Universal land-air amphibious robot dual-mode active-disturbance-rejection control method

Abstract

The invention discloses a universal dual-mode active-disturbance-rejection control method for an air-ground amphibious robot. The method comprises the steps that kinematics modeling is conducted on a flight end and a ground end according to the structure of the air-ground amphibious robot; carrying out ground-air double-end common disturbance pre-modeling; constructing an extended state observer; detecting a current motion mode; weighting a nonlinear least square model, compensating disturbance information, and estimating total disturbance in real time; and a bimodal active-disturbance-rejection controller is designed, disturbance is compensated in real time, and bimodal active-disturbance-rejection control over the air-ground amphibious robot is achieved. According to the method, information of all motion mode models and common disturbance of the air-ground amphibious robot are preprocessed, the improved active-disturbance-rejection control method is applied under dual modes, the adaptive capacity of the air-ground amphibious robot to external disturbance and complex environments during reconnaissance work is greatly improved, and the method has universality in control over the air-ground amphibious robots of various mechanical structures.

Description

Technical Field

[0001] The present invention belongs to the technical field of amphibious robot control, and in particular to a universal dual-mode auto-disturbance rejection control method for land and air amphibious robots. Background Art

[0002] Currently, reconnaissance missions are characterized by highly variable and complex environments. Reconnaissance robots are gradually replacing manual reconnaissance as a primary means of battlefield reconnaissance and post-disaster rescue. To adapt to complex environments and flexibly respond to various emergencies, amphibious reconnaissance robots capable of operating in a variety of environments are becoming a research hotspot. While existing amphibious reconnaissance robots possess certain bimodal motion capabilities, they often suffer from poor terrain adaptability and insufficient control precision, hindering the smooth progress of reconnaissance operations.

[0003] Current amphibious robots have diverse mechanical structures, and the flight and ground systems often use different algorithms for control. These control algorithms have inconsistent performance, lack consistency, and lack multi-structure, dual-modal adaptability. There is an urgent need for a control algorithm for amphibious robots that can adapt to various external disturbances and complex environments. Summary of the Invention

[0004] The purpose of the present invention is to provide a control algorithm for an amphibious robot that can adapt to various external disturbances and complex environments in order to address the existing technical problems of existing amphibious robots.

[0005] The technical solution to achieve the purpose of the present invention is: on the one hand, a universal dual-modal active disturbance rejection control method for an amphibious robot is provided, the method comprising:

[0006] Step 1: Conduct dual-modal modeling of the land and air amphibious robot;

[0007] Step 2: Pre-modeling of external disturbances at both land and air terminals;

[0008] Step 3: Construct an extended state observer;

[0009] Step 4: Real-time detection of motion patterns;

[0010] Step 5: Based on the outputs of Step 1, Step 2, and Step 4, disturbance information compensation is performed, and then a disturbance-compensated extended state observer is obtained based on the disturbance information compensation;

[0011] Step 6: Based on the observation results of the extended state observer after disturbance compensation, a dual-mode active disturbance rejection controller is designed to realize the dual-mode active disturbance rejection control of the land and air amphibious robot.

[0012] Furthermore, the dual-modal modeling of the land-air amphibious robot described in step 1 specifically includes:

[0013] Step 1-1: Based on the mechanical structure of the amphibious robot, perform force analysis and kinematic analysis on each motion unit to obtain dynamic parameters and kinematic parameters;

[0014] Step 1-2: perform dual-mode modeling based on dynamic parameters and kinematic parameters, wherein the dual-mode includes a flight mode and a ground mode.

[0015] Furthermore, the pre-modeling of the external disturbance at both land and air terminals described in step 2 specifically includes:

[0016] Step 2-1: Pre-modeling of flight-end disturbances:

[0017]

[0018] Where μ is the gradient coefficient, V is the gust wind speed, V max is the maximum gust wind speed, G is the gust transfer function, t0 is the start time, t is the end time, and s represents the complex frequency variable in the transfer function;

[0019] Step 2-2: Pre-modeling of ground disturbance:

[0020]

[0021] Where, F L 、F R Represents the resistance on the left and right sides of the road, f L 、f R are the resistance coefficients on the left and right sides of the road respectively, and m is the weight of the air-ground amphibious robot.

[0022] Furthermore, step 3 of constructing the extended state observer specifically includes:

[0023] Step 3-1, establish the Lumberg observer equation:

[0024]

[0025] in, represents the total disturbance differential term, f is the total disturbance, A, B, C, L, E are the coefficient matrices of the observer equation, is the state of the observation system, is the first-order differential of the observation system state, u is the observation system input, and y is the observation system output;

[0026] Design the extended state observer:

[0027]

[0028] Step 3-2, determine the parameters of the extended state observer:

[0029] Ensure that the extended state observer is fully converged, make the (A-LC) eigenvalue located in the negative half plane, introduce the observer bandwidth ω0>0, configure the characteristic equation poles to be located on -ω0, and obtain the characteristic equation λ(s):

[0030] λ(s)=|sI-(A-LC)|=(s+ω o ) 3 =0

[0031] The extended state observer gain matrix is obtained from this

[0032] Furthermore, the real-time detection of the motion pattern in step 4 specifically includes:

[0033] Step 4-1, first level of detection

[0034] (1) Detecting the air pressure at the current location of the amphibious reconnaissance robot. If the difference between the air pressure and the ground pressure is greater than a preset threshold, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1. Otherwise, execute (2).

[0035] (2) obtaining the vertical acceleration of the current amphibious reconnaissance robot, and determining whether the difference between the vertical acceleration and the negative value of the gravitational acceleration is within a preset deviation range; if so, determining that the current motion mode is the ground mode, and setting the current mode check bit α to 0; otherwise, determining that the current motion mode is the flight mode, and setting the current mode check bit α to 1;

[0036] Step 4-2, second test

[0037] Detect the current dominant control signal. If the ground motion control unit continuously outputs a signal, the current motion mode is determined to be the ground mode, and the current mode check bit α is set to 0. If the flight motion control unit continuously outputs a signal, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1.

