A robot control method

By constructing a solution model and optimizing the algorithm to adjust the thrust magnitude and direction of the thrust component, the problem of traditional wall-climbing robots having difficulty adhering on non-ferromagnetic curved surfaces was solved, and stable movement with low energy consumption was achieved.

CN118625848BActive Publication Date: 2024-11-05HEBEI UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411111013.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-11-05
Estimated Expiration
2044-08-14

AI Technical Summary

Technical Problem

Traditional wall-climbing robots struggle to achieve effective adhesion on non-ferromagnetic curved surfaces. Existing duct-type adhesion methods cannot automatically adjust the direction and magnitude of thrust according to the surface conditions, resulting in high energy consumption or difficulty in movement.

Method used

By constructing a solution model, using optimization algorithms to adjust the thrust magnitude and direction of the thrust components, and combining particle swarm optimization to obtain the target thrust value and attitude angle, the automatic adjustment of the thrust components is achieved.

Benefits of technology

This reduces the robot's energy consumption on non-ferromagnetic curved surfaces and improves the stability and efficiency of movement on uneven surfaces.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118625848B_ABST
    Figure CN118625848B_ABST
Patent Text Reader

Abstract

The present invention provides a robot control method, which belongs to the field of robot automatic control technology. The robot includes a first platform, a mobile wheel group and a thrust assembly. They are respectively used to drive the first platform to move and provide thrust to the first platform. The method includes: constructing a conversion relationship between thrust and a second posture angle according to the robot's acceleration, first posture angle, total gravity, and friction coefficient with a curved wall, and obtaining a solution model. According to the solution model, an optimization algorithm is used to optimize the second posture angle and thrust size to obtain a target second posture angle and a target thrust value. Finally, the thrust assembly of the robot is controlled to move to the target second posture angle, and thrust is provided according to the target thrust value. In this way, the direction and size of the thrust can be automatically adjusted according to the actual inclination angle of the curved wall, so that the robot can run at a lower power while maintaining normal operation, thereby reducing energy consumption.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention generally relates to the technical field of robot automatic control, and in particular to a robot and a control method thereof. Background Art

[0002] Traditional wall-climbing robots achieve the wall-climbing effect by electromagnetically adsorbing ferromagnetic walls with electromagnetic devices. With the advancement of industrial service processes, non-ferromagnetic large components are increasingly used in the construction, petrochemical, transportation and other industries, resulting in frequent problems such as wind turbine blade breakage, bridge pier crack maintenance, and high-altitude falls on artificial vertical walls in recent years.

[0003] The adsorption mechanism is the core component of the wall-climbing robot. When the wall is a non-ferromagnetic surface, the traditional adsorption method can no longer achieve the adsorption effect and it is difficult to fit the curved surface. Once the adsorption mechanism cannot be adsorbed to the wall, the movement of the wall-climbing robot will not fit the wall surface, further invalidating the auxiliary functions of the robot. Its design and optimization are important challenges for the wall-climbing robot to adapt to the adsorption of non-ferromagnetic curved surfaces.

[0004] The solution in the prior art is to use a ducted fan as the adsorption mechanism, and use the fan to provide thrust to make the robot fit the wall, so that the wall-climbing robot can perform adsorption operations on non-ferromagnetic surfaces. This method can well solve the drawbacks of adsorption medium limitation, low load capacity, vacuum leakage, high cost and difficult design control.

[0005] However, existing duct-type adsorption methods often provide a constant thrust, and the thrust direction is generally fixed to the robot itself. When a wall-climbing robot moves on a curved wall, the thrust may be too large and consume additional power; or the thrust may be too small to support the movement of the wall-climbing robot on the curved wall. Therefore, it is urgent to find a control method that can automatically adapt to adjust the thrust direction and thrust size according to the surface conditions. Summary of the invention

[0006] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a robot control method.

