module, it drives the actuators of the fixed-wing UAV, including the servo and engine throttle. The actuator drive module has the ability of fast response and high-precision control, and can accurately convert the control instructions into the actions of the actuator, thereby achieving the adjustment of the UAV's attitude and flight status.

2. The fixed-wing UAV trajectory tracking control system based on interference observer according to claim 1 is characterized in that: The specific implementation of the interference observer module is: Constructing a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, defining the state variables as the UAV's attitude, velocity, and position, and the input variables as control inputs, including rudder angle and throttle opening. The output variables are measurable sensor data. Design disturbance observer: Based on the state-space model, use the sliding mode disturbance observer or extended state observer method to design the disturbance observer; Adaptive parameter adjustment: The disturbance observer uses an adaptive algorithm to adjust the observation parameters. The adaptive algorithm adjusts the observer's gain parameters in real time based on the disturbance estimation error and changes in the system state variables to improve the disturbance observer's adaptability to different interference environments and flight conditions.

3. The fixed-wing UAV trajectory tracking control system based on interference observer according to claim 1 is characterized in that: The specific implementation steps of the trajectory planning module are as follows: Mission objective analysis: Analyzes mission objective information from the ground control station or pre-set mission objective information to determine the UAV's starting position, target position, mission type, and other mission-related requirements; Environmental modeling: Use map information, obstacle information, and weather data to model the UAV's flight environment, build an environmental model including flight no-fly zones, obstacle distribution, and weather conditions (information to provide environmental constraints for trajectory planning; Initial trajectory generation: Using a search algorithm such as the Rapidly Exploring Random Tree (RRT) algorithm or the A* algorithm, an initial feasible trajectory is generated from the starting position to the target position, taking into account environmental constraints and the dynamic limitations of the UAV. During the initial trajectory generation process, the optimization goal is to minimize the flight distance or flight time, while ensuring that the trajectory avoids obstacles and no-fly zones. Trajectory Optimization: The initial trajectory is optimized using an optimization-based approach. The optimization objectives include minimizing flight energy consumption, maximizing mission completion efficiency, and smoothing the trajectory to reduce the UAV's attitude changes. During the optimization process, the trajectory is adjusted and optimized to obtain the final reference trajectory, taking into account the UAV's dynamic model, flight constraints, and interference information estimated by the interference observer.

4. The fixed-wing UAV trajectory tracking control system based on interference observer according to claim 1 is characterized in that: The specific design method of the controller module is as follows: Model predictive control part: Prediction model establishment: Based on the dynamic model of the fixed-wing UAV, a prediction model for model predictive control is established. The prediction model predicts the state of the UAV at multiple moments in the future, including attitude, speed, and position, based on the current system state and future control inputs. Objective function definition: Define the objective function of the model predictive control. The objective function usually includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the UAV and the reference trajectory. The control input change term is used to limit the rate of change of the control input to ensure system stability and control smoothness. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements. Rolling optimization solution: In each control cycle, model predictive control solves the optimization problem based on the prediction model and objective function to obtain the optimal control input sequence for multiple future moments; Sliding surface design: Based on the trajectory tracking error of the fixed-wing UAV, a sliding surface is designed. When moving on the sliding surface, it can achieve stable tracking of the reference trajectory and has a certain degree of robustness. Sliding mode control law design: Based on the motion of the sliding surface, a sliding mode control law is designed. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface; the switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface. Chattering suppression: Improve the sliding mode control law using methods such as the boundary layer method or integral sliding mode control. The boundary layer method sets a boundary layer near the sliding surface and makes the sliding mode control law continuous within the boundary layer, reducing the frequency of switching control and thus reducing chattering; Integral sliding mode control introduces an integral term to enable the system to be on the sliding mode surface at the initial moment, thus avoiding the chattering problem of traditional sliding mode control in the initial stage. Composite control strategy fusion: Model predictive control and sliding mode control are combined to form a composite control strategy. In the composite control strategy, model predictive control is responsible for generating the optimal control input sequence based on the system's predicted state and reference trajectory to achieve accurate trajectory tracking; sliding mode control provides additional control for system uncertainties and interferences, ensuring that the system can still operate stably in the presence of interferences.