[0038] Step 4-3, determine whether the detection results of step 4-1 and step 4-2 are consistent. If they are consistent, output the motion mode detection result and the mode check bit. Otherwise, return to execute step 4-1 and repeat until the preset detection number limit is reached.

[0039] Furthermore, the specific process of step 5 includes:

[0040] Step 5-1: Robot motion sample data collection

[0041] Based on the kinematic modeling and disturbance pre-modeling in steps 1 and 2, various common external disturbances are applied one by one in the absence of external disturbances, ground mode, and flight mode, and the robot motion deviation data under the corresponding disturbances is collected;

[0042] Step 5-2: Establish a total disturbance model for estimating the motion state of the amphibious robot due to various disturbances and model uncertainties:

[0043] y i =f(x i ,β)

[0044] Where x i represents the i-th external disturbance or internal uncertainty, y i Represents x i The impact on the robot motion parameters; the internal uncertainty represents the uncertain parameters in the kinematic modeling; β represents the model parameters;

[0045] Step 5-2, weighted nonlinear least squares fitting:

[0046] According to the robot motion sample data collected in step 5-1 and the estimated total disturbance model, the influence of each disturbance and model uncertainty on the robot motion stability is evaluated, and the corresponding weights are determined to obtain the weighted matrix ω n is the weight of the nth external disturbance or internal uncertainty, and n is the total number of external disturbances or internal uncertainties;

[0047] Based on the weighted matrix, the total disturbance model is obtained by fitting using the weighted nonlinear least squares method:

[0048]

[0049] Among them, β represents the model parameters, S(β) is the total disturbance fitting, ω i is the weight of the i-th external disturbance or internal uncertainty;

[0050] The Gauss-Newton method is used to iteratively obtain the optimal model parameter β;

[0051] Step 5-3, estimate the total disturbance compensation:

[0052] Based on the motion mode detection results of step 4, the disturbances in non-current motion modes are eliminated, the corresponding weights are set to 0, and the non-influence weight elimination matrix M(α) is obtained, and the compensation gain matrix N is obtained:

[0053] N=αS(β)M(α)

[0054] Wherein, α is the mode check bit corresponding to the motion mode detection result. If the motion mode is ground mode, the mode check bit α=0; if the motion mode is flight mode, the mode check bit α=1;

[0055] Based on this, the improved state observer after disturbance compensation is obtained:

[0056]

[0057] Furthermore, in step 6, the dual-mode active disturbance rejection controller compensates for disturbances in real time and performs rapid error response to solve the algorithm control variable based on the real-time estimated value of the extended state observer after disturbance compensation. Specifically, the steps include:

[0058] Step 6-1, establish the state error feedback control law:

[0059] Design the controller:

[0060]

[0061] Among them, u0 is the control quantity when disturbance compensation is not considered, represents the total disturbance of the robot estimated in real time by the extended state observer after disturbance compensation, b is the control input gain, and u is the control quantity output by the controller; when the extended state observer is estimated accurately, that is, When the original amphibious robot motion control system is changed to a double integrator series type: f is the total disturbance;

[0062] The PD control law is used as the state error feedback control law:

[0063]

[0064] Where k p , k d is the controller parameter, is the state estimate of the amphibious robot system observed by the extended state observer after disturbance compensation, is the estimated value of the state differential of the amphibious robot system observed by the extended state observer after disturbance compensation, U is the control input, is the derivative of the control input;

[0065] Considering that the extended state observer can observe the actual state, the state estimation value of the land-air amphibious robot system is similar to the real state quantity of the system, and the state differential estimation value of the land-air amphibious robot system is similar to the differential of the real state quantity of the system, that is,

[0066] The motion control system of the amphibious robot is designed as a critically damped second-order system to ensure that the system has no overshoot and has fast response capability:

[0067]

[0068] Where, ω c >0 indicates the controller bandwidth, which is used to adjust the response speed and stability of the system;

[0069] Step 6-2, solve the control quantity:

[0070] Calculated according to the state error feedback control law:

[0071]

[0072] Then the control quantity u is:

[0073]

[0074] Where U represents the system control input, is the differential of the control input, v1 represents the input of the state error feedback control law, and v2 represents the differential of the input of the state error feedback control law;

[0075] The accuracy is determined by the observer bandwidth ω0, and the control gain is determined by the controller bandwidth ω c Decide;

[0076] By comparing ω0 and ω c Parameter tuning is performed to make the error between the extended state observer and the controller converge to 0 after disturbance compensation.

[0077] On the other hand, a universal dual-mode active disturbance rejection control system for an amphibious reconnaissance robot is provided, the system comprising:

[0078] The first module is used for dual-modal modeling of land and air amphibious robots;

[0079] The second module is used to pre-model external disturbances at both land and air ends;

[0080] The third module is used to construct an extended state observer;

[0081] The fourth module is used to detect motion patterns in real time;

[0082] a fifth module, configured to perform disturbance information compensation based on outputs of the first module, the second module, and the fourth module, and then obtain a disturbance-compensated extended state observer based on the disturbance information compensation;

[0083] The sixth module is used to design a dual-modal active disturbance rejection controller based on the observation results of the extended state observer after disturbance compensation, so as to realize the dual-modal active disturbance rejection control of the land and air amphibious robot.

[0084] On the other hand, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the general dual-modal self-anti-disturbance control method for the land-air amphibious robot is implemented.

[0085] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the universal dual-modal self-disturbance rejection control method of the land-air amphibious robot is implemented.

[0086] Compared with the prior art, the present invention has the following significant advantages:

[0087] (1) For amphibious robots with different mechanical structures, mathematical models are pre-established and a unified algorithm control is adopted, which has universality in the dual-mode control of amphibious robots.