[0007] The present invention provides a robot control method, wherein the robot comprises a first platform, and a moving wheel group and a thrust assembly installed on the first platform; the moving wheel group is used to drive the first platform to move along a curved wall; the thrust assembly is used to provide thrust to the first platform so that the moving wheel group abuts against the curved wall; the thrust provided by the thrust assembly is adjustable in size and direction;

[0008] Methods include:

[0009] Obtaining a solution model; the solution model is used to represent the relationship between the robot's acceleration, the first attitude angle, the robot's total gravity, the friction coefficient of the curved wall, the magnitude of the thrust provided by the thrust assembly, and the second attitude angle; the first attitude angle is the attitude angle of the robot in the world coordinate system; the second attitude angle is the attitude angle of the thrust provided by the thrust assembly in the robot body coordinate system;

[0010] Get the robot's real-time acceleration, real-time first attitude angle, total gravity and friction coefficient at the current moment;

[0011] Substituting the real-time acceleration, the real-time first attitude angle, the total gravity and the friction coefficient into the solution model, and using the optimization algorithm to optimize the magnitude of the thrust and the second attitude angle, to obtain a target thrust value and a target second attitude angle; the target thrust value is the minimum value to ensure that the robot's moving wheel group abuts against the curved wall; the target second attitude angle is the second attitude angle of the thrust when the magnitude of the thrust is equal to the target thrust value;

[0012] The thrust assembly is controlled to move to the target second attitude angle, and thrust is provided according to the target thrust value.

[0013] According to the technical solution provided by the present invention, the step of obtaining the solution model includes:

[0014] Determine, according to the first posture angle, the total gravity and the acceleration, the components of the resultant force acting on the robot along each axis of the robot body coordinate system to obtain a plurality of resultant force components;

[0015] Determine the components of the external force applied to the robot along each axis of the robot body coordinate system according to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust, and the friction coefficient, and obtain multiple force components;

[0016] For each axis of the robot body coordinate system, the resultant force component is respectively set equal to the force component to obtain the solution model.

[0017] According to the technical solution provided by the present invention, the steps of determining the components of the external force applied to the robot along each axis of the robot body coordinate system according to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust and the friction coefficient, and obtaining multiple force components include:

[0018] Determine, according to the first posture angle and the total gravity, the components of the total gravity along each axis of the robot body coordinate system to obtain a plurality of total gravity components;

[0019] According to the second posture angle and the magnitude of the thrust, components of the thrust along each axis of the robot body coordinate system are determined to obtain a plurality of thrust components;

[0020] According to the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the supporting force of the curved wall on the robot along each axis of the robot body coordinate system to obtain a plurality of supporting force components;

[0021] According to the friction coefficient, the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the friction force between the curved wall and the robot along each axis of the robot body coordinate system to obtain a plurality of friction force components;

[0022] The vector sums of the total gravity component, thrust component, support force component and friction force component along each axis of the robot body coordinate system are calculated to obtain a plurality of the force components.

[0023] According to the technical solution provided by the present invention, the optimization algorithm is a particle swarm algorithm;

[0024] The step of optimizing the thrust value and the second attitude angle by using an optimization algorithm to obtain a target thrust value and a target second attitude angle comprises:

[0025] S41: constructing a target optimization function according to the thrust;

[0026] S42: setting a search space for the second posture angle; setting the number of particles of the particle swarm algorithm, the initial position and initial velocity of each particle in the search space, and the maximum number of iterations; setting a termination condition according to the target optimization function;

[0027] S43: according to the current position of each particle in the search space, obtain the corresponding i-th second posture angle; the initial value of i is 1;

[0028] S44: Substituting the i-th second attitude angle into the solution model to calculate the i-th thrust;

[0029] S45: Substituting the i-th thrust into the target optimization function to calculate and obtain the i-th target optimization value;

[0030] S46: Update the position of each particle; update the speed of each particle;

[0031] S47: Determination: When the termination condition or the maximum number of iterations is reached, the historical optimal solution is used as the target second attitude angle and the target thrust value; otherwise, the value i is increased by one, and steps S43 to S47 are repeated.

[0032] According to the technical solution provided by the present invention, the method of updating the position of each particle includes:

[0033] Arrange the i-th target optimization values ​​corresponding to the current positions of all particles from small to large to obtain the i-th arrangement order;

[0034] Move the particles corresponding to the i-th target optimization value of the first set proportion at the front in the i-th arrangement order to the position of the particle with the largest i-th target optimization value corresponding to the current position;

[0035] Use the particle swarm position update formula to update the positions of all particles.

