Control system of wall-climbing robot based on bionic gait control

Abstract

The application provides a control system of a wall-climbing robot based on bionic gait control. The method comprises the following steps: a diagonal foot asynchronous timing control module determines the release duration of each magnetic foot of the wall-climbing robot based on the asynchronous linkage characteristics of the gecko's diagonal feet and a pre-trained neural network model, and determines the asynchronous timing of the wall-climbing robot according to the release duration of each magnetic foot; a body posture self-adaptive adjustment module determines the target posture of the wall-climbing robot according to the posture angle deviation of the wall-climbing robot; a bionic obstacle avoidance collaborative control module determines an obstacle avoidance path according to the position of a target point, and controls the wall-climbing robot to perform obstacle avoidance movement in combination with the asynchronous timing of the wall-climbing robot and the target posture. In this way, by imitating the asynchronous linkage characteristics of the gecko's diagonal feet, the timing of the foot end of the wall-climbing robot, the body posture and the obstacle avoidance action are dynamically controlled, so as to improve the operation stability of the wall-climbing robot.

Description

Technical Field

[0001] This invention relates to the field of intelligent robots, and more specifically to a control system for a wall-climbing robot based on biomimetic gait control. Background Technology

[0002] In existing technologies, multi-legged wall-climbing robots generally adopt a "fixed-time symmetrical control" strategy, resulting in poor adaptability to complex curved surfaces, unstable adhesion, and weak obstacle-crossing ability. Although the biological gait characteristics of geckos are recognized, current technologies lack specific means to translate them into effective and engineerable robot control schemes. Summary of the Invention

[0003] This application aims to provide a control system for a wall-climbing robot based on biomimetic gait control, so as to improve the operational stability of the climbing robot.

[0004] Firstly, a control system for a wall-climbing robot based on biomimetic gait control is provided, the system comprising: The diagonal leg asynchronous timing control module is used to determine the release duration of each magnetic foot of the wall-climbing robot based on the asynchronous linkage characteristics of the diagonal legs of the gecko and a pre-trained neural network model, and to determine the asynchronous timing of the wall-climbing robot according to the release duration of each magnetic foot. The asynchronous linkage characteristics of the diagonal legs of the gecko include the wall curvature, the body tilt angle, the distance to the target point, and the magnetic foot adsorption pressure. The target point represents the endpoint of the motion planning of the wall-climbing robot. The body posture adaptive adjustment module is used to determine the target posture of the wall-climbing robot based on the posture angle deviation of the wall-climbing robot; The biomimetic obstacle-avoidance cooperative control module is used to determine the obstacle avoidance path based on the position of the target point, and control the wall-climbing robot to perform obstacle avoidance movements in combination with the asynchronous timing of the wall-climbing robot and the target posture.

[0005] Optionally, the pre-trained neural network model can be represented as:

[0006] in, Indicates the release time of the magnetic foot, For network parameters, C represents the wall curvature, D represents the fuselage tilt angle, and D represents the distance from the target point. Indicates the magnetic foot adsorption pressure. This indicates the preset maximum release time.

[0007] Optionally, the fuselage attitude adaptive adjustment module is also configured to: If the absolute value of the difference between the current attitude angle and the target attitude angle of the wall-climbing robot is greater than a preset angle threshold, the wall-climbing robot is controlled to adjust its attitude; or If the absolute value of the distance deviation between the wall-climbing robot and the target point is greater than a preset distance threshold, the wall-climbing robot is controlled to adjust its posture. The posture adjustment includes foot adjustment and torso adjustment.

[0008] Optionally, the fuselage attitude adaptive adjustment module is also configured to: Get the current wall inclination angle; Determine the horizontal reference attitude angle based on the current wall tilt angle; The target attitude angle is determined based on the horizontal reference attitude angle, the preset curvature compensation coefficient, the preset distance deviation compensation coefficient, the wall curvature, and the distance deviation from the target point.

