A method, system, device and medium for controlling the flight of an unmanned aerial vehicle (UAV).

Abstract

This application discloses a method, system, device, and medium for unmanned aerial vehicle (UAV) flight control. The method includes: performing attitude control processing on the measured state and desired state using a model predictive control-based attitude controller to obtain a control output; performing disturbance observation and compensation processing on the measured state and control output using a linear extended state observer to obtain a disturbance compensation value; performing a flip switching judgment processing on the control output and disturbance compensation value according to a preset flip threshold to obtain a flip switching judgment result; when the flip switching judgment result is a flip switching, updating the desired state to generate a flip control command for flight control of the UAV; when the flip switching judgment result is normal control, superimposing the control output according to the disturbance compensation value to determine the execution command for flight control of the UAV. The embodiments of this application can improve flight stability and safety and can be widely applied in the field of UAV control technology.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a UAV flight control method, system, device and medium. Background Technology

[0002] In the fields of modern automation and remote control, quadcopter aircraft (drones) are widely used in various tasks such as surveillance, agriculture, and rescue operations due to their superior maneuverability and flexibility. However, these applications often require quadcopter aircraft to operate stably in complex and dynamically changing environments, especially when faced with sudden and severe disturbances (such as strong winds, sudden changes in airflow, and force interactions). Such disturbances can cause the aircraft to lose control or even crash, resulting in safety risks and economic losses. In related technologies, quadcopter control systems employ a PID-based method for drone flight control. However, under extreme or unpredictable environmental conditions, the PID-based method struggles to provide effective control, leading to drone loss of control and affecting the stability and safety of drone flight. In summary, the technical problems existing in related technologies need to be improved. Summary of the Invention

[0003] The main objective of this application is to provide a method, system, device, and medium for controlling unmanned aerial vehicle (UAV) flight, which can improve the stability and safety of UAV flight.

[0004] To achieve the above objectives, one aspect of this application proposes a flight control method for unmanned aerial vehicles (UAVs), the method comprising:

[0005] Obtain the actual and expected states of the drone;

[0006] The measured state and the desired state are processed by an attitude controller based on model predictive control to obtain a control output.

[0007] The measured state and the control output are subjected to disturbance observation and compensation processing by a linear extended state observer to obtain the disturbance compensation value.

[0008] The control output and the disturbance compensation value are subjected to a flip-over switching judgment process based on a preset flip-over threshold to obtain a flip-over switching judgment result.

[0009] When the flip switching judgment result is flip switching, the desired state is updated and calculated to generate a flip control command to control the flight of the UAV.

[0010] When the flip-switch judgment result is normal control, the control output is superimposed based on the disturbance compensation value to determine the execution command to control the UAV in flight.

[0011] In some embodiments, the attitude control processing of the measured state and the desired state by a model predictive control-based attitude controller to obtain a control output includes:

[0012] The measured state and the desired state are subjected to trajectory generation processing to obtain a spherical reference trajectory;

[0013] Based on the spherical reference trajectory, model predictive control processing is performed to obtain the control output.

[0014] In some embodiments, the trajectory generation process of the measured state and the desired state to obtain a spherical reference trajectory includes:

[0015] Get default parameters;

[0016] The default parameters are iteratively generated based on the fastest control synthesis function to obtain a set of interpolation coefficients.

[0017] Based on the set of interpolation coefficients, spherical trajectory geometric interpolation is performed on the measured state and the desired state to obtain a spherical reference trajectory.

[0018] In some embodiments, the step of performing model predictive control processing based on the spherical reference trajectory to obtain a control output includes:

[0019] For each point of the spherical reference trajectory, a lookup table is generated.

[0020] The measured state is obtained, and the measured state is subjected to coordinate transformation to obtain the attitude vector;

[0021] A sliding window search is performed on the lookup table based on the attitude vector to obtain a reference trajectory segment;

[0022] The reference trajectory segment is subjected to attitude control optimization processing using a predictive model to obtain the control output.

[0023] In some embodiments, the step of performing disturbance observation and compensation processing on the measured state and the control output through a linear extended state observer to obtain a disturbance compensation value includes:

[0024] The system estimate is obtained by estimating the angular velocity and the extended state through the control output of the linear extended state observer;

[0025] The system estimate is processed by error calculation based on the measured state to obtain the disturbance compensation value.

[0026] In some embodiments, when the flip switching determination result is a flip switching, updating the desired state and generating a flip control command to control the UAV's flight includes:

[0027] The desired state is updated to obtain the flipped state;

[0028] The measured state and the flip state are processed by an attitude controller based on model predictive control to obtain a flip control output.

[0029] The measured state and the flip control output are perturbed and compensated by a linear extended state observer to obtain the flip perturbation compensation value.

[0030] The flip control output is superimposed based on the flip disturbance compensation value to determine the flip control command for flight control of the UAV.

[0031] In some embodiments, before obtaining the measured state and desired state of the drone, the process further includes integrating the drone, specifically including:

[0032] The UAV integrates a three-dimensional electronic speed controller and a forward and reverse propeller;

[0033] The brushless motor is driven by the three-dimensional electronic speed controller to rotate the drone in both forward and reverse directions.

[0034] The reversible propellers provide stable and consistent thrust as the UAV rotates in both directions.

