Vertical guide rail climbing robot and track, sliding mode and air pressure control method thereof

Abstract

The invention discloses a vertical guide rail climbing robot and a track, sliding mode and air pressure control method thereof. The vertical guide rail climbing robot comprises a supporting base, a movement module arranged on the supporting base and a center control module. The motion module comprises two symmetrically arranged negative pressure track systems, and the two negative pressure track systems are used for being adsorbed to the two sides of the guide rail correspondingly to form a double-side adsorption climbing structure; the negative-pressure crawler system comprises a driving unit, crawler sheets connected through pin shafts and a negative-pressure suction cup unit. The negative pressure suction cup units are arranged on the crawler belt pieces and are used for being adsorbed on the guide rails; the driving unit is arranged on the supporting base and used for driving the crawler belt pieces to rotate so that the robot can climb on the guide rails. The central control module is used for controlling the driving unit and the negative pressure suction cup unit. The problems that an existing climbing robot is limited in applicable guide rail type, insufficient in climbing stability, large in guide rail surface damage and low in control precision in a complex environment can be solved.

Description

Technical Field

[0001] The present application belongs to the field of robotics technology, and in particular relates to a vertical guide rail climbing robot and its trajectory, sliding mode, and air pressure control method. Background Art

[0002] Currently, high-altitude and vertical operations (such as elevator guide rail inspection and cable-stayed bridge maintenance) rely primarily on manual labor, which is associated with low efficiency and high risk. While existing wall-climbing robots can partially replace manual labor, they still have significant limitations: First, their suction methods are limited. For example, electromagnetic suction only works on metal surfaces, while traditional suction-cup robots are primarily designed for tubular structures and struggle to adapt to unusual shapes like T- and V-shaped guide rails. Second, their motion stability is insufficient. Wheeled or tracked robots are prone to slipping while climbing, and hard contact surfaces can scratch the guide rails. Third, their control precision is low and they lack the ability to adapt to dynamic environments, making them prone to loss of control, particularly in the event of air pressure leaks or external disturbances.

[0003] In the prior art, CN114013525A proposes a suction-cup tracked wall-climbing robot that achieves movement by controlling negative pressure with a blocking block. However, this robot is only applicable to flat walls and fails to address the multi-directional disturbance suppression and variable width adaptation issues encountered during rail climbing. Furthermore, its mechanical negative pressure control struggles to cope with air pressure fluctuations in complex environments, resulting in unstable suction force.

[0004] To address the above problems, there is an urgent need for a new climbing robot that can adapt to various guide rail types and has intelligent pressure regulation and anti-interference control. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a vertical guide rail climbing robot and its trajectory, sliding mode, and air pressure control method, so as to solve the problems mentioned in the above background technology, such as the limited applicable guide rail types, insufficient climbing stability, severe damage to the guide rail surface, and low control accuracy in complex environments.

[0006] To achieve the above objectives, this application adopts the following technical solutions: In a first aspect, a vertical guide rail climbing robot is provided, comprising a support base, a motion module disposed on the support base, and a central control module; The motion module includes two symmetrically arranged negative pressure crawler systems, which are respectively used to be adsorbed on both sides of the guide rail to form a double-sided adsorption climbing structure; The negative pressure crawler system includes a drive unit, crawler shoes connected by a pin shaft, and a negative pressure suction cup unit; The negative pressure suction cup unit is provided on the track shoe, and the negative pressure suction cup unit is used to be adsorbed on the guide rail; The driving unit is arranged on the supporting base, and the driving unit is used to drive the crawler shoe to rotate so as to enable the robot to climb on the guide rail; The central control module is used to control the driving unit and the negative pressure suction cup unit.

[0007] In a possible implementation, the driving unit includes a driving motor, two rotating shafts and a driving wheel; The driving wheels are coaxially arranged on the rotating shaft, and the track shoes are wound around the two driving wheels; The rotating shafts are rotatably arranged on the supporting bases respectively, and one end of the rotating shaft is connected to the driving motor, and the driving motor is used to drive the rotating shaft to rotate.

[0008] In a possible implementation, the negative pressure suction cup unit includes an air pump, a negative pressure box, an anti-leakage unit, and a plurality of vacuum suction cup units; The negative pressure box is located between the driving wheels, and a negative pressure cavity is provided in the hollow of the negative pressure box. A dovetail groove is provided on the side of the negative pressure box close to the track, and a plurality of air extraction holes are provided on the bottom wall of the dovetail groove; The negative pressure chamber is connected to the air pump, and the air pump is used to keep the negative pressure chamber in a negative pressure state; The anti-leakage units are respectively arranged on the negative pressure boxes, and the anti-leakage units are arranged in communication with the air extraction holes, and the anti-leakage units are used to keep the corresponding air extraction holes in a blocked state; The vacuum suction cup units are respectively arranged on the track shoes, one end of the vacuum suction cup unit is used to be adsorbed on the guide rail, and the other end is used to slide in the dovetail groove; When the vacuum suction cup unit is close to the corresponding anti-leakage unit, the corresponding air extraction hole is in an open state.

[0009] In one possible implementation, the central control module includes a controller, a laser radar, and a sensor: The laser radar is used to measure the distance between the robot and the guide rail; The controller is connected to the driving motor and the air pump through signal lines, and is used to control the flow rate change rate of the air pump and the motion state of the robot respectively.

[0010] In a possible implementation, the anti-leakage unit includes an air-stop core and an air-stop spring; The air-stop core and the air-stop spring are both arranged in the air-extraction hole, one end of the air-stop spring is connected to the bottom wall of the air-extraction hole, and the other end is connected to the air-stop core. The air-stop spring enables the air-stop core to block the air-extraction hole, and the part of the air-stop core extending out of the air-extraction hole is located in the dovetail groove.