[0009] Optionally, the target attitude angle is expressed as:

[0010] in, Indicates the target attitude angle. This represents the curvature compensation coefficient. This represents the distance deviation compensation coefficient. Indicates the horizontal reference attitude angle. This indicates the distance deviation from the target point.

[0011] Optionally, the fuselage attitude adaptive adjustment module is also configured to: The adsorption pressure at each foot of the wall-climbing robot is detected in real time. If the adsorption pressure at any foot of the wall-climbing robot is less than a preset rated value, the adsorption time at that foot is extended.

[0012] Optionally, the biomimetic obstacle-crossing cooperative control module is also configured to: Obtain the height of the obstacle and the coordinates of the target point; The lifting height of the obstacle-side foot of the wall-climbing robot is determined based on the height of the obstacle and a preset safety margin. The obstacle-side feet of the wall-climbing robot are controlled to move sequentially according to the lifting height, and the non-obstacle-side feet of the wall-climbing robot are controlled to increase the suction force according to a preset ratio to complete the obstacle avoidance action.

[0013] Based on the aforementioned control system for a wall-climbing robot using biomimetic gait control, the diagonal-leg asynchronous timing control module, based on the asynchronous linkage characteristics of a gecko's diagonal legs and a pre-trained neural network model, determines the release duration of each magnetic foot of the wall-climbing robot. Based on the release duration of each magnetic foot, it determines the asynchronous timing of the wall-climbing robot. The asynchronous linkage characteristics of the gecko's diagonal legs include the wall curvature, body tilt angle, distance to the target point, and magnetic foot adsorption pressure. The target point represents the endpoint of the wall-climbing robot's motion planning. The body posture adaptive adjustment module determines the target posture of the wall-climbing robot based on its posture angle deviation. The biomimetic obstacle-crossing cooperative control module determines the obstacle avoidance path based on the position of the target point and controls the wall-climbing robot's obstacle-avoidance movement by combining the asynchronous timing of the wall-climbing robot and the target posture. Thus, by mimicking the asynchronous linkage characteristics of a gecko's diagonal legs, the timing of the robot's feet, body posture, and obstacle-crossing actions are dynamically controlled to improve the operational stability of the wall-climbing robot. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the control system of the wall-climbing robot based on biomimetic gait control provided in the embodiments of this application; Figure 2 This is a flowchart of a diagonal asynchronous timing control module provided in a specific embodiment of this application; Figure 3 This is a flowchart of a fuselage attitude adaptive adjustment module provided in a specific embodiment of this application; Figure 4 This is a flowchart of a specific embodiment of the biomimetic obstacle-crossing cooperative control module provided in this application. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0016] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0017] The control system of the wall-climbing robot based on biomimetic gait control provided in this application will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0018] Please see Figure 1 This is a schematic diagram of the control system of a wall-climbing robot based on biomimetic gait control provided in an embodiment of this application. Figure 1 As shown, the control system of this biomimetic gait control-based wall-climbing robot includes: The diagonal leg asynchronous timing control module is used to determine the release duration of each magnetic foot of the wall-climbing robot based on the asynchronous linkage characteristics of the gecko's diagonal legs and a pre-trained neural network model, and to determine the asynchronous timing of the wall-climbing robot according to the release duration of each magnetic foot. The asynchronous linkage characteristics of the gecko's diagonal legs include the wall curvature, the body tilt angle, the distance to the target point, and the magnetic foot adsorption pressure. The target point represents the endpoint of the wall-climbing robot's motion planning.

[0019] In the embodiments of this application, the diagonal leg asynchronous timing control module can mimic the asynchronous linkage characteristics of the diagonal legs of a gecko and dynamically determine the optimal adsorption / release timing of each leg through a pre-trained neural network model.

[0020] In one specific embodiment, the neural network model is a feedforward neural network containing one hidden layer (10 neurons), with the hidden layer using the ReLU activation function and the output layer using the Sigmoid activation function. The mathematical expression of this neural network model is:

[0021] in, Indicates the release time of the magnetic foot, For network parameters, C represents the wall curvature, D represents the fuselage tilt angle, and D represents the distance from the target point. Indicates the magnetic foot adsorption pressure. This indicates the preset maximum release time.