[0035] To achieve the above objectives, another aspect of this application proposes a drone flight control system, the system comprising:

[0036] The first module is used to obtain the actual and expected states of the UAV;

[0037] The second module is used to perform attitude control processing on the measured state and the desired state through a model predictive control-based attitude controller to obtain a control output;

[0038] The third module is used to perform disturbance observation and compensation processing on the measured state and the control output through a linear extended state observer to obtain the disturbance compensation value.

[0039] The fourth module is used to perform a flip-over switching judgment process on the control output and the disturbance compensation value according to a preset flip-over threshold, and obtain a flip-over switching judgment result.

[0040] The fifth module is used to update the desired state when the flip switching judgment result is flip switching, and generate a flip control command to control the flight of the UAV.

[0041] The sixth module is used to perform superposition processing on the control output based on the disturbance compensation value when the flip-switch judgment result is normal control, and determine the execution command to perform flight control on the UAV.

[0042] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0043] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0044] The embodiments of this application include at least the following beneficial effects: This application provides a UAV flight control method, system, device, and medium. This solution uses a model predictive control-based attitude controller to perform attitude control processing on the measured state and the desired state to obtain a control output. It can control the UAV by combining model prediction and omnidirectional attitude control, achieving effective control in all attitude directions and enhancing the robustness of control, further reducing the risk of UAV crashes. In addition, this solution also uses a linear extended state observer to perform disturbance observation and compensation processing on the measured state and control output to obtain disturbance compensation values, enabling real-time estimation and compensation of disturbances inside and outside the system. This improves the stability of UAV flight control. Furthermore, this solution can perform flip switching judgment processing on the control output and disturbance compensation value according to the preset flip threshold to obtain the flip switching judgment result. When the flip switching judgment result is flip switching, the desired state is updated and calculated to generate flip control commands to control the UAV flight. When the flip switching judgment result is normal control, the control output is superimposed according to the disturbance compensation value to determine the execution command to control the UAV flight. Thus, the flip switching strategy effectively judges whether external disturbances exceed the UAV's tolerance range, enhancing the stability and safety of UAV flight control. Attached Figure Description

[0045] Figure 1 This is a flowchart of a drone flight control method provided in an embodiment of this application;

[0046] Figure 2 This is a schematic diagram of a spherical reference trajectory provided in an embodiment of this application;

[0047] Figure 3 This is a schematic diagram of a flip-to-switch provided in an embodiment of this application;

[0048] Figure 4 This is a schematic diagram of the structure of an unmanned aerial vehicle (UAV) flight control system provided in an embodiment of this application;

[0049] Figure 5 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of systems and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0051] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0052] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0054] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0055] Two-way thrust quadcopter: This is an advanced quadcopter drone design that generates thrust both upwards and downwards. This design allows the drone to perform more complex flight maneuvers, such as high-speed flips and inverted flight. The two-way thrust system is achieved by improving the propulsion system of traditional quadcopters, which typically includes counter-rotating propellers and an advanced electronic speed controller (ESC).

[0056] Sudden, severe disturbances refer to sudden and intense environmental changes or external forces occurring within a very short period of time. These disturbances can severely impact the stability and performance of an aircraft. Examples include strong winds, sudden changes in airflow, and physical collisions. Such disturbances pose a significant challenge to the aircraft's control system, requiring specialized designs to ensure safe and effective responses.

[0057] Active Disturbance Rejection Control (ADRC) is a control strategy whose main purpose is to actively identify and resist external and internal disturbances to a system. ADRC improves system robustness by estimating the total disturbances to the system and compensating for these disturbances in real time within the control law. This method is applicable to various dynamic systems, especially under unstable environmental conditions or with unknown disturbances. The core algorithm of ADRC is the Extended State Observer (ESO).

[0058] Extended State Observer (ESO): ESOs are used to estimate and compensate for internal system dynamics and external disturbances in real time. By requiring only system input and output information, ESOs can provide accurate estimates of the system state, including uncertain model dynamics and external disturbances. They improve control performance by compensating for these estimated disturbances in the feedback. The application of ESOs enhances the robustness and adaptability of control systems, enabling them to exhibit better performance in the face of uncertainties and external disturbances.

[0059] Geometric control is a control method based on differential geometry and Lie group theory, used to handle nonlinear control systems, such as quadcopters. Geometric control considers the system's geometry and characteristics, effectively addressing attitude and position control problems. It avoids the singularity problems of conventional methods, providing more accurate and stable control, especially in complex space operations.

[0060] Model Predictive Control (MPC) is an advanced control strategy that predicts the system's behavior over a future period based on its dynamic model and optimizes the current control inputs. MPC solves a rolling time-domain optimization problem, updating the control strategy in real time to adapt to environmental changes and system constraints. This approach is particularly effective when considering multiple input and output variables and system constraints, and it holds great promise for applications in robotics, autonomous driving, and other fields.