[0088] (2) The common disturbances that the amphibious robot is susceptible to in the two motion modes of the land and air are considered in advance, and pre-modeling is performed to provide prior knowledge for disturbance control.

[0089] (3) The motion state of the amphibious robot is detected in real time, and the influence of model uncertainty and external disturbances on state observation in other modes is eliminated according to the current state, thereby effectively improving the accuracy of the anti-disturbance control.

[0090] (4) Pre-collect disturbance and model uncertainty impact data, assign weights according to the degree of influence through the nonlinear least squares method, fit it into a pre-estimated total disturbance model for compensation, obtain an improved extended state observer, and conduct targeted observations of the system's internal state and external disturbances.

[0091] (5) Based on the observation data of the improved extended state observer, a dual-mode active disturbance rejection controller is designed to simplify the design of the feedback control law. The system control has no overshoot and is fast enough.

[0092] (6) The robot's dual-mode control stability and control accuracy are considered in a coordinated manner, eliminating the tedious single-mode independent control process and significantly improving the consistency of the control of the land and air amphibious robot.

[0093] The present invention is further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 The figure is a flow chart of a dual-modal anti-disturbance control method for a land-air amphibious robot in one embodiment.

[0095] Figure 2 The figure is an overall structural diagram of the land-air amphibious reconnaissance robot in one embodiment.

[0096] Figure 3The figure is a structural diagram of a four-rotor flight frame of an amphibious reconnaissance robot in one embodiment.

[0097] Figure 4 The figure is a structural diagram of a tracked chassis of an amphibious reconnaissance robot in one embodiment.

[0098] Reference numerals: quadrotor flight frame 1; crawler chassis 2; main controller 3; flight control cabin 4; image transmission device 5; chassis motion control box 6; triangular track wheel 7; flight duct 11; flight motor 12; three-blade propeller 13; electronic speed controller 14; environmental perception module 15; flight shock absorber plate 16; camera adjustment screw 17; image transmission camera 18; tripod base 21; tensioning mechanism 22; rubber crawler 23; driving wheel 24; tensioning wheel 25; supporting wheel 26; DETAILED DESCRIPTION

[0099] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0100] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0101] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features specified as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but this must be based on the fact that ordinary technicians in this field can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0102] In one embodiment, combined Figure 1 , provides a general dual-mode active disturbance rejection control method for an amphibious robot, the method comprising:

[0103] Step 1: Conduct dual-modal modeling of the land and air amphibious robot;

[0104] Step 2: Pre-modeling of external disturbances at both land and air terminals;

[0105] Step 3: Construct an extended state observer;

[0106] Step 4: Real-time detection of motion patterns;

[0107] Step 5: Based on the outputs of Step 1, Step 2, and Step 4, disturbance information compensation is performed, and then a disturbance-compensated extended state observer is obtained based on the disturbance information compensation;

[0108] Step 6: Based on the observation results of the extended state observer after disturbance compensation, a dual-mode active disturbance rejection controller is designed to realize the dual-mode active disturbance rejection control of the land and air amphibious robot.

[0109] Furthermore, in one embodiment, the dual-modal modeling of the land-air amphibious robot described in step 1 specifically includes:

[0110] Step 1-1: Based on the mechanical structure of the amphibious robot, perform force analysis and kinematic analysis on each motion unit to obtain dynamic parameters and kinematic parameters;

[0111] Step 1-2: perform dual-mode modeling based on dynamic parameters and kinematic parameters, wherein the dual-mode includes a flight mode and a ground mode.

[0112] Furthermore, in one embodiment, the pre-modeling of the external disturbance at both land and air terminals described in step 2 specifically includes:

[0113] Step 2-1: Pre-modeling of flight-end disturbances:

[0114]

[0115] Where μ is the gradient coefficient, V is the gust wind speed, V max is the maximum gust wind speed, G is the gust transfer function, t0 is the start time, t is the end time, and s represents the complex frequency variable in the transfer function;

[0116] Step 2-2: Pre-modeling of ground disturbance:

[0117]

[0118] Where, F L 、F R Represents the resistance on the left and right sides of the road, f L 、f R are the resistance coefficients on the left and right sides of the road respectively, and m is the weight of the air-ground amphibious robot.

[0119] Furthermore, in one embodiment, the step 3 of constructing the extended state observer specifically includes:

[0120] Step 3-1, establish the Lumberg observer equation:

[0121]

[0122] in, represents the total disturbance differential term, f is the total disturbance, A, B, C, L, E are the coefficient matrices of the observer equation, is the state of the observation system, is the first-order differential of the observation system state, u is the observation system input, and y is the observation system output;

[0123] Design the extended state observer:

[0124]

[0125] Step 3-2, determine the parameters of the extended state observer:

[0126] Ensure that the extended state observer is fully converged, make the (A-LC) eigenvalue located in the negative half plane, introduce the observer bandwidth ω0>0, configure the characteristic equation poles to be located on -ω0, and obtain the characteristic equation λ(s):

[0127] λ(s)=|sI-(A-LC)|=(s+ω o ) 3 =0

[0128] The extended state observer gain matrix is obtained from this

[0129] Furthermore, in one embodiment, the real-time detection of the motion pattern in step 4 specifically includes:

[0130] Step 4-1, first level of detection

[0131] (1) Detecting the air pressure at the current location of the amphibious reconnaissance robot. If the difference between the air pressure and the ground pressure is greater than a preset threshold, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1. Otherwise, execute (2).

[0132] (2) obtaining the vertical acceleration of the current amphibious reconnaissance robot, and determining whether the difference between the vertical acceleration and the negative value of the gravitational acceleration is within a preset deviation range; if so, determining that the current motion mode is the ground mode, and setting the current mode check bit α to 0; otherwise, determining that the current motion mode is the flight mode, and setting the current mode check bit α to 1;

[0133] Step 4-2, second test

[0134] Detect the current dominant control signal. If the ground motion control unit continuously outputs a signal, the current motion mode is determined to be the ground mode, and the current mode check bit α is set to 0. If the flight motion control unit continuously outputs a signal, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1.