[0011] In a possible implementation, the vacuum suction cup unit includes a rubber suction cup, an air extraction pipe, a distance adjustment spring, and an air extraction valve; The contact surface between the rubber suction cup and the guide rail is a circular protrusion; The air extraction pipe is arranged on the track shoe, one end of the air extraction pipe is connected to the base of the rubber suction cup, and the other end passes through the track shoe and is connected to the air extraction valve; The distance-adjusting spring is sleeved on the exhaust pipe, and the two ends of the distance-adjusting spring respectively abut against the top surface of the rubber suction cup and the bottom surface of the track shoe; The air extraction valve is slidably arranged in the dovetail groove, and an air extraction groove is arranged on one side of the air extraction valve close to the air extraction hole, and the air extraction groove is arranged along the length direction of the dovetail groove; The air extraction valve is used to allow the air-stop core to extend into the air extraction hole when the track piece moves in the dovetail groove, so that the air extraction groove is connected with the air extraction hole.

[0012] In a possible implementation, the fitting clearance between the dovetail groove and the air extraction valve is ≤0.1 mm, and a sealing rubber gasket is provided on the fitting surface.

[0013] In a second aspect, a trajectory control method for a vertical guide rail climbing robot is provided, comprising the following steps: S1: Data fusion and state estimation: Collect data through lidar and sensors, and use Kalman filter algorithm to fuse: ; in, is the state estimate at the previous moment, is the Kalman gain, is the current observation value, H is the observation matrix; S2: Adaptive control algorithm optimization: According to the position error e(t)= - , update the parameters through the adaptive control gain adjustment formula:

[0014] in, is the learning rate (range 0.01-0.1).

[0015] In a third aspect, a sliding mode control method for a vertical guide rail climbing robot is provided, comprising the following steps: Define the sliding surface: ; The control law is: ; Through the Lyapunov function Optimize weight coefficient And control gain K to suppress trajectory chattering.

[0016] In a fourth aspect, a method for controlling air pressure of a vertical guide rail climbing robot is provided, comprising the following steps: S1: System state equation modeling: ; in is the ambient atmospheric pressure, is the cavity pressure, is the cavity volume, is the air pump flow rate, To control the change rate of the flow rate of the input vacuum pump; S2: Adaptive extended Kalman filter parameter identification: The sliding window method updates the measurement noise and uses the residual Dynamically adjust measurement noise:

[0017] Where m is the sliding window length (range 5-10); The forgetting factor method updates the process noise and uses the new information Dynamically adjust process noise:

[0018] in, is the forgetting factor, ranging from 0.9 to 0.99; S3: Nonlinear Model Predictive Control Solution: Cost function construction:

[0019] in is the target stable air pressure value, They are tracking performance weight, energy consumption weight and output stability weight respectively; Constraints: Control Input That is, the change rate of the air pump flow rate should be controlled within arrive Between, take the virtual input , construct the constraint function:

[0020] The cost function is updated as:

[0021] Where R is the constraint penalty coefficient.

[0022] Compared with the prior art, this application has the following beneficial effects: The present application provides a vertical guide rail climbing robot, which exhibits many significant technical effects in structural design. The symmetrical adsorption structure formed by the double-sided negative pressure crawler system significantly improves the climbing stability and ensures the safety of high-altitude operations by balancing the forces on both sides of the guide rail. The modular design divides the robot into a support base, a motion module, and a central control module. Each module is independent and coordinated, which not only improves the efficiency of disassembly and maintenance, but also adapts to guide rails of different widths by flexibly adjusting the module spacing, significantly enhancing the adaptability to special-shaped guide rails such as T-type and V-type. In addition, the non-contact adsorption of the negative pressure suction cup unit breaks through the limitations of traditional electromagnetic adsorption and can be adapted to guide rails of various materials such as metal, wood, and concrete, avoiding scratches on the surface of the guide rail by the wheeled structure, and can expand the application scenarios to fields such as old metal guide rail inspection and wooden board maintenance. The overall structural design effectively solves the core problems of poor stability, narrow scope of application, and large guide rail damage in the existing technology, providing a more reliable solution for high-altitude operation automation.

[0023] In one possible implementation, the robot achieves stable drive on a vertical guide rail through the coordinated design of a drive motor, rotating shaft, and drive wheels. The drive motor operates at low speed, cooperating with the rotating shaft and drive wheels to drive the track shoes, reducing inertial impact and preventing slippage. The two drive wheels are coaxially arranged on the rotating shaft, balancing force on both sides of the track, reducing the risk of deviation and improving motion stability. The rotating shaft is mounted on a support base and connected to the drive motor, creating a stable structure that ensures effective transmission of driving force, thereby improving the stability and reliability of the robot's climbing, and resolving the issue of insufficient motion stability in existing technologies.

[0024] In one possible implementation, the vacuum pump cooperates with the negative pressure chamber to form a stable negative pressure environment. Combined with the sealing function of the anti-leakage unit, the vacuum hole can be accurately opened when the vacuum suction cup unit approaches, thereby realizing dynamic adsorption and avoiding continuous leakage; the sliding matching design of the dovetail groove and the vacuum suction cup unit not only provides guidance for the movement of the track piece and ensures the accuracy of the adsorption position, but also reduces disturbances by limiting the freedom of movement and improves the adsorption reliability; the distributed layout and time-sharing opening mechanism of multiple vacuum suction cup units ensure that the negative pressure chamber maintains stable air pressure during the adsorption process, balances the adsorption force, and thereby enhances the robot's climbing stability and environmental adaptability on the vertical guide rail, effectively solving the problem of unstable adsorption of traditional suction cups and susceptibility to rail structure restrictions. In one possible implementation, the central control module integrates lidar, sensors, and a controller to create an intelligent perception and precise control system. The lidar measures the lateral distance between the robot and the guide rail in real time with an accuracy of ±1mm, providing critical data for trajectory adjustment. Sensors monitor state parameters such as air pressure and speed in real time, integrating this data with the lidar data through the controller's built-in Kalman filter algorithm to achieve multi-source information complementation. Using an adaptive control algorithm, the controller dynamically adjusts the drive motor speed and the rate of change of the vacuum pump flow rate, maintaining a tracking error of ≤1mm and controlling pressure fluctuations in the negative pressure chamber to within ±1kPa.

