A quadrupedal gecko robot and control system therefor

Abstract

The application provides a four-legged crawling gecko robot and a control system thereof. The four-legged crawling gecko robot comprises a trunk structure, four symmetrically distributed leg structures, and four foot suction cups. The four-legged crawling gecko robot further comprises a motor driving system. A damping turntable is used as a waist joint for connecting a front trunk and a rear trunk of the trunk structure, so that the robot swings left and right. A longitudinal servo steering engine is used as a shoulder joint for connecting the trunk structure and the leg structure, so as to control the lifting and falling of the leg structure. A first transverse servo steering engine is used as an elbow joint for connecting the longitudinal servo steering engine and the leg structure, so as to drive the movement of the leg structure. A second transverse servo steering engine is used as a wrist joint for connecting the leg structure and the foot suction cup, so as to adjust the foot landing point position of the foot suction cup. The four-legged crawling gecko robot can realize flexible movement in a complex environment and maintain the stability of movement in a vertical plane.

Description

Technical Field

[0001] This application belongs to the field of robot control technology, and in particular relates to a quadrupedal crawling gecko robot and its control system. Background Technology

[0002] With the rapid development of robotics technology, exploring the efficient locomotion mechanisms of organisms in nature and applying them to robot design has become a popular research area. Geckos, renowned for their exceptional climbing abilities, offer valuable inspiration for biomimetic robot design through their unique attachment system and flexible locomotion patterns. However, achieving stable movement of gecko robots in complex environments, especially in vertical planes and effectively avoiding obstacles, requires solving numerous technical challenges. Traditional robots, when performing climbing tasks, are often limited by fixed attachment devices and complex control systems, making it difficult to exhibit the flexibility and adaptability of geckos in changing environments. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a quadrupedal crawling gecko robot and its control system. The torso structure uses driven axes to simulate the rotational movement of a gecko's waist to enhance the robot's movement flexibility. Each leg structure has one longitudinal servo motor and two lateral servo motors, which can realize different gaits to adapt to the robot's movement in different environments. A CPG control model is constructed using a Hopf oscillator to provide a stable rhythmic signal for the robot's movement. A barometric pressure sensor is used to introduce feedback signals to realize gait control of the robot on complex terrain with obstacles. The output robot foot trajectory control signal is further optimized using curve fitting methods to ensure the stability of the robot's movement in the vertical plane.

[0004] This application provides a quadrupedal crawling gecko robot, including a torso structure, four symmetrically distributed leg structures, and four foot suction cups; it also includes a motor drive system.

[0005] The damping turntable, acting as a waist joint (joint0), is used to connect the front and rear trunks of the torso structure, enabling the robot to swing left and right.

[0006] The longitudinal servo motor, acting as shoulder joint 1, is used to connect the torso structure and the leg structure to control the raising and lowering of the leg structure.

[0007] The first lateral servo motor, acting as elbow joint 2, is used to connect the longitudinal servo motor to the leg structure in order to drive the movement of the leg structure.

[0008] The second lateral servo motor, acting as a wrist joint 3, is used to connect the leg structure and the foot suction cup to adjust the landing point position of the foot suction cup.

[0009] Furthermore, it also includes an air pump drive system:

[0010] A vacuum pump is used to remove air from the foot suction cups and, in conjunction with the opening and closing of the solenoid valve, enables the foot suction cups to be attracted and detached.

[0011] A pressure sensor is used to monitor the adsorption state of the foot suction cup and output a high-level signal indicating successful adsorption, or a low-level signal indicating unsuccessful adsorption.

[0012] The foot suction cup, vacuum pump, and solenoid valve are connected using a three-way conduit.

[0013] Furthermore, the longitudinal servo motor is a 15kg.cm PWM servo motor, suitable for lifting and lowering the leg structure with low torque;

[0014] The first and second lateral servo motors are 45 kg·cm servo motors, which are suitable for the crawling movement of the robot with high torque in the vertical plane;

[0015] The rotation angle range of the first lateral servo motor is set to [-45°, 45°], and the rotation angle range of the second lateral servo motor is set to [0°, 135°], so that the leg structures do not collide during rotation.

[0016] This application also provides a control system for a quadrupedal crawling gecko robot, applied to the motor drive system and air pump drive system of the robot as described above; the control system includes:

[0017] The CPG control module is used to output periodic initial control signals based on the input CPG initial configuration parameters;

[0018] The control processing module is used to formulate the robot's gait switching and stride adjustment strategies, generate foot movement trajectories based on the initial control signal, and output feedback signals to the robot based on the real-time received monitoring signals of the foot suction cups.