[0135] Step 4-3, determine whether the detection results of step 4-1 and step 4-2 are consistent. If they are consistent, output the motion mode detection result and the mode check bit. Otherwise, return to execute step 4-1 and repeat until the preset detection number limit is reached.

[0136] Furthermore, in one embodiment, the specific process of step 5 includes:

[0137] Step 5-1: Robot motion sample data collection

[0138] Based on the kinematic modeling and disturbance pre-modeling in steps 1 and 2, various common external disturbances are applied one by one in the absence of external disturbances, ground mode, and flight mode, and the robot motion deviation data under the corresponding disturbances is collected;

[0139] Step 5-2: Establish a total disturbance model for estimating the motion state of the amphibious robot due to various disturbances and model uncertainties:

[0140] y i =F(x i ,β)

[0141] Where x i represents the i-th external disturbance or internal uncertainty, y i Represents x i The impact on the robot motion parameters; the internal uncertainty represents the uncertain parameters in the kinematic modeling; β represents the model parameters;

[0142] Step 5-2, weighted nonlinear least squares fitting:

[0143] According to the robot motion sample data collected in step 5-1 and the estimated total disturbance model, the influence of each disturbance and model uncertainty on the robot motion stability is evaluated, and the corresponding weights are determined to obtain the weighted matrix ω n is the weight of the nth external disturbance or internal uncertainty, and n is the total number of external disturbances or internal uncertainties;

[0144] Based on the weighted matrix, the total disturbance model is obtained by fitting using the weighted nonlinear least squares method:

[0145]

[0146] Among them, β represents the model parameters, S(β) is the total disturbance fitting, ω i is the weight of the i-th external disturbance or internal uncertainty;

[0147] The Gauss-Newton method is used to iteratively obtain the optimal model parameter β;

[0148] Step 5-3, estimate the total disturbance compensation:

[0149] Based on the motion mode detection results of step 4, the disturbances in non-current motion modes are eliminated, the corresponding weights are set to 0, and the non-influence weight elimination matrix M(α) is obtained, and the compensation gain matrix N is obtained:

[0150] N=αS(β)M(α)

[0151] Wherein, α is the mode check bit corresponding to the motion mode detection result. If the motion mode is ground mode, the mode check bit α=0; if the motion mode is flight mode, the mode check bit α=1;

[0152] Based on this, the improved state observer after disturbance compensation is obtained:

[0153]

[0154] Furthermore, in one embodiment, in step 6, the dual-mode active disturbance rejection controller compensates for disturbances in real time and performs rapid error response to solve the algorithm control variable based on the real-time estimated value of the extended state observer after disturbance compensation. Specifically, the steps include:

[0155] Step 6-1, establish the state error feedback control law:

[0156] Design the controller:

[0157]

[0158] Among them, u0 is the control quantity when disturbance compensation is not considered, represents the total disturbance of the robot estimated in real time by the extended state observer after disturbance compensation, b is the control input gain, and u is the control quantity output by the controller; when the extended state observer is estimated accurately, that is, When the original amphibious robot motion control system is changed to a double integrator series type: f is the total disturbance;

[0159] The PD control law is used as the state error feedback control law:

[0160]

[0161] Where k p , k d is the controller parameter, is the state estimate of the amphibious robot system observed by the extended state observer after disturbance compensation, is the estimated value of the state differential of the amphibious robot system observed by the extended state observer after disturbance compensation, U is the control input, is the derivative of the control input;

[0162] Considering that the extended state observer can observe the actual state, the state estimation value of the land-air amphibious robot system is similar to the real state quantity of the system, and the state differential estimation value of the land-air amphibious robot system is similar to the differential of the real state quantity of the system, that is,

[0163] The motion control system of the amphibious robot is designed as a critically damped second-order system to ensure that the system has no overshoot and has fast response capability:

[0164]

[0165] Where, ω c >0 indicates the controller bandwidth, which is used to adjust the response speed and stability of the system;

[0166] Step 6-2, solve the control quantity:

[0167] Calculated according to the state error feedback control law:

[0168]

[0169] Then the control quantity u is:

[0170]

[0171] Where U represents the system control input, is the differential of the control input, v1 represents the input of the state error feedback control law, and v2 represents the differential of the input of the state error feedback control law;

[0172] The accuracy is determined by the observer bandwidth ω0, and the control gain is determined by the controller bandwidth ω c Decide;

[0173] By comparing ω0 and ω c Parameter tuning is performed to make the error between the extended state observer and the controller converge to 0 after disturbance compensation.

[0174] In one embodiment, a universal dual-mode active disturbance rejection control system for an amphibious robot is provided, the system comprising:

[0175] The first module is used for dual-modal modeling of land and air amphibious robots;

[0176] The second module is used to pre-model external disturbances at both land and air ends;

[0177] The third module is used to construct an extended state observer;

[0178] The fourth module is used to detect motion patterns in real time;

[0179] a fifth module, configured to perform disturbance information compensation based on outputs of the first module, the second module, and the fourth module, and then obtain a disturbance-compensated extended state observer based on the disturbance information compensation;

[0180] The sixth module is used to design a dual-modal active disturbance rejection controller based on the observation results of the extended state observer after disturbance compensation, so as to realize the dual-modal active disturbance rejection control of the land and air amphibious robot.

[0181] Regarding the specific definition of the universal dual-modal auto-disturbance rejection control system for land and air amphibious robots, please refer to the definition of the universal dual-modal auto-disturbance rejection control method for land and air amphibious robots above, which will not be repeated here. The various modules in the above-mentioned universal dual-modal auto-disturbance rejection control system for land and air amphibious robots can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0182] In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the general dual-modal self-disturbance rejection control method for a land-air amphibious robot when executing the computer program.