[0022] During the training of the neural network model, data acquisition is performed first. A high-speed motion capture system is used to capture data on geckos at various curvatures (0-5). Gait sequences of walking on surfaces with inclination angles (0-90°) were recorded, including the timing of each foot lifting and landing. The environmental parameters at each time step are also considered. Supervised learning is employed, with mean squared error (MSE) as the primary loss function, and a stability penalty term is added. ,in This represents the maximum fluctuation in other foot adsorption pressures during foot release. The network parameters are optimized using gradient descent to optimize the network output. It can not only mimic the gait of a gecko, but also suppress the shaking of the aircraft.

[0023] During the wall-climbing robot's movement, the system controller acquires sensor data in real time, including but not limited to wall curvature, robot tilt angle, distance to the target point, and magnetic foot adsorption pressure. This data is input into a pre-trained neural network to calculate the ideal release time for the two pairs of diagonal feet: left front / right rear and right front / left rear. The control system then controls the diagonal feet to perform a "release-move-adhere" sequence asynchronously, ensuring that at least three feet are always in an adsorbed state for stable walking. When the distance to the target D ≤ 0.3m, the network automatically outputs a shorter release time. (e.g., 0.1-0.5 seconds) to achieve fine adjustment and precise positioning.

[0024] The body posture adaptive adjustment module is used to determine the target posture of the wall-climbing robot based on the posture angle deviation of the wall-climbing robot.

[0025] In this embodiment, the body posture adaptive adjustment module can mimic the torso deformation of a gecko and adjust the body posture in real time through closed-loop control. Specifically, it can determine the magnetic foot joint angle adjustment command based on the current posture angle, wall curvature, the difference in adsorption pressure of each foot, and the distance deviation from the target point, so that it fits the wall surface.

[0026] First, the target attitude angle can be calculated based on the horizontal reference attitude angle, preset curvature compensation coefficient, preset distance deviation compensation coefficient, wall curvature, and distance deviation from the target point. The target attitude angle can be expressed as:

[0027] in, Indicates the target attitude angle. This represents the curvature compensation coefficient. This represents the distance deviation compensation coefficient. Indicates the horizontal reference attitude angle. This represents the distance deviation from the target point. The curvature compensation coefficient and the distance deviation compensation coefficient can be determined through regression analysis of gecko movement data.

[0028] If the absolute value of the difference between the current attitude angle and the target attitude angle of the wall-climbing robot is greater than a preset angle threshold, the robot is controlled to adjust its attitude. In a specific example, the absolute value of the difference between the current attitude angle and the target attitude angle being greater than the preset angle threshold can be expressed as: .

[0029] Alternatively, if the absolute value of the distance deviation between the wall-climbing robot and the target point is greater than a preset distance threshold, the wall-climbing robot can be controlled to adjust its posture. In a specific example, the absolute value of the distance deviation between the wall-climbing robot and the target point being greater than the preset distance threshold can be expressed as: .

[0030] Posture adjustment can include, but is not limited to, foot adjustment and torso adjustment. Foot adjustment can be achieved by finely adjusting the angle of the magnetic foot joints to shift the body's center of gravity towards the side requiring stronger adhesion. Torso adjustment can be achieved by controlling the middle of the body to make the overall body conform more closely to the curve of the wall.

[0031] During the posture adjustment process of the wall-climbing robot, if suction pressure is detected at any foot end... If the rated pressure value is reached, the adsorption time of that foot will be extended immediately, and the adjacent foot ends will be instructed to compensate by increasing the contact force or fine-tuning the angle, forming a closed loop of adsorption force to avoid positioning deviation.

[0032] The biomimetic obstacle-avoidance cooperative control module is used to determine the obstacle avoidance path based on the position of the target point, and control the wall-climbing robot to perform obstacle avoidance movements in combination with the asynchronous timing of the wall-climbing robot and the target posture.