[0019] The trajectory planning module is used to fit the foot movement trajectory to ensure continuous change in velocity and acceleration, and to convert the fitted foot movement trajectory into joint control signals for output.

[0020] The leg joints include: shoulder joint 1, elbow joint 2, and wrist joint 3;

[0021] The robot's gait includes: triped gait, diagonal gait, and stationary turning gait.

[0022] Furthermore, when outputting periodic initial control signals, the CPG control module is specifically used for:

[0023] A symmetrically coupled network consisting of four Hopf oscillators is used to output a reference signal that enables the movement of each leg structure of the robot.

[0024] The reference signal is mapped to the initial control signal for each leg joint and foot suction cup;

[0025] The initial control signals include: leg lift signal, leg drop signal, leg movement signal, foot adsorption signal, and foot detachment signal.

[0026] Furthermore, when outputting feedback signals to the foot suction cup, the control processing module is specifically used for:

[0027] The movement phases of each leg structure are determined; the movement phases include: the support phase and the swing phase;

[0028] Acquire the first real-time monitoring signal of the foot suction cup connected to the leg structure during the support phase;

[0029] If the first real-time monitoring signal is a high-level signal, then proceed to the next step;

[0030] If the first real-time monitoring signal is a low-level signal, then output a first feedback signal to control the leg structure to lift up and fall again, and reacquire the second real-time monitoring signal for the foot suction cup;

[0031] If the second real-time monitoring signal is a high-level signal, then proceed to the next step;

[0032] If the second real-time monitoring signal is a low-level signal, a second feedback signal is output to adjust the landing point position of the foot suction cup, and a third real-time monitoring signal for the foot suction cup is acquired again until the foot suction cup is successfully attached.

[0033] Furthermore, the trajectory planning module fits the foot movement trajectory in the following way:

[0034] The foot motion trajectory is fitted using a quartic quasi-uniform B-spline function as shown in the following formula, resulting in the fitted foot motion trajectory P(u):

[0035]

[0036] In the formula, P i As control points, n+1 points at equal intervals are selected on the foot movement trajectory, and the curve degree is k; B i,k (u) is a B-spline basis function, usually defined using the Cox-deBoor recursive formula;

[0037] The Cox-deBoor recursive formula is as follows:

[0038]

[0039] In the formula, u i For a non-decreasing sequence, u i The values ​​are as follows:

[0040]

[0041] Among them, the non-decreasing sequence u i The node vector U is formed, and the expression for U is as follows:

[0042] U = {u0, u1, u2, ..., u} n+k ,u n+k+1 ,u n+k+2}

[0043] The quadrupedal crawling gecko robot and its control system provided in this application have the following technical effects: 1) The torso structure uses a driven axis to simulate the rotational movement of the gecko's waist to enhance the robot's movement flexibility; each leg structure has one longitudinal servo motor and two lateral servo motors, which can realize different gaits to adapt to the robot's movement in different environments; 2) A CPG control model is constructed using a Hopf oscillator to provide a stable rhythmic signal for the robot's movement; and a pressure sensor is used to introduce feedback signals to realize gait control of the robot on complex terrain with obstacles; 3) The output robot foot trajectory control signal is further optimized using a curve fitting method to ensure the stability of the robot's movement in the vertical plane. Attached Figure Description

[0044] Figure 1 This paper shows an overall structural diagram of the quadrupedal crawling gecko robot provided in an embodiment of this application;

[0045] Figure 2 A diagram showing the correspondence between the joints of the quadrupedal gecko robot provided in the embodiments of this application is shown;

[0046] Figure 3 This paper shows an abstract structural diagram of a single-leg linkage of a quadrupedal gecko robot provided in an embodiment of this application.

[0047] Figure 4 This illustration shows a motion space diagram of a robotic arm, analyzed using the Monte Carlo method and enclosed by four circles, provided in an embodiment of this application.

[0048] Figure 5 This application provides a schematic diagram and phase diagram of the tripedal gait of a quadrupedal crawling gecko robot.

[0049] Figure 6 This paper shows a pseudo-diagonal gait diagram and phase diagram of the quadrupedal crawling gecko robot provided in an embodiment of this application;

[0050] Figure 7 This paper shows a schematic diagram and phase diagram of the stationary turning gait of the quadrupedal gecko robot provided in an embodiment of this application;

[0051] Figure 8 This illustration shows a schematic diagram of the mutual coupling of three Hopf oscillators provided in an embodiment of this application;

[0052] Figure 9 A schematic diagram of the CPG control module of the quadrupedal gecko robot provided in an embodiment of this application is shown;

[0053] Figure 10 This application provides a diagram showing the correspondence between CPG joint control signals according to an embodiment.

