Controller self-adaption method and device under complex working condition, storage medium and electronic equipment
By dynamically adjusting the model parameters and controllable gain of the fire-fighting UAV control system, the control stability problem under complex operating conditions is solved, and the drone is accurately controlled and safely flown under different disturbance environments are achieved.
Patent Information
- Application Number
- CN202510111584.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-23
- Publication Date
- 2025-05-30
AI Technical Summary
Under complex operating conditions, existing PID control algorithms cannot effectively ensure the safety and stability of fire-fighting drones, especially when facing different types of disturbances and control cycle changes.
The model parameters and controllable gain are dynamically adjusted based on the prediction error signal of the real controlled system and the controlled system model until the prediction error signal is less than the preset threshold, the target controller is generated, and the controller parameters are updated to control the target object.
It realizes stable control of fire-fighting drones under complex working conditions, overcomes different types of disturbances, meets the requirements of precise hovering and precise landing, and improves the reliability and safety of the drone platform.
Smart Images
Figure CN120065722A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of automation control technology, and in particular, to a controller adaptation method, device, storage medium, and electronic device under complex working conditions. Background Art
[0002] Large tethered fire-fighting drones need to operate between high-rise buildings and buildings, and the required control stability is self-evident. However, in the actual working environment, there are often various forms of interference sources for tethered drones, and such interference will pose a threat to the safety and stability of the drones.
[0003] Under complex working conditions, the proportional, integral, and derivative PID control algorithms in the related art cannot always keep the dynamic response characteristic indexes within the expected range when the drone faces different types of disturbances and different stages of the control period, and cannot ensure that the fire-fighting drone is always under safe and stable attitude control.
[0004] Based on this, there is an urgent need for a drone control method that can cope with complex working conditions to improve the safety and reliability of fire-fighting drones. Summary of the Invention
[0005] The purpose of this application is to provide a controller adaptation method, device, storage medium, and electronic device under complex working conditions, which can enable the fire-fighting drone to overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, so as to improve the reliability and safety of the drone platform and provide guarantee for safe flight operation.
[0006] This application provides a controller adaptation method under complex working conditions, including: Dynamically adjusting the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generating a target controller based on the model parameters of the controlled system model; replacing the controller parameters of the previous sampling period with the parameters of the target controller, and using the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
[0007] Optionally, dynamically adjusting the model parameters of the controlled system based on the prediction error signal between the real controlled system and the controlled system model includes: calculating a first output of the real controlled system at the next moment based on the input and output of the real controlled system at the current moment, and predicting a second output at the next moment based on the model parameters of the controlled system model at the current moment; calculating the difference between the first output and the second output to obtain the prediction error signal between the real controlled system and the controlled system model at the next moment.
[0008] Optionally, dynamically adjusting the model parameters of the controlled system based on the prediction error signal between the real controlled system and the controlled system model includes: if the prediction error signal at the next moment is less than the threshold range, then the model parameters of the controlled system model are not updated; otherwise, the model parameters of the controlled system model are updated.
[0009] Optionally, dynamically adjusting the controllable gain based on the prediction error signal between the real controlled system and the controlled system model includes: calculating the optimal adaptation rate of the controllable gain at the current moment based on the prediction error signals and the controllable gain at each moment, and using the gradient descent algorithm; updating the controllable gain at the current moment based on the optimal adaptation rate of the controllable gain at the current moment to obtain the updated controllable gain; wherein, the adaptation rate of the controllable gain is used to minimize the sum of the squares of the error accumulation of the prediction error signal.
[0010] Optionally, dynamically adjusting the model parameters of the controlled system based on the prediction error signal between the real controlled system and the controlled system model includes: calculating a measurement observation signal based on the input of the real controlled system and the output of the real controlled system, and calculating the adaptive gain matrix at the current moment based on the measurement observation signal; calculating the adaptive gain matrix at the next moment based on the matrix inversion lemma and the adaptive gain matrix at the current moment; updating the model parameters of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model at the next moment, the adaptive gain matrix at the next moment, the updated controllable gain, and the model parameters of the controlled system model at the current moment to obtain the updated model parameters.
