A ducted flying platform semi-dynamic obstacle physical interaction control method

Abstract

This invention relates to the field of unmanned aerial vehicle (UAV) control technology, specifically a physical interaction control method for semi-dynamic obstacles on a ducted flight platform. The method includes: establishing a closed-loop state model of the ducted flight platform incorporating physical interaction reaction forces; designing an adaptive ideal system for obstacle characteristics; calculating real-time stabilization compensation control quantities for physical interaction; and combining this with the flight platform's own real-time state information to achieve adaptive adjustment and stable passage of the ducted flight platform through semi-dynamic obstacles with different physical characteristics. This invention achieves adaptive control of the platform for different physical characteristics of semi-dynamic obstacles, such as rebound and damping, without the need for force sensors or control law switching. This ensures the stability of the platform during semi-dynamic obstacle passage, expands the platform's maneuverability and operational range, and enables stable passage under five conditions: small rebound with small damping, medium rebound with small damping, large rebound with small damping, small rebound with large damping, and large rebound with large damping.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle control technology, and more specifically to a semi-dynamic obstacle physical interaction control method for a ducted flight platform. Background Technology

[0002] Currently, unmanned aerial platforms can be divided into open rotorcraft platforms and ducted rotorcraft platforms. Both mainly operate in open spaces and have the ability to avoid obstacles when encountering environmental obstructions.

[0003] Unlike static obstacles such as walls and tables, and dynamic obstacles such as flocks of birds, obstacles such as doors and windows remain static when no external force is applied, but can exhibit dynamic characteristics such as opening under the active external force applied by the unmanned aerial platform; these are called semi-dynamic obstacles. Figure 1 As shown, through semi-dynamic obstacle physical interaction control of the unmanned aerial platform, the unmanned aerial platform can actively pass through doors and windows, breaking through the conventional obstacle avoidance limitations of unmanned aerial platforms and expanding the maneuver space and operating range.

[0004] Compared to open rotorcraft platforms, ducted rotorcraft platforms have a duct that expands the rotor, allowing direct contact with the environment through the duct ring without the need for additional mechanisms. This makes them ideal for implementing semi-dynamic obstacle physical interaction.

[0005] The key to semi-dynamic obstacle physical interaction control of ducted flight platforms lies in ensuring stability and precise tracking of a given crossing trajectory during the crossing process. Stability refers to the stability of the ducted flight platform under the reaction force of semi-dynamic obstacles. However, due to the varying and wide-ranging physical characteristics of different doors and windows, such as rebound and damping, which cannot be precisely known in advance, it is impossible to design a single physical interaction control strategy in advance. This poses a significant challenge to the stability of the ducted flight platform during door and window crossings and to the precise tracking of the crossing trajectory. Specifically, during the process of the ducted flight platform crossing doors and windows: When facing small rebounds, small damping doors and windows, the reaction force is usually a single transient impact. It is necessary to ensure the stability of the ducted flight platform under transient impact reaction force and the accuracy of subsequent trajectory tracking. Under this condition, the instantaneous transient impact mainly affects stability, while the subsequent process mainly affects the accuracy of trajectory tracking. When faced with doors or windows exhibiting large rebound and low damping, or vice versa, the reaction force exhibits pulse-like characteristics. The rapid compensation of this reaction force by the ducted flight platform can lead to uncontrollable oscillating rebound forces. This is because traditional adaptive controllers need to estimate the external force by observing changes in the flight platform's motion state, then calculate the compensation control quantity, and finally apply it to the flight platform to generate a response. This process inherently involves a response delay. When the reaction force changes rapidly in a pulse-like manner, there is a phase deviation between the timing of the compensation control quantity's application to the flight platform and the actual timing of the external force's application. This results in the flight platform still experiencing compensation in the same direction after the reaction force disappears, leading to over-response. To correct this over-response, the controller performs a callback. If the callback coincides with a new pulse reaction force, it further exacerbates the flight platform's oscillations and deteriorates trajectory tracking performance. When faced with large rebounds, high-damping doors and windows, the reaction force on a ducted flight platform is continuous. If the ducted flight platform does not compensate for the reaction force in time, it may retreat or even become unstable under the action of external forces.

[0006] Existing technologies lack a physical interaction method that can adaptively control the different physical characteristics of semi-dynamic obstacles, such as rebound and damping, making it difficult to guarantee the stability of the ducted flight platform during the entire process of traversing semi-dynamic obstacles and the accurate tracking of the traverse trajectory. Summary of the Invention

[0007] In view of this, the present invention provides a semi-dynamic obstacle physical interaction control method for ducted flight platforms, which aims to solve the problems of ducted flight platforms having difficulty in adaptively adjusting control strategies and insufficient stability during the crossing process when crossing semi-dynamic obstacles with different physical characteristics such as rebound and damping. This improves the adaptive control capability of ducted flight platforms for semi-dynamic obstacles with different physical characteristics, ensures the stability of the crossing process, and expands the maneuverability and operating range of ducted flight platforms.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a semi-dynamic obstacle physical interaction control method for a ducted flight platform, comprising the following steps: Establish a closed-loop state model of a ducted flight platform that includes physical interaction reaction forces; Based on the closed-loop state model of the flight platform, design an adaptive ideal system for obstacle characteristics; Based on the obstacle characteristic adaptive ideal system, calculate the real-time stabilization compensation control quantity of physical interaction; At every moment, the physical interaction control law is solved and implemented to drive the flight platform to achieve stable control during the process of traversing semi-dynamic obstacles.

[0009] In one specific feasible implementation, establishing a closed-loop state model of the ducted flight platform that includes physical interaction reaction forces includes: Establish a dynamic model of the ducted flight platform; Based on the dynamic model of the ducted flight platform, vertical speed controllers, roll attitude controllers, pitch attitude controllers, and yaw attitude controllers are designed for the vertical, roll, pitch, and yaw channels respectively. A lateral speed controller and a longitudinal speed controller are designed for the lateral and longitudinal channels, respectively, on the outer ring of the roll attitude controller and pitch attitude controller. Based on the dynamic model of the ducted flight platform and the designed attitude controller and velocity controller, the closed-loop state models of the lateral and longitudinal channels of the flight platform are obtained.

[0010] In a specific feasible implementation, the closed-loop state model of the lateral passage of the flight platform is as follows: The closed-loop state model of the longitudinal passage of the flight platform is as follows: in, for Laplace form, For the lateral velocity of the flight platform, for Laplace form, For the desired lateral velocity of the flight platform, Let be the closed-loop transfer function of the lateral passage of the flight platform. The physical interaction reaction force experienced by the flight platform in the lateral direction The resulting equivalent disturbance The equivalent input loss in the lateral channel caused by the change in rotor aerodynamic characteristics due to obstacle constraints during the physical interaction process of the flight platform; for Laplace form, For the longitudinal velocity of the flight platform, for Laplace form, For the desired longitudinal velocity of the flight platform, Let be the closed-loop transfer function of the longitudinal channel of the flight platform. The physical interaction reaction force experienced by the flight platform along the longitudinal direction The resulting equivalent disturbance The equivalent input loss in the longitudinal channel is caused by the change in rotor aerodynamic characteristics due to obstacle constraints during the physical interaction process of the flight platform.

[0011] In one specific feasible implementation, the design of an adaptive ideal system for obstacle characteristics based on the closed-loop state model of the flight platform includes: Based on the closed-loop state model of the flight platform, the physical interaction control law of the flight platform is defined as follows: Using this as input to the closed-loop state model of the flight platform, a closed-loop state model of the flight platform considering semi-dynamic obstacle physical interaction control is obtained. Design an ideal system that adapts to obstacle characteristics.

[0012] In one specific feasible implementation, the design barrier characteristic adaptive ideal system is: in, for Laplace form, , The first gain of the ideal system is adapted to the obstacle characteristics. The second gain of the ideal system is adapted to the obstacle characteristics. The third gain of the ideal system is adapted to the obstacle characteristics. The fourth gain for an ideal system that adapts to obstacle characteristics. The physical interaction equivalent perturbation of the obstacle-adaptive ideal system. For the condition values ​​of a piecewise variable ideal system, This is a frequency normalization index.

[0013] In one specific implementation scheme, the step of designing an adaptive ideal system for obstacle characteristics based on the closed-loop state model of the flight platform further includes: Based on the adaptive ideal system of obstacle characteristics, the closed-loop state model of the flight platform considering semi-dynamic obstacle physical interaction control is re-represented in state-space form; Design a state estimator for the system. Based on the flight platform closed-loop state model and the state estimator, the flight platform closed-loop state error is obtained. .

[0014] In one specific implementation scheme, the calculation of the real-time stabilization compensation control quantity for physical interaction based on the obstacle characteristic adaptive ideal system includes: Design an adaptive ideal system based on obstacle characteristics. and The update law is in, , They are respectively , rate of change, For the first gain of the update law, Represents the projection operator. For the second gain of the update law, This refers to the closed-loop state error of the flight platform.