[0054] Figure 11 A schematic diagram of the feedback action provided in an embodiment of this application is shown;

[0055] Figure 12 This paper shows a force analysis diagram of the quadrupedal gecko robot provided in this embodiment when its right front leg and left hind leg are supporting legs.

[0056] Figure 13 This paper shows the velocity and acceleration curves of the foot motion trajectory directly generated by the initial control signal output by the CPG control module provided in this embodiment of the application in the XYZ plane.

[0057] Figure 14 The fitted foot motion trajectory diagram and velocity and acceleration change diagram provided in the embodiments of this application are shown. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this technical solution clearer, the following detailed description, in conjunction with specific embodiments, further illustrates this technical solution. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of this technical solution.

[0059] Example 1: Quadrupedal Gecko Robot

[0060] Please see Figure 1 , Figure 1 The diagram shown is an overall structural diagram of the quadrupedal gecko robot provided in an embodiment of this application. Figure 1 As shown in (a), the quadrupedal gecko robot includes a torso structure, four symmetrically distributed leg structures, and four foot suction cups.

[0061] The quadrupedal gecko robot also includes a motor drive system: a damping turntable as a waist joint (joint 0) connects the front and rear torsos of the body structure to allow the robot to swing left and right; a longitudinal servo motor as a shoulder joint (joint 1) connects the body structure and the leg structure to control the lifting and lowering of the leg structure; a first lateral servo motor as an elbow joint (joint 2) connects the longitudinal servo motor and the leg structure to drive the movement of the leg structure; and a second lateral servo motor as a wrist joint (joint 3) connects the leg structure and the foot suction cups to adjust the landing point of the foot suction cups.

[0062] In this structure, to mimic the waist movement of a gecko, the torso is divided into front and rear parts, connected by a damping turntable as a driven shaft in the middle. This allows the robot's waist to swing left and right during movement. The waist structure design enhances the robot's mobility, enabling it to turn in place. The robot's leg structure is as follows... Figure 1 As shown in (b), the four leg structures are symmetrically distributed, and each leg has three active degrees of freedom. Unlike the two vertical axes and one horizontal axis of the conventional quadruped robot leg structure, the legs of this application adopt a structural design of one vertical axis and two horizontal axes. The vertical axis only controls the lifting and lowering of the leg, while the horizontal axis is responsible for driving the robot to move and adjusting the position of the footing point. This design makes the robot's feet wobble less and helps the body's center of mass to be close to the crawling surface to reduce the overturning moment. Figure 1 (c) shows the suction cups on the gecko robot's feet. This application uses a self-designed pneumatic flexible biomimetic octopus suction cup. This suction cup is characterized by its small size and strong suction force, enabling stable adhesion while supporting the weight of the gecko robot. See here. Figure 2 The diagram showing the correspondence between the joints of each leg of the quadrupedal gecko robot provided in the embodiments of this application.

[0063] The longitudinal servo motor is a 15 kg·cm PWM servo motor, suitable for lifting and lowering the leg structure with small torque; the first and second lateral servo motors are 45 kg·cm servo motors, suitable for the robot's crawling movement with large torque in the vertical plane.

[0064] In this structure, since the shoulder joint 1 only needs to bear the lifting and lowering of a single leg and the torque change is small, a 15kg.cm PWM servo motor is used. However, when crawling on a vertical plane, the elbow joint 2 and wrist joint 3 need to bear up to half of the robot's overall weight, so a 45kg.cm servo motor with greater torque is selected.

[0065] As an example, the motor parameters are shown in Table 1 below:

[0066] Table 1. Motor Parameter Table

[0067]

[0068]

[0069] The rotation angle range of the first lateral servo motor is set to [-45°, 45°], and the rotation angle range of the second lateral servo motor is set to [0°, 135°], so that the leg structures do not collide during rotation.

[0070] In this structure, since the four legs of the robot in this application have the same structure, taking the right front leg as an example, ignoring the shoulder joint of the leg, the leg can be abstracted as a planar two-degree-of-freedom link. Please refer to [link to relevant documentation]. Figure 3 The diagram shown is an abstract structural diagram of a single-leg linkage of a quadrupedal gecko robot. A Cartesian coordinate system is established with the center point of axis 1 as the origin. The coordinates of the leg landing point are:

[0071]

[0072] In the formula, (x t ,y t ) represents the coordinates of the landing point, L1 and L2 represent the lengths of link 1 and 2 respectively, and θ1 and θ2 represent the rotation angles of link 1 and 2.