[0011] This application also provides a controller adaptation device under complex working conditions, including: A parameter update module, which is used to dynamically adjust the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model, and generate a target controller based on the model parameters of the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold; An object control module, which is used to replace the controller parameters of the previous sampling period with the parameters of the target controller, and use the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
[0012] Optionally, the parameter update module is specifically configured to calculate a first output of the real controlled system at the next moment based on the input and output of the real controlled system at the current moment, and predict a second output at the next moment based on the model parameters of the controlled system model at the current moment; The parameter update module is specifically further configured to calculate the difference between the first output and the second output to obtain the prediction error signal between the real controlled system and the controlled system model at the next moment.
[0013] Optionally, the parameter update module is specifically configured to: if the prediction error signal at the next moment is less than the threshold range, do not update the model parameters of the controlled system model, otherwise, update the model parameters of the controlled system model.
[0014] Optionally, the parameter update module is specifically configured to calculate the optimal adaptation rate of the controllable gain at the current moment based on the prediction error signals and controllable gains at each moment, and use the gradient descent algorithm; the parameter update module is specifically further configured to update the controllable gain at the current moment based on the optimal adaptation rate of the controllable gain at the current moment to obtain the updated controllable gain; wherein, the adaptation rate of the controllable gain is used to: minimize the sum of the squares of the error accumulation of the prediction error signal.
[0015] Optionally, the parameter update module is specifically configured to calculate a measurement observation signal based on the input of the real controlled system and the output of the real controlled system, and calculate an adaptive gain matrix at the current moment based on the measurement observation signal; the parameter update module is further specifically configured to calculate the adaptive gain matrix at the next moment based on the matrix inversion lemma and the adaptive gain matrix at the current moment; the parameter update module is further specifically configured to update the model parameters of the controlled system model based on the prediction error signal of the real controlled system and the controlled system model at the next moment, the adaptive gain matrix at the next moment, the updated controllable gain, and the model parameters of the controlled system model at the current moment, so as to obtain the updated model parameters.
[0016] The present application also provides a computer program product, including a computer program / instructions, which when executed by a processor, implement the steps of the controller adaptation method under complex working conditions as described in any one of the above.
[0017] The present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps of the controller adaptation method under complex working conditions as described in any one of the above.
[0018] The present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the controller adaptation method under complex working conditions as described in any one of the above.
[0019] The controller adaptation method, device, storage medium, and electronic device provided by the present application dynamically adjust the model parameters and controllable gain of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generate a target controller based on the model parameters of the controlled system model; replace the controller parameters of the previous sampling period with the parameters of the target controller, and use the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and precise landing of the drone, improve the reliability and safety of the drone platform, and provide guarantee for safe flight operation. Description of the Drawings
[0020] To more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.
[0021] Figure 1 is one of the schematic flowcharts of the controller adaptation method under complex working conditions provided by the present application; Figure 2 is the second of the schematic flowcharts of the controller adaptation method under complex working conditions provided by the present application; Figure 3 is the schematic structural diagram of the controller adaptation device under complex working conditions provided by the present application; Figure 4 is the schematic structural diagram of the electronic device provided by the present application. Detailed implementation manners
[0022] To make the objectives, technical solutions, and advantages of the present application clearer, the following will clearly and completely describe the technical solutions in the present application in conjunction with the accompanying drawings in the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.
[0023] The terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application can be implemented in an order different from those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.
[0024] Large tethered fire-fighting drones need to operate between high-rise buildings. The control stability required is self-evident. However, in the actual working environment, there are often various forms of interference sources for tethered drones. During the takeoff and landing phases, different directions and intensities of turbulent flows will be generated between buildings at different heights; during the start-stop and working phases of the fire-fighting system, the sudden influx and stop of fire-fighting liquid in the vertical fire-fighting pipeline hundreds of meters long, and the disturbances brought about by sudden changes in the pressure, flow rate, mixing ratio, etc. of the fire-fighting liquid during the fire-fighting system spraying operation are all complex and strong interference sources. Especially when the drone is very close to the building or when the drone lands on the takeoff and landing platform, such interference will pose a threat to the safety and stability of the drone.
[0025] In such a complex environment, the dynamic response characteristic indexes obtained by the PID control algorithm in the related technology when the drone faces different types of disturbances and different stages of the control cycle cannot always be maintained within the expected range. Based on this, an adaptive method is needed that can identify the controlled system model and perform local optimization of the controller parameters based on the new model and the requirements of the system dynamic response indexes to meet the control requirements of different disturbances and each control stage.