5. A fixed-wing UAV trajectory tracking control method based on interference observer, characterized in that: The following steps are involved: Data collection steps: Use sensors such as an inertial measurement unit, global positioning system, barometric altimeter, and airspeed meter to collect the fixed-wing UAV's attitude, position, velocity, and acceleration data in real time. Preprocess the collected data, including filtering and denoising operations, to improve the accuracy and reliability of the data. Interference estimation steps: Based on the dynamic model of the fixed-wing UAV, a disturbance observer is used to estimate the external disturbance acting on the UAV in real time. By analyzing and processing sensor data, the disturbance observer separates the external disturbance from the system state variables to obtain the interference estimate. The disturbance observer uses an adaptive algorithm to automatically adjust the observation parameters according to the changes in the UAV's flight state to improve the accuracy of the interference estimation. Trajectory planning: Receive mission objective information from the ground control station or pre-set mission objective information, combine it with the UAV's current position and attitude information, and plan a flight trajectory that meets the mission requirements and flight constraints. The trajectory planning process includes mission objective analysis, environment modeling, initial trajectory generation, and trajectory optimization. The rapid exploration random tree (RRT) algorithm or A* algorithm search algorithm is used to generate the initial feasible trajectory. The initial trajectory is then optimized through nonlinear programming (NLP) or model predictive control (MPC) optimization methods to obtain the final reference trajectory. Controller design and calculation steps: Based on the disturbance information estimated by the disturbance observer and the reference trajectory generated by the trajectory planning module, a controller is designed to achieve precise trajectory tracking control of the fixed-wing UAV. The controller adopts a composite control strategy based on model predictive control (MPC) and sliding mode control. The model predictive control part uses the UAV's dynamic model to predict the system state at multiple moments in the future and calculates the optimal control input sequence by optimizing the objective function. The sliding mode control part designs the sliding surface and sliding mode control law based on the trajectory tracking error to improve the robustness of the system. In the controller design process, the UAV's dynamic constraints, control input constraints, and the disturbance information estimated by the disturbance observer are considered. By reasonably adjusting the control parameters, the controller can achieve optimal control performance under different flight conditions. Actuator driving steps: According to the control instructions output by the controller module, the actuators of the fixed-wing UAV, including the servo and engine throttle, are driven. The actuator driving module converts the control instructions into the actions of the actuators, thereby adjusting the UAV's attitude and flight status. During the actuator driving process, the actions of the actuators are monitored and feedback is provided to ensure that the actuators can accurately execute the control instructions.

6. The fixed-wing UAV trajectory tracking control method based on interference observer according to claim 5 is characterized in that: In the interference estimation step, a sliding mode interference observer or an extended state observer is used to perform interference estimation. The specific steps are as follows: Construct a state-space model: Convert the fixed-wing UAV's dynamic equations into a state-space form, define state variables, input variables, and output variables, and treat external disturbances as uncertain inputs to the system in the state-space model. Design disturbance observer: Taking the sliding mode disturbance observer as an example, design the sliding surface and sliding mode control law based on the state space model; In the design of the sliding surface, the state variables and disturbance estimation errors of the system are considered, and the sliding mode control law is adjusted according to the movement of the sliding surface. By adjusting the parameters of the sliding mode control law, the output of the observer can quickly track the disturbance signal in the system. Adaptive parameter adjustment: The adaptive algorithm adjusts the observer gain parameters in real time according to the disturbance estimation error and the changes of system state variables.