[0025] A trajectory control method for a vertical rail-climbing robot uses lidar and sensor data fusion, combined with adaptive and sliding-mode control algorithms, to achieve precise trajectory optimization. Kalman filtering improves state estimation accuracy, while adaptive control dynamically adjusts control gains to minimize trajectory tracking error to ≤1mm. Sliding-mode control, combined with Lyapunov stability analysis, rapidly suppresses external disturbances, maintaining a jitter amplitude of ≤±0.5mm. This significantly improves motion stability and detection accuracy in complex environments, effectively counteracting the effects of uneven rail surfaces and airflow disturbances, ensuring a linear climbing deviation of ≤2mm / m on the robot's vertical rails.

[0026] A sliding mode control method for a vertical guide rail climbing robot enhances the robustness of the robot's motion by defining a multivariable sliding mode surface and optimizing the control law. A control strategy based on error differential equations and Lyapunov functions converges the system state to the sliding mode surface within a finite time, with a response time of ≤0.3s to disturbances such as impacts on guide rail joints. High-order sliding mode control principles are introduced to suppress chattering, improve control input smoothness, and reduce mechanical wear. Weight coefficients are used to balance various error components, keeping trajectory chatter within a certain range and ensuring stable operation of the robot in changing environments.

[0027] A method for controlling air pressure in a vertical rail-climbing robot achieves precise and stable pressure in the negative pressure chamber by combining an adaptive extended Kalman filter with nonlinear model predictive control. The AEKF dynamically updates noise parameters to address large variations in the magnitude of system variables, achieving an air pressure estimation error of ±1 kPa and a leakage coefficient identification error of ≤5%. The NMPC optimizes the control sequence based on the C / GMRES algorithm, making it suitable for embedded master control and effectively ensuring the robot's stable suction force and operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the overall structure of a vertical guide rail climbing robot provided in this application; Figure 2 A schematic diagram of a vertical guide rail climbing robot provided in this application from another perspective; Figure 3 A partial cross-sectional view of a vertical guide rail climbing robot provided in this application; Figure 4 A schematic diagram of the overall structure of a negative pressure box provided in this application; Figure 5 This is a schematic structural diagram of a negative pressure box provided in this application from another perspective; Figure 6 A schematic diagram of the overall structure of a negative pressure suction cup unit provided in this application; Figure 7 for Figure 4 Schematic diagram from another perspective; Figure 8 A schematic diagram of a vertical guide rail climbing robot in motion at a specific moment provided in this application; Figure 9 A schematic diagram of the motion state of a vertical guide rail climbing robot provided in this application at another moment; Figure 10 A schematic diagram of a negative pressure chamber air pressure control structure provided in this application; Figure 11 A schematic diagram of AEKF air pressure tracking provided by this application; Figure 12 A schematic diagram of flow tracking of an AEKF vacuum pump provided in this application; Figure 13 A schematic diagram of AEKF leakage coefficient tracking provided by this application; Figure 14 A schematic diagram of EKF air pressure tracking provided by this application; Figure 15 A schematic diagram of EKF vacuum pump flow tracking provided in this application; Figure 16 A schematic diagram of EKF leakage coefficient tracking provided in this application.

[0029] The accompanying drawings are as follows: 1. Support base; 2. Motion module; 3. Central control module; 31. Controller; 32. Laser radar; 4. Negative pressure track system; 41. Drive unit; 411. Drive motor; 412. Two rotating shafts; 413. Drive wheel; 42. Track shoe; 43. Negative pressure suction cup unit; 431. Air pump; 432. Negative pressure box; 44. Anti-leakage unit; 441. Air stop core; 442. Air stop spring; 45. Vacuum suction cup unit; 451. Rubber suction cup; 452. Air extraction pipe; 453. Adjustable distance spring; 454. Air extraction valve; 457. Air extraction groove; 458. Rubber gasket; 459. Adjustable distance bolt; 46. Negative pressure chamber; 47. Dovetail groove; 48. Air extraction hole. DETAILED DESCRIPTION

[0030] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope. Therefore, the drawings and description are to be regarded as illustrative in nature and not restrictive.

[0031] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0032] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0033] In this application, unless otherwise expressly specified or limited, terms such as "installed," "connected," "connect," and "fixed" should be understood in a broad sense. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection, or communication; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on specific circumstances.

[0034] In this application, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Moreover, a first feature being "above," "above," and "above" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below," "below," and "below" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0035] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0036] Adaptive extended Kalman filter (AEKF); Nonlinear model predictive control (NMPC); Extended Kalman filter (EKF).

[0037] like Figure 1 、 Figure 2 and Figure 3 As shown, the present application provides a vertical guide rail climbing robot, which may include a support base 1, a motion module 2 and a central control module 3 arranged on the support base 1.

[0038] The support base 1 can be designed as a lightweight bow-shaped frame and made of aluminum alloy.

[0039] The motion module 2 includes two symmetrically arranged negative pressure track systems 4. The two negative pressure track systems 4 are respectively used to be adsorbed on both sides of the guide rail to form a double-sided adsorption climbing structure.

[0040] The two negative pressure crawler systems 4 are symmetrically distributed along the central axis of the guide rail and can be adapted to special-shaped guide rails such as T-shaped and V-shaped.

[0041] The negative pressure crawler system 4 may include a driving unit 41 , crawler shoes 42 connected by a pin shaft, and a negative pressure suction cup unit 43 .

[0042] The negative pressure suction cup unit 43 is disposed on the track shoe 42 , and the negative pressure suction cup unit 43 is used for adsorbing on the guide rail.

[0043] The driving unit 41 is disposed on the supporting base 1 . By adjusting the driving unit 41 , the driving unit 41 can drive the track shoe 42 to rotate, so as to enable the robot to climb on the guide rail.

[0044] The central control module 3 is arranged on the support base 1 . By adjusting the central control module 3 , the central control module 3 can control the driving unit 41 to work, and can also control the adsorption strength of the negative pressure suction cup unit 43 on the guide rail.