[0073] To ensure the robotic leg rotates without collisions, we define θ1∈[-45,45] and θ2∈[0,135]. The Monte Carlo method is used to analyze the robotic arm's motion space, and the results are as follows: Figure 4 (a). It is easy to see that the motion space of the two-degree-of-freedom linkage robotic arm is composed of Figure 4 (b) The enclosed space is formed by the four circles shown. Formula 2 below is the equation of the four circles involved:

[0074]

[0075] In the formula, θ 1min θ 1max The minimum and maximum values ​​that θ1 can reach, θ 2min θ 2max The minimum and maximum values ​​that θ2 can reach.

[0076] The quadrupedal gecko robot also includes an air pump drive system: a vacuum pump, used to expel air from the foot suction cups, and in conjunction with the opening and closing of the solenoid valve, to achieve the adsorption and detachment of the foot suction cups; a pressure sensor, used to monitor the adsorption status of the foot suction cups, and output a high-level signal indicating successful adsorption, or a low-level signal indicating unsuccessful adsorption; wherein, the foot suction cups, vacuum pump and solenoid valve are connected by a three-way conduit.

[0077] In this structure, four foot suction cups are driven by a small air pump. A three-way conduit connects the foot suction cups, a vacuum pump, and a solenoid valve. The vacuum pump removes air from the foot suction cups to achieve adsorption. The solenoid valve switches between open and closed states to enable rapid adsorption and detachment of the foot suction cups. An air pressure sensor monitors the adsorption status of the foot suction cups to achieve closed-loop control.

[0078] As an example, the parameters of the vacuum pump are shown in Table 2 below:

[0079] Table 2. Vacuum Pump Parameter Table

[0080] parameter Parameter value weight 100g size 32*32*65mm Operating voltage 5-12V Operating current ＜0.45A Maximum vacuum -450mmHg Control method Electronic control

[0081] Example 2: Control System for a Quadrupedal Gecko Robot

[0082] First, let's introduce the movement state of the quadrupedal gecko robot:

[0083] During the movement of the quadrupedal gecko robot, the movement of each leg structure is divided into two phases depending on the state of motion:

[0084] 1) Support Phase: To ensure the robot's leg structure is stably supported on the climbing surface, a vacuum pump is activated to extract air from the foot suction cups, while a solenoid valve closes to prevent external air from entering, thus creating and maintaining a vacuum state inside the foot suction cups. This state allows the foot suction cups to firmly adhere to the climbing surface, achieving relative stillness between the leg structure and the climbing surface, providing a stable support foundation for the robot.

[0085] 2) Swinging Phase: In this phase, the vacuum pump stops working while the solenoid valve opens, allowing external air to enter the foot suction cups, thus balancing the air pressure inside and outside the suction cups. This change weakens the adhesion between the foot suction cups and the climbing surface, eventually causing them to detach, allowing the robot's leg structure to swing freely, preparing for the next movement or posture adjustment.

[0086] Based on the alternating switching of the support and swing phases of the leg structure, this application designs three gaits for a quadrupedal gecko robot that can be used for movement under different environmental conditions: tripedal gait, pseudo-diagonal gait, and stationary turning gait. Specifically:

[0087] 1) Triped gait: such as Figure 5The diagram shows the tripedal gait and phase diagram of the quadrupedal crawling gecko robot provided in this application, with the robot's lumbar joint (joint0) remaining fixed. In the first stage, the robot sequentially swings its leg structures to change the landing point, thus moving the leg structures forward. During this process, except for the currently moving leg structure, the foot suction cups connected to the other three leg structures remain in an adhesive state to ensure the overall stability of the robot. In the second stage, after all four leg structures have reached their predetermined positions, the foot suction cups connected to the four leg structures remain in an adhesive state. The robot moves forward by synchronously and with equal amplitude swinging of the four leg structures. During this process, it is necessary to ensure that the swinging movements of the two left leg structures and the two right leg structures are symmetrical in amplitude and timing.

[0088] 2) Pseudo-diagonal gait: such as Figure 6 The diagram shows a pseudo-diagonal gait and phase diagram of the quadrupedal gecko robot provided in this embodiment. The pseudo-diagonal gait, a simplified form of the tripedal gait, exhibits a unique gait pattern during robot movement. In this gait, the robot's four leg structures swing synchronously but in groups. Two diagonal leg structures (i.e., the right front leg and the left hind leg, or the left front leg and the right hind leg) are in the same motion phase. Specifically, when the right front leg and the left hind leg maintain stable contact with the contact surface (i.e., in an adsorption state) and swing to propel the robot forward, the left front leg and the right hind leg are in a detached state, swinging to switch to a new foothold. Subsequently, the roles of these two sets of leg structures alternate; the left front leg and the right hind leg enter a supporting state, while the right front leg and the left hind leg enter the detachment phase, repeating this cycle. This alternating switching reduces the three movements of the tripedal gait to one, increasing the robot's movement speed.