[0036] According to the technical solution provided by the present invention, the first set ratio is greater than or equal to zero and less than or equal to one, and increases monotonically with the increase of the number of iterations.

[0037] According to the technical solution provided by the present invention, the first set ratio increases monotonically with the increase of the number of iterations as follows:

[0038] When the number of iterations is less than or equal to half of the maximum number of iterations, the first set ratio is equal to 50%;

[0039] When the number of iterations is greater than half of the maximum number of iterations, the first set ratio is equal to the ratio of the current number of iterations to the maximum number of iterations.

[0040] According to the technical solution provided by the present invention, the step of controlling the thrust assembly to move to the target second attitude angle includes:

[0041] Obtain the current posture of the thrust assembly in the robot body coordinate system to obtain the third posture angle;

[0042] According to the second attitude angle and the third attitude angle, calculating the angle of rotation of the thrust assembly from the third attitude angle to the second attitude angle along the positive direction of each degree of freedom, and obtaining the rotation angle of each degree of freedom;

[0043] When the rotation angle is greater than 180 degrees, a compensation angle is rotated in the opposite direction of the degree of freedom; the compensation angle is equal to the difference between 360 degrees and the rotation angle.

[0044] According to the technical solution provided by the present invention, the method of providing thrust according to the target thrust value includes:

[0045] When the acceleration of the robot is equal to zero, the target thrust value is rounded up to obtain a thrust adjustment value;

[0046] The thrust assembly is controlled to provide a thrust having the magnitude of the thrust adjustment value.

[0047] The beneficial effects of the present invention are:

[0048] The robot includes a first platform, and a moving wheel group and a thrust assembly installed thereon; the moving wheel group is used to drive the first platform to move, and the thrust assembly is used to provide thrust to the first platform. The method includes: according to the current acceleration of the robot, the first attitude angle in the world coordinate system, the total gravity, and the friction coefficient with the curved wall, the relationship between the thrust provided by the thrust assembly and the second attitude angle of the thrust direction in the robot body coordinate system is constructed to obtain a solution model. In order to find the best combination of thrust and second attitude angle to make the robot power lower, the second attitude angle and thrust size are optimized using an optimization algorithm to obtain a target second attitude angle and a target thrust value. Finally, the thrust assembly of the robot is controlled to move to the target second attitude angle, and thrust is provided according to the target thrust value. This method can automatically adjust the direction and size of the thrust according to the actual inclination angle of the curved wall, so that the robot can run at a lower power and reduce energy consumption while maintaining normal operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0050] Figure 1 This is a schematic diagram of the robot's structure;

[0051] Figure 2 This is the main view of the robot;

[0052] Figure 3 This is a schematic diagram of the forces acting on the robot when it contacts a curved wall;

[0053] Among them: 1. first platform; 2. connecting plate; 3. telescopic device; 4. first driving device; 5. omnidirectional wheel; 6. fan; 7. second driving device; 8. second platform; 9. third driving device; 10. curved wall. DETAILED DESCRIPTION

[0054] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0055] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0056] Please refer to Figure 1-2 , the present invention provides a robot control method.

[0057] The robot includes a first platform 1 and a moving wheel group and a thrust assembly installed on the first platform 1; the moving wheel group is used to abut against the curved wall 10, driving the first platform 1 to move along the curved wall 10; the thrust assembly is used to provide thrust to the first platform 1 and the moving wheel group, so that the moving wheel group abuts against the curved wall 10; the thrust size and direction provided by the thrust assembly are adjustable.

[0058] The mobile wheel set comprises:

[0059] A connecting plate 2, wherein the connecting plate 2 is rotatably mounted on the first platform 1;

[0060] A telescopic device 3, one end of which is rotatably mounted on the first platform 1, and the other end of which is rotatably mounted on the connecting plate 2;

[0061] A first driving device 4, wherein the first driving device 4 is fixedly mounted on the connecting plate 2;

[0062] an omnidirectional wheel 5, wherein the omnidirectional wheel 5 is fixedly connected to a driving shaft of the first driving device 4; the first driving device 4 is used to drive the omnidirectional wheel 5 to rotate;

[0063] The first driving device 4 is used to drive the connecting plate 2 and the omnidirectional wheel 5 to rotate relative to the first platform 1 , and adjust the axial direction of the omnidirectional wheel 5 to adapt to the curved wall 10 .