[0045] In the embodiment of the present application, the vertical guide rail climbing robot demonstrates significant technical effects in many aspects of structural design. The symmetrical adsorption structure formed by the double-sided negative pressure crawler system 4 significantly improves the climbing stability and ensures the safety of high-altitude operations by balancing the forces on both sides of the guide rail. The modular design divides the robot into a support base 1, a motion module 2, and a central control module 3. Each module is independent and collaborative, which not only improves the efficiency of disassembly and maintenance, but also significantly enhances the adaptability to special-shaped guide rails such as T-shaped and V-shaped by flexibly adjusting the module spacing to adapt to guide rails of different widths. In addition, the non-contact adsorption of the negative pressure suction cup unit 43 breaks through the limitations of traditional electromagnetic adsorption and can be adapted to guide rails of various materials such as metal, wood, and concrete, avoiding scratches on the surface of the guide rail by the wheeled structure, and can expand the application scenarios to fields such as old metal guide rail detection and wooden board maintenance. The overall structural design effectively solves the core problems of poor stability, narrow scope of application, and large guide rail damage in the existing technology, providing a more reliable solution for high-altitude operation automation.

[0046] In a possible embodiment, the driving unit 41 may include a driving motor 411 , two rotating shafts 412 and a driving wheel 413 .

[0047] The driving wheels 413 are coaxially arranged on the rotating shaft, and the driving wheels 413 and the rotating shaft are connected by a key. The track shoe 42 is wound around the two driving wheels 413 so that the track shoe 42 can move with the rotation of the driving wheels 413. One end of the rotating shaft is fixed to the support base 1 through a bearing, and the other end of the rotating shaft is connected to the driving shaft of the driving motor 411. By adjusting the driving motor 411, the driving motor 411 can drive the rotating shaft to rotate.

[0048] Optionally, the drive motor 411 can be set to a 24V / 3.2A DC reduction motor, which can output a speed of 16rpm, controlling the linear speed of the track shoe 42 at 0.1m / s, effectively avoiding slipping or adsorption failure caused by speed fluctuations during climbing.

[0049] In the embodiment of the present application, the coordinated design of the drive motor 411, the rotating shaft, and the drive wheel 413 achieves stable driving of the robot on the vertical guide rail. The drive motor 411 operates in a low-speed mode, cooperating with the rotating shaft and the drive wheel 413 to drive the track shoe 42 to rotate, thereby reducing inertial impact and preventing slippage. The two drive wheels 413 are coaxially arranged on the rotating shaft, which balances the force on both sides of the track, reduces the risk of deviation, and improves the stability of movement. The rotating shaft is mounted on the support base 1 and connected to the drive motor 411, resulting in a stable structure, ensuring the effective transmission of driving force, thereby improving the stability and reliability of the robot's climbing, and solving the problem of insufficient movement stability in the prior art.

[0050] In one possible embodiment, Figure 4 and Figure 5 As shown, the negative pressure suction cup unit 43 may include an air pump 431 , a negative pressure box 432 , an anti-leakage unit 44 and a plurality of vacuum suction cup units 45 .

[0051] The negative pressure box 432 can be fixed to the support base 1 by bolts and is located between the two driving wheels 413. The negative pressure box 432 is hollow and defines a negative pressure chamber 46. A dovetail groove 47 is defined on the side of the negative pressure box 432 close to the track and distributed along the direction of track movement. A plurality of air extraction holes 48 are connected and provided on the bottom wall of the dovetail groove 47.

[0052] Optionally, there are 10 air extraction holes 48 , which are evenly distributed in the middle of the negative pressure box 432 .

[0053] The negative pressure chamber 46 is connected to the air pump 431. The inner wall of the negative pressure chamber 46 is provided with a rubber gasket 458. The rubber gasket 458 can improve the airtightness of the negative pressure box 432. By adjusting the air pump 431, the air pump 431 can make the negative pressure chamber 46 in a negative pressure state.

[0054] The anti-leakage units 44 are respectively arranged on the negative pressure box 432, and the anti-leakage units 44 are connected and arranged in the air extraction holes 48. The anti-leakage units 44 can keep the corresponding air extraction holes 48 in a blocked state.

[0055] The vacuum suction cup units 45 are respectively provided on the track pieces 42 . One end of the vacuum suction cup units 45 is used for adsorbing on the guide rail, and the other end is used for sliding in the dovetail groove 47 .

[0056] When the vacuum suction cup unit 45 is close to the corresponding anti-leakage unit 44 , the corresponding air extraction hole 48 is in an open state, so that the vacuum suction cup unit 45 can be adsorbed on the guide rail.

[0057] In the embodiment of the present application, the vacuum pump 431 cooperates with the negative pressure chamber 46 to form a stable negative pressure environment. Combined with the sealing function of the anti-leakage unit 44, the vacuum hole 48 can be accurately opened when the vacuum suction cup unit 45 approaches, thereby realizing dynamic adsorption and avoiding continuous leakage; the sliding cooperation design of the dovetail groove 47 and the vacuum suction cup unit 45 not only provides guidance for the movement of the track piece 42 and ensures the accuracy of the adsorption position, but also reduces disturbances by limiting the freedom of movement and improves the adsorption reliability; the distributed layout and time-sharing opening mechanism of multiple vacuum suction cup units 45 ensure that the negative pressure chamber 46 maintains stable air pressure during the adsorption process, balances the adsorption force, and thereby enhances the climbing stability and environmental adaptability of the robot on the vertical guide rail, effectively solving the problem that the traditional suction cup adsorption is unstable and easily restricted by the guide rail structure.

[0058] In a possible embodiment, the central control module 3 may include a controller 31 , a laser radar 32 , and a sensor; The laser radar 32 can measure the distance between the robot and the guide rail.

[0059] Optionally, the laser radar 32 is installed at the front center of the support base 1, with a measurement range of 0.1-10m, an accuracy of ±1mm, and scans the lateral distance between the robot and the guide rail at a frequency of 20Hz.

[0060] The controller 31 is connected to the driving motor 411 and the air pump 431 through signal lines, and is used to control the flow rate change rate of the air pump 431 and the motion state of the robot.

[0061] Specifically, the controller 31 may be an industrial-grade PLC that supports RS485 communication protocol and has built-in adaptive control and sliding mode control algorithm programs, and can process multi-sensor data in real time and output control instructions.