[0089] 3) Stationary turning gait: such as Figure 7 The diagram shows the in-situ turning gait and phase diagram of the quadrupedal gecko robot provided in this embodiment. The in-situ turning gait can be seen as the robot rotating independently around a central driven axis in two parts: front and rear. When the robot turns left, the right front leg and left hind leg are active swinging legs, swinging back in the direction of turning to rotate the main body of the robot, while the left front leg and right hind leg act as supporting legs and do not swing. When the robot turns right, the swinging leg and supporting leg states switch, with the left front leg and right hind leg becoming active swinging legs, and the right front leg and left hind leg becoming supporting legs. Furthermore, it is easy to see from the robot's rotation process that the swing amplitude of the robot's elbow joint joint2 directly reflects its turning angle, which simplifies the complexity of turning control.

[0090] like Figure 7As shown, in a stationary turning gait, the robot's position after rotation is constrained by the swing distance of each leg. Therefore, it is necessary to calculate the relationship between the joint swing angle and the robot's landing point after rotation. Specifically:

[0091] Establish a Cartesian coordinate system with the robot's center point as the origin. R1 is the radius of the robot's circumcircle, H1 is the distance between the robot's front and rear legs, H2 is the width of the robot's main body, L1 is the length of the robot's elbow link, R2 is the distance from the robot's wrist joint (joint3) to the center point of the foot suction cup, and θ... R Let θ be the turning angle of the robot. i Let θ be the initial angle of the robot's turning elbow joint joint2. t Let θ be the angle of the elbow joint joint2 after it swings during robot turning, and θ be the angle during rotation. t =θ i -θ R The relationship between the landing point (x, y) of the rotating foot and the robot's rotation angle is as follows:

[0092]

[0093] This application utilizes a quadrupedal crawling gecko robot control system to achieve changes in the motion state of a single leg structure and switching of the robot's gait.

[0094] The structure and function of the quadrupedal gecko robot's control system will be introduced next:

[0095] The control system for the quadrupedal gecko robot includes:

[0096] The CPG control module outputs periodic initial control signals based on the input CPG initial configuration parameters. The control processing module formulates gait switching and stride adjustment strategies for the robot, generates foot motion trajectories based on the initial control signals, and outputs feedback signals to the robot based on real-time monitoring signals of the foot suction cups. The trajectory planning module fits the foot motion trajectory to ensure continuous changes in velocity and acceleration, and converts the fitted foot motion trajectory into joint control signals for output. The leg joints include: shoulder joint (joint1), elbow joint (joint2), and wrist joint (joint3). The robot's gait includes: tripedal gait, diagonal gait, and stationary turning gait.

[0097] Furthermore, when outputting periodic initial control signals, the CPG control module is specifically used for:

[0098] A symmetrically coupled network consisting of four Hopf oscillators is used to output a reference signal that enables the movement of each leg structure of the robot.

[0099] The reference signal is mapped to the initial control signal for each leg joint and foot suction cup;

[0100] The initial control signals include: leg lift signal, leg drop signal, leg movement signal, foot adsorption signal, and foot detachment signal.

[0101] Specifically, compared to other types of oscillators, the Hopf oscillator is favored for its simplicity, reliability, and adaptability. By adjusting its key parameters, such as following speed, limit cycle radius, and system natural frequency, the robot's gait characteristics can be easily altered to adapt to different motion requirements and environmental changes. Therefore, a central pattern generator based on a Hopf oscillator not only provides stable gait control for the robot but also endows it with the ability to flexibly adapt to various complex environments.

[0102] Next, we will introduce the idea of ​​using a Hopf oscillator to build a CPG control module:

[0103] The Hopf oscillator is a nonlinear oscillator, and its coupling model is described by the ordinary differential equation shown in equation (4):

[0104]

[0105] In the formula, x i ,y i It is the state variable of the i-th oscillator, and also the output signal of the oscillator; α is the first derivative of x and y with respect to time; α is the convergence weight, controlling the convergence speed of the oscillator; μ is the amplitude coefficient, used to control the amplitude of the oscillator; w represents the natural frequency of the oscillator; p xn and p yn This is the coupling term of the oscillators, representing a total of n oscillators coupled to each other; It is the phase difference between oscillator i and oscillator j.

[0106] Here, as Figure 8 The diagram shows three Hopf oscillators coupled together according to an embodiment of this application. The output state variables of the three oscillators are {x1 y1}, {x2 y2}, and {x3 y3}, respectively, where p n (q i )=[p xn (i) p yn (i)] T This represents the coupling term between oscillator n and oscillator i; the three oscillators influence each other through the coupling term, causing their state variables to have a fixed phase difference.