[0064] The thrust assembly comprises:

[0065] The fan 6 is installed on the transmission assembly; the fan 6 is used to provide thrust for the first platform 1 and the moving wheel group; the transmission assembly is used to drive the fan 6 to rotate and adjust the thrust direction.

[0066] A second driving device 7, wherein the second driving device 7 is fixedly mounted on the first platform 1;

[0067] A second platform 8, wherein the second platform 8 is fixedly connected to the driving shaft of the second driving device 7; the second platform 8 is parallel to the driving shaft of the second driving device 7;

[0068] A third driving device 9, wherein the third driving device 9 is fixedly mounted on the second platform 8; a driving axis of the third driving device 9 and a driving axis of the second driving device 7 are perpendicular to each other;

[0069] The fan 6 is fixedly connected to the driving shaft of the third driving device 9, and the rotation axis of the fan 6 is perpendicular to the driving shaft of the third driving device 9;

[0070] The second driving device 7 is used to drive the second platform 8, the third driving device 9 and the fan 6 to rotate around the driving shaft of the second driving device 7; the third driving device 9 is used to drive the fan 6 to rotate around the driving shaft of the third driving device 9.

[0071] Methods include:

[0072] Obtaining a solution model; the solution model is used to represent the relationship between the robot's acceleration, the first attitude angle, the robot's total gravity, the friction coefficient of the curved wall, the magnitude of the thrust provided by the thrust assembly, and the second attitude angle; the first attitude angle is the attitude angle of the robot in the world coordinate system; the second attitude angle is the attitude angle of the thrust provided by the thrust assembly in the robot body coordinate system;

[0073] Get the robot's real-time acceleration, real-time first attitude angle, total gravity and friction coefficient at the current moment;

[0074] Substituting the real-time acceleration, the real-time first attitude angle, the total gravity and the friction coefficient into the solution model, and using the optimization algorithm to optimize the magnitude of the thrust and the second attitude angle, to obtain a target thrust value and a target second attitude angle; the target thrust value is the minimum value to ensure that the robot's moving wheel group abuts against the curved wall; the target second attitude angle is the second attitude angle of the thrust when the magnitude of the thrust is equal to the target thrust value;

[0075] The thrust assembly is controlled to move to the target second attitude angle, and thrust is provided according to the target thrust value.

[0076] Specifically, the acceleration of the robot can be measured by an accelerometer installed on the first platform.

[0077] The first attitude angle includes: the robot's roll angle , Pitch angle and yaw angle .

[0078] Since the thrust assembly has two degrees of freedom in this embodiment, the second posture angle includes: a first degree of freedom angle and an unknown second degree of freedom angle, which respectively represent the angles along the two degrees of freedom between the current posture of the thrust assembly and the initial posture.

[0079] The friction coefficient μ is obtained by searching the database: the specific value of the friction coefficient μ is obtained by querying the friction coefficient between the curved wall of the material and the moving wheel set. Or it can be obtained by actual measurement: when the moving wheel set and the curved wall slide relative to each other, the ratio of the pressure between the two and the friction force on the robot is equal to the friction coefficient μ.

[0080] The world coordinate system includes X, Y, and Z axes. The robot body coordinate system includes Xc, Yc, and Zc axes.

[0081] Specifically, based on the robot's current acceleration, the first attitude angle in the world coordinate system, the total gravity, and the friction coefficient with the curved wall, the relationship between the thrust provided by the thrust assembly and the second attitude angle of the thrust direction in the robot's body coordinate system is constructed to obtain a solution model.

[0082] In order to find the best combination of thrust and second attitude angle to make the robot power lower, the target optimization function is first constructed according to the thrust, and the second attitude angle is optimized using the optimization algorithm. After obtaining the target second attitude angle, it is substituted into the solution model to calculate the target thrust value.