[0026] In view of the above technical problems existing in the related technology, the embodiment of the present application provides a controller adaptive method under complex working conditions. This method can give an optimized controller adapted to the controlled system model, replace the original controller parameters with the optimized controller parameters, and have the control strategy of the control system for the control output under the new parameters. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, improve the reliability and safety of the drone platform, and provide guarantee for safe flight operations.
[0027] Exemplarily, as Figure 1 shown, the input of the mathematical model of the real controlled system is , and the output is . We need to find the parameters of this real system. We establish a model that can simulate its working principle. This model is an adaptive discrete model (i.e., the above-mentioned controlled system model), and the parameters of this model can be adjusted. We input the same input signal to this model and the real controlled system (i.e., the above-mentioned real controlled system). As Figure 1 shown, as a zero-order hold, Z.O.H. (Zero-Order Hold) filters out the high-frequency components in the input signal on the one hand, making the output signal smoother, and on the other hand, converts the discrete sampling signal into an approximate continuous signal, and works together with the controlled object Plant and the A / D converter to jointly construct the actual controlled system. represents the perturbation suffered by the real model (i.e., the above-mentioned random perturbation data). The real output of the system is measured by sensors. The model output represents an estimated state, and the difference between the real output and the model output constitutes the prediction error signal. This error signal is then input into the PAA (Parameter Adaptive Algorithm), and the PAA algorithm will, according to the magnitude and direction of the prediction error signal, on the one hand, dynamically adjust the parameters of the controlled system model to make it close to the dynamic characteristics of the real object, and on the other hand, adjust the controllable gain to compensate for the influence brought by the gain drift change of the controlled system model. The adjustment goal is to make the error between the output of the controlled system model and the output of the actual system as small as possible, and the compensated gain is consistent with the actual one, so as to achieve the consistency of the two.
[0028] Next, in combination with the accompanying drawings, the controller adaptation method under complex working conditions provided by the embodiments of the present application will be described in detail through specific embodiments and their application scenarios.
[0029] As Figure 2 shown, a controller adaptation method under complex working conditions provided by an embodiment of the present application is applied to a controller system. The controller system includes: a real controlled system and a controlled system model. The method may include the following steps 201 and 202: Step 201, dynamically adjust the model parameters and controllable gain of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model. After the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, generate a target controller based on the model parameters of the controlled system model.
[0030] Wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random perturbation data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
[0031] Exemplarily, when the output error between the controlled system model and the real controlled system reaches an acceptable threshold, we consider that the controlled system model can accurately represent the real controlled system, and its parameters match the parameters of the real controlled system. At this time, the obtained parameters are the required real system parameters. Using these adaptive parameters, the mathematical model of the real system can be established. Once the parameters of the current controlled system model are determined, we can immediately design a corresponding controller according to the controlled system model and update its parameters to replace the controller parameters of the previous sampling period, so as to provide the system with the latest control instructions.
[0032] Specifically, in the above step 201, the step of updating the model parameters of the controlled system model may further include the following steps 201a1 to 201a3: Step 201a1: Calculate the first output of the real controlled system at the next moment based on the input and output of the real controlled system at the current moment, and calculate the second output at the next moment predicted based on the model parameters of the controlled system model at the current moment.
[0033] Exemplarily, the above second output is: the prior prediction value at the next moment predicted based on the model parameters of the controlled system model at the current moment.
[0034] Step 201a2: Calculate the difference between the first output and the second output to obtain the prediction error signal of the real controlled system and the controlled system model at the next moment.
[0035] Step 201a3: If the prediction error signal at the next moment is less than the threshold range, do not update the model parameters of the controlled system model; otherwise, update the model parameters of the controlled system model.
[0036] Exemplarily, as Figure 1 shown, represents the output of the real controlled system, represents the output of the controlled system model, represents that the difference between the actual output and the estimated value is the prediction error signal, represents the prior error.
[0037] Exemplarily, based on the following formula (1), it can be known that when is less than the threshold range (i.e., the above threshold range), it is considered that the controlled system model is stable with the actual situation and no update is made; otherwise, update the model parameters of the controlled system model: (Formula (1)) Exemplarily, when the system is disturbed, the gain of the controlled object may change, and this change is generally unobservable. Therefore, we need to use the controllable gain to compensate for the influence brought by the drift of the actual object gain. We need to determine the adaptive rate of the controllable gain to minimize the sum of the squares of the error accumulation.
[0038] Specifically, in the above step 201, the step of updating the controllable gain may further include the following steps 201b1 and 201b2: Step 201b1: Based on the prediction error signals and the controllable gain at each moment, and using the gradient descent algorithm, calculate the optimal adaptive rate of the controllable gain at the current moment.
[0039] Step 201b2: Update the controllable gain at the current moment based on the optimal adaptive rate of the controllable gain at the current moment to obtain the updated controllable gain.
[0040] Among them, the adaptive rate of the controllable gain is used to: minimize the sum of squares of the error accumulation of the prediction error signal.
[0041] Exemplarily, the controllable gain The adaptive rate C can be expressed by the following formula (2): (Formula (2)) After that, use the gradient descent algorithm to obtain The optimal adaptive rate, which can be specifically expressed by the following formula (3): (Formula (3)) Among them, Is the step size of the actual application algorithm, which can be adjusted according to the actual application frequency. Here, 0.05 can be taken. The updated Can be expressed by the following formula (4): (Formula (4)) Specifically, in the above step 201, the step of updating the model parameters of the controlled system model may further include the following steps 201c1 to 201c3: Step 201c1: Calculate the measurement observation signal based on the input of the actual controlled system and the output of the actual controlled system, and calculate the adaptive gain matrix at the current moment based on the measurement observation signal.
[0042] Step 201c2: Calculate the adaptive gain matrix at the next moment based on the matrix inversion lemma and the adaptive gain matrix at the current moment.
[0043] Step 201c3: Update the model parameters of the controlled system model based on the prediction error signal at the next moment between the actual controlled system and the controlled system model, the adaptive gain matrix at the next moment, the updated controllable gain, and the model parameters of the controlled system model at the current moment to obtain the updated model parameters.
[0044] Exemplarily, use To represent the model parameter vector, Is the measurement observation signal, which is a column vector and includes the control quantity of the controlled system And the output of the controlled system . In the formula is the prior prediction error, whose value is equal to the difference between the actual value of the system at the next moment and the prior prediction value at the next moment predicted according to the estimated parameters at the current moment. The time-domain discrete model of the controlled system is set as . Among them, the prior prediction data is . The prior prediction error can be obtained as , which can be specifically expressed by the following Formula Five, Formula Six and Formula Seven: (Formula Five) (Formula Six) (Formula Seven) Among them, is the adaptive gain matrix, which is related to the measurement vector each time and can be expressed as: According to the matrix inversion lemma of linear algebra: , the adaptive gain at the next moment is obtained. It can be expressed by the following Formula Eight: (Formula Eight) Step 202: Replace the controller parameters of the previous sampling period with the parameters of the target controller, and use the controller with updated parameters to control the target object.
[0045] Exemplarily, using the above iterative vector and controllable gain , the latest parameter estimation model can be obtained, which can be specifically expressed by the following Formula Nine: (Formula Nine) In the above formula is the column vector of the parameters of the controlled system to be estimated , and the elements in the vector are the numerator and denominator coefficients of the controlled system transfer function converted into a discrete transfer function. is the newly estimated parameter at time t + 1.
[0046] It should be noted that the advantage of the adaptive algorithm in this embodiment is that it only depends on the data at the current moment to predict the quantity at the next moment, without looking back at the data of all past moments. In addition, the unobservable gain of the actual controlled object can be compensated according to the controllable gain. The design of this algorithm ensures that the denominator is always non-zero, which means that on the premise that the parameters themselves are stable, the iterative process will not get out of control or diverge. The entire system follows the above algorithm for iterative update to obtain the prediction parameters of the controlled system in real time.
[0047] The controller adaptation method under complex working conditions provided by the embodiments of the present application dynamically adjusts the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generates a target controller based on the model parameters of the controlled system model; replaces the controller parameters of the previous sampling period with the parameters of the target controller, and uses the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, improve the reliability and safety of the drone platform, and provide guarantee for safe flight operation.
[0048] It should be noted that for the controller adaptation method under complex working conditions provided by the embodiments of the present application, the execution subject can be a controller adaptation device under complex working conditions, or a control module in the controller adaptation device under complex working conditions for executing the controller adaptation method under complex working conditions. In the embodiments of the present application, taking the controller adaptation device under complex working conditions executing the controller adaptation method under complex working conditions as an example, the controller adaptation device under complex working conditions provided by the embodiments of the present application is described.
[0049] It should be noted that in the embodiments of the present application, the controller adaptation method under complex working conditions shown in the above-mentioned various method drawings is exemplarily described by taking one drawing in the embodiments of the present application as an example. Specifically, when implemented, the controller adaptation method under complex working conditions shown in the above-mentioned various method drawings can also be implemented in combination with any other combinable drawings schemed in the above embodiments, which will not be elaborated here.
[0050] The controller adaptation device under complex working conditions provided by the present application is described below, and the following description can be correspondingly referred to the controller adaptation method under complex working conditions described above.
[0051] Figure 3 is a schematic structural diagram of the controller adaptation device under complex working conditions provided by the embodiments of the present application, as Figure 3 shown, and specifically includes: A parameter update module 301, configured to dynamically adjust the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model, and generate a target controller based on the model parameters of the controlled system model after the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold; An object control module 302, configured to replace the controller parameters of the previous sampling period with the parameters of the target controller, and control the target object by using the controller with updated parameters; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random perturbation data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
[0052] Optionally, the parameter update module 301 is specifically configured to calculate a first output of the real controlled system at the next moment based on the input and output of the real controlled system at the current moment, and predict a second output at the next moment based on the model parameters of the controlled system model at the current moment; The parameter update module 301 is further specifically configured to calculate a difference between the first output and the second output to obtain a prediction error signal between the real controlled system and the controlled system model at the next moment.
[0053] Optionally, the parameter update module 301 is specifically configured to: if the prediction error signal at the next moment is less than a threshold range, do not update the model parameters of the controlled system model, otherwise, update the model parameters of the controlled system model.
[0054] Optionally, the parameter update module 301 is specifically configured to calculate an optimal adaptation rate of the controllable gain at the current moment based on the prediction error signals and controllable gains at each moment, and use the gradient descent algorithm; the parameter update module 301 is further specifically configured to update the controllable gain at the current moment based on the optimal adaptation rate of the controllable gain at the current moment to obtain an updated controllable gain; wherein, the adaptation rate of the controllable gain is used to minimize the sum of the squares of the error accumulations of the prediction error signals.
[0055] Optionally, the parameter update module 301 is specifically configured to calculate a measurement observation signal based on the input of the real controlled system and the output of the real controlled system, and calculate an adaptive gain matrix at the current moment based on the measurement observation signal; the parameter update module 301 is further specifically configured to calculate an adaptive gain matrix at the next moment based on the matrix inversion lemma and the adaptive gain matrix at the current moment; the parameter update module 301 is further specifically configured to update the model parameters of the controlled system model based on the prediction error signal of the real controlled system and the controlled system model at the next moment, the adaptive gain matrix at the next moment, the updated controllable gain, and the model parameters of the controlled system model at the current moment, so as to obtain updated model parameters.
[0056] The controller adaptive device under complex working conditions provided by this application dynamically adjusts the model parameters and controllable gain of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model. After the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, a target controller is generated based on the model parameters of the controlled system model; the controller parameters of the previous sampling period are replaced with the parameters of the target controller, and the target object is controlled by using the controller with updated parameters; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, improve the reliability and safety of the drone platform, and provide guarantee for safe flight operation.
[0057] Figure 4 An entity structure diagram of an electronic device is exemplified, such as Figure 4As shown in the figure, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 complete communication with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the controller adaptation method under complex working conditions. The method includes: dynamically adjusting the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generating a target controller based on the model parameters of the controlled system model; replacing the controller parameters of the previous sampling period with the parameters of the target controller, and using the controller with updated parameters to control the target object; where the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random perturbation data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of perturbations, meet the requirements of precise hovering operation and accurate landing of the drone, improve the reliability and safety of the drone platform, and provide guarantee for safe flight operations.
[0058] In addition, when the logic instructions in the above-mentioned memory 430 are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.
[0059] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the controller adaptation method under complex working conditions provided by the above-mentioned various methods. The method includes: dynamically adjusting the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generating a target controller based on the model parameters of the controlled system model; replacing the controller parameters of the previous sampling period with the parameters of the target controller, and using the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, so as to improve the reliability and safety of the drone platform and provide guarantee for safe flight operation.
[0060] On another aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it realizes the controller adaptation method under complex working conditions provided by the above-mentioned various methods. The method includes: dynamically adjusting the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generating a target controller based on the model parameters of the controlled system model; replacing the controller parameters of the previous sampling period with the parameters of the target controller, and using the controller with updated parameters to control the target object; wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system further includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model. In this way, the fire-fighting drone can overcome different types of disturbances, meet the requirements of precise hovering operation and accurate landing of the drone, so as to improve the reliability and safety of the drone platform and provide guarantee for safe flight operation.
[0061] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative effort.
[0062] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A controller adaptive method under complex working conditions, characterized in that: Applied to a controller system, the controller system comprising: a real controlled system and a controlled system model; The method comprises: Dynamically adjusting the model parameters and controllable gain of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and generating a target controller based on the model parameters of the controlled system model; Replacing the controller parameters of the previous sampling period with the parameters of the target controller, and controlling the target object using the controller with updated parameters; Wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system also includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
2. The method according to claim 1, characterized in that The dynamically adjusting the model parameters of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model comprises: Calculating a first output of the real controlled system at a next moment based on the input and output of the real controlled system at a current moment, and predicting a second output at a next moment based on model parameters of the controlled system model at the current moment; The difference between the first output and the second output is calculated to obtain a prediction error signal between the real controlled system and the controlled system model at the next moment.
3. The method according to claim 2, characterized in that The dynamically adjusting the model parameters of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model comprises: If the prediction error signal at the next moment is smaller than the threshold range, the model parameters of the controlled system model are not updated; otherwise, the model parameters of the controlled system model are updated.
4. The method according to claim 1, characterized in that: The dynamically adjusting the controllable gain based on the prediction error signal between the real controlled system and the controlled system model comprises: Based on the prediction error signal and controllable gain at each moment, the optimal adaptive rate of the controllable gain at the current moment is calculated using the gradient descent algorithm; The controllable gain at the current moment is updated based on the optimal adaptive rate of the controllable gain at the current moment to obtain an updated controllable gain; The adaptive rate of the controllable gain is used to minimize the square sum of the accumulated errors of the prediction error signal.
5. The method according to any one of claims 2 to 4, characterized in that The dynamically adjusting the model parameters of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model comprises: Based on the input of the real controlled system and the output of the real controlled system, a measurement observation signal is calculated, and based on the measurement observation signal, an adaptive gain matrix at a current moment is calculated; Based on the matrix inversion lemma and the adaptive gain matrix at the current moment, the adaptive gain matrix at the next moment is calculated; Based on the predicted error signal between the real controlled system and the controlled system model at the next moment, the adaptive gain matrix at the next moment, the updated controllable gain and the model parameters of the controlled system model at the current moment, the model parameters of the controlled system model are updated to obtain updated model parameters.
6. A controller adaptive device under complex working conditions, characterized in that: Applied to a controller system, the controller system comprising: a real controlled system and a controlled system model; The device comprises: A parameter updating module, used for dynamically adjusting the model parameters and controllable gains of the controlled system model based on the prediction error signal between the real controlled system and the controlled system model, until the prediction error signal between the real controlled system and the controlled system model is less than a preset error threshold, and then generating a target controller based on the model parameters of the controlled system model; An object control module, used to replace the controller parameters of the previous sampling period with the parameters of the target controller, and control the target object using the controller with updated parameters; Wherein, the real controlled system and the controlled system model have the same input, and the input of the real controlled system also includes random disturbance data; the controllable gain is used to adjust the output of the controlled system model to reduce the output error between the real controlled system and the controlled system model.
7. The device according to claim 6, characterized in that The parameter updating module is specifically used to calculate the first output of the real controlled system at the next moment based on the input and output of the real controlled system at the current moment, and to predict the second output at the next moment based on the model parameters of the controlled system model at the current moment; The parameter updating module is specifically used to calculate the difference between the first output and the second output to obtain a prediction error signal between the real controlled system and the controlled system model at the next moment.
8. The device according to claim 7, characterized in that The parameter updating module is specifically used for not updating the model parameters of the controlled system model if the prediction error signal at the next moment is smaller than a threshold range, and otherwise, updating the model parameters of the controlled system model.
9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the controller self-adaptation method under complex working conditions as claimed in any one of claims 1 to 5 are implemented.
10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the controller self-adaptation method under complex working conditions as described in any one of claims 1 to 5 are implemented.