[0107] The quadrupedal gecko robot of this application has three active degrees of freedom for each leg. Therefore, a symmetrically coupled network composed of four Hopf oscillators is used to output a reference signal for the movement of each leg. This reference signal is then mapped to preliminary control signals for the three joints of the leg and the foot air pump, ultimately yielding the following result: Figure 9 The diagram shown is a schematic of the CPG control module of the quadrupedal crawling gecko robot provided in this embodiment of the application.

[0108] Actual testing revealed that when the swing leg switches to the supporting leg, if the leg suction cup falls directly to a position parallel to the climbing surface, a complete seal may not be formed between the suction cup and the climbing surface. By excessive downward pressure, the air between the suction cup and the climbing surface can be expelled, forming an initial seal. This initial seal is a prerequisite for the suction cup to generate effective suction force. At the same time, after the suction cup is attached, the leg needs to return to a position parallel to the suction plane to ensure that each supporting leg is at the same height from the climbing surface. To achieve the action of excessive downward pressure of the leg and return to the original position, the leg-lifting signal output by the CPG network is used as the trigger signal, and two superimposed sine signals are used as the actual waveform output signal. The modified oscillator model has a mapping relationship between the rotation angle of the three joints of a single leg and the corresponding oscillator output signal as shown in the following equation (5):

[0109]

[0110] In the formula, θ h and θ k The rotation angles of elbow joint 2 and wrist joint 3 are A, respectively. h A k θ represents the range of change in the angle between the two joints. M ε(t) represents the rotation angle of the shoulder joint (joint1), ε(t) represents the unit step signal, and H represents the suction cup air pump enable control signal. A value of 1 indicates that the suction cup is closed when it is suctioned, and a value of 0 indicates that the suction cup is released and detached.

[0111] As an example, such as Figure 10 The diagram shown is a mapping of CPG joint control signals provided in an embodiment of this application.

[0112] Furthermore, when outputting feedback signals to the foot suction cup, the control processing module is specifically used for:

[0113] The movement phases of each leg structure are determined; the movement phases include: the support phase and the swing phase;

[0114] Acquire the first real-time monitoring signal of the foot suction cup connected to the leg structure during the support phase;

[0115] If the first real-time monitoring signal is a high-level signal, then proceed to the next step;

[0116] If the first real-time monitoring signal is a low-level signal, then output a first feedback signal to control the leg structure to lift up and fall again, and reacquire the second real-time monitoring signal for the foot suction cup;

[0117] If the second real-time monitoring signal is a high-level signal, then proceed to the next step;

[0118] If the second real-time monitoring signal is a low-level signal, a second feedback signal is output to adjust the landing point position of the foot suction cup, and a third real-time monitoring signal for the foot suction cup is acquired again until the foot suction cup is successfully attached.

[0119] Next, we will introduce the feedback mechanism for dealing with uneven ground contact at the footing points of the four groups of crawling gecko robots:

[0120] To ensure the stability of the robot's movement, this application adds a feedback action; for details, please refer to [link / reference needed]. Figure 11 The diagram shown is a feedback action illustration provided in an embodiment of this application. Air pressure sensors are installed on the four suction cups on the robot's feet to provide real-time feedback for closed-loop control; when the legs switch from a swaying state to a supported state, an air pump extracts air from the suction cups, causing the suction cups to adhere to a flat surface, such as... Figure 11 As shown in (a), at this time we need to detect whether the air pressure value inside the suction cup has reached the safety threshold. If the air pressure value inside the support leg suction cup reaches the safety threshold, the air pressure sensor outputs a high-level signal. If it fails to reach the safety threshold, the air pressure sensor outputs a low-level signal. By detecting the sensor signal, the control processing layer will activate two actions to help the robot walk smoothly on the climbing surface.

[0121] If the air pressure sensor outputs a low-level signal, it indicates that the suction cup has failed to adhere successfully to the flat surface. In this case, the suction leg will be controlled to lift again and repeat the lowering action. If the air pressure value still does not reach the target after lowering again, it can be determined that the landing surface of the leg is uneven, causing the suction cup to not fully adhere to the ground. Figure 11 (b) At this point, the control processing layer will adjust the output signal waveform to change the swing angle of the robot's legs and guide the robot to reselect a suitable landing point; if the air pressure sensor outputs a high-level signal, it can be inferred that the suction cup has successfully adhered to the plane and the robot will normally execute the next action.

[0122] The four-group crawling gecko robot control system of this application, by introducing a feedback mechanism, enables the robot to avoid obstacles in complex terrain.