[0183] For the specific limitations of each step, please refer to the limitations of the general dual-modal active disturbance rejection control method for amphibious robots mentioned above, which will not be repeated here.

[0184] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the universal dual-modal anti-disturbance control method of the land-air amphibious robot is implemented.

[0185] For the specific limitations of each step, please refer to the limitations of the general dual-modal active disturbance rejection control method for amphibious robots mentioned above, which will not be repeated here.

[0186] In one embodiment, combined Figures 2 to 4A new amphibious reconnaissance robot employing the general dual-modal active disturbance rejection control method for amphibious robots is provided. The robot comprises a quadrotor frame 1, a tracked chassis 2, a main controller 3, a flight control cabin 4, an image transmission device 5, a chassis motion control box 6, and triangular track wheels 7. The quadrotor frame 1 employs a ducted structure, with the flight control cabin 4 fixed above the frame 1, the main controller 3 mounted within the flight control cabin 4, and the image transmission device 5 fixed in front of the frame 1. The tracked chassis 2 is fixed below the frame 1, with a chassis motion control box 6 mounted in the center. Two triangular track wheels 7 are rigidly connected to the chassis motion control box 6 and fixed to either side of the tracked chassis 2. The image transmission device 5 is used for real-time video capture and transmission.

[0187] Furthermore, in some embodiments, the ducted structure includes flight ducts 11 symmetrically distributed on both sides of the quadrotor flight frame 1 to reduce aerodynamic losses and suppress flight noise.

[0188] Furthermore, in some embodiments, a flight motor 12 is installed inside the flight duct 11; the flight motor 12 is electrically connected to an electronic speed regulator 14 fixed below the quadcopter flight frame 1, and a three-blade propeller 13 is fixed above the rotating shaft of the flight motor 12.

[0189] Furthermore, in some embodiments, the flight control cabin 4 is internally equipped with an environmental perception module 15 and a main controller 3. The main controller 3 centrally manages the motion logic control, land and flight mode switching, and executes the control variable solution of the active disturbance rejection control algorithm (the algorithm of the present invention) in real time. The environmental perception module 15 includes sensors such as an inertial measurement unit and a Beidou positioning module, which are used to measure the robot's posture in real time, obtain high-precision position information, and provide key data to the main controller 3 to ensure flight stability.

[0190] Furthermore, in some embodiments, a flight shock absorbing plate 16 is installed above the flight control cabin 4 .

[0191] Furthermore, in some embodiments, the image transmission device 5 acquires image information via an image transmission camera 18 and wirelessly transmits the image information to the monitoring terminal. The image transmission camera 18 is rigidly connected to the flight damping plate 16, and the shooting angle is adjusted by, but not limited to, a camera adjustment screw 17.

[0192] Furthermore, in some embodiments, the triangular track wheel 7 is connected to the track drive motor via the drive wheel 24 .

[0193] Furthermore, in some embodiments, the triangular track wheel includes a triangular base frame 21, a tensioning mechanism 22, a rubber track 23, a drive wheel 24, a tensioning wheel 25, and a supporting roller 26; the triangular base frame 21 serves as the main frame, and the top of the base frame is rigidly connected to the drive wheel 24. The middle of the base frame is hinged to the tensioning mechanism 22 and the tensioning wheel 25 on both sides through a connecting shaft. The bottom of the base frame is hinged to the supporting roller 26; the tensioning mechanism 22 is provided with a ball screw, which is connected to the tensioning wheel 25. The tensioning force is adjusted by adjusting the axial position of the ball screw, and the tensioning wheel 25 is pushed toward the rubber track 23 to ensure that the triangular track maintains appropriate tension during operation; the rubber track 23 is continuously engaged with the drive wheel 24 through multiple belt lugs at the top to ensure non-slip traction under harsh road conditions. Here, the tension is adjusted by the tensioning wheels 25 on both sides, and the supporting wheels 26 are arranged at the bottom to guide the movement of the track and reduce the rolling resistance; the driving wheel 24 is connected to the track drive motor inside the chassis motion control box 6, and the driving force of the driving wheels 24 on both sides is adjusted by the motor speed, and the differential drive realizes rapid steering.

[0194] Furthermore, in some embodiments, the supporting wheels 26 are composed of at least three sets of rigidly hinged rubber wheels, which engage with the belt lugs of the rubber track 23 to evenly distribute the weight of the robot on the rubber track 23.

[0195] In one embodiment, the new amphibious reconnaissance robot is controlled by using the dual-modal active disturbance rejection control method of the amphibious robot according to the present invention, specifically including:

[0196] Step 1: Conduct dual-modal modeling of the land and air amphibious robot;

[0197] The robot's double-end dynamic equation in step 1 has the characteristics of underactuation, strong coupling, and nonlinearity. A mathematical model is established based on the Newton-Euler equation to establish the mathematical model of the amphibious robot. The specific process includes:

[0198] Step 1-1, establish the mathematical model of the quadrotor flight end:

[0199]

[0200] Where M is the mass of the aircraft body; L is the distance from the center of mass to the center of the motor; U i (i=1,2,3,4) is the force condition of each channel during the robot movement; I x ,I y ,I z Respectively represent the moment of inertia of the corresponding axis; J r is the rotor moment of inertia; g represents the acceleration due to gravity; k i (i=1,2,3,4,5,6) represents the air resistance coefficient;

[0201] The robot model posture coupling term is regarded as an internal disturbance and compensated through an extended state observer, and the state space equation is established accordingly;

[0202] The input of the system is U i (i=1,2,3,4), select the system state variables:

[0203]

[0204] The system output is: represents the linear velocity of the robot in the earth coordinate system, Indicates the angular velocity of the robot in the body coordinate system.