[0062] The sensor may include an air pressure sensor for measuring the air pressure in the negative pressure chamber 46 with an accuracy of ±0.1 kPa; A speed encoder may also be included to monitor the speed of the driving wheel 413 with a resolution of 0.1 rpm; It may also include a tilt sensor to detect the posture of the fuselage with an accuracy of ±0.5°, and is connected to the controller 31 through an analog input module.

[0063] In this embodiment of the present application, the central control module 3 integrates a laser radar 32, sensors, and controller 31 to form an intelligent perception and precision control system. The laser radar 32 measures the lateral distance between the robot and the guide rail in real time with an accuracy of ±1mm, providing key data for trajectory adjustment. The sensors monitor state parameters such as air pressure and speed in real time, integrating this data with the laser radar 32 data through the Kalman filter algorithm built into the controller 31 to achieve multi-source information complementation. Based on an adaptive control algorithm, the controller 31 dynamically adjusts the speed of the drive motor 411 and the flow rate change rate of the vacuum pump 431, ensuring that the robot's trajectory tracking error is ≤1mm and that the pressure fluctuation in the negative pressure chamber 46 is controlled within ±1kPa.

[0064] This design not only improves the robot's real-time perception of the guide rail environment, but also achieves coordinated optimization of motion state and adsorption force through a closed-loop control link, effectively resisting external disturbances and ensuring stable climbing under complex guide rail conditions, solving the problems of low control accuracy and insufficient environmental adaptability in existing technologies.

[0065] In a possible embodiment, the anti-leakage unit 44 may include an air-stop core 441 and an air-stop spring 442 .

[0066] The air-stop core 441 and the air-stop spring 442 are both arranged in the air-exhaust hole 48. One end of the air-stop spring 442 is integrally connected to the bottom wall of the air-exhaust hole 48, and the other end is integrally connected to the end of the air-stop core 441. The air-stop spring 442 enables the air-stop core 441 to block the air-exhaust hole 48, and the part of the air-stop core 441 extending out of the air-exhaust hole 48 is located in the dovetail groove 47. The air-stop core 441 can block the air-exhaust hole 48 in a natural state.

[0067] In this embodiment, the anti-leakage unit 44 achieves dynamic sealing and efficient communication of the air extraction hole 48 through a purely mechanical structure: the elastic force of the air stop spring 442 normally blocks the air stop core 441, preventing continuous leakage from the negative pressure chamber 46. When the vacuum suction cup unit 45 moves with the track shoe 42 to the air extraction hole 48, the air extraction valve 454 presses against the inclined surface of the air stop core 441, compressing the spring to quickly open the air extraction hole 48, establishing air communication between the negative pressure chamber 46 and the suction cup. This design requires no additional power, reduces energy loss, and avoids potential delays or malfunctions.

[0068] In one possible embodiment, Figure 6 and Figure 7 As shown, the vacuum suction cup unit 45 may include a rubber suction cup 451 , an air extraction pipe 452 , a distance adjustment spring 453 and an air extraction valve 454 .

[0069] The rubber suction cup 451 can be made of silicone rubber with a Shore hardness of 50A. The contact surface is a circular protrusion with a diameter of 15mm and a height of 3mm. It is connected to one end of the exhaust pipe 452 through an M3 thread. The surface roughness of the protrusion Ra≤1.6μm ensures sealing when in contact with the guide rail.

[0070] The air extraction pipe 452 is disposed on the track piece 42 . One end of the air extraction pipe 452 is connected to the base of the rubber suction cup 451 , and the other end passes through the track piece 42 and is connected to the air extraction valve 454 .

[0071] Optionally, the exhaust pipe 452 is made of aluminum alloy with an inner diameter of 2 mm, an outer diameter of 4 mm, and a length of 10 mm. One end is inserted into the center hole of the base of the rubber suction cup 451 and sealed with anaerobic glue, and the other end passes through the preset through hole of the track shoe 42 and is threadedly connected to the air inlet at the bottom of the exhaust valve 454.

[0072] The distance-adjusting spring 453 is sleeved on the exhaust pipe 452 , and two ends of the distance-adjusting spring 453 respectively abut against the top surface of the rubber suction cup 451 and the bottom surface of the track shoe 42 .

[0073] In addition, a distance-adjusting bolt 459 is provided at one end of the exhaust pipe 452 , and the length of the distance-adjusting spring 453 can be changed by rotating the distance-adjusting bolt 459 .

[0074] Optionally, the diameter of the steel wire of the adjustable spring 453 can be 0.8mm, the outer diameter is 6mm, the free length is 10mm, the compression range is 2-5mm, the elastic coefficient is 5N / m, and it is sleeved on the outside of the exhaust pipe 452, with the two ends respectively abutting the top surface of the rubber suction cup 451 and the bottom surface of the track shoe 42, and can adapt to the guide rail width change of ±0.5mm.

[0075] The exhaust valve 454 is slidably disposed within the dovetail groove 47. An exhaust groove 457 is provided on one side of the exhaust valve 454 near the exhaust hole 48. The exhaust groove 457 is disposed along the length of the dovetail groove 47. As the exhaust valve 454 moves along with the track shoe 42 within the dovetail groove 47, the air stop core 441 extends into the exhaust hole 48, thereby connecting the exhaust groove 457 to the exhaust hole 48.

[0076] Optionally, the main body of the exhaust valve 454 is made of polytetrafluoroethylene, with a trapezoidal cross-section, which is gap-matched with the dovetail groove 47. A slender exhaust groove 457 is processed on the inside, and the bottom of the groove is 0.5 mm away from the top surface of the exhaust hole 48 to ensure that the exhaust groove 457 is connected to the exhaust hole 48 when the air-stop core 441 is compressed and retracted.

[0077] like Figure 8 and Figure 9 As shown, when the track piece 42 drives the vacuum cup unit 45 to move along the dovetail groove 47, the trapezoidal slope of the air extraction valve 454 gradually approaches the air stop core 441. The air stop core 441 extends out of the air extraction hole 48 under the action of the spring force, blocking the air extraction hole 48, and isolating the negative pressure chamber 46 from the outside.

[0078] When the trapezoidal slope of the air extraction valve 454 contacts the air-stop core 441, the slope of the air extraction valve 454 begins to squeeze the air-stop core 441, the compression of the air-stop spring 442 gradually increases, the air-stop core 441 retracts, and the air-stop core 441 no longer blocks the air extraction hole 48. The air extraction hole 48 is connected to the air extraction groove 457, thereby realizing the connection between the rubber suction cup 451 and the air extraction pipe 452, and the negative pressure chamber 46 extracts air from the rubber suction cup 451 through the air extraction pipe 452 to form an adsorption force.

[0079] When the exhaust valve 454 continues to move, the inclined surface of the exhaust valve 454 contacts the next air-stop core 441, and the inclined surface of the exhaust valve 454 begins to squeeze the next air-stop core 441. As a result, the next exhaust hole 48 is connected to the exhaust pipe 452 and the rubber suction cup 451, and then air is exhausted to form an adsorption force.

[0080] Therefore, when the rubber suction cups 451 move along with the track shoe 42 , the rubber suction cups 451 on a line of the track shoe 42 are all in an adsorption state.

[0081] When the inclined surface of the air extraction valve 454 passes through the air-stop core 441 , the air-stop core 441 blocks the corresponding air extraction hole 48 under the elastic force of the air-stop spring 442 .

[0082] In the embodiment of the present application, the vacuum suction cup unit 45 significantly improves the robot's adsorption stability and environmental adaptability through innovative structural design: the circular raised contact surface of the rubber suction cup 451 can fit tightly against the subtle bumps on the surface of the guide rail, enhancing the sealing effect, improving the adsorption force while avoiding scratching the guide rail; the adjustable spring 453 is sleeved on the exhaust pipe 452, and can automatically adapt to changes in the width of the guide rail through elastic deformation, buffering horizontal disturbances and reducing body vibration; the coordination of the exhaust valve 454 and the dovetail groove 47 and the design of the exhaust groove 457 ensure that the exhaust hole 48 is opened and closed accurately, realizing dynamic connection between the negative pressure chamber 46 and the suction cup, and combining the time-sharing exhaust strategy to effectively control air pressure fluctuations and ensure stable adsorption force. The overall design solves the problems of unstable adsorption and high precision requirements of traditional suction cups through the collaboration of multiple components, enabling the robot to climb stably in a variety of guide rail environments.

[0083] In a possible embodiment, the fitting clearance between the dovetail groove 47 and the air extraction valve 454 is ≤0.1 mm, and a sealing rubber gasket is provided on the fitting surface.

[0084] In the embodiment of the present application, by controlling the fitting clearance between the dovetail groove 47 and the exhaust valve 454 to ≤0.1mm and arranging a sealing rubber gasket on the fitting surface, the sealing performance and movement accuracy of the negative pressure system are significantly improved: precise gap control combined with the elastic sealing of the rubber gasket ensures stable adsorption force and avoids unstable climbing caused by leakage; at the same time, strict clearance fit provides a single-degree-of-freedom guide for the exhaust valve 454, so that its swing error is controlled within an acceptable range, ensuring that the exhaust groove 457 and the exhaust hole 48 are precisely aligned, thereby improving the gas path conduction efficiency; in addition, the flexible buffering and wear-resistant material properties of the rubber gasket can adapt to the guide rail processing errors and harsh environments, reduce component wear, extend the service life of the robot, and effectively solve the problems of large leakage and large guide deviation of the traditional structure.

[0085] A trajectory control method for a vertical guide rail climbing robot comprises the following steps: S1: Analyze and calculate the operating environment and preset system status data in real time, and iteratively obtain the current optimal system status and corresponding parameters. The specific steps are as follows: Data fusion: The Kalman filter fusion algorithm is used to integrate the information from different sensors. The standard form of Kalman filter is: ; in, is the state estimate at the last moment, is the Kalman gain, is the current observation value, is the observation matrix; Real-time state estimation and error correction: Based on the fused sensor data, the robot's current position and velocity are estimated using a motion model. Assuming the robot moves along a track, the robot's motion state is described using a classical dynamics model: , ; in, , is the position of the robot on the two-dimensional plane, It's the robot in time The speed of time, is the angle the robot is facing.

[0086] The data is collected by LiDAR 32 and sensors and fused using the Kalman filter algorithm: ; in, is the state estimate at the previous moment, is the Kalman gain, is the current observation value, H is the observation matrix, and the state estimation includes the position, velocity, and angle of the robot.

[0087] S2: Adaptive control algorithm optimization: Adaptive control algorithm: Adaptive control is a control strategy that adjusts the error according to real-time estimation. It achieves optimal path tracking and defect detection by dynamically adjusting the control gain based on the error. The target dynamic behavior of the reference model is expressed as: ; in, is the acceleration of the reference model.

[0088] The real-time dynamic behavior of the robot is expressed as: ; in, is the current position error, is the control gain, is the acceleration generated by the control system (adjusted according to the error calculated in real time), and the control gain Dynamically adjust according to the current error. Suppose the current error of the robot is , the control gain is adaptively adjusted using the following formula: ,in, is the learning rate (adjusting the gain to gradually reduce the system error).

[0089] Real-time analysis and calculation: Analyze the robot's operating status and adjust the robot's control strategy based on the current system state parameters and external environment parameters. Utilize adaptive control algorithms to iteratively calculate the current optimal system state and corresponding parameters based on the current operating status and environmental adaptability analysis results.

[0090] In the embodiment of the present application, this method achieves precise optimization of the robot's trajectory by fusing data from the laser radar 32 and sensors, combined with an adaptive control algorithm and a sliding mode control algorithm. Kalman filtering improves state estimation accuracy, and adaptive control dynamically adjusts control gains to minimize trajectory tracking error to ≤1mm. Sliding mode control, combined with Lyapunov stability analysis, rapidly suppresses external disturbances, achieving a jitter amplitude of ≤±0.5mm. This significantly improves motion stability and detection accuracy in complex environments, effectively counteracting the effects of uneven rail surfaces and airflow disturbances, ensuring that the robot's linear climbing deviation on vertical rails is ≤2mm / m.

[0091] Considering the high demands placed on the robot's motion control capabilities by the variability of its operating environment, a method for optimizing the robot's trajectory controller design using an adaptive sliding mode control algorithm is proposed. During robot motion, the robot uses a laser radar (LIDAR) and related sensors to collect relevant data along its trajectory. Using an adaptive control algorithm, the robot analyzes and calculates the operating environment and preset system states in real time, iteratively determining the optimal system state and corresponding parameters. These parameters are then fed into the sliding mode control algorithm in real time. The sliding mode switching function and related parameters are calculated using an error differential equation and Lyapunov stability constraints. The trajectory controller is then optimized using the sliding mode control algorithm. This method enables the robot to identify the environment and system parameters or states while autonomously adjusting and optimizing the trajectory controller. This allows the control system to change its motion state based on interference factors, effectively counteracting the adverse effects of unexpected excitations during the robot's motion. This improves the robot's motion stability and related detection accuracy in complex environments.

[0092] At the same time, in order to address the chattering phenomenon caused by the discontinuity of the control law function U in the traditional sliding mode controller, the high-order sliding mode control principle (Super-Twisting sliding mode control) can be used to optimize the sliding mode controller. In the trajectory tracking control of the robot, a fixed-time observer is used to observe and estimate the external disturbance and joint state, and feed it back to the adaptive sliding mode controller for tracking control, which can suppress the chattering phenomenon to a certain extent and ensure the relative stability of the system state.

[0093] The robot uses the error differential equation and Lyapunov stability constraints to obtain the sliding mode switching function and related parameters. A sliding mode control method for a vertical guide rail climbing robot includes the following steps: Assume the initial state of the robot is , define the target trajectory , the error variable is: , , , ; Through the kinematic and dynamic models of the robot: , , , ,in, is the robot angular velocity, is the robot linear acceleration, then the error change rate is: , , , ; Define the Lyapunov function , , calculate the time derivative of the Lyapunov function: ; Define the sliding surface , ,in, , , , It is a positive weight coefficient used to balance the influence of each error component. The goal of the sliding mode controller is to enable the system state to enter and remain on the sliding surface. The control law is as follows: ,in, is the control gain, It is a negative sign function.

[0094] The design of the sliding mode controller is optimized through error differential equation and Lyapunov stability analysis. The specific optimization process includes: adjusting the sliding surface coefficient , , , , optimize the control gain , to avoid chattering.

[0095] In the embodiment of the present application, the robustness of the robot motion is enhanced by defining a multivariable sliding surface and optimizing the control law. The control strategy designed based on the error differential equation and the Lyapunov function can make the system state converge to the sliding surface within a limited time, and the response time to disturbances such as the impact of the guide rail joint is ≤0.3s; the high-order sliding mode control principle is introduced to suppress vibration, improve the smoothness of the control input, reduce mechanical wear, and at the same time balance the various error components through weight coefficients to control the trajectory vibration within a certain range, ensuring the stable operation of the robot in a changing environment. A method for controlling air pressure of a vertical guide rail climbing robot comprises the following steps: According to simulation experiments, the leakage of the negative pressure track negative pressure chamber during operation can be approximately expressed as:

[0096] in is the system leakage coefficient, which changes slightly and can be approximately considered to satisfy:

[0097] According to the law of conservation of mass, the pressure in the negative pressure cavity has the following relationship:

[0098] Therefore, the system state equation can be obtained: ; in is the ambient atmospheric pressure, is the cavity pressure, is the cavity volume, is the air pump flow rate, To control the flow rate change rate of the vacuum pump input.

[0099] Adaptive Extended Kalman Filter Parameter Identification: 1) Perform Kalman filtering: Prior predictions:

[0100] Correction:

[0101] (2) Adaptive step, forming an adaptive extended Kalman filter: Based on the sliding window method, the residual is used to update the measurement noise and the residual is defined is the difference between the measured value and the EKF posterior estimate, that is,

[0102] The measurement noise is dynamically adjusted by the sliding window method:

[0103] Where m is the sliding window length (range 5-10); The forgetting factor method updates the process noise and uses the new information Dynamically adjust process noise:

[0104] in, is the forgetting factor, ranging from 0.9 to 0.99; S3: Nonlinear Model Predictive Control Solution: Cost function construction:

[0105] in is the target stable air pressure value, They are tracking performance weight, energy consumption weight and output stability weight respectively; Constraints: Control Input That is, the change rate of the air pump flow rate should be controlled within arrive Between, take the virtual input , construct the constraint function:

[0106] The cost function is updated as:

[0107] Where R is the constraint penalty coefficient.

[0108] In the embodiments of this application, Figure 10 As shown in the figure, a method for controlling air pressure in a vertical rail climbing robot combines AEKF with NMPC to achieve precise and stable pressure in the negative pressure chamber. AEKF dynamically updates noise parameters to address the large magnitude differences in system variables, achieving an air pressure estimation error of ±1 kPa and a leakage coefficient identification error of ≤5%. NMPC optimizes the control sequence based on the C / GMRES algorithm, making it suitable for embedded master control and effectively ensuring the stability of the robot's suction force and operational efficiency.

[0109] In addition, Figure 11-16 As shown: Verified by MATLAB simulation, the AEKF algorithm has good parameter identification effect in this negative pressure system. Although some similar identification algorithms may have inaccurate identification or even divergence due to large magnitude differences between system variables (in the International System of Units, the cavity air pressure is 10 to the sixth power, the air pump flow is 10 to the -4 power, and the leakage coefficient is 10 to the -9 power), it can still achieve good tracking effect. The tracking effect is also improved compared with the standard EKF before optimization, with smaller fluctuations and higher accuracy.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that modifications to the technical solutions described in the aforementioned embodiments, or equivalent replacement of some or all of the technical features therein, do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A vertical guide rail climbing robot, characterized in that: It comprises a support base (1), a motion module (2) and a central control module (3) arranged on the support base (1); The motion module (2) includes two symmetrically arranged negative pressure crawler systems (4), and the two negative pressure crawler systems (4) are respectively used to be adsorbed on both sides of the guide rail to form a double-sided adsorption climbing structure; The negative pressure crawler system (4) comprises a drive unit (41), crawler shoes (42) connected by a pin shaft, and a negative pressure suction cup unit (43); The negative pressure suction cup unit (43) is arranged on the track shoe (42), and the negative pressure suction cup unit (43) is used to be adsorbed on the guide rail; The driving unit (41) is arranged on the supporting base (1), and the driving unit (41) is used to drive the crawler shoe (42) to rotate, so as to enable the robot to climb on the guide rail; The central control module (3) is used to control the driving unit (41) and the negative pressure suction cup unit (43).

2. The vertical guide rail climbing robot according to claim 1, characterized in that: The driving unit (41) includes a driving motor (411), two rotating shafts (412) and a driving wheel (413); The driving wheels (413) are coaxially arranged on the rotating shaft, and the track shoes (42) are wound around the two driving wheels (413); The rotating shafts are rotatably arranged on the supporting base (1), and one end of the rotating shaft is connected to the driving motor (411). The driving motor (411) is used to drive the rotating shaft to rotate.

3. The vertical guide rail climbing robot according to claim 1, characterized in that: The negative pressure suction cup unit (43) includes an air pump (431), a negative pressure box (432), an anti-leakage unit (44), and a plurality of vacuum suction cup units (45); The negative pressure box (432) is located between the driving wheels (413), and a negative pressure chamber (46) is provided in the hollow of the negative pressure box (432). A dovetail groove (47) is provided on a side of the negative pressure box (432) close to the track, and a plurality of air extraction holes (48) are provided on the bottom wall of the dovetail groove (47). The negative pressure chamber (46) is connected to the air pump (431), and the air pump (431) is used to keep the negative pressure chamber (46) in a negative pressure state; The anti-leakage units (44) are respectively arranged on the negative pressure boxes (432), and the anti-leakage units (44) are arranged in communication with the air extraction holes (48), and the anti-leakage units (44) are used to keep the corresponding air extraction holes (48) in a blocked state; The vacuum suction cup units (45) are respectively arranged on the track pieces (42), one end of the vacuum suction cup units (45) is used for adsorbing on the guide rail, and the other end is used for sliding in the dovetail groove (47); When the vacuum suction cup unit (45) is close to the corresponding anti-leakage unit (44), the corresponding air extraction hole (48) is in an open state.

4. The vertical guide rail climbing robot according to claim 3, characterized in that: The central control module (3) includes a controller (31), a laser radar (32) and a sensor: The laser radar (32) is used to measure the distance between the robot and the guide rail; The controller (31) is respectively connected to the driving motor (411) and the air pump (431) via signal lines for communication, and is used to control the flow rate change rate of the air pump (431) and the motion state of the robot.

5. The vertical guide rail climbing robot according to claim 3, characterized in that: The anti-leakage unit (44) includes an air-stop core (441) and an air-stop spring (442); The air-stop core (441) and the air-stop spring (442) are both arranged in the air-extraction hole (48), one end of the air-stop spring (442) is connected to the bottom wall of the air-extraction hole (48), and the other end is connected to the air-stop core (441), the air-stop spring (442) enables the air-stop core (441) to block the air-extraction hole (48), and the portion of the air-stop core (441) extending out of the air-extraction hole (48) is located in the dovetail groove (47).

6. The vertical guide rail climbing robot according to claim 5, characterized in that: The vacuum suction cup unit (45) includes a rubber suction cup (451), an air extraction pipe (452), a distance adjustment spring (453) and an air extraction valve (454); The contact surface between the rubber suction cup (451) and the guide rail is a circular protrusion; The air extraction pipe (452) is arranged on the track piece (42), one end of the air extraction pipe (452) is connected to the base of the rubber suction cup (451), and the other end passes through the track piece (42) and is connected to the air extraction valve (454); The distance adjustment spring (453) is sleeved on the exhaust pipe (452), and the two ends of the distance adjustment spring (453) respectively abut against the top surface of the rubber suction cup (451) and the bottom surface of the track shoe (42); The air extraction valve (454) is slidably arranged in the dovetail groove (47), and an air extraction groove (457) is provided on one side of the air extraction valve (454) close to the air extraction hole (48), and the air extraction groove (457) is arranged along the length direction of the dovetail groove (47); The air extraction valve (454) is used to allow the air stop core (441) to extend into the air extraction hole (48) during the movement of the track piece (42) in the dovetail groove (47), so that the air extraction groove (457) is connected to the air extraction hole (48).

7. The vertical guide rail climbing robot according to claim 5, characterized in that: The matching clearance between the dovetail groove (47) and the air extraction valve (454) is ≤0.1 mm, and a sealing rubber gasket is provided on the matching surface.

8. A trajectory control method for a vertical guide rail climbing robot according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Data fusion and state estimation: The data collected by the laser radar (32) and sensors are fused using the Kalman filter algorithm: ; in, is the state estimate at the previous moment, is the Kalman gain, is the current observation value, H is the observation matrix; S2: Adaptive control algorithm optimization: According to the position error e(t)= - , update the parameters through the adaptive control gain adjustment formula: in, is the learning rate (range 0.01-0.1).

9. A sliding mode control method for a vertical guide rail climbing robot according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Define the sliding surface: ; S2: The control law is: ; S3: Through Lyapunov function Optimize weight coefficient And control gain K to suppress trajectory chattering.

10. A method for controlling air pressure of a vertical guide rail climbing robot according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: System state equation modeling: ； in is the ambient atmospheric pressure, is the cavity pressure, is the cavity volume, is the flow rate of the air pump (431), To control the flow rate change rate of the input air pump (431); S2: Adaptive extended Kalman filter parameter identification: The sliding window method updates the measurement noise and uses the residual Dynamically adjust measurement noise: Where m is the sliding window length (range 5-10); The forgetting factor method updates the process noise and uses the new information Dynamically adjust process noise: in, is the forgetting factor, ranging from 0.9 to 0.99; S3: Nonlinear Model Predictive Control Solution: Cost function construction: in is the target stable air pressure value, They are tracking performance weight, energy consumption weight and output stability weight respectively; Constraints: Control Input That is, the flow rate change rate of the air pump (431) should be controlled within arrive Between, take the virtual input , construct the constraint function: The cost function is updated as: Where R is the constraint penalty coefficient.

Images