[0061] In related technologies, quadcopter aircraft are often required to operate stably in complex and dynamically changing environments, especially when faced with sudden and severe disturbances (such as strong winds, sudden changes in airflow, and force interactions). Such disturbances can cause the aircraft to lose control or even crash, resulting in safety risks and economic losses. Traditional quadcopter control systems, such as PID-based methods, can handle these challenges to some extent, but their effectiveness is often limited under extreme or unpredictable environmental conditions. Therefore, developing a robust control system that can effectively cope with such severe disturbances is particularly important. Currently, most traditional quadcopter UAV attitude control is based on small-angle assumptions, which are effective for routine flight missions. However, when it comes to large-angle attitude adjustments, especially in cases of drastic environmental changes and complex flight missions, the limitations of this traditional method become apparent.

[0062] In view of this, this application provides a UAV flight control method, system, device, and medium. This scheme uses a model predictive control-based attitude controller to process the measured and desired states to obtain control outputs. It combines model prediction and omnidirectional attitude control to control the UAV, achieving effective control in all attitude directions and enhancing control robustness, further reducing the risk of UAV crashes. In highly dynamic flight environments, achieving precise large-angle attitude control while maintaining stability has become one of the key challenges in the development of quadcopter UAV technology. This application provides a possibility for this requirement through a control strategy based on geometric methods. This method, based on nonlinear control theory, considers the overall dynamics and geometric characteristics of the quadcopter, providing a method for precise control in the entire attitude space. By applying geometric control, the UAV can achieve highly stable and flexible operation in three-dimensional space, especially when facing severe disturbances such as strong winds or sudden airflow changes. The introduction of geometric control methods opens up new possibilities for precise attitude control and stable operation of quadcopter UAVs under extreme conditions, greatly improving the application range and performance of the aircraft.

[0063] Furthermore, this scheme also uses a linear extended state observer (ESO) to perform disturbance observation and compensation processing on the measured state and control output, obtaining disturbance compensation values. This enables real-time estimation and compensation of disturbances inside and outside the system, improving the stability of UAV flight control. Since the main advantages of ADRC lie in its flexibility and robustness, it becomes an ideal choice for solving control problems in highly uncertain and dynamically changing environments. The basic principle of ADRC is to actively estimate and compensate for internal and external disturbances. It utilizes an extended state observer (ESO) to estimate the unknown dynamics of the system and external disturbances, and uses these estimates in the control strategy to counteract the effects of the disturbances. A key advantage of this method is that it does not rely on precise knowledge of the system model, making it particularly useful in complex systems where the model is uncertain or difficult to model accurately.

[0064] Furthermore, this scheme can perform flip switching judgment processing on the control output and disturbance compensation value according to a preset flip threshold to obtain the flip switching judgment result. When the flip switching judgment result is flip switching, the desired state is updated and calculated to generate flip control commands for flight control of the UAV. When the flip switching judgment result is normal control, the control output is superimposed according to the disturbance compensation value to determine the execution command for flight control of the UAV. Thus, the flip switching strategy effectively judges whether external disturbances exceed the UAV's tolerance range, enhancing the stability and safety of UAV flight control. This application further improves the performance of quadcopter by introducing bidirectional thrust technology. The bidirectional thrust quadcopter has made significant improvements to the design of traditional quadcopter aircraft. This improved system can output upward or downward thrust, significantly enhancing the maneuverability and agility of the UAV. This means that compared with traditional unidirectional thrust quadcopter aircraft, the bidirectional thrust model can track more complex and diverse flight trajectories, such as rapid aerial flips and inverted flight. These capabilities make the UAV more adaptable when performing complex tasks or operating in variable environments. Meanwhile, two-way thrust aircraft can generate greater speed differences, thus providing greater attitude control torque. In the event of sudden and severe disturbances, such as strong winds or sudden changes in airflow, this characteristic provides powerful dynamic support for the UAV, helping it to quickly and effectively adjust its attitude to maintain stability. This enhanced control capability is crucial for improving the stability and reliability of aircraft in complex or extreme environments.

[0065] In summary, this application combines bidirectional thrust technology, ADRC, and geometry-based control methods for flight control of a novel quadcopter, which is of great significance for improving the stability and safety of the aircraft in extreme environments. It can not only enhance the aircraft's application capabilities in complex environments, but also reduce potential safety risks and economic losses.

[0066] Figure 1 This is an optional flowchart of a drone flight control method provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0067] Step S101: Obtain the actual and expected states of the UAV;

[0068] Step S102: The measured state and the desired state are processed by an attitude controller based on model predictive control to obtain a control output.

[0069] Step S103: The measured state and the control output are subjected to disturbance observation and compensation processing by a linear extended state observer to obtain the disturbance compensation value;

[0070] Step S104: Perform a flip-over switching judgment process on the control output and the disturbance compensation value according to the preset flip-over threshold to obtain the flip-over switching judgment result;

[0071] Step S105: When the flip switching judgment result is flip switching, the desired state is updated and calculated to generate a flip control command to control the flight of the UAV.

[0072] Step S106: When the flip-switch judgment result is normal control, the control output is superimposed according to the disturbance compensation value to determine the execution command to control the UAV for flight.