[0205] The state space expression of the mathematical model of the quadrotor flight end is obtained:

[0206]

[0207] Step 1-2, mathematical modeling of the track ground side:

[0208] Simplify the tracked robot model into a two-wheel differential drive robot model:

[0209]

[0210] where v l 、v r Track speed on both sides, d LR is the virtual wheel spacing, v c and ω c are the linear velocity and angular velocity of the robot's center of mass, respectively.

[0211] Step 2: Pre-modeling of external disturbances at both land and air terminals. This includes:

[0212] Step 2-1: Pre-modeling of common disturbances on the flight side:

[0213] Gusts are a common meteorological phenomenon characterized by dramatic changes in wind speed over a short period of time. They occur suddenly and briefly, exhibiting significant transient and irregular characteristics. When an amphibious robot is in flight mode, these sudden, short-lived strong winds can significantly impact the robot's stability, control performance, and safety. Analysis has found that the mechanism of gust action is highly similar to the characteristics of the cosine function, and can therefore be described using the following formula:

[0214]

[0215] Where: μ is the gradient coefficient, V is the gust wind speed, V maxis the maximum gust wind speed, G is the gust transfer function, t0 is the start time, and t is the end time;

[0216] Step 2-2: Pre-modeling of common disturbances on the ground side:

[0217] In the ground driving mode of the amphibious robot, the road environment it faces is complex and changeable, and the physical properties of different road sections vary significantly, including but not limited to road hardness, friction resistance coefficient, and adhesion coefficient. Due to the uneven spatial distribution of these physical properties, the road friction resistance experienced by the motion units on both sides of the robot during driving varies. Establish a road resistance model on both sides:

[0218]

[0219] Among them: f L 、f R is the road resistance coefficient, and m is the weight of the amphibious robot.

[0220] Step 3: Construct an extended state observer;

[0221] The method of the present invention expands the disturbance effect f affecting the output of the controlled object into a new state variable, and uses a special feedback mechanism to establish an extended state observer of the disturbance effect; specifically, it includes:

[0222] Step 3-1, establish the Lumberg observer equation:

[0223]

[0224] in, represents the total disturbance differential term, f is the total disturbance, A, B, C, L, E are the coefficient matrices of the observer equation, is the state of the observation system, is the first-order differential of the observation system state, u is the observation system input, and y is the observation system output;

[0225] Design the extended state observer:

[0226]

[0227] Step 3-2, determine the parameters of the extended state observer:

[0228] Ensure that the extended state observer is fully converged, make the (A-LC) eigenvalue located in the negative half plane, introduce the observer bandwidth ω0>0, configure the characteristic equation poles to be located on -ω0, and obtain the characteristic equation λ(s):

[0229] λ(s)=|sI-(A-LC)|=(s+ω o ) 3 =0

[0230] The extended state observer gain matrix is obtained from this

[0231] Step 4: Real-time motion pattern detection; specifically including:

[0232] Step 4-1, first level of detection

[0233] (1) Detecting the air pressure at the current location of the amphibious reconnaissance robot. If the difference between the air pressure and the ground pressure is greater than a preset threshold, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1. Otherwise, execute (2).

[0234] (2) obtaining the vertical acceleration of the current amphibious reconnaissance robot, and determining whether the difference between the vertical acceleration and the negative value of the gravitational acceleration is within a preset deviation range; if so, determining that the current motion mode is the ground mode, and setting the current mode check bit α to 0; otherwise, determining that the current motion mode is the flight mode, and setting the current mode check bit α to 1;

[0235] Step 4-2, second test

[0236] Detect the current dominant control signal. If the ground motion control unit continuously outputs a signal, the current motion mode is determined to be the ground mode, and the current mode check bit α is set to 0. If the flight motion control unit continuously outputs a signal, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1.

[0237] Step 4-3, determine whether the detection results of step 4-1 and step 4-2 are consistent. If they are consistent, output the motion mode detection result and the mode check bit. Otherwise, return to execute step 4-1 and repeat until the preset detection number limit is reached.

[0238] Step 5: Based on the outputs of steps 1, 2, and 4, disturbance information compensation is performed, and then based on the disturbance information compensation, a disturbance-compensated extended state observer is obtained. The specific process includes:

[0239] Step 5-1: Robot motion sample data collection

[0240] Based on the kinematic modeling and disturbance pre-modeling in steps 1 and 2, various common external disturbances are applied one by one in the absence of external disturbances, ground mode, and flight mode, and the robot motion deviation data under the corresponding disturbances is collected;

[0241] Step 5-2: Establish a total disturbance model for estimating the motion state of the amphibious robot due to various disturbances and model uncertainties:

[0242] y i =f(xi ,β)

[0243] Where x i represents the i-th external disturbance or internal uncertainty, y i Represents x i The impact on the robot motion parameters; the internal uncertainty represents the uncertain parameters in the kinematic modeling; β represents the model parameters;

[0244] Step 5-2, weighted nonlinear least squares fitting:

[0245] According to the robot motion sample data collected in step 5-1 and the estimated total disturbance model, the influence of each disturbance and model uncertainty on the robot motion stability is evaluated, and the corresponding weights are determined to obtain the weighted matrix ω n is the weight of the nth external disturbance or internal uncertainty, and n is the total number of external disturbances or internal uncertainties;

[0246] Based on the weighted matrix, the total disturbance model is obtained by fitting using the weighted nonlinear least squares method:

[0247]

[0248] Among them, β represents the model parameters, S(β) is the total disturbance fitting, ω i is the weight of the i-th external disturbance or internal uncertainty;

[0249] The Gauss-Newton method is used to iteratively obtain the optimal model parameter β. The incremental equation is:

[0250] (J T WJ)Δβ=J T Wr

[0251] Where J is the Jacobian matrix, W is the diagonal weight matrix, and r is the residual vector;

[0252] Step 5-3, estimate the total disturbance compensation:

[0253] Based on the motion mode detection results of step 4, the disturbances in non-current motion modes are eliminated, the corresponding weights are set to 0, and the non-influence weight elimination matrix M(α) is obtained, and the compensation gain matrix N is obtained:

[0254] N=αS(β)M(α)

[0255] Wherein, α is the mode check bit corresponding to the motion mode detection result. If the motion mode is ground mode, the mode check bit α=0; if the motion mode is flight mode, the mode check bit α=1;

[0256] Based on this, the improved state observer after disturbance compensation is obtained:

[0257]

[0258] Step 6: Based on the observation results of the disturbance-compensated extended state observer, a dual-mode active disturbance rejection controller is designed to implement dual-mode active disturbance rejection control for the land and air amphibious robot. Specifically, it includes:

[0259] Step 6-1, establish the state error feedback control law:

[0260] Design the controller:

[0261]

[0262] Among them, u0 is the control quantity when disturbance compensation is not considered, represents the total disturbance of the robot estimated in real time by the extended state observer after disturbance compensation, b is the control input gain, and u is the control quantity output by the controller; when the extended state observer is estimated accurately, that is, When the original amphibious robot motion control system is changed to a double integrator series type: f is the total disturbance;

[0263] The PD control law is used as the state error feedback control law:

[0264]

[0265] Where k p , k d is the controller parameter, is the state estimate of the amphibious robot system observed by the extended state observer after disturbance compensation, is the estimated value of the state differential of the amphibious robot system observed by the extended state observer after disturbance compensation, U is the control input, is the derivative of the control input;

[0266] Considering that the extended state observer can observe the actual state, the state estimation value of the land-air amphibious robot system is similar to the real state quantity of the system, and the state differential estimation value of the land-air amphibious robot system is similar to the differential of the real state quantity of the system, that is,

[0267] The motion control system of the amphibious robot is designed as a critically damped second-order system to ensure that the system has no overshoot and has fast response capability:

[0268]

[0269] Where, ω c >0 indicates the controller bandwidth, which is used to adjust the response speed and stability of the system;c The introduction of k p , k d Dual parameter tuning, converting complex parameter tuning problems into bandwidth adjustment problems;

[0270] Step 6-2, solve the control quantity:

[0271] Calculated according to the state error feedback control law:

[0272]

[0273] Then the control quantity u is:

[0274]

[0275] Where U represents the system control input, is the differential of the control input, v1 represents the input of the state error feedback control law, and v2 represents the differential of the input of the state error feedback control law;

[0276] The accuracy is determined by the observer bandwidth ω0, and the control gain is determined by the controller bandwidth ω c Decide;

[0277] The control method is to adjust ω0 and ω c Perform parameter tuning to make the observer and controller errors converge to 0. Based on the estimation of the improved extended state observer in different modes of the land and air amphibious robot, it is concluded that high-precision estimation values can be provided in both modes, making the robot's dual-mode control sufficiently accurate and fast.

[0278] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only illustrative of the principles of the present invention. Without departing from the spirit and scope of the present invention, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A universal dual-mode active disturbance rejection control method for amphibious robots, characterized in that: The method comprises: Step 1: Conduct dual-modal modeling of the land and air amphibious robot; Step 2: Pre-modeling of external disturbances at both land and air terminals; Step 3: Construct an extended state observer; Step 4, detecting motion patterns in real time; Step 5: Based on the outputs of Step 1, Step 2, and Step 4, disturbance information compensation is performed, and then a disturbance-compensated extended state observer is obtained based on the disturbance information compensation; Step 6: Based on the observation results of the extended state observer after disturbance compensation, a dual-mode active disturbance rejection controller is designed to realize the dual-mode active disturbance rejection control of the land and air amphibious robot.

2. The universal dual-mode active disturbance rejection control method for land and air amphibious robots according to claim 1 is characterized in that: Step 1 is to perform dual-modal modeling of the land-air amphibious robot, specifically including: Step 1-1: Based on the mechanical structure of the amphibious robot, perform force analysis and kinematic analysis on each motion unit to obtain dynamic parameters and kinematic parameters; Step 1-2: perform dual-mode modeling based on dynamic parameters and kinematic parameters, wherein the dual-mode includes a flight mode and a ground mode.

3. The universal dual-mode active disturbance rejection control method for land and air amphibious robots according to claim 1 is characterized in that: Step 2 describes the pre-modeling of external disturbances at both land and air terminals, specifically including: Step 2-1: Pre-modeling of flight-end disturbances: Where μ is the gradient coefficient, V is the gust wind speed, V max is the maximum gust wind speed, G is the gust transfer function, t0 is the start time, t is the end time, and s represents the complex frequency variable in the transfer function; Step 2-2: Pre-modeling of ground disturbance: Where, F L 、F R Represents the resistance on the left and right sides of the road, f L 、f R are the resistance coefficients on the left and right sides of the road respectively, and m is the weight of the air-ground amphibious robot.