[0083] Finally, the robot's thrust component is controlled to move to the target second attitude angle, and thrust is provided according to the target thrust value. This method can automatically adjust the direction and magnitude of the thrust according to the actual inclination angle of the curved wall, so that the robot can operate at a lower power while maintaining normal operation, thereby reducing energy consumption.

[0084] Furthermore, the step of obtaining the solution model includes:

[0085] Determine, according to the first posture angle, the total gravity and the acceleration, the components of the resultant force acting on the robot along each axis of the robot body coordinate system to obtain a plurality of resultant force components;

[0086] Determine the components of the external force applied to the robot along each axis of the robot body coordinate system according to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust, and the friction coefficient, and obtain multiple force components;

[0087] For each axis of the robot body coordinate system, the resultant force component is respectively set equal to the force component to obtain the solution model.

[0088] Specifically, the magnitude of the resultant force is equal to the product of the acceleration and the mass of the robot. After the acceleration direction and magnitude are detected, the resultant force components can be calculated based on the rotation matrix. The rotation matrix is ​​used to transform the direction in the world coordinate system to the robot body coordinate system. The resultant force is equal to the vector sum of the external forces acting on the robot. When the acceleration is zero, the vector sum of the external forces acting on the robot is zero.

[0089] Further, according to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust and the friction coefficient, the components of the external force applied to the robot along each axis of the robot body coordinate system are determined, and the step of obtaining multiple force components includes:

[0090] Determine, according to the first posture angle and the total gravity, the components of the total gravity along each axis of the robot body coordinate system to obtain a plurality of total gravity components;

[0091] According to the second posture angle and the magnitude of the thrust, components of the thrust along each axis of the robot body coordinate system are determined to obtain a plurality of thrust components;

[0092] According to the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the supporting force of the curved wall on the robot along each axis of the robot body coordinate system to obtain a plurality of supporting force components;

[0093] According to the friction coefficient, the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the friction force between the curved wall and the robot along each axis of the robot body coordinate system to obtain a plurality of friction force components;

[0094] The vector sums of the total gravity component, thrust component, support force component and friction force component along each axis of the robot body coordinate system are calculated to obtain a plurality of the force components.

[0095] Specifically, the acceleration and the total gravity are converted to the robot body coordinate system by using a rotation matrix to obtain an acceleration component and a total gravity component; the acceleration component includes the components of the acceleration along each axis of the robot body coordinate system; the total gravity component includes the components of the total gravity along each axis of the robot body coordinate system;

[0096] According to the second posture angle, the thrust is transformed into the robot body coordinate system to obtain a thrust component; the thrust component is an unknown quantity, including components of the thrust along each axis of the robot body coordinate system;

[0097] The support force of the curved wall 10 on the robot is calculated according to the vector sum of the total gravity component along the first direction and the thrust component along the first direction being equal to the acceleration component along the first direction; the first direction is the normal of the point of contact with the robot on the curved wall 10, which is parallel to one of the coordinate axes of the robot body coordinate system;

[0098] The friction force of the curved wall 10 on the robot is calculated according to the supporting force and the friction coefficient;

[0099] The product of the robot mass and each of the acceleration components is respectively set to be equal to the total gravity component, the thrust component, the components of the support force along each axis of the robot body coordinate system, and the components of the friction force along each axis of the robot body coordinate system, and the solution model is obtained by joint calculation.

[0100] Specifically, according to the second posture angle and the magnitude of the thrust, determining the components of the thrust along each axis of the robot body coordinate system, and obtaining the multiple thrust components comprises:

[0101] A first rotation matrix and a second rotation matrix are obtained according to the second posture angle; the second posture angle includes an unknown first degree of freedom angle and an unknown second degree of freedom angle; the first rotation matrix and the second rotation matrix both include a first degree of freedom angle and a second degree of freedom angle;

[0102] The components of the thrust along each axis of the world coordinate system are listed in a matrix form to obtain a first thrust matrix;

[0103] Performing matrix multiplication on the second rotation matrix, the first rotation matrix and the first thrust matrix in order to obtain a second thrust matrix;

[0104] Respectively represent each element in the second thrust matrix as a component of the thrust along each axis in the robot coordinate system to obtain the thrust component.

[0105] Specifically, total gravity, acceleration, thrust, support force and friction are all vectors.