7. The fixed-wing UAV trajectory tracking control method based on interference observer according to claim 5 is characterized in that: In the trajectory planning step, an optimization-based method is used to optimize the trajectory. The specific steps are as follows: Define optimization objectives: Based on mission requirements and flight performance requirements, define the objective function for trajectory optimization. The objective function includes minimizing flight time, flight energy consumption, and maximizing mission completion efficiency. Establish an optimization model: including the objective function, constraints and decision variables. The decision variables include the position, posture and speed of the trajectory points. Solve the optimization problem: Utilize a numerical solution algorithm to adjust and optimize the trajectory, taking into account the interference information estimated by the interference observer, to obtain the optimal reference trajectory that meets the mission requirements and flight constraints.

8. The fixed-wing UAV trajectory tracking control method based on interference observer according to claim 5 is characterized in that: In the controller design and calculation steps, the specific implementation steps of the model predictive control part are as follows: Predictive model establishment: Based on the dynamic model of the fixed-wing UAV, a predictive model for model predictive control is established. The predictive model can predict the state of the UAV at multiple moments in the future based on the current system state and future control inputs; Objective function definition: Define the objective function of model predictive control. The objective function includes a tracking error term and a control input change term. The tracking error term is used to measure the deviation between the current state of the drone and the reference trajectory. The control input change term is used to limit the rate of change of the control input. The weight coefficient of the objective function is adjusted according to the actual flight requirements and system performance requirements. Rolling optimization solution: In each control cycle, based on the prediction model and objective function, the optimal control input sequence for multiple future moments is obtained by solving the optimization problem. The optimization problem is solved numerically using the quadratic programming (QP) algorithm. During the solution process, the UAV's dynamic constraints, control input constraints, and disturbance information estimated by the disturbance observer are considered to ensure the feasibility and effectiveness of the optimization results. Only the first control variable in the optimized control input sequence is applied to the UAV system. In the next control cycle, prediction and optimization are performed again to achieve rolling optimization control.

9. The fixed-wing UAV trajectory tracking control method based on interference observer according to claim 5 is characterized in that: In the controller design and calculation steps, the specific implementation steps of the sliding mode control part are as follows: Sliding surface design: Design a sliding surface based on the trajectory tracking error of the fixed-wing UAV. A linear sliding surface design method is used, using the linear combination of position error and attitude error as the sliding surface variable. The design of the sliding surface should enable the system to stably track the reference trajectory when moving on the sliding surface and have a certain degree of robustness; Sliding mode control law design: Design a sliding mode control law based on the motion of the sliding surface. The sliding mode control law consists of an equivalent control part and a switching control part. The equivalent control part is used to offset the deterministic part of the system, so that the system maintains stable motion on the sliding surface. The switching control part is used to overcome the uncertainty and interference of the system, so that the system can quickly converge to the sliding surface. Based on the interference information estimated by the interference observer, the switching control gain is adjusted to ensure that the system maintains good control performance in the presence of interference. Chattering suppression: To suppress chattering in sliding mode control, methods such as the boundary layer method or integral sliding mode control are used to improve the sliding mode control law. The boundary layer method sets a boundary layer near the sliding surface and makes the sliding mode control law continuous within the boundary layer, thus reducing the frequency of switching control. Integral sliding mode control introduces the integral term, which enables the system to be on the sliding surface at the initial moment, avoiding the chattering problem of traditional sliding mode control in the initial stage.

10. The fixed-wing UAV trajectory tracking control method based on interference observer according to claim 5, characterized in that: In the actuator driving step, the action of the actuator is monitored and feedback is provided, which is specifically implemented as follows: Sensor feedback: Sensors, including servo angle sensors and throttle position sensors, are installed on the actuators of fixed-wing drones to monitor the actuator's motion status in real time. The data collected by the sensors is fed back to the controller module, which adjusts the control instructions based on the feedback data to ensure that the actuator can accurately execute the control instructions. Fault diagnosis and fault-tolerant control: Real-time monitoring and fault diagnosis of the working status of the actuator are carried out. When a fault in the actuator is detected, the fault-tolerant control strategy is used to adjust the system.

Images