4. The universal dual-mode active disturbance rejection control method for land and air amphibious robots according to claim 1, characterized in that: Step 3 of constructing the extended state observer specifically includes: Step 3-1, establish the Lumberg observer equation: in, represents the total disturbance differential term, f is the total disturbance, A, B, C, L, E are the coefficient matrices of the observer equation, is the state of the observation system, is the first-order differential of the observation system state, u is the observation system input, and y is the observation system output; Design the extended state observer: Step 3-2, determine the parameters of the extended state observer: Ensure that the extended state observer is fully converged, make the (A-LC) eigenvalue located in the negative half plane, introduce the observer bandwidth ω0>0, configure the characteristic equation poles to be located on -ω0, and obtain the characteristic equation λ(s): λ(s)=|sI-(A-LC)|=(s+ω o ) 3 =0 The extended state observer gain matrix is obtained from this 5. The universal dual-mode active disturbance rejection control method for land and air amphibious robots according to claim 1 is characterized in that: The real-time detection of motion patterns in step 4 specifically includes: Step 4-1, first level of detection (1) Detecting the air pressure at the current location of the amphibious reconnaissance robot. If the difference between the air pressure and the ground pressure is greater than a preset threshold, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1. Otherwise, execute (2). (2) obtaining the vertical acceleration of the current amphibious reconnaissance robot, and determining whether the difference between the vertical acceleration and the negative value of the gravitational acceleration is within a preset deviation range; if so, determining that the current motion mode is the ground mode, and setting the current mode check bit α to 0; otherwise, determining that the current motion mode is the flight mode, and setting the current mode check bit α to 1; Step 4-2, second test Detect the current dominant control signal. If the ground motion control unit continuously outputs a signal, the current motion mode is determined to be the ground mode, and the current mode check bit α is set to 0. If the flight motion control unit continuously outputs a signal, the current motion mode is determined to be the flight mode, and the current mode check bit α is set to 1. Step 4-3, determine whether the detection results of step 4-1 and step 4-2 are consistent. If they are consistent, output the motion mode detection result and the mode check bit. Otherwise, return to execute step 4-1 and repeat until the preset detection number limit is reached.

6. The universal dual-mode active disturbance rejection control method for an amphibious robot according to claim 4, characterized in that: The specific process of step 5 includes: Step 5-1: Robot motion sample data collection Based on the kinematic modeling and disturbance pre-modeling in steps 1 and 2, various common external disturbances are applied one by one in the absence of external disturbances, ground mode, and flight mode, and the robot motion deviation data under the corresponding disturbances is collected; Step 5-2: Establish a total disturbance model for estimating the motion state of the amphibious robot due to various disturbances and model uncertainties: y i =f(x i ,b) Where x i represents the i-th external disturbance or internal uncertainty, y i Represents x i The results of the impact on the robot's motion parameters; the internal uncertainty represents the uncertain parameters in the kinematic modeling; β represents the model parameters; Step 5-2, weighted nonlinear least squares fitting: According to the robot motion sample data collected in step 5-1 and the estimated total disturbance model, the influence of each disturbance and model uncertainty on the robot motion stability is evaluated, and the corresponding weights are determined to obtain the weighted matrix ω n is the weight of the nth external disturbance or internal uncertainty, and n is the total number of external disturbances or internal uncertainties; Based on the weighted matrix, the total disturbance model is obtained by fitting using the weighted nonlinear least squares method: Among them, β represents the model parameters, S(β) is the total disturbance fitting, ω i is the weight of the i-th external disturbance or internal uncertainty; The Gauss-Newton method is used to iteratively obtain the optimal model parameter β; Step 5-3, estimate the total disturbance compensation: Based on the motion mode detection results of step 4, the disturbances in non-current motion modes are eliminated, the corresponding weights are set to 0, and the non-influence weight elimination matrix M(α) is obtained, and the compensation gain matrix N is obtained: N=αS(β)M(α) Wherein, α is the mode check bit corresponding to the motion mode detection result. If the motion mode is ground mode, the mode check bit α=0; if the motion mode is flight mode, the mode check bit α=1; Based on this, the improved state observer after disturbance compensation is obtained:

7. The universal dual-mode active disturbance rejection control method for land and air amphibious robots according to claim 1 is characterized in that: In step 6, the dual-mode active disturbance rejection controller compensates for disturbances in real time and performs rapid error response based on the real-time estimated value of the extended state observer after disturbance compensation, thereby solving the algorithm control variable. Specifically, the following steps are performed: Step 6-1, establish the state error feedback control law: Design the controller: Among them, u0 is the control quantity when disturbance compensation is not considered, represents the total disturbance of the robot estimated in real time by the extended state observer after disturbance compensation, b is the control input gain, and u is the control quantity output by the controller; when the extended state observer is estimated accurately, that is, When the original amphibious robot motion control system is changed to a double integrator series type: f is the total disturbance; The PD control law is used as the state error feedback control law: Where k p , k d is the controller parameter, is the state estimate of the amphibious robot system observed by the extended state observer after disturbance compensation, is the estimated value of the state differential of the amphibious robot system observed by the extended state observer after disturbance compensation, U is the control input, is the derivative of the control input; Considering that the extended state observer can observe the actual state, the state estimation value of the land-air amphibious robot system is similar to the real state quantity of the system, and the state differential estimation value of the land-air amphibious robot system is similar to the differential of the real state quantity of the system, that is, The motion control system of the amphibious robot is designed as a critically damped second-order system to ensure that the system has no overshoot and has fast response capability: Where, ω c >0 indicates the controller bandwidth, which is used to adjust the response speed and stability of the system; Step 6-2, solve the control quantity: Calculated according to the state error feedback control law: Then the control quantity u is: Where U represents the system control input, is the differential of the control input, v1 represents the input of the state error feedback control law, and v2 represents the differential of the input of the state error feedback control law; The accuracy is determined by the observer bandwidth ω0, and the control gain is determined by the controller bandwidth ω c Decide; By comparing ω0 and ω c Parameter tuning is performed to make the error between the extended state observer and the controller converge to 0 after disturbance compensation.

8. A universal dual-mode active disturbance rejection control system for amphibious and land robots based on the method according to any one of claims 1 to 7, characterized in that: The system comprises: The first module is used for dual-modal modeling of land and air amphibious robots; The second module is used to pre-model external disturbances at both land and air ends; The third module is used to construct an extended state observer; The fourth module is used to detect motion patterns in real time; a fifth module, configured to perform disturbance information compensation based on outputs of the first module, the second module, and the fourth module, and then obtain a disturbance-compensated extended state observer based on the disturbance information compensation; The sixth module is used to design a dual-modal active disturbance rejection controller based on the observation results of the extended state observer after disturbance compensation, so as to realize the dual-modal active disturbance rejection control of the land and air amphibious robot.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images