[0106] The rotation matrix between the robot body coordinates and the world coordinate system is:

[0107]

[0108] in, represents the rotation matrix, is the robot's roll angle, is the pitch angle, is the yaw angle.

[0109] The matrix of the robot's total gravity in the world coordinate system is expressed as:

[0110]

[0111] in, G The matrix representation of the total gravity in the world coordinate system, G x Represents the component of the total gravity along the X-axis of the world coordinate system, G y Represents the component of total gravity along the Y axis of the world coordinate system, G z Represents the component of total gravity along the Z axis of the world coordinate system, G 0 represents the magnitude of the total gravity, and the negative sign represents the direction along the negative direction of the coordinate axis.

[0112] The way to transform the total gravity into the robot body coordinate system is: ;

[0113] in, G cThe matrix representation of the total gravity in the robot body coordinate system is expressed as follows:

[0114]

[0115] Specifically, the first thrust matrix is ​​a matrix representation of the thrust in the robot body coordinate system when the thrust assembly is in the initial posture, specifically:

[0116]

[0117] in, F’ represents the first thrust matrix, F p Indicates the magnitude of the thrust, and the negative sign indicates that the direction is along the negative direction of the coordinate axis.

[0118] According to the second posture angle, the thrust is transformed into the robot body coordinate system as follows:

[0119]

[0120] Specifically expand to get the second thrust matrix:

[0121]

[0122] F c represents the second thrust, M x represents the first rotation matrix, M y represents the second rotation matrix, F xc represents the component of thrust along the Xc axis of the robot body coordinate system, F yc represents the component of thrust along the Yc axis of the robot body coordinate system, F zc It represents the component of thrust along the Zc axis of the robot body coordinate system; represents the first degree of freedom angle, Represents the second degree of freedom angle.

[0123] First rotation matrix M x and the second rotation matrix M y They are:

[0124]

[0125] In some implementations, if the acceleration is zero, no transformation is required and the components along each axis are directly substituted into 0.

[0126] If the acceleration is not zero, the method is the same as above: first write the acceleration as a matrix representation in world coordinates, and then use the rotation matrix Transform to the robot body coordinate system. Each element in the resulting matrix represents the component of acceleration along each axis of the robot body coordinate system.

[0127] Specifically, refer to Figure 3 , G The total gravity, F Indicates thrust, N Indicates support force, f The circle represents the robot as a mass point, and the triangle represents the curved wall 10.

[0128] The support force is perpendicular to the normal of the curved wall at the contact point of the robot. Since the robot body coordinate system is consistent with the robot itself, the robot fits the curved wall. Therefore, the Zc axis of the robot body coordinate system is parallel to the support force direction, that is, the first direction. It can be seen that the components of the support force in the Xc axis and Yc axis in the robot body coordinate system are both 0, and the component of the Zc axis is not 0.

[0129] Thus, the vector sum of the support force of the curved wall on the robot, the total gravity component along the first direction, and the thrust component along the first direction is equal to the acceleration component along the first direction. Finally, the calculation formula of the support force can be obtained. Due to the existence of unknown parameters, the support force cannot be directly calculated at present.

[0130] Specifically, the friction force is less than or equal to the product of the friction coefficient and the support force. In this embodiment, the thrust can be made smaller by calculating that the friction force is equal to the product of the friction coefficient and the support force. The magnitude of the friction force can be calculated based on the support force. Since the support force can only be derived from the calculation formula, the magnitude of the friction force can also only be obtained from the calculation formula.

[0131] Since the friction force is perpendicular to the support force, the friction force is in the XcOYc plane, and the components along the Xc axis and the Yc axis are unknown.

[0132] Specifically, let the product of the robot mass and each acceleration component be equal to the total gravity component, the thrust component, the support force component along each axis of the robot body coordinate system, and the friction force component along each axis of the robot body coordinate system, and obtain the sum of the components of the three axes. At the same time, combined with the calculation formula of friction force, the matrix representation of thrust in the robot body coordinate system, and the conversion relationship between each thrust component and thrust size, 8 equations are obtained, which are:

[0133]

[0134] in, F xc , Fyc , F zc They represent the components of thