A low yaw angle velocity quadrotor single rotor failure tolerant control method

Abstract

This invention discloses a fault-tolerant control method for a quadrotor with low yaw rate under single rotor failure, belonging to the field of unmanned aerial vehicle (UAV) fault-tolerant control. The method includes: analyzing the problem of strong coupling between the yaw channel and pitch / roll channels in traditional quadrotor UAVs after complete single rotor failure, which easily leads to high-speed spin; selecting a parallel layout configuration to reduce channel coupling; establishing a dynamic model under complete single rotor failure conditions, modeling the pull loss caused by rotor failure as an equivalent system control quantity disturbance, and uniformly merging it into a lumped disturbance; constructing a disturbance observation and compensation mechanism to achieve real-time estimation and dynamic feedforward compensation of the disturbance; utilizing the bidirectional pull characteristics of the rotor to allow the compensated control quantity to break through the unidirectional pull constraint, redistributing the control quantity of the remaining healthy rotor under failure conditions to form a new dynamic equilibrium. This invention can maintain a low yaw rate under single rotor failure conditions, ensuring trajectory tracking accuracy and improving system safety and controllability.

Description

Technical Field

[0001] This invention relates to the field of fault-tolerant control technology for unmanned aerial vehicles (UAVs), and in particular to a fault-tolerant control method for single-rotor failure of a quadcopter with low yaw rate. Background Technology

[0002] Quadrotor drones, as a typical multi-rotor aircraft, have been widely used in aerial photography, logistics transportation, power line inspection, and emergency rescue due to their advantages such as simple structure, vertical takeoff and landing, and good hovering performance. Their basic control principle involves adjusting the speed of the four rotors to change the thrust generated by each rotor, thereby achieving independent control of the aircraft's pitch, roll, and yaw attitude channels, as well as altitude. Under normal flight conditions, traditional cross-shaped or X-shaped quadrotors can generate the required control torque through differential speed adjustment, achieving stable six-degree-of-freedom motion control. However, when a rotor completely loses its thrust output due to motor stall, blade damage, or ESC malfunction during flight, the system faces the severe challenge of partial actuator failure. This has become a key technical problem restricting the safety and reliability of quadrotor flight.

[0003] To address the problem of complete failure of a single quadrotor, various fault-tolerant control schemes have been proposed in existing technologies. One typical approach is to maintain the controllability of the aircraft in roll, pitch, and trajectory directions by redistributing the thrust commands of the remaining three rotors after a failure. However, this type of method usually abandons active control of the yaw channel. Since the yaw motion of a quadrotor system is driven by rotor anti-torque differential, after a single rotor failure, the anti-torques of the remaining rotors cannot cancel each other out. The aircraft will experience continuous yaw rotation under the strong coupling of roll, pitch, and yaw channels, thus operating in a degenerate flight state with high-speed spin characteristics. Another improved scheme adopts an active fault-tolerant control approach, which involves real-time detection and isolation of the failed rotor through a fault diagnosis module, and then dynamically reconstructing the control allocation matrix based on the diagnosis results, so that the remaining healthy rotors are rematched with the desired control torque. In addition, some researchers, starting from controllability theory, have analyzed the controllability conditions of the system after a failure within the framework of positive controllability theory, considering the physical constraint that the rotor can only generate unidirectional thrust (i.e., positive thrust), and designed fault-tolerant control laws accordingly. Some technical solutions also improve the system's control redundancy after a failure by adding redundant actuators (such as additional spare rotors or tilting mechanisms).

[0004] However, the aforementioned existing technologies still have the following drawbacks:

[0005] (1) In terms of safety, the traditional fault-tolerant control method that abandons yaw channel control causes the aircraft to continue spinning at high speed after a failure, which not only seriously affects the accuracy of track tracking, but also poses a potential threat to the safety of payload equipment and near-human operation scenarios.

[0006] (2) The active fault-tolerant method that relies on fault diagnosis and control allocation reconfiguration has a control performance that is highly dependent on the real-time performance and accuracy of fault diagnosis and isolation. In engineering implementation, it may still be affected by factors such as fault perception error, detection delay and transient discontinuity of control switching, making it difficult to guarantee robustness.

[0007] (3) Due to the unidirectional positive tensile force constraint of the rotor, the traditional controllability analysis method (such as the Popov-Belevitch-Hautus condition) is no longer applicable to the system analysis after the fault and must be re-discussed under the framework of positive controllability theory, which increases the complexity of control design.

[0008] (4) The solution of improving controllability after failure by adding redundant actuators will bring additional hardware size, weight and cost burden, and increase the complexity of system design, which is not conducive to the requirements of lightweight and low-cost applications. Summary of the Invention

[0009] The technical problem to be solved by this invention is to provide a fault-tolerant control method for single rotor failure of a quadcopter with low yaw rate, so as to solve the system safety and controllability problems after single rotor failure and suppress high yaw rate in fault-tolerant flight; at the same time, the method is deployed by implementing the method through PX4 and GAZEBO software in-loop simulation, which shortens the debugging cycle and reduces development cost.

[0010] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0011] A fault-tolerant control method for a single rotor of a quadcopter with low yaw rate includes the following steps:

[0012] Step 1: Analyze the problem of strong coupling between the yaw channel and the pitch and roll channels of traditional quadcopter UAVs after the complete failure of a single rotor, which easily leads to high-speed spin. By analyzing the system coupling mechanism, a parallel layout configuration is selected to reduce the coupling effect between channels.

[0013] Step 2: Under the parallel layout configuration, establish a dynamic model under the condition of complete failure of a single rotor, model the loss of tension caused by rotor failure as equivalent to the system control quantity disturbance, and uniformly merge it into the lumped disturbance.

[0014] Step 3: Construct a disturbance observation and compensation mechanism to observe and estimate the lumped disturbance online, and compensate the estimated value into the control quantity in a dynamic feedforward manner;

[0015] Step 4: Utilize the bidirectional tension characteristics of the rotor to enable the control quantity after feedforward compensation to break through the traditional unidirectional tension constraint. Distribute the control quantity in the failure state to allow the remaining healthy rotor to form a new dynamic equilibrium state, thereby maintaining a low yaw rate and ensuring track tracking accuracy in the failure condition.

[0016] A further improvement of the technical solution of this invention lies in the following specific implementation process: The trajectory controller generates a virtual control quantity based on the desired trajectory, and combines it with the desired altitude value as the input of the fault-tolerant controller; the fault-tolerant controller calculates the nominal control quantity based on the input; equivalent disturbance modeling is performed for single rotor failure, and the disturbance is incorporated into the system lumped disturbance, and the disturbance is observed and compensated online through a disturbance observation and compensation mechanism; the final control quantity is obtained after feedforward compensation, and the desired tension value of each rotor is calculated according to the control allocation matrix; the rotor responds to the desired tension value according to its bidirectional tension characteristics, thereby realizing fault-tolerant control.

[0017] A further improvement of the technical solution of the present invention is that: the fault-tolerant controller uses the estimated value of the disturbance observer to perform feedforward compensation to obtain the final control quantity after compensation, and calculates the expected tension control quantity of each rotor according to the control allocation matrix relationship; the feedforward compensation is to superimpose the equivalent disturbance estimate output by the disturbance observer into the nominal control quantity in the form of negative feedback.

[0018] A further improvement of the technical solution of the present invention is that: in step 1, the parallel layout configuration is: the traditional cross layout is changed to a diagonal layout in opposite directions, and the four rotors are arranged in pairs and parallel, so that the coupling relationship between the yaw channel and the pitch and roll channels changes, thereby reducing unnecessary yaw torque after a single rotor fails; the rotors on the two arms in the same direction have opposite blade rotation directions, so that the anti-torque generated by the rotors on the same side cancels each other out.

[0019] A further improvement of the technical solution of the present invention is that: in step 1, the bidirectional tension is: breaking through the traditional rotor unidirectional positive tension constraint, enabling the healthy rotor to have positive and negative tension output capabilities, expanding the control quantity after the fault can be set, and providing a physical basis for torque redistribution.

[0020] A further improvement to the technical solution of this invention is as follows: In step 2, the loss of tension caused by rotor failure is modeled as an equivalent system control quantity disturbance and uniformly merged into the lumped disturbance; specifically, the impact of rotor failure on the system is modeled as an equivalent control quantity disturbance, the fault part is separated, and a health matrix and a fault matrix are defined. The diagonal elements of the fault matrix represent the degree of failure of each rotor, and the health matrix is ​​the difference between the allocation matrix and the fault matrix. Under the condition that the rotor fault information is unknown, the system still allocates according to the original allocation matrix, thereby expressing the equivalent disturbance of rotor failure on the system as the relationship between the health matrix and the actual allocation quantity.

[0021] A further improvement of the technical solution of the present invention is that: in step 3, the construction of the disturbance observation and compensation mechanism specifically involves: constructing a fault-tolerant mechanism that does not rely on fault diagnosis and isolation to avoid the reconstruction of the control allocation matrix; and using a disturbance observer to estimate and compensate for the equivalent control disturbance caused by rotor failure as an expanded state in real time.

[0022] A further improvement of the technical solution of the present invention is that: in the parallel layout configuration, the yaw torque is not coupled with the resultant pull force after the failure of a single rotor, thereby avoiding the generation of high-speed spin after the failure of a single rotor.

[0023] A further improvement of the technical solution of the present invention is that the method is implemented on the software-in-the-loop simulation platform of PX4 open-source flight control firmware and Gazebo physics and dynamics engine, and completes the deployment of controllers, disturbance observers and modification of aircraft parameters.

[0024] A further improvement to the technical solution of this invention is as follows: the software-in-the-loop simulation platform is implemented as follows: by modifying the sdf file in PX4, the visualization effect of the parallel layout configuration in the GAZEBO physical simulation platform is realized, and the rotor tension coefficient and speed range are modified; by modifying the rotor drive source code in the PX4 firmware, the modification of the reverse torque direction of any rotor and the removal of the unidirectional tension constraint of any rotor are realized; a custom module is created through the PX4 firmware to complete the deployment of the controller and disturbance observer; the system state published to the uORB message pool after the PX4 firmware data is fused is used as the input of the controller and disturbance observer; the system control quantity after compensation by the controller and disturbance observer is allocated to the four rotors through a fixed control allocation matrix and normalized data processing, as the input of the rotor desired tension control quantity, and finally sent to the GAZEBO side via UDP packetization.

[0025] The technological advancements achieved by this invention due to the adoption of the above technical solutions are as follows:

[0026] 1. This invention reconstructs the physical architecture of a quadcopter UAV by combining a parallel-layout fuselage structure with a rotor actuator capable of bidirectional tension. This weakens the coupling effect between the yaw channel and the pitch and roll channels after a single rotor fails at the hardware level. Combined with a low yaw rate fault-tolerant control method, it can maintain a low yaw rate even under extreme sudden conditions of complete single rotor failure, avoiding the risk of loss of control caused by high-speed spin in traditional methods. This effectively ensures flight safety, payload equipment safety, and safety in near-human operation scenarios.

[0027] 2. This invention addresses the asymmetric dynamic state and extreme disturbances caused by single rotor failure by constructing a fault-tolerant mechanism that does not rely on fault diagnosis and isolation, avoiding the reconstruction of the control allocation matrix. By utilizing a disturbance observer, the equivalent control disturbance caused by rotor failure is estimated and compensated in real time as an extended state. This avoids the performance degradation caused by fault perception errors, detection delays, and transient discontinuities in control switching in traditional methods, enabling the system to have rapid response capability and strong robustness to sudden faults.

[0028] 3. This invention only requires adjusting the body layout and selecting a rotor with bidirectional tensile characteristics, without adding redundant actuators, thus avoiding additional hardware scale, weight and cost burden. The system design complexity is low, which is conducive to lightweight and low-cost applications.

[0029] 4. This invention is based on the PX4 open-source flight control firmware and the GAZEBO physics and dynamics engine to build a software-in-the-loop simulation platform, complete the deployment of the controller and disturbance observer and the modification of aircraft parameters. It can efficiently verify the effectiveness and reliability of the fault-tolerant control algorithm in the simulation environment, greatly shorten the UAV debugging cycle, and reduce the cost and risk of actual flight testing. Attached Figure Description

[0030] 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a schematic diagram of the parallel layout configuration of the quadcopter provided in an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram of the inertial coordinate system and the body coordinate system in an embodiment of the present invention;

[0033] Figure 3 This is a block diagram of the control system structure provided in the embodiments of the present invention;

[0034] Figure 4 This is a comparison diagram of the takeoff tracking effects of two configurations provided in the embodiments of the present invention;

[0035] Figure 5 This is the Euler angle response diagram of a rotor with a parallel layout in a hovering state in an embodiment of the present invention.

[0036] Figure 6 This is a normalized control quantity diagram of the rotor in a hovering state during a parallel layout, as shown in this embodiment of the invention.

[0037] Figure 7 This is a diagram showing the yaw rate of a rotor in a hovering state during a sudden failure in a parallel configuration according to an embodiment of the present invention.

[0038] Figure 8 This is an equivalent control disturbance diagram for sudden rotor failure in a hovering state in a parallel layout according to an embodiment of the present invention.

[0039] Figure 9 This is a comparison diagram of the three-dimensional spatial orthogonal projection distances of the two configurations in trajectory tracking in an embodiment of the present invention;

[0040] Figure 10 This is a diagram of the normalized control quantity of the rotors in the parallel layout for track tracking after the failure of any single rotor in an embodiment of the present invention.

[0041] Figure 11 This is a diagram showing the yaw rate of a parallel arrangement for track tracking after the failure of any single rotor in an embodiment of the present invention.

[0042] Figure 12 This is an equivalent control disturbance diagram for track tracking in a parallel layout according to an embodiment of the present invention after the failure of any single rotor. Detailed Implementation

[0043] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes 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 process, method, product or device.

[0044] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments:

[0045] like Figures 1 to 3 As shown, a fault-tolerant control method for a single rotor of a quadcopter with low yaw rate includes the following steps:

[0046] Step 1: Analyze the problem of strong coupling between the yaw channel and the pitch and roll channels of traditional quadcopter UAVs after the complete failure of a single rotor, which easily leads to high-speed spin. By analyzing the system coupling mechanism, a parallel layout configuration is selected to reduce the coupling effect between channels.

[0047] Specifically, such as Figure 1As shown, the parallel propeller configuration (PPC) reconstructs the physical structure of the quadcopter, changing the traditional alternating propeller configuration (APC) to a diagonal, counter-rotating configuration. The four rotors are no longer arranged symmetrically in a front-to-back, left-to-right pattern, but rather in pairs, arranged in parallel (similar to a double-row parallel arrangement). The purpose is to change the coupling relationship between the yaw channel and the pitch and roll channels, thereby reducing unnecessary yaw moments in the event of a single rotor failure. By modifying the sdf file in PX4, the visualization of the parallel configuration in the GAZEBO physics simulation platform is achieved, along with modifications to the rotor thrust coefficient and speed range.

[0048] Bidirectional tension: Breaking through the traditional unidirectional positive tension constraint of rotors, this feature enables healthy rotors to output tension in both directions, expands the control quantity set after a fault, and provides a physical basis for torque redistribution. By modifying the rotor drive source code in the PX4 firmware, it is possible to modify the reverse torque direction of any rotor and remove the unidirectional tension constraint of any rotor.

[0049] Step 2: Under the parallel layout configuration, establish a dynamic model under the condition of complete single rotor failure. The tension loss caused by rotor failure is modeled as an equivalent system control disturbance and uniformly merged into the lumped disturbance; specifically, as follows... Figure 2 As shown:

[0050] 2.1 Coordinate System Establishment: To facilitate the establishment of the dynamic model and subsequent control and analysis, an inertial coordinate system is selected. With body coordinate system .

[0051] 2.2 Dynamic Modeling: Assuming the body is a rigid body and the inertia matrix is ​​approximately a diagonal matrix. ,and Based on this, the quadrotor dynamics model is established as follows:

[0052] (1)

[0053] in, and Let QUAV be the position and velocity in the inertial coordinate system; Using principal axis vector Describe the projection in the inertial coordinate system, which satisfies , This is the gravitational acceleration vector in the inertial coordinate system. For the angular velocity of the machine system, For the mass of a quadcopter, This represents the total thrust generated by the four propellers. It is a three-axis torque.

[0054] Rotation matrix as follows:

[0055] (2)

[0056] in,

[0057]

[0058] Select The allocation matrix is ​​as follows:

[0059] (3)

[0060] in, Let the desired thrust vectors of the four rotors be... The length of the boom. This is the anti-torque coefficient.

[0061] The state vectors are selected as follows:

[0062] (4)

[0063] The control variables are selected as follows:

[0064] (5)

[0065] 2.3 Rotor Failure Equivalent Disturbance Modeling: The impact of rotor failure on the system is modeled as an equivalent control disturbance. To isolate the faulty component, let... Define the health matrix , .

[0066] To avoid fault diagnosis relying on dynamic control allocation and reconfiguration, the fault-tolerant control method proposed in this invention is unknown to rotor fault information, so the system will still allocate according to equation (3). At this time, the equivalent disturbance to the system due to rotor failure is as follows:

[0067] (6)

[0068] The equivalent disturbance control quantities are selected as follows:

[0069] (7)

[0070] Step 3: Construct a disturbance observation and compensation mechanism to observe and estimate the lumped disturbance online, and compensate the estimated value into the control quantity in a dynamic feedforward manner;

[0071] Specifically, a fault-tolerant mechanism independent of fault diagnosis and isolation is constructed to avoid the reconstruction of the control allocation matrix. A disturbance observer is used to estimate and compensate for the equivalent control disturbance caused by rotor failure as an expanded state in real time. A custom module is created using PX4 firmware to complete the deployment of the controller and disturbance observer. The system state, after being fused with PX4 firmware data and published to the uORB message pool, serves as the input to the controller and disturbance observer. The system control quantities compensated by the controller and disturbance observer are then allocated to the four rotors through a fixed control allocation matrix and normalized data processing, serving as the input to the rotor's desired tension control quantity. Finally, the data is packaged and sent to the GAZEBO side via UDP.

[0072] 3.1 Track Controller Section:

[0073] The mathematical model separating position and velocity from equation (1) is as follows:

[0074] (8)

[0075] in, , , , .

[0076] Based on feedback linearization and backstepping, the trajectory controller part is specifically as follows:

[0077] (9)

[0078] in, This indicates the desired trajectory and desired speed. , To control design parameters.

[0079] 3.2 Disturbance Observer Design:

[0080] According to equation (1), the height and attitude model is obtained as follows:

[0081] (10)

[0082] in, For additive equivalent perturbation terms, The matrix part is specifically as follows:

[0083] (11)

[0084] The constructed perturbation observer is as follows:

[0085] (12)

[0086] in, express The estimated value, Indicates the estimation error. express The estimated value " represents the dot product of vectors, Represents the hyperbolic tangent function. and Design parameters for the estimator.

[0087] 3.3 Fault-tolerant controller design:

[0088] The controller is designed based on feedback linearization and backstepping, specifically as follows:

[0089] (13)

[0090] The disturbance rejection and fault tolerance controller proposed in this invention utilizes The feedforward compensation is obtained as follows:

[0091] (14)

[0092] Step 4: Utilize the bidirectional tension characteristics of the rotor to enable the control quantity after feedforward compensation to break through the traditional unidirectional tension constraint. Distribute the control quantity in the failure state to allow the remaining healthy rotor to form a new dynamic equilibrium state, thereby maintaining a low yaw rate and ensuring track tracking accuracy in the failure condition.

[0093] The specific implementation process of this method is as follows: the trajectory controller generates a virtual control quantity based on the desired trajectory, and combines it with the desired altitude value as the input of the fault-tolerant controller; the fault-tolerant controller calculates the nominal control quantity based on this input; equivalent disturbance modeling is performed for single rotor failure, and the disturbance is incorporated into the system lumped disturbance, and the disturbance is observed and compensated online through a disturbance observation and compensation mechanism; the final control quantity is obtained after feedforward compensation, and the desired tension value of each rotor is calculated according to the control allocation matrix; the rotor responds to the desired tension value according to its bidirectional tension characteristics, thereby realizing fault-tolerant control.

[0094] like Figure 3 As shown, this illustrates the signal propagation direction between modules, the hierarchy and connection relationships between controllers, and the data interaction between the GAZEBO simulation platform and the controllers and disturbance observers deployed in the PX4 firmware. Based on the PX4 open-source flight control firmware and the GAZEBO physics and dynamics engine software-in-the-loop simulation platform, the deployment of controllers and disturbance observers, as well as the modification of aircraft parameters, were completed. Specifically:

[0095] (1) The trajectory controller obtains a virtual trajectory based on the desired trajectory. The high expectation is used as the expected input of the fault-tolerant controller;

[0096] (2) Fault-tolerant controller based on With high expectations, to obtain nominal control. ;

[0097] (3) Model the failure of a single rotor, analyze the equivalent disturbance effect of the lack of tension on the control quantity, incorporate it into the system lumped disturbance, and establish a disturbance observation and compensation mechanism based on this to conduct online observation and compensation for system disturbances.

[0098] (4) Equivalent control quantity disturbance After feedforward compensation, the final control quantity of the system is obtained. and according to the control allocation matrix The relationship is used to obtain the desired tension control amount for each rotor;

[0099] (5) Since the rotor has bidirectional tension characteristics, it can respond to bidirectional desired tension control quantity, thereby realizing fault-tolerant control.

[0100] Regarding low yaw rates, the analysis is as follows:

[0101] The parallel configuration (PPC) design is key to achieving non-high-speed spin passive fault-tolerant control. Hovering flight provides a preliminary, intuitive explanation of this. Taking the complete failure of rotor 2 as an example, rotors 1 and 3 need to accelerate forward to double the thrust and maintain altitude. Meanwhile, rotor 4 essentially stops spinning, thus maintaining hover. Furthermore, because the blade structures of rotors 1 and 3 are opposite, the resulting counter-torque cancels each other out, preventing high-speed spin in the PPC after rotor failure.

[0102] The yaw channel torque model is extracted from the traditional crossover layout (APC) and the PPC, specifically as follows:

[0103] (15)

[0104] The direct cause of yaw spin is Without loss of generality, we will use the complete failure of rotor number 1 of the PPC as an example. The tensile force it can provide at this point... The corresponding control quantity for:

[0105] (16)

[0106] By performing an equivalent transformation on the above equation, the control requirements can be expressed as follows:

[0107] (17)

[0108] According to the fault tolerance mechanism analysis, if ,but The actual tension distributed to the rotor is:

[0109] (18)

[0110] in, Therefore, the yaw torque for both configurations under the condition of complete single rotor failure is:

[0111] (19)

[0112] Because after a single rotor completely fails, the yaw torque of the APC is directly coupled with the resultant pull force, and Therefore, APC will generate high-speed spin after a single rotor completely fails. In contrast, PPC structurally avoids the coupling relationship between a single rotor completely failing and the resultant force, thus enabling fault-tolerant control at a lower yaw rate.

[0113] To verify the effectiveness and superiority of the fault-tolerant control method for a single rotor of a quadcopter with low yaw rate proposed in this invention, the applicant conducted multiple sets of simulation experiments. The specific experimental conditions and results are as follows:

[0114] The parameters of the aircraft model used in the experiment, as well as the parameters of the controller and disturbance observer, are shown in Tables 1 and 2.

[0115] Table 1 Key parameters of the model

[0116]

[0117] Table 2 Parameter Table for Controller and Disturbance Observer

[0118]

[0119] 1. Verification of high-altitude tracking capability:

[0120] Figure 4 This is a schematic diagram illustrating the comparison results of the method described in this invention in altitude tracking. For example... Figure 4 As shown, the double-dash line and the dotted-dash line correspond to the height tracking response of two different configurations under fault-free operating conditions, respectively; the dark black line represents the height tracking response of the parallel layout under the failure condition of rotor No. 2. Figure 4 It can be seen that the control method described in this invention has accurate height tracking capability under normal operating conditions, and can still effectively track the height expectation after rotor failure. This shows that the method can maintain good height control performance under both normal and fault conditions, and verifies the robustness of the fault-tolerant control framework described in this invention on the vertical channel.

[0121] 2. Verification of attitude stabilization during hovering:

[0122] Figure 5The Euler angle response of the system described in this invention after a sudden rotor failure in a hovering state is shown below. Figure 5 As shown, under the proposed control method, the system can be quickly corrected after any single rotor failure and can regain stability within a finite time. This result demonstrates that the fault-tolerant control method for low yaw rate quadrotor single rotor failure provided by this invention does not rely on fault diagnosis and detection mechanisms. It can quickly suppress attitude disturbances through real-time compensation by a disturbance observer at the moment of rotor failure, enabling the aircraft to remain stable in hover.

[0123] 3. Verification of rotor control quantity allocation in fault-tolerant control:

[0124] Figure 6 This refers to the normalized control quantity of the rotor after a sudden single-rotor failure in a hovering state, as described in this invention. Figure 6 As shown in the figure, the expected changes in the control quantities of the remaining three healthy rotors after the failure of any single rotor are illustrated. Experimental results demonstrate that the control method proposed in this invention can adjust the output of the remaining rotors in real time through a control allocation strategy after the failure of any single rotor, thereby establishing a new equilibrium relationship in the system and achieving fault-tolerant control. Since the center of mass of the machine body is biased towards rotors 1 and 2, the expected control quantities of the diagonal rotors will differ slightly in fault-tolerant control. This reflects that the control allocation method described in this invention can adaptively adjust the tension output of each rotor according to the actual physical parameters to compensate for the effects of asymmetric mass distribution.

[0125] 4. Verification of yaw rate suppression effect:

[0126] Figure 7 This demonstrates the yaw rate of the method described in this invention after encountering a sudden single-rotor failure while hovering. For example... Figure 7 As shown, the control method described in this invention can effectively suppress the yaw rate of the airframe after any single rotor fails. In contrast to the degradation state of conventional quadrotors in the prior art, which exhibit continuous high-speed spinning after abandoning yaw channel control following single rotor failure, this invention, through the synergistic effect of parallel layout configuration and bidirectional thrust control distribution, significantly suppresses the yaw rate to a low level, thus verifying the significant advantages of this invention in ensuring flight safety.

[0127] 5. Performance verification of the disturbance observer estimation:

[0128] Figure 8 The observations demonstrate the equivalent control disturbance after a sudden single-rotor failure in a hovering state, as described in this invention. Figure 8As shown, the disturbance observer in the control method of this invention can estimate the equivalent disturbance caused by rotor failure in real time and accurately, and the rapid convergence characteristic of its estimate verifies the effectiveness of the observation. This observation result further illustrates that the dynamic feedforward compensation mechanism of this invention can respond promptly to sudden faults, providing a reliable basis for the accurate compensation of subsequent control quantities.

[0129] 6. Verification of 3D trajectory tracking performance:

[0130] Figure 9 The vertical distance error of the method described in this invention in three-dimensional trajectory tracking is demonstrated. For example... Figure 9 As shown, under the control method described in this invention, the system's trajectory tracking accuracy is not significantly affected by single rotor failure, and the overall trajectory does not deviate or diverge significantly. This result demonstrates that the method described in this invention not only performs well in hover point stabilization tasks but also possesses excellent fault tolerance in dynamic trajectory tracking tasks, ensuring mission continuity for the aircraft under failure conditions.

[0131] 7. Verification of rotor control quantities and yaw rate under multiple operating conditions:

[0132] Figures 10 to 12 This further demonstrates the comprehensive performance of the method described in this invention under arbitrary single rotor failure conditions. Wherein:

[0133] Figure 10 The invention demonstrates the rotor expected normalized control quantity under arbitrary single rotor failure conditions using the method described in this invention. It shows that under different failure modes, the control allocation strategy described in this invention can reallocate the tension command according to the real-time state of the current healthy rotor, and the output of each remaining rotor is reasonable and without saturation anomalies.

[0134] Figure 11 The invention demonstrates the yaw rate of the airframe in three-dimensional trajectory tracking under arbitrary single rotor failure conditions using the method described in this invention. This further verifies the invention's ability to consistently suppress yaw rate under different failure scenarios, and further proves the versatility and reliability of the method.

[0135] Figure 12 The invention demonstrates the observation of equivalent control disturbances under arbitrary single rotor failure conditions using the method described in this invention. The results show that the disturbance observer can achieve fast and accurate estimation of equivalent optimal motion under various single rotor failure modes, exhibiting good adaptability and robustness.

[0136] In summary, the fault-tolerant control method for low yaw rate quadrotor single rotor failure provided by this invention, through the collaborative design of parallel layout configuration and bidirectional thrust rotor, disturbance observation and feedforward dynamic compensation mechanism, and control allocation strategy that does not rely on fault diagnosis, can effectively suppress the yaw rate of the aircraft, maintain stability and track tracking coarseness under extreme conditions of complete failure of any single rotor. It has outstanding advantages such as high safety, strong robustness, low cost and short development cycle, and is suitable for fault-tolerant control engineering applications of various quadrotor UAVs.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A low yaw angle velocity quadrotor single rotor failure tolerant control method, characterized in that, Includes the following steps: Step 1: Analyze the problem of strong coupling between the yaw channel and the pitch and roll channels of traditional quadcopter UAVs after the complete failure of a single rotor, which easily leads to high-speed spin. By analyzing the system coupling mechanism, a parallel layout configuration is selected to reduce the coupling effect between channels. Step 2: Under the parallel layout configuration, establish a dynamic model under the condition of complete failure of a single rotor, model the loss of tension caused by rotor failure as equivalent to the system control quantity disturbance, and uniformly merge it into the lumped disturbance. Step 3: Construct a disturbance observation and compensation mechanism to observe and estimate the lumped disturbance online, and compensate the estimated value into the control quantity in a dynamic feedforward manner; Step 4: Utilize the bidirectional tension characteristics of the rotor to enable the control quantity after feedforward compensation to break through the traditional unidirectional tension constraint. Distribute the control quantity in the failure state to allow the remaining healthy rotor to form a new dynamic equilibrium state, thereby maintaining a low yaw rate and ensuring track tracking accuracy in the failure condition.

2. The low yaw rate four-rotor single-rotor failure tolerant control method according to claim 1, characterized in that, The specific implementation process of this method is as follows: the trajectory controller generates a virtual control quantity based on the desired trajectory, and combines it with the desired altitude value as the input of the fault-tolerant controller; the fault-tolerant controller calculates the nominal control quantity based on the input; equivalent disturbance modeling is performed for single rotor failure, the disturbance is incorporated into the system lumped disturbance, and the disturbance is observed and compensated online through the disturbance observation and compensation mechanism; The final control quantity is obtained after feedforward compensation, and the expected tension value of each rotor is calculated based on the control allocation matrix. The rotor responds to the desired tension value based on its bidirectional tension characteristics, thus achieving fault-tolerant control.

3. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 2, characterized in that, The fault-tolerant controller uses the estimated value from the disturbance observer for feedforward compensation to obtain the final control quantity after compensation, and calculates the expected tension control quantity of each rotor according to the control allocation matrix relationship; the feedforward compensation is to superimpose the equivalent disturbance estimate output by the disturbance observer back into the nominal control quantity in the form of negative feedback.

4. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, In step 1, the parallel layout configuration is as follows: the traditional cross layout is changed to a diagonal layout in opposite directions, with the four rotors arranged in pairs and in parallel, so that the coupling relationship between the yaw channel and the pitch and roll channels changes, thereby reducing unnecessary yaw moment after a single rotor fails. The rotors on the two arms in the same direction have opposite blade rotation directions, so that the counter-torques generated by the rotors on the same side cancel each other out.

5. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, In step 1, the bidirectional tension is: breaking through the traditional unidirectional positive tension constraint of the rotor, enabling the healthy rotor to have the ability to output positive and negative tension, expanding the control quantity after the fault can be collected, and providing a physical basis for torque redistribution.

6. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, In step 2, the loss of tension caused by rotor failure is modeled as an equivalent system control disturbance and uniformly merged into a lumped disturbance. Specifically, the impact of rotor failure on the system is modeled as an equivalent control disturbance, the fault part is separated, and a health matrix and a fault matrix are defined. The diagonal elements of the fault matrix represent the degree of failure of each rotor, and the health matrix is ​​the difference between the allocation matrix and the fault matrix. Even when rotor fault information is unknown, the system still allocates according to the original allocation matrix. Thus, the equivalent disturbance of rotor failure to the system can be expressed as the relationship between the health matrix and the actual allocation.

7. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, In step 3, the construction of the disturbance observation and compensation mechanism specifically involves: constructing a fault-tolerant mechanism that does not rely on fault diagnosis and isolation to avoid the reconstruction of the control allocation matrix; and using a disturbance observer to estimate and compensate for the equivalent control disturbance caused by rotor failure as an expanded state in real time.

8. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, In the parallel layout configuration, the yaw torque is not coupled with the resultant pull force after a single rotor fails, thereby avoiding high-speed spin after a single rotor fails.

9. The method for fault-tolerant control of a single rotor failure in a quadcopter with low yaw rate according to claim 1, characterized in that, The method is implemented on a software-in-the-loop simulation platform based on the PX4 open-source flight control firmware and the GAZEBO physics and dynamics engine, and completes the deployment of the controller, disturbance observer and modification of the aircraft parameters.

10. A fault-tolerant control method for a single rotor failure of a quadcopter with low yaw rate according to claim 9, characterized in that, The software-in-the-loop simulation platform is implemented as follows: by modifying the sdf file in PX4, the visualization effect of the parallel layout configuration in the GAZEBO physical simulation platform is realized, and the rotor tension coefficient and speed range are modified. By modifying the rotor drive source code in the PX4 firmware, the modification of the reverse torque direction of any rotor and the removal of the unidirectional tension constraint of any rotor are realized. A custom module is created using PX4 firmware to complete the deployment of the controller and disturbance observer. The system state, which is published to the uORB message pool after the PX4 firmware data is fused, is used as the input to the controller and disturbance observer. The system control quantity after compensation by the controller and disturbance observer is distributed to the four rotors through a fixed control allocation matrix and after normalization data processing, as the input of the rotor's expected tension control quantity. Finally, it is sent to the GAZEBO side via UDP packetization.

Images