[0033] In this embodiment, the biomimetic obstacle-crossing cooperative control module can mimic the gecko's intelligent obstacle-crossing strategy of "probing-adjusting-crossing". This module can be divided into a probing phase, an adjustment phase, and an obstacle-crossing phase.

[0034] Probe phase: The height of obstacles ahead is identified using a laser rangefinder. The visual positioning module continuously tracks the coordinates of the target point.

[0035] Adjustment phase: The obstacle-crossing path is planned using algorithms, and the lift height of the foot on the obstacle side is calculated.

[0036]

[0037] At the same time, the non-obstacle side foot is instructed to increase its suction force to 160% of the original suction force to resist the overturning moment generated by lifting the foot.

[0038] Obstacle-crossing phase: The controlled feet on the obstacle side are lifted step-by-step in the order of "forefoot → middle foot → hind foot" to cross the obstacle. The key point is that only one foot is lifted in each step, while the other feet remain attached. During each lifting and landing step, the diagonal foot asynchronous timing control module and the fuselage attitude adaptive adjustment module work simultaneously. The diagonal foot asynchronous timing control module dynamically adjusts the timing of the other feet to maintain a stable rhythm, while the fuselage attitude adaptive adjustment module fine-tunes the fuselage attitude and the angles of the other feet in real time to ensure the fuselage remains stable and the target is not lost during obstacle crossing.

[0039] The control system of the wall-climbing robot based on biomimetic gait control described above utilizes a diagonal-leg asynchronous timing control module. This module, based on the asynchronous linkage characteristics of a gecko's diagonal legs and a pre-trained neural network model, determines the release duration of each magnetic foot of the robot. Based on the release duration of each magnetic foot, it determines the asynchronous timing of the robot. The gecko's diagonal-leg asynchronous linkage characteristics include wall curvature, body tilt angle, distance to the target point, and magnetic foot adsorption pressure. The target point represents the endpoint of the robot's motion planning. A body posture adaptive adjustment module determines the target posture of the robot based on its posture angle deviation. A biomimetic obstacle-crossing cooperative control module determines the obstacle avoidance path based on the target point's position and, combined with the robot's asynchronous timing and target posture, controls the robot's obstacle-avoidance movement. Thus, by mimicking the asynchronous linkage characteristics of a gecko's diagonal legs, the timing of the robot's feet, body posture, and obstacle-crossing actions are dynamically controlled, improving the operational stability of the climbing robot.

[0040] Figure 2 This is a flowchart of a diagonal asynchronous timing control module provided in a specific embodiment of this application. Figure 2 As shown, the control flow of the diagonal asynchronous timing control module may include: Module initialization; Sensor data inputs: wall curvature C, fuselage tilt angle The distance D from the target point and the magnetic foot adsorption pressure ; Distance determination: Does it satisfy D≤0.3m? Neural network calculation: Calculate the release time of each leg while ensuring D ≤ 0.3m; Timing control: Diagonal feet execute release-move-adhesion asynchronously; Output instructions: timing of magnetic foot adsorption / release.

[0041] Figure 3 This is a flowchart of a fuselage attitude adaptive adjustment module provided in a specific embodiment of this application. Figure 3 As shown, the process of the fuselage attitude adaptive adjustment module may include: Module initialization; Sensor data inputs: current attitude angle, wall curvature C, difference in adsorption pressure at each foot, distance deviation. ; Attitude deviation judgment: Does it meet the requirements? or

[0042] Calculate the ideal attitude angle (i.e., the target attitude angle in this application): ; Adjustments performed: foot joint angle adjustment + bending of the flexible linkage of the fuselage; Real-time feedback: Detects adsorption pressure, extends adsorption time, and adjusts adjacent feet; Output command: Magnetic foot joint angle adjustment command.

[0043] Figure 4 This is a flowchart of a biomimetic obstacle-crossing cooperative control module provided in a specific embodiment of this application. Figure 3 As shown, the flow of the biomimetic obstacle-crossing cooperative control module may include: Module initialization; Obstacle detection: Laser rangefinders identify obstacle heights ; Exploration phase: Identify obstacle types and their impact on the path; Adjustment phase: Planning obstacle-crossing routes and calculating lift heights. ; Collaborative control: Non-obstacle foot measurement increases adsorption force by up to 160%; During the obstacle crossing phase: the feet on the obstacle side are raised and lifted in the order of "forefoot-middle foot-hind foot" to cross the obstacle; Return to normal mode: Obstacle crossing completed, return to normal walking mode.

[0044] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0046] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0047] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0048] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0049] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0050] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0051] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed by the present invention.

Claims

1. A control system for a wall-climbing robot based on biomimetic gait control, characterized in that, The control system includes: The diagonal leg asynchronous timing control module is used to determine the release duration of each magnetic foot of the wall-climbing robot based on the asynchronous linkage characteristics of the diagonal legs of the gecko and a pre-trained neural network model, and to determine the asynchronous timing of the wall-climbing robot according to the release duration of each magnetic foot. The asynchronous linkage characteristics of the diagonal legs of the gecko include the wall curvature, the body tilt angle, the distance to the target point, and the magnetic foot adsorption pressure. The target point represents the endpoint of the motion planning of the wall-climbing robot. The body posture adaptive adjustment module is used to determine the target posture of the wall-climbing robot based on the posture angle deviation of the wall-climbing robot; The biomimetic obstacle-avoidance cooperative control module is used to determine the obstacle avoidance path based on the position of the target point, and control the wall-climbing robot to perform obstacle avoidance movements in combination with the asynchronous timing of the wall-climbing robot and the target posture.

2. The system according to claim 1, characterized in that, The pre-trained neural network model is represented as follows: in, Indicates the release time of the magnetic foot, For network parameters, C represents the wall curvature, D represents the fuselage tilt angle, and D represents the distance from the target point. Indicates the magnetic foot adsorption pressure. This indicates the preset maximum release time.

3. The system according to claim 1, characterized in that, The fuselage attitude adaptive adjustment module is also configured to: If the absolute value of the difference between the current attitude angle and the target attitude angle of the wall-climbing robot is greater than a preset angle threshold, the wall-climbing robot is controlled to adjust its attitude; or If the absolute value of the distance deviation between the wall-climbing robot and the target point is greater than a preset distance threshold, the wall-climbing robot is controlled to adjust its posture. The posture adjustment includes foot adjustment and torso adjustment.

4. The system according to claim 3, characterized in that, The fuselage attitude adaptive adjustment module is also configured to: Get the current wall inclination angle; Determine the horizontal reference attitude angle based on the current wall tilt angle; The target attitude angle is determined based on the horizontal reference attitude angle, the preset curvature compensation coefficient, the preset distance deviation compensation coefficient, the wall curvature, and the distance deviation from the target point.

5. The system according to claim 3, characterized in that, The target attitude angle is expressed as: in, Indicates the target attitude angle. This represents the curvature compensation coefficient. This represents the distance deviation compensation coefficient. Indicates the horizontal reference attitude angle. This indicates the distance deviation from the target point.

6. The system according to claim 1, characterized in that, The fuselage attitude adaptive adjustment module is also configured to: The adsorption pressure at each foot of the wall-climbing robot is detected in real time. If the adsorption pressure at any foot of the wall-climbing robot is less than a preset rated value, the adsorption time at that foot is extended.

7. The system according to claim 1, characterized in that, The biomimetic obstacle-crossing cooperative control module is also configured to: Obtain the height of the obstacle and the coordinates of the target point; The lifting height of the obstacle-side foot of the wall-climbing robot is determined based on the height of the obstacle and a preset safety margin. The obstacle-side feet of the wall-climbing robot are controlled to move sequentially according to the lifting height, and the non-obstacle-side feet of the wall-climbing robot are controlled to increase the suction force according to a preset ratio to complete the obstacle avoidance action.

Images