[0123] Furthermore, the trajectory planning module fits the foot movement trajectory in the following way:

[0124] The foot motion trajectory is fitted using a quartic quasi-uniform B-spline function as shown in the following formula, resulting in the fitted foot motion trajectory P(u):

[0125]

[0126] In the formula, P i As control points, n+1 points at equal intervals are selected on the foot movement trajectory, and the curve degree is k; B i,k (u) is a B-spline basis function, usually defined using the Cox-deBoor recursive formula;

[0127] The Cox-deBoor recursive formula is as follows:

[0128]

[0129] In the formula, u i For a non-decreasing sequence, u i The values ​​are as follows:

[0130]

[0131] Among them, the non-decreasing sequence u i The node vector U is formed, and the expression for U is as follows:

[0132] U = {u0, u1, u2, ..., u} n+k ,u n+k+1 ,u n+k+2}

[0133] Next, we will introduce the design philosophy of the trajectory planning module:

[0134] Please see Figure 12 The diagram shown in this application illustrates the force analysis of the quadrupedal gecko robot when its right front and left hind legs are supporting legs. When the robot is climbing a vertical wall, with its right front and left hind legs in a supporting state and its left front and right hind legs in a swinging state, the force distribution of the supporting legs in a stationary state is as follows: Figure 12 As shown in (a), the two supporting legs are subjected to a vertically downward component of the robot's gravity F along the link direction. N Frictional force F in the opposite direction S .

[0135] When the robot changes from a stationary state to a moving state, such as Figure 12 As shown in (b), the two supporting legs are stationary relative to the ground. By swinging the joints of the supporting legs downwards, the robot's center of gravity changes from stationary to moving upwards. Let the instantaneous acceleration of the upward movement at this moment be a, and the maximum frictional force provided by the legs be F. Smax ,have:

[0136]

[0137] In the formula, m' is the robot weight component acting on a single supporting leg, μ is the coefficient of friction between the foot suction cup and the climbing surface, and N is the suction force provided by the foot suction cup when it is in a vacuum state.

[0138] It is evident that the supporting legs, in addition to bearing the robot's weight, also need to withstand the acceleration forces generated during the robot's movement. Excessive acceleration increases the adhesion burden on the foot suction cups, potentially exceeding the frictional limit provided by the suction cups, causing unstable adhesion or even detachment. Therefore, during foot trajectory planning, it is crucial to ensure smooth acceleration along the foot trajectory to avoid abrupt accelerations that could negatively impact the suction cup adhesion.

[0139] Therefore, to ensure the stability of robot walking, the foot trajectory design should consider:

[0140] 1) Maintaining smooth and continuous motion: During movement, it is essential to ensure that the mechanism moves smoothly and continuously, avoiding significant vertical fluctuations, horizontal tilting, or forward and backward vibrations. This stability helps prevent the robot from losing stability due to uncoordinated movements and reduces errors caused by fluctuations, ensuring that the robot smoothly swings to the predetermined landing point.

[0141] 2) Smoothness of foot movement trajectory: During the movement, it is necessary to ensure that the velocity and acceleration of the foot trajectory in the three directions of movement are smooth and continuous, so as to avoid the robot from generating excessively high instantaneous acceleration force during movement, which would affect the suction cup adsorption stability.

[0142] Figure 13 The figure shows the velocity and acceleration curves of the foot motion trajectory directly generated by the initial control signal output by the CPG control module provided in this embodiment of the application in the XYZ plane. It can be seen that the velocity and acceleration curves are not smooth, and the acceleration curve has obvious abrupt changes.

[0143] Therefore, this application uses a quasi-uniform B-spline curve to fit the foot motion trajectory. The quasi-uniform B-spline curve has third-order derivative continuity, which satisfies the requirement of a smooth acceleration curve without abrupt changes. The fitted foot motion trajectory and velocity and acceleration variation diagrams are shown below. Figure 14 As shown in (a), (b), and (c), it can be seen that the fitted curve can achieve a smooth transition of velocity and acceleration curves at specific positions while ensuring that the trajectory remains basically unchanged.

[0144] The above content is only a preferred embodiment of the present invention. For those skilled in the art, many changes can be made in the specific implementation and application scope based on the ideas of the present invention. As long as these changes do not depart from the concept of the present invention, they all fall within the protection scope of this patent.