[0073] Steps S101 to S106 of this embodiment first acquire the measured state and desired state of the UAV. Then, an attitude controller based on model predictive control processes the measured state and desired state to obtain a control output. This predictive control, based on predictions over a future period, provides the current control quantity, better addressing sudden situations such as strong instantaneous disturbances. Combined with a coordinate-free attitude control method that avoids singularities, a real-time target attitude tracking trajectory conforming to rigid body rotational dynamics is proposed. By tracking the reference trajectory, effective control in all attitude directions can be efficiently achieved, particularly suitable for situations with large attitude errors caused by strong disturbances. This embodiment uses a linearly extended state observer to perform disturbance observation and compensation processing on the measured state and control output to obtain disturbance compensation values. By integrating Active Disturbance Rejection Control (ADRC) and geometric control, the accuracy of disturbance observation is significantly improved. ADRC's advantage lies in its ability to estimate and compensate for disturbances inside and outside the system in real time, while geometric control provides a coordinate-free attitude representation and control method that avoids singularity problems. Therefore, the method of this application is not only applicable to minor disturbances in normal flight conditions, but also effectively addresses sudden and severe disturbances encountered in high-speed flight or complex weather conditions. Finally, the control output and the disturbance compensation value are subjected to flip switching judgment processing according to a preset flip threshold to obtain the flip switching judgment result. The embodiment of this application provides a larger thrust difference through bidirectional thrust to achieve flip switching, thereby providing greater torque and enhancing the UAV's resistance to severe disturbances. Secondly, this thrust capability enables the UAV to quickly switch between two stable states: normal hovering and inverted hovering. Inverted flight itself is also a stable state, which increases the flexibility of the control strategy. When subjected to impact or other disturbances, the UAV can quickly switch to the stable state closest to the current state. This strategy significantly reduces the time and energy loss required to adjust to a stable state, effectively improving the UAV's responsiveness and safety in complex environments.

[0074] In step S101 of some embodiments, the measured state of the UAV can be obtained through sensors. This measured state includes the UAV's real-time position, attitude, heading angle, speed, etc. Alternatively, the desired state of the UAV can be obtained through other means by the controller. This desired state is the position that the UAV is expected to reach, and the attitude, speed, angular velocity, etc., when reaching that position, etc., and is not limited to these.

[0075] In step S102 of some embodiments, the attitude control processing of the measured state and the desired state by the attitude controller based on model predictive control to obtain the control output includes:

[0076] The measured state and the desired state are subjected to trajectory generation processing to obtain a spherical reference trajectory;

[0077] Based on the spherical reference trajectory, model predictive control processing is performed to obtain the control output.

[0078] In this embodiment, a controller combining model prediction and geometric control, i.e., a model prediction control-based attitude controller, outputs the desired motor speed (i.e., control output) by inputting the measured state and the desired state. The underlying driver first converts the desired motor speed into a weak electrical control signal, and the electronic speed controller then drives the motor to rotate according to the control signal. This embodiment utilizes geometric control, specifically for coordinate-independent attitude tracking problems, which refers to accurately controlling a UAV to follow a predetermined attitude without relying on a specific coordinate system. This method can be compared to playing a game from a first-person perspective: just as a player controls actions directly from the character's first-person viewpoint, the UAV adjusts its attitude based on its own "viewpoint" and position. By using geometric representations such as quaternions or direction cosine matrices for the target and current attitudes, a global and continuous attitude description is provided for the UAV, effectively avoiding problems such as gimbal lock in Euler angle representation. This embodiment considers the unique rotational dynamics characteristics of quadcopter UAVs, including limitations on angular velocity and angular acceleration, thereby providing a smooth and continuous rotational path for the UAV and generating a spherical reference trajectory. The control input is determined by solving the MPC problem based on the spherical reference trajectory, controlling the UAV to reach the desired direction, and finally outputting the control output. This embodiment combines model predictive control and geometric attitude control to generate a real-time target attitude tracking trajectory that conforms to the rigid body rotational dynamics characteristics. It effectively solves the cumulative error problem that may arise in continuous rotation using traditional Euler angle representation, and achieves effective control in all attitude directions, especially suitable for large attitude error situations under strong disturbances. Furthermore, by predicting the UAV's state over a future period, the robustness of the control is enhanced, further reducing the risk of UAV crashes.

[0079] In some embodiments, the trajectory generation process of the measured state and the desired state to obtain a spherical reference trajectory includes:

[0080] Get default parameters;

[0081] The default parameters are iteratively generated based on the fastest control synthesis function to obtain a set of interpolation coefficients.

[0082] Based on the set of interpolation coefficients, spherical trajectory geometric interpolation is performed on the measured state and the desired state to obtain a spherical reference trajectory.

[0083] In this embodiment, default parameters are obtained, including dt, r, and h0. r directly determines the maximum angular acceleration of the reference trajectory, dt is the sampling period, and h0 is the filtering parameter, typically 3 to 10 times dt. An appropriately sized filtering parameter can make the trajectory smoother. In this embodiment, the spherical reference trajectory is initialized based on the default parameters during the first program run. The trajectory is also regenerated when the default parameters are updated in the parameter tuning interface. (Refer to...) Figure 2 Then, the default parameters are iteratively calculated using the fastest control synthesis function fhan() to generate interpolation coefficients t. Then, spherical reference trajectory points are generated according to the geometric principle of spherical trajectory interpolation. The spherical reference trajectory is obtained by connecting the spherical reference trajectory points.

[0084] In some embodiments, the step of performing model predictive control processing based on the spherical reference trajectory to obtain a control output includes:

[0085] For each point of the spherical reference trajectory, a lookup table is generated.

[0086] The measured state is obtained, and the measured state is subjected to coordinate transformation to obtain the attitude vector;

[0087] A sliding window search is performed on the lookup table based on the attitude vector to obtain a reference trajectory segment;

[0088] The reference trajectory segment is subjected to attitude control optimization processing using a predictive model to obtain the control output.

[0089] In this embodiment, each point of the spherical reference trajectory is added to an empty table to generate a lookup table. Then, the UAV attitude R in the world coordinate system is converted to the desired attitude R. d The attitude vector is obtained by representing the reference frame. Then, the prediction compensation of the prediction model is obtained. Based on the attitude vector and the prediction step size N, the reference trajectory segment is retrieved from the lookup table. Specifically, the index k is retrieved from the lookup table based on the attitude vector. Starting from the determined index k, a spherical trajectory reference segment of length N is read forward. This spherical trajectory reference segment is used as the desired state of the MPC. The attitude control optimization process is performed on the reference trajectory segment by the prediction model. By optimizing the control input, the attitude error and angular velocity error are minimized. Considering the constraints of system dynamics and control input, the optimal attitude control problem is obtained. The expression of this optimal attitude control problem is as follows:

[0090]

[0091] In the formula, k represents the current prediction time of the UAV, k+1 represents the next prediction time of the UAV, the first row represents the attitude vector update, and does not directly include the angular acceleration term. The second and third rows describe the update of the angular velocity Ω in the pitch and roll channels (the first two dimensions), respectively, including the control input u. rot The influence of the disturbance term ζ. ζ in the fourth line considers the predefined desired angular velocity and the external disturbance. The impact on angular velocity. This application's embodiments define an optimization objective function aimed at minimizing the attitude error e. b3 and angular velocity error e Ω The weighted sum of squares, and the control input u rot The weighted sum of squares. This is achieved by setting the weight matrix Q. b3 Q Ω and Q ur This is used to adjust the relative importance of different parts. Simultaneously, the error of the final state is also adjusted through the weight matrix P. b3 and P Ω Weighting is applied to ensure that the final attitude and angular velocity are close to the desired values. This application's embodiments achieve precise control of the aircraft's attitude by considering accurate modeling of system dynamics and the rational construction of the optimization problem. By optimizing the control input, attitude and angular velocity errors can be effectively reduced, control performance improved, and the physical constraints of the control input satisfied.

[0092] In step S103 of some embodiments, the perturbation observation and compensation processing of the measured state and the control output through a linear extended state observer to obtain a perturbation compensation value includes:

[0093] The system estimate is obtained by estimating the angular velocity and the extended state through the control output of the linear extended state observer;

[0094] The system estimate is processed by error calculation based on the measured state to obtain the disturbance compensation value.

[0095] In this embodiment, angular velocity and extended state estimation are performed using the control output of the linear extended state observer to obtain the system estimate. The principle formula for disturbance observation and compensation is as follows:

[0096]

[0097] In the formula, the first row shows the estimated angular velocity. The update mechanism. Here, This represents the estimated angular velocity at the next time step, while This is the estimated value at the current time step. The update process consists of two parts: one part is based on the control input u. rotand moment of inertia The direct impact; another part is the estimation error. After a proportional gain of 2l w The adjusted value is Ω(k), where Ω(k) is the actual angular velocity. In this way, the observer can dynamically adjust the estimate to reduce the difference between the actual and estimated values. The second line of the equation describes the extended state, i.e., the additional torque τ caused by the unknown disturbance. ξ The estimation and update process. Here, τ ξ (k+1) represents the estimate for the next time step, while τ ξ (k) is the estimated value at the current time step. The update mechanism takes into account the estimation error of the angular velocity. and a proportional gain This approach reflects the direct impact of such errors on the extended state estimation. This design allows the ESO to track and compensate for the effects caused by unknown perturbations.

[0098] Therefore, by employing ESO, the embodiments of this application can not only provide real-time estimation of the current state of the system (such as angular velocity), but also estimate and compensate for the parts of the system dynamics that are not directly measured (such as additional torque caused by external disturbances), thereby enhancing the robustness of the control system to unknown disturbances. This characteristic makes the design particularly valuable when dealing with complex dynamic systems, especially in maintaining high performance and stability in the face of internal and external disturbances.

[0099] In step S104 of some embodiments, the control output and the disturbance compensation value are subjected to a flip-over switching judgment process according to a preset flip-over threshold to obtain a flip-over switching judgment result.

[0100] In this embodiment of the application, the flipping threshold includes the maximum angular acceleration threshold Ω. max Among the motor speed thresholds, the maximum angular acceleration threshold is an empirical value, set to 60 rad / s in the experiment. 2 Due to the rotational inertia I of the drone v It is a constant, therefore it can be equivalently converted to the maximum torque threshold of ESO. When the maximum torque threshold observed by ESO exceeds IvΩ... max When the motor speed threshold is reached, it is considered one of the triggering conditions of the actual system. The motor speed threshold is a hardware-related threshold, which is the speed range of different motors when executing the same normalized throttle command (a decimal between 0 and 1). In this embodiment, it is set to 90% of the maximum speed. That is, when the motor speed reaches 90% of the maximum speed, it is considered that the power of the UAV is no longer sufficient to resist the disturbance, and then the flip strategy is triggered, resulting in a flip switching judgment result. The flip switching judgment result includes flip switching and normal control.

[0101] In step S105 of some embodiments, when the flip switching determination result is flip switching, updating the desired state and generating a flip control command to control the UAV's flight includes:

[0102] The desired state is updated to obtain the flipped state;

[0103] The measured state and the flip state are processed by an attitude controller based on model predictive control to obtain a flip control output.

[0104] The measured state and the flip control output are perturbed and compensated by a linear extended state observer to obtain the flip perturbation compensation value.

[0105] The flip control output is superimposed based on the flip disturbance compensation value to determine the flip control command for flight control of the UAV.

[0106] In the embodiments of this application, reference is made to Figure 3 When the flip switching judgment result is flip switching, a flip attitude of 180° is generated. Then, based on this new flip attitude, a reference trajectory segment of length N is re-found. MPC recalculates the control variable u based on this new reference and the current attitude value. rot This yields the flip control output. Simultaneously, the ESO re-estimates the disturbance τ based on the new control command and the measured conditions. ξ The roll disturbance compensation value is obtained. Finally, the re-estimated roll disturbance compensation value is added to the roll control output to obtain the roll control command executed by the UAV, thereby controlling the flight of the UAV.

[0107] In some embodiments, before obtaining the measured state and desired state of the drone, the process further includes integrating the drone, specifically including:

[0108] The UAV integrates a three-dimensional electronic speed controller and a forward and reverse propeller;

[0109] The brushless motor is driven by the three-dimensional electronic speed controller to rotate the drone in both forward and reverse directions.

[0110] The reversible propellers provide stable and consistent thrust as the UAV rotates in both directions.

[0111] In this embodiment, a 3D Electronic Speed ​​Controller (3D ESC) and reversible propellers are integrated into the UAV. The 3D ESC drives a brushless motor to rotate in both directions (ordinary ESCs only support unidirectional motor rotation), thereby generating the required thrust in both directions through the propellers. This capability is crucial for improving the aircraft's maneuverability and adaptability, especially when facing severe interference or performing complex flight maneuvers. Furthermore, the reversible propellers ensure consistent aerodynamic efficiency during both forward and reverse rotation of the UAV. This design guarantees stable and consistent thrust regardless of the propeller's rotation direction, which is essential for the aircraft's stability and performance under various flight conditions.

[0112] The solutions of this application embodiment will be described in detail and explained below with reference to specific application examples:

[0113] This application embodiment can be applied to UAV flight control scenarios. First, it integrates a three-dimensional electronic speed controller and reversible propellers to enable UAV roll control. Then, it acquires the UAV's measured state through sensors and the desired state through a controller. Finally, it inputs the measured and desired states into a model predictive control-based attitude controller for geometric attitude control. This is achieved by generating a coordinate-invariant global reference trajectory covering the attitude vector b3 from the initial attitude [0, ∈, ∈-1]. T Reach the target pose [0, 0, 1] TThe addition of ∈ facilitates the correct initiation of spherical interpolation. ∈ is a positive floating-point number close to 0, used to prevent numerical computational anomalies (division by zero errors) and mitigate noise effects; typically, 1e-4 or smaller is acceptable, depending on the precision of the computational platform. Furthermore, the use of the fhan function ensures smooth interpolation between vectors that conforms to angular acceleration constraints, setting a predefined dynamic for attitude tracking. The spherical trajectory covers the range from -π / 2 to π / 2, ensuring the reference trajectory remains available throughout the flipping operation, providing reliable guidance for attitude tracking. Moreover, since the reference trajectory segments are obtained through a lookup table (LKT), which is only regenerated when adjusting tracking dynamics, this is a computationally efficient method for obtaining reference trajectories. This is also a highly parameterized method, allowing new LKT generation within a single control cycle to ensure attitude control conforms to the main dynamic characteristics of UAV rotation, while avoiding the complex nonlinear calculations inherent in rotation problems. This ensures that the quadcopter UAV determines the control variables by solving the MPC problem, controlling the UAV to reach the desired direction, ultimately obtaining the control output. The control output is the desired motor speed. The underlying driver first converts the desired motor speed into a low-voltage control signal, and the electronic speed controller then drives the motor to rotate according to the control signal. Then, the Extended State Observer (ESO) estimates the total disturbance currently experienced by the UAV based on the measured state, desired state, and control output. The total disturbance includes internal and external disturbances; internal disturbances are modeling errors, while external disturbances include gusts, collisions, and other external interference. The total disturbance is superimposed on the control output to compensate for the disturbance. This embodiment uses a flip flight strategy to cope with severe disturbances exceeding the conventional compensation range. When the UAV encounters extreme disturbances such as sudden collisions, a preset flip threshold is used to switch the control output and disturbance compensation value, estimating the disturbance magnitude. If the maximum power output cannot cope with the disturbance, the UAV will compliantly perform a 180-degree flip to achieve inverted flight; otherwise, normal flight control is maintained. On the one hand, it can preserve the drone's power, ensuring that the drone remains controllable under sudden and intense interference (preserving power means preserving controllability). On the other hand, it can avoid direct impact damage to the drone caused by interference, thus enhancing the drone's maneuverability and adaptability in emergency situations.

[0114] Please see Figure 4 This application also provides a drone flight control system that can implement the above-described drone flight control method. The system includes:

[0115] The first module 401 is used to acquire the actual and expected states of the UAV;

[0116] The second module 402 is used to perform attitude control processing on the measured state and the desired state through a model predictive control-based attitude controller to obtain a control output;

[0117] The third module 403 is used to perform disturbance observation and compensation processing on the measured state and the control output through a linear extended state observer to obtain a disturbance compensation value.

[0118] The fourth module 404 is used to perform a flip-over switching judgment process on the control output and the disturbance compensation value according to a preset flip-over threshold, and obtain a flip-over switching judgment result.

[0119] The fifth module 405 is used to update the desired state when the flip switching judgment result is flip switching, and generate a flip control command to control the flight of the UAV.

[0120] The sixth module 406 is used to perform superposition processing on the control output according to the disturbance compensation value when the flip-switch judgment result is normal control, and determine the execution command to perform flight control on the UAV.

[0121] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0122] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned UAV flight control method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0123] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0124] Please see Figure 5 , Figure 5 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0125] The processor 501 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0126] The memory 502 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 502 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 502 and is called and executed by the processor 501 using the UAV flight control method of the embodiments of this application.

[0127] The input / output interface 503 is used to implement information input and output;

[0128] The communication interface 504 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0129] Bus 505 transmits information between various components of the device (e.g., processor 501, memory 502, input / output interface 503, and communication interface 504);

[0130] The processor 501, memory 502, input / output interface 503, and communication interface 504 are connected to each other within the device via bus 505.

[0131] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described UAV flight control method.

[0132] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0133] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0134] This application provides a method, system, device, and medium for UAV flight control. This solution uses a model predictive control-based attitude controller to process the measured and desired states, obtaining a control output. It combines model prediction and omnidirectional attitude control to effectively control the UAV in all attitude directions, enhancing control robustness and further reducing the risk of UAV crashes. Furthermore, this solution uses a linearly extended state observer to observe and compensate for disturbances in the measured state and control output, obtaining disturbance compensation values. This allows for real-time estimation and compensation of internal and external disturbances, improving the stability of UAV flight control. Moreover, this solution performs a flip-switch judgment process on the control output and disturbance compensation values ​​based on a preset flip threshold, obtaining a flip-switch judgment result. When the flip-switch judgment result is a flip switch, the desired state is updated and a flip control command is generated to control the UAV. When the flip-switch judgment result is normal control, the control output is superimposed based on the disturbance compensation value to determine the execution command for flight control. This flip-switch strategy effectively determines whether external disturbances exceed the UAV's tolerance range, enhancing the stability and safety of UAV flight control.

[0135] This scheme significantly improves the accuracy of disturbance observation by combining Active Disturbance Rejection Control (ADRC) and geometric control. This integrated method can estimate and compensate for disturbances inside and outside the system in real time, and is equally effective even in the face of sudden and severe disturbances encountered during high-speed flight or under complex weather conditions. Furthermore, employing a geometric manifold-based attitude control method, this scheme proposes a real-time target attitude tracking trajectory that conforms to the characteristics of rigid body rotational dynamics. This method effectively solves the problem of cumulative errors that may arise in continuous rotation using traditional Euler angle representation and can achieve effective control in all attitude directions, especially suitable for large attitude error situations under strong disturbances. One of the core aspects of the attitude layer control in this application is model predictive control, which has the ability to predict the UAV's state over a future period, enhancing control robustness and further reducing the risk of UAV crashes. This invention utilizes the bidirectional thrust capability of a quadcopter to enhance the control torque range, generating a larger control torque in extreme cases to cope with sudden and severe disturbances. In addition, the bidirectional thrust capability endows the quadcopter UAV with roll flight capability, which improves the quadcopter UAV's adaptability when encountering disturbances exceeding the compensation range. This not only enhances its maneuverability and adaptability, but also significantly improves the safety and reliability of the aircraft under sudden and severe interference.

[0136] In summary, this invention significantly improves upon existing technologies in terms of interference observation, attitude control, and response to severe interference for quadcopter aircraft, providing more reliable and efficient technical support for the high-risk environment applications of quadcopter UAVs.

[0137] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0138] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0139] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0140] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0141] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0142] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0143] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0144] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0145] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0147] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for controlling flight of a UAV, the method comprising: The method comprises: acquiring a measured state and a desired state of a UAV; performing attitude control processing on the measured state and the desired state by an attitude controller based on model predictive control to obtain a control output; performing disturbance observation and compensation processing on the measured state and the control output by a linear extended state observer to obtain a disturbance compensation value; performing flip switching judgment processing on the control output and the disturbance compensation value according to a preset flip threshold to obtain a flip switching judgment result; when the flip switching judgment result is flip switching, performing update calculation processing on the desired state to generate a flip control instruction for flight control of the UAV; when the flip switching judgment result is normal control, performing superposition processing on the control output according to the disturbance compensation value to determine an execution instruction for flight control of the UAV; when the flip switching judgment result is flip switching, performing update calculation processing on the desired state to generate a flip control instruction for flight control of the UAV, comprising: updating the desired state to obtain a flip state; performing attitude control processing on the measured state and the flip state by an attitude controller based on model predictive control to obtain a flip control output; performing disturbance observation and compensation processing on the measured state and the flip control output by a linear extended state observer to obtain a flip disturbance compensation value; wherein the expression of disturbance observation and compensation is as follows: ； In the formula, the first row shows the estimated angular velocity. Update mechanism This represents the estimated angular velocity at the next time step. This represents the estimated value at the current time step. The update process consists of two parts: one part is based on the control input. and moment of inertia The direct impact; another part is the estimation error. After a proportional gain Adjustments after adjustment, here This is the actual angular velocity. The second line of the formula describes the extended state, i.e., the additional torque caused by the unknown disturbance. The estimation update process, here, This represents the estimated value for the next time step, while This is an estimate of the current time step; the update mechanism takes into account the estimation error of the angular velocity. and a proportional gain This reflects the direct impact of errors on extended state estimation; performing superposition processing on the flip control output according to the flip disturbance compensation value to determine a flip control instruction for flight control of the UAV.

2. The method of claim 1, wherein, The attitude control processing on the measured state and the desired state by the attitude controller based on model predictive control to obtain a control output comprises: performing trajectory generation processing on the measured state and the desired state to obtain a spherical reference trajectory; performing model predictive control processing on the spherical reference trajectory to obtain a control output.

3. The method of claim 2, wherein, The trajectory generation processing on the measured state and the desired state to obtain a spherical reference trajectory comprises: acquiring default parameters; performing iterative generation processing on the default parameters according to a fastest control synthesis function to obtain a set of interpolation coefficients; performing spherical trajectory geometric interpolation processing on the measured state and the desired state according to the set of interpolation coefficients to obtain a spherical reference trajectory.

4. The method of claim 2, wherein, The model predictive control processing on the spherical reference trajectory to obtain a control output comprises: performing construction processing on each spherical reference trajectory point of the spherical reference trajectory to generate a lookup table; acquiring the measured state, performing coordinate conversion processing on the measured state to obtain an attitude vector; performing sliding window lookup processing on the lookup table according to the attitude vector to obtain a reference trajectory segment; performing attitude control optimization processing on the reference trajectory segment by a prediction model to obtain a control output.

5. The method of claim 1, wherein, The disturbance observation and compensation processing on the measured state and the control output by the linear extended state observer to obtain a disturbance compensation value comprises: The angular velocity estimation and the extended state estimation are performed on the control output of the linear extended state observer to obtain a system estimation value; An error calculation process is performed on the system estimation value according to the measured state to obtain a disturbance compensation value.

6. The method according to any one of claims 1 to 5, characterized in that, Before the measured state and the expected state of the UAV are obtained, the UAV is integrated, specifically including: The three-dimensional electronic speed controller and the reversible propeller are integrated on the UAV; The brushless motor is driven by the three-dimensional electronic speed controller to rotate the UAV in the forward and reverse directions; The reversible propeller provides stable and consistent thrust when the UAV rotates in the forward and reverse directions.

7. A drone flight control system, characterized in that, The system comprises: A first module for obtaining the measured state and the expected state of the UAV; A second module for performing attitude control processing on the measured state and the expected state by the attitude controller based on model predictive control to obtain a control output; A third module for performing disturbance observation and compensation processing on the measured state and the control output by the linear extended state observer to obtain a disturbance compensation value; A fourth module for performing flip switching judgment processing on the control output and the disturbance compensation value according to a preset flip threshold to obtain a flip switching judgment result; A fifth module for performing update calculation processing on the expected state when the flip switching judgment result is flip switching to generate a flip control instruction for flight control of the UAV; A sixth module for performing superposition processing on the control output according to the disturbance compensation value when the flip switching judgment result is normal control to determine an execution instruction for flight control of the UAV; When the flip switching judgment result is flip switching, the expected state is updated to obtain a flip state, the measured state and the flip state are subjected to attitude control processing by the attitude controller based on model predictive control to obtain a flip control output, the measured state and the flip control output are subjected to disturbance observation and compensation processing by the linear extended state observer to obtain a flip disturbance compensation value, the flip disturbance compensation value is used to perform superposition processing on the flip control output to determine a flip control instruction for flight control of the UAV. The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method of any one of claims 1 to 6 when executing the computer program. The computer program is executed by the processor to implement the method of any one of claims 1 to 6. ​ ​ ； In the formula, the first row shows the estimated angular velocity. Update mechanism This represents the estimated angular velocity at the next time step. This represents the estimated value at the current time step. The update process consists of two parts: one part is based on the control input. and moment of inertia The direct impact; another part is the estimation error. After a proportional gain Adjustments after adjustment, here This is the actual angular velocity. The second line of the formula describes the extended state, i.e., the additional torque caused by the unknown disturbance. The estimation update process, here, This represents the estimated value for the next time step, while This is an estimate of the current time step; the update mechanism takes into account the estimation error of the angular velocity. and a proportional gain This reflects the direct impact of errors on extended state estimation; ​ 8. An electronic device, comprising: ​ 9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. ​

Images