[0015] In a specific feasible implementation, according to and Calculate the real-time stabilization compensation control quantity of physical interaction. for: .

[0016] In a specific feasible implementation, the step of solving and implementing the physical interaction control law at each moment to drive the flight platform to achieve stable control during the process of traversing semi-dynamic obstacles includes: At every moment, the control quantity is real-time stabilized and compensated based on physical interaction. and the expected lateral and longitudinal speeds of the flight platform Calculate the physical interaction control law of the flight platform Physical interaction control law As input to the closed-loop state model of the flight platform, and combined with the attitude controller and velocity controller in the closed-loop state model of the flight platform, the flight platform is driven to achieve stable control during the process of crossing semi-dynamic obstacles.

[0017] In one specific implementation scheme, the frequency normalization index Depend on The calculations were performed using a sliding window method, where RMS was calculated. The window length is 70. The window length is 50. This is a preset positive decimal to prevent calculation overflow when the denominator is zero. The value range is 10 -6 ~10 -5 The constants between.

[0018] Compared with existing technologies, the ducted flight platform physical interaction control method of this invention enables the ducted flight platform to actively traverse semi-dynamic obstacles, breaking through the limitations of conventional obstacle avoidance, expanding maneuverability and operational range. By establishing a closed-loop state model of the ducted flight platform including physical interaction reaction forces, designing an adaptive ideal system for obstacle characteristics, calculating real-time stability-enhancing compensation control quantities for physical interaction, and combining the flight platform's own real-time state information, the method achieves adaptive adjustment and stable traversal of the ducted flight platform against semi-dynamic obstacles with different physical characteristics, effectively improving the stability of the traversal process. It has the following beneficial effects: This invention enables adaptive control of the platform based on its real-time status information, without the need for force sensors or control law switching, to address different physical characteristics of semi-dynamic obstacles such as rebound and damping. This ensures the stability of the platform during the passage of semi-dynamic obstacles and expands the platform's maneuverability and operational range.

[0019] This invention can adapt to semi-dynamic obstacles with a wide range of physical characteristics, and can achieve stable crossing in five situations: small rebound with small damping, medium rebound with small damping, large rebound with small damping, small rebound with large damping, and large rebound with large damping. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1 This is a schematic diagram illustrating the process of a ducted flight platform actively traversing a semi-dynamic obstacle (gate).

[0022] Figure 2 This is an overall flowchart of a semi-dynamic obstacle physical interaction control method for a ducted flight platform as described in this invention.

[0023] Figure 3 To design a flowchart.

[0024] Figure 4 This is a flowchart of the implementation process.

[0025] Figure 5 The diagrams show the effects of four controllers on semi-dynamic obstacles with small rebound (spring0.001) and small damping (damping0.001). (a) shows the reaction force of the four controllers, (b) shows the x-axis velocity of the four controllers, (c) shows the pitch attitude of the four controllers, and (d) shows the time constant of the obstacle characteristic adaptive ideal system.

[0026] Figure 6 The diagrams show the effects of four controllers on semi-dynamic obstacles with medium rebound (spring 0.02) and low damping (damping 0.001). (a) shows the reaction force of the four controllers, (b) shows the x-axis velocity of the four controllers, (c) shows the pitch attitude of the four controllers, and (d) shows the time constant of the obstacle characteristic adaptive ideal system.

[0027] Figure 7 The diagrams show the effects of four controllers under semi-dynamic obstacles with large rebound (spring0.3) and small damping (damping0.001). (a) shows the reaction force of the four controllers, (b) shows the x-axis velocity of the four controllers, (c) shows the pitch attitude of the four controllers, and (d) shows the time constant of the obstacle characteristic adaptive ideal system.

[0028] Figure 8 The diagrams show the effects of four controllers on semi-dynamic obstacles with small rebound (spring0.001) and large damping (damping0.2). (a) shows the reaction force of the four controllers, (b) shows the x-axis velocity of the four controllers, (c) shows the pitch attitude of the four controllers, and (d) shows the time constant of the obstacle characteristic adaptive ideal system.

[0029] Figure 9 The diagrams show the effects of four controllers under semi-dynamic obstacles with large rebound (spring0.2) and large damping (damping0.1). (a) shows the reaction force of the four controllers, (b) shows the x-axis velocity of the four controllers, (c) shows the pitch attitude of the four controllers, and (d) shows the time constant of the obstacle characteristic adaptive ideal system. Detailed Implementation

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] The present invention discloses a semi-dynamic obstacle physical interaction control method for a ducted flight platform, such as... Figure 2 As shown, it includes the following steps: ① Establish a closed-loop state model of the ducted flight platform that includes physical interaction reaction forces; The dynamic model of the ducted jet platform is established as follows: in, , , These are the longitudinal, lateral, and vertical accelerations of the flight platform, respectively. , , These are the longitudinal, lateral, and vertical velocities of the flight platform, respectively. , , These are the roll, pitch, and yaw accelerations of the flight platform, respectively. , , These are the roll, pitch, and yaw angular velocities of the flight platform, respectively. , These are the roll angle and pitch angle of the flight platform, respectively. For the quality of the flight platform, , , These are the moments of inertia in the roll, pitch, and yaw directions of the flight platform, respectively. It is the acceleration due to gravity. For the lift of the flight platform rotor, , , These are the roll, pitch, and yaw control moments of the flight platform, respectively. , These are the physical interaction reaction forces that the flight platform experiences along the longitudinal and lateral directions, respectively. Based on the dynamic model of a ducted flight platform, vertical speed controllers, roll attitude controllers, pitch attitude controllers, and yaw attitude controllers are designed for the vertical, roll, pitch, and yaw channels respectively: Vertical speed controller Roll attitude controller Pitch Attitude Controller Yaw attitude controller in, For the yaw angle of the flight platform, , , , These are the desired vertical velocity, desired roll angle, desired pitch angle, and desired yaw angle of the flight platform, respectively. The control gain is tuned using bandwidth matching and pole placement methods, with the desired closed-loop dynamic characteristics of each channel as the target. Vertical channel first control gain Vertical channel second control gain , The desired vertical velocity response time constant (ranging from 0.2 to 0.5 s); First control gain of the roll channel Second control gain of the roll channel , The desired roll-off response time constant (range 0.1–0.3 s). The damping ratio is 0.7 to 1.0. Pitch channel first control gain Pitch channel second control gain , The desired pitch response time constant (range 0.1–0.3 s). The damping ratio is 0.7 to 1.0. Yaw channel first control gain yaw channel second control gain , The desired yaw response time constant (range 0.15–0.4 s). The damping ratio is 0.7 to 1.0. The lateral and longitudinal speed controllers are designed for the lateral and longitudinal channels respectively on the outer rings of the roll attitude controller and pitch attitude controller. in, , These represent the desired lateral velocity and desired longitudinal velocity of the flight platform, respectively, and the first control gain of the lateral channel. Second control gain of the lateral channel , The desired lateral velocity response time constant (range 0.3–0.6 s). Damping ratio (range 0.7–1.0); first control gain of the longitudinal channel. Longitudinal channel second control gain , The desired longitudinal velocity response time constant (range 0.3–0.6 s). The damping ratio is 0.7 to 1.0; the control outputs of the lateral speed controller and the longitudinal speed controller are the reference inputs of the roll attitude controller and the pitch attitude controller, respectively. Based on the dynamic model of the ducted flight platform and the designed attitude controller and velocity controller, the closed-loop state models of the lateral channel and the longitudinal channel of the flight platform are obtained as follows: in, for Laplace form, for Laplace form, Let be the closed-loop transfer function of the lateral passage of the flight platform. The physical interaction reaction force experienced by the flight platform in the lateral direction The resulting equivalent disturbance The equivalent input loss in the lateral channel is caused by changes in rotor aerodynamic characteristics due to obstacle constraints during the physical interaction process of the flight platform. for Laplace form, for Laplace form, Let be the closed-loop transfer function of the longitudinal channel of the flight platform. The physical interaction reaction force experienced by the flight platform along the longitudinal direction The resulting equivalent disturbance The equivalent input loss in the longitudinal channel caused by the change in rotor aerodynamic characteristics due to obstacle constraints during the physical interaction process of the flight platform; The closed-loop state models of the lateral and longitudinal passages of the flight platform have the same mathematical form. In steps ② and ③, based on the closed-loop state models of the lateral and longitudinal passages of the flight platform, the same design process is used to construct the physical interaction control laws of the lateral and longitudinal passages. For simplicity, the closed-loop state models of the lateral and longitudinal passages are uniformly expressed as: in, , , , , ; ② Based on the closed-loop state model of the flight platform, design an adaptive ideal system with obstacle characteristics. The design process is as follows: Figure 3 As shown; Figure 3 The first four items correspond to step ①, and the fifth, sixth, and seventh items correspond to steps ② through ④ respectively. Based on the closed-loop state model of the flight platform, the physical interaction control law of the flight platform is defined as follows: Using this as input to the flight platform closed-loop state model, the resulting flight platform closed-loop state model considering semi-dynamic obstacle physical interaction control is as follows: in, for Laplace form, , The input to the original closed-loop state model of the flight platform is the desired lateral and longitudinal velocities of the flight platform. Laplace form, Real-time stabilization and compensation control for physical interaction Laplace form, Calculated from steps ② and ③, To control the gain, ,in This is the cutoff frequency of the low-pass filter. The value range is between 30 and 50; The design of an ideal system that adapts to obstacle characteristics is as follows: in, The time constant of the ideal system is adapted to the obstacle characteristics. Laplace form, , The first gain of the ideal system is adapted to the obstacle characteristics. The second gain of the ideal system is adapted to the obstacle characteristics. The third gain of the ideal system is adapted to the obstacle characteristics. The fourth gain of the adaptive ideal system for obstacle characteristics is determined by the desired ideal tracking time and tuned using the pole placement method. Specifically, for a given desired ideal tracking time... (Value range 0.3~0.5s), define the system reference response frequency band. ( (The time constant conversion factor, with a value ranging from 3 to 4), based on the dynamic order distribution of the ideal system's state equations, the following explicit mapping relationship is established using the characteristic equation coefficient matching principle: Obstacle characteristics adaptive ideal system first gain Second gain of an adaptive ideal system with barrier characteristics Third gain of an adaptive ideal system with obstacle characteristics Fourth gain of an adaptive ideal system with barrier characteristics ;in, Offline calibration coefficients to match the rotational inertia of the flight platform and the contact stiffness of the obstacle. The value ranges from 0.0015 to 0.0025. The value ranges from 0.01 to 0.03. The value range is between 0.2 and 0.3. The value range is between 0.00015 and 0.0002. Preferably, during the tuning process, the calibration coefficients can be fine-tuned through offline simulation to compensate for the actual interaction nonlinearity, ensuring that the output trajectory of the ideal system meets the preset dynamic response requirements; The physical interaction equivalent perturbation of the obstacle-adaptive ideal system, and The relationship between them is: in, for The time domain representation, for The estimated value can be obtained from steps ② and ③; The time constant of the ideal system is adapted to the obstacle characteristics. The condition value ranges from -2 to -1. As a frequency normalization index, by The calculations were performed using a sliding window method, where RMS was calculated. The window length is 70. The window length is 50. This is a preset positive decimal to prevent calculation overflow when the denominator is zero. The value range is 10 -6 ~10 -5 The constants between; for The rate of change; Flight platform closed-loop state error Calculate using the following steps: a. Based on the obstacle characteristics of the adaptive ideal system, the closed-loop state model of the flight platform considering semi-dynamic obstacle physical interaction control can be re-expressed as: All variables have been defined above; b. The system state estimator in design step a is: , , They are respectively , , The time-domain representation of the estimated value, for The differential; c. Based on the closed-loop state model of the flight platform in a and the state estimator in b, obtain the closed-loop state error of the flight platform. for: for The time-domain representation; ③ Based on the obstacle characteristic adaptive ideal system, calculate the real-time stabilization compensation control quantity of physical interaction; Design an adaptive ideal system based on obstacle characteristics. and The update law is in, , They are respectively , rate of change, This is the first gain of the update law, with a value ranging from -30 to 0. The projection operator is represented by the formula. The calculation yielded, where , This is the projection tolerance limit, typically taken between 0.01 and 0.1. for Upper limit of value, Representation function gradient, express norm, This is the second gain of the update law, and its value ranges from 1 to 2. according to and Calculate the real-time stabilization compensation control quantity of physical interaction. for: ④ For example Figure 4 As shown, the physical interaction control law is solved and implemented at each moment to drive the flight platform to achieve stable control during the process of traversing semi-dynamic obstacles: At each moment, based on the physical interaction real-time stabilization compensation control quantity obtained in step ③, and the expected lateral and longitudinal speeds of the flight platform Calculate the physical interaction control law of the flight platform Physical interaction control law As input to the closed-loop state model of the flight platform, and combined with the attitude controller and velocity controller in the closed-loop state model of the flight platform in step ①, the flight platform is driven to achieve stable control during the process of traversing semi-dynamic obstacles; here, the calculation will be... As the input to the closed-loop state model, which is described above, the input to the speed controller is used, and the output of the speed controller is used as the input to the attitude controller. The basic control principle is as follows: During the physical interaction between the flight platform and the semi-dynamic obstacle, the different physical characteristics of the semi-dynamic obstacle, such as rebound and damping, cause closed-loop state errors of the flight platform. and rate of change Real-time changes are observed by designing an adaptive ideal system based on obstacle characteristics, and estimating the equivalent perturbation value of the physical interaction of the adaptive ideal system based on obstacle characteristics. and frequency normalization index Adaptive adjustment of physical interaction real-time stabilization compensation control quantity The expected lateral and longitudinal speeds of the flight platform Based on this, real-time stabilization compensation control quantities are added through physical interaction. This yields a dynamically adjusted physical interaction control law. It is applied to the flight platform to enable adaptive adjustment and stable passage through semi-dynamic obstacles with different physical characteristics during the physical interaction process of the flight platform.

[0032] The following are specific examples: The controller parameter settings proposed in this invention are as follows: =-2, =0.07, =1.6, =0.06, =30, To verify that the method proposed in this invention can achieve stable passage through semi-dynamic obstacles with different physical characteristics such as rebound and damping, the control effects under five conditions—small rebound + small damping, medium rebound + small damping, large rebound + small damping, small rebound + large damping, and large rebound + large damping—are presented. The condition classification is mainly based on three types of reaction forces: single impact force, oscillation force, and continuous contact force. Regardless of the material or installation method of doors and windows, from a dynamic perspective, they must belong to one of these three reaction force types. Five typical combinations of rebound and damping were then selected to cover these three reaction force types.

[0033] The simulation physical model of the opening and closing door in this invention is as follows: ,in Let be the moment of inertia of the door. It is a fixed value, determined by the mass distribution of the door body. The damping coefficient (N*m / (deg / s)) characterizes the physical properties of damping. The rotational stiffness coefficient (N*m / deg) characterizes the physical properties of springback. The opening angle of the door. The torque applied to the door hinge.

[0034] Simulation results show that the method proposed in this invention ( Figures 5 to 9 The VAC controller (referred to as VAC) can achieve stable crossing in all five scenarios, adapting to semi-dynamic obstacles with wide variations in physical characteristics. For comparison, the PD controller, PID controller, and conventional adaptive controller are also presented. Figures 5 to 9As shown in the figure (denoted as CAC), while these controllers can achieve stable crossing in some or all situations, they cannot adapt to all situations. The difference between conventional adaptive controllers and the controller of this invention is that they lack the "obstacle characteristic adaptation" component; the rest of the controller structure is the same. It is a constant value (0.0018 is taken according to the actual debugging), and all other parameters are the same.

[0035] In the five simulation results below, the y-axis label Fx (N) represents the reaction force on the flight platform, x_rate (m / s) represents the UAV's x-axis velocity, and theta (rad) represents the UAV's pitch angle. The time constant is the adaptive ideal system time constant for obstacle characteristics.

[0036] Small rebound (spring 0.001), small damping (damping 0.001) Depend on Figure 5 The external forces, velocity, attitude, and controller parameters of the flight platform shown are as follows. It can be seen that when the flight platform traverses a semi-dynamic obstacle with low rebound and low damping, it will experience a brief, sudden collision force, and all four controllers can successfully pass through. Figure 5 (a) shows the reaction forces experienced by different controllers; Figure 5 (b) For the x-axis velocity of different controllers, the recovery speed of the four controllers for tracking the expected reference velocity trajectory is similar, but the PID controller is slower to recover speed when leaving the obstacle due to the integral action; Figure 5 (c) Pitch attitude of different controllers; Figure 5 (d) is the time constant diagram of the VAC of the obstacle-adaptive ideal system.

[0037] Medium rebound (spring 0.02), small damping (damping 0.001) Depend on Figure 6 The external forces, velocity, attitude, and controller parameters of the flight platform shown are as follows. It can be seen that when the flight platform traverses a semi-dynamic obstacle with rebound and low damping, multiple collision forces will occur, and all four controllers can successfully pass through. Figure 6 (a) shows the reaction forces experienced by different controllers; Figure 6 (b) For different controller x-axis velocities, the CAC controller will exhibit large oscillations when leaving a semi-dynamic obstacle, resulting in a decrease in control quality. The VAC controller has lower speed fluctuation amplitude and frequency than the CAC, thus avoiding large oscillations. Figure 6 (c) shows the pitch attitude of different controllers. The pitch attitude of the CAC controller oscillates significantly and is unstable, while the pitch attitude of the VAC controller is relatively stable. Figure 6(d) is the time constant diagram of the VAC of the obstacle-adaptive ideal system.

[0038] Large rebound (spring 0.3), small damping (damping 0.001) Depend on Figure 7 The external forces, velocity, attitude, and controller parameters of the flight platform shown are as follows. It is known that when the flight platform traverses a semi-dynamic obstacle with high rebound and low damping, it will be subjected to a force of high-frequency collision in the early stage and continuous contact in the later stage. The PD controller cannot pass under the large reaction force and stops moving forward, while the other three types of controllers can pass successfully. Figure 7 (a) shows the reaction forces experienced by different controllers; Figure 7 (b) For different controllers with x-axis velocities, the PID controller has a slower crossing time and its control quality deteriorates when leaving the semi-dynamic obstacle, resulting in a sudden surge. The CAC controller performs better than the PID controller, but there are small oscillations during the process. The VAC controller can ensure fast crossing and the oscillations are significantly reduced compared to the CAC controller during the process. Figure 7 (c) Pitch attitude of different controllers; Figure 7 (d) is the time constant diagram of the VAC of the obstacle-adaptive ideal system.

[0039] Small rebound (spring 0.001), large damping (damping 0.2) Depend on Figure 8 The external forces, velocity, attitude, and controller parameters of the flight platform shown are as follows. It can be seen that when the flight platform traverses a semi-dynamic obstacle with low rebound and high damping, it will be subjected to a force that initially involves high-frequency collisions and later involves small-amplitude continuous contact. All four controllers can successfully navigate this obstacle. Figure 8 (a) shows the reaction force on different controllers. Because the damping is related to the angular velocity of the door, and the flight platform collides with the door multiple times in the early stage, the angular velocity of the door will tend to stabilize in the later stage. Therefore, the flight platform will be subjected to a force of high-frequency collision in the early stage and small-amplitude continuous contact in the later stage throughout the process. Figure 8 (b) For different controllers with x-axis velocities, the PD controller has a slower recovery speed for tracking the expected reference trajectory, the PID controller has a slower speed recovery when leaving the gate, the CAC controller exhibits large, high-frequency, and continuous oscillations, and the control process is unstable, while the VAC controller has both a faster reference trajectory tracking recovery speed and can avoid large, high-frequency oscillations. Figure 8 (c) Pitch attitude of different controllers. The CAC controller has a larger attitude oscillation frequency throughout the process, while the VAC controller ensures attitude stability. Figure 8 (d) is the time constant diagram of the VAC of the obstacle-adaptive ideal system.

[0040] High springback (spring 0.2), high damping (damping 0.1) Depend on Figure 9 The external forces, velocity, attitude, and controller parameters of the flight platform shown are as follows. It is known that when the flight platform traverses a semi-dynamic obstacle with high rebound and high damping, it will be subjected to a force of high-frequency collision in the early stage and continuous contact in the later stage. The PD controller cannot pass under the large reaction force and stops moving, while the other three types of controllers can pass successfully. Figure 9 (a) shows the reaction forces experienced by different controllers; Figure 9 (b) For the x-axis velocity of different controllers, the PID controller can pass successfully, but the crossing time is slow, and the control quality deteriorates when leaving the semi-dynamic obstacle, with a sudden surge, slow recovery speed, and slow speed return when leaving the obstacle. The CAC controller can pass successfully, but there is high-frequency oscillation, which is more dangerous. The VAC controller can ensure fast crossing and maintain stability during the crossing process. Figure 9 (c) Pitch attitudes of different controllers. The CAC controller shows large oscillations in attitude, while the VAC controller shows smaller oscillation frequency and amplitude than the CAC, and the attitude is more stable. Figure 9 (d) is the time constant diagram of the VAC of the obstacle-adaptive ideal system.

[0041] The various embodiments described in this specification are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling physical interaction of a ducted flying platform semi-dynamic obstacle, characterized in that, Includes the following steps: Establish a closed-loop state model of a ducted flight platform that includes physical interaction reaction forces; Based on the closed-loop state model of the flight platform, design an adaptive ideal system for obstacle characteristics; Based on the obstacle characteristic adaptive ideal system, calculate the real-time stabilization compensation control quantity of physical interaction; At every moment, the physical interaction control law is solved and implemented to drive the flight platform to achieve stable control during the process of traversing semi-dynamic obstacles.

2. The ducted flying platform semi-dynamic obstacle physical interaction control method according to claim 1, characterized in that, The establishment of a closed-loop state model for a ducted flight platform that includes physical interaction reaction forces includes: Establish a dynamic model of the ducted flight platform; Based on the dynamic model of the ducted flight platform, vertical speed controllers, roll attitude controllers, pitch attitude controllers, and yaw attitude controllers are designed for the vertical, roll, pitch, and yaw channels respectively. A lateral speed controller and a longitudinal speed controller are designed for the lateral and longitudinal channels, respectively, on the outer ring of the roll attitude controller and pitch attitude controller. Based on the dynamic model of the ducted flight platform and the designed attitude controller and velocity controller, the closed-loop state models of the lateral and longitudinal channels of the flight platform are obtained.

3. The method of claim 2, wherein, The closed-loop state model of the lateral passage of the flight platform is as follows: The closed-loop state model of the longitudinal passage of the flight platform is as follows: wherein, is the Laplace form of is the lateral velocity of the flight platform, is the Laplace form of is the desired lateral velocity of the flight platform, is the closed-loop transfer function of the lateral channel of the flight platform, is the physical interaction force acting on the flight platform in the lateral direction resulting equivalent disturbance, is the equivalent input loss of the lateral channel of the flight platform caused by the change of the rotor aerodynamic characteristics due to the obstacle constraint during the physical interaction process of the flight platform. for Laplace form, For the longitudinal velocity of the flight platform, for Laplace form, For the desired longitudinal velocity of the flight platform, Let be the closed-loop transfer function of the longitudinal channel of the flight platform. The physical interaction reaction force experienced by the flight platform along the longitudinal direction The resulting equivalent disturbance The equivalent input loss in the longitudinal channel is caused by the change in rotor aerodynamic characteristics due to obstacle constraints during the physical interaction process of the flight platform.

4. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 1, characterized in that, The design of an adaptive ideal system for obstacle characteristics based on the closed-loop state model of the flight platform includes: Based on the closed-loop state model of the flight platform, the physical interaction control law of the flight platform is defined as follows: Using this as input to the closed-loop state model of the flight platform, a closed-loop state model of the flight platform considering semi-dynamic obstacle physical interaction control is obtained. Design an ideal system that adapts to obstacle characteristics.

5. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 4, characterized in that, The design barrier characteristic adaptive ideal system is: in, for Laplace form, , The first gain of the ideal system is adapted to the obstacle characteristics. The second gain of the ideal system is adapted to the obstacle characteristics. The third gain of the ideal system is adapted to the obstacle characteristics. The fourth gain for an ideal system that adapts to obstacle characteristics. The physical interaction equivalent perturbation of the obstacle-adaptive ideal system. For the condition values ​​of a piecewise variable ideal system, This is a frequency normalization index.

6. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 5, characterized in that, The design of an adaptive ideal system for obstacle characteristics based on the closed-loop state model of the flight platform also includes: Based on the adaptive ideal system of obstacle characteristics, the closed-loop state model of the flight platform considering semi-dynamic obstacle physical interaction control is re-represented in state-space form; Design a state estimator for the system. Based on the flight platform closed-loop state model and the state estimator, the flight platform closed-loop state error is obtained. .

7. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 1, characterized in that, The adaptive ideal system based on obstacle characteristics calculates the real-time stabilization compensation control quantity for physical interaction, including: Design an adaptive ideal system based on obstacle characteristics. and The update law is in, , They are respectively , rate of change, For the first gain of the update law, Represents the projection operator. For the second gain of the update law, This refers to the closed-loop state error of the flight platform.

8. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 7, characterized in that, according to and Calculate the real-time stabilization compensation control quantity of physical interaction. for: 。 9. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 1, characterized in that, The process of solving and implementing the physical interaction control law at each moment to drive the flight platform to achieve stable control during the process of traversing semi-dynamic obstacles includes: At every moment, the control quantity is real-time stabilized and compensated based on physical interaction. and the expected lateral and longitudinal speeds of the flight platform Calculate the physical interaction control law of the flight platform Physical interaction control law As input to the closed-loop state model of the flight platform, and combined with the attitude controller and velocity controller in the closed-loop state model of the flight platform, the flight platform is driven to achieve stable control during the process of crossing semi-dynamic obstacles.

10. The semi-dynamic obstacle physical interaction control method for a ducted flight platform according to claim 5, characterized in that, The frequency normalization index Depend on The calculations were performed using a sliding window method, where RMS was calculated. The window length is 70. The window length is 50. This is a preset positive decimal to prevent calculation overflow when the denominator is zero. The value range is 10 -6 ~10 -5 The constants between.

Images