Claims

1. A quadrupedal gecko robot, comprising a torso structure, four symmetrically distributed leg structures, and four foot suction cups; characterized in that, It also includes the motor drive system: The damping turntable, acting as a waist joint (joint0), is used to connect the front and rear trunks of the torso structure, enabling the robot to swing left and right. The longitudinal servo motor, acting as shoulder joint 1, is used to connect the torso structure and the leg structure to control the raising and lowering of the leg structure. The first lateral servo motor, acting as elbow joint 2, is used to connect the longitudinal servo motor to the leg structure in order to drive the movement of the leg structure. The second lateral servo motor, acting as a wrist joint 3, is used to connect the leg structure and the foot suction cup to adjust the landing point position of the foot suction cup. It also includes an air pump drive system: A vacuum pump is used to remove air from the foot suction cups and, in conjunction with the opening and closing of the solenoid valve, enables the foot suction cups to be attracted and detached. A pressure sensor is used to monitor the adsorption state of the foot suction cup and output a high-level signal indicating successful adsorption, or a low-level signal indicating unsuccessful adsorption. The foot suction cup, vacuum pump, and solenoid valve are connected by a three-way conduit. It also includes a control system applied to the motor drive system and the air pump drive system: The CPG control module is used to output periodic initial control signals based on the input CPG initial configuration parameters; The control processing module is used to formulate the robot's gait switching and stride adjustment strategies, generate foot movement trajectories based on the initial control signal, and output feedback signals to the robot based on the real-time received monitoring signals of the foot suction cups. The trajectory planning module is used to fit the foot movement trajectory to ensure continuous change in velocity and acceleration, and to convert the fitted foot movement trajectory into joint control signals for output. The leg joints include: shoulder joint 1, elbow joint 2, and wrist joint 3. The robot's gait includes: tripedal gait, diagonal gait, and stationary turning gait; When outputting a periodic initial control signal, the CPG control module is specifically used for: A symmetrically coupled network consisting of four Hopf oscillators is used to output a reference signal that enables the movement of each leg structure of the robot. The reference signal is mapped to the initial control signal for each leg joint and foot suction cup; The initial control signals include: leg lift signal, leg drop signal, leg movement signal, foot adsorption signal, and foot detachment signal; To achieve the action of excessive leg compression and return to the original position, the leg-lifting signal output by the CPG network is used as the trigger signal, and two superimposed sine signals are used as the actual waveform output signals. The mapping relationship between the rotation angles of the three joints of a single leg and the corresponding oscillator output signals in the modified oscillator model is shown below: ； In the formula, and These are the rotation angles of the elbow joint (joint2) and the wrist joint (joint3), respectively. , This represents the range of change in the angle between the two joints. The rotation angle of the shoulder joint joint1. For unit step signal, This is the enable control signal for the suction cup air pump. A value of 1 indicates that the suction cup is closed and sucking air in, while a value of 0 indicates that the suction cup is released and detached. It is the first The state variables of an oscillator are also the output signals of the oscillator.

2. The quadrupedal crawling gecko robot as described in claim 1, characterized in that, The longitudinal servo motor is a 15kg.cm PWM servo motor, suitable for lifting and lowering the leg structure with low torque; The first and second lateral servo motors are 45 kg·cm servo motors, which are suitable for the crawling movement of the robot with high torque in the vertical plane; Wherein, the rotation angle range of the first horizontal servo motor is set as follows: The second lateral servo motor rotation angle range is This is to prevent the leg structures from colliding during rotation.

3. The quadrupedal crawling gecko robot as described in claim 1, characterized in that, When outputting a feedback signal to the foot suction cup, the control processing module is specifically used for: The movement phases of each leg structure are determined; the movement phases include: the support phase and the swing phase; Acquire the first real-time monitoring signal of the foot suction cup connected to the leg structure during the support phase; If the first real-time monitoring signal is a high-level signal, then proceed to the next step; If the first real-time monitoring signal is a low-level signal, then output a first feedback signal to control the leg structure to lift up and fall again, and reacquire the second real-time monitoring signal for the foot suction cup; If the second real-time monitoring signal is a high-level signal, then proceed to the next step; If the second real-time monitoring signal is a low-level signal, a second feedback signal is output to adjust the landing point position of the foot suction cup, and a third real-time monitoring signal for the foot suction cup is acquired again until the foot suction cup is successfully attached.

4. The quadrupedal crawling gecko robot as described in claim 1, characterized in that, The trajectory planning module fits the foot movement trajectory in the following way: The foot motion trajectory is fitted using a quartic quasi-uniform B-spline function as shown in the following formula to obtain the fitted foot motion trajectory. : ； In the formula, As control points, select equidistant points on the foot movement trajectory. There are points, and the curve degree is... ; For B-spline basis functions, they are usually defined using the Cox-deBoor recursive formula; The Cox-deBoor recursive formula is as follows: ； In the formula, It is a non-decreasing sequence. The values ​​are as follows: ； Among them, the non-decreasing sequence Constructing node vectors , The expression is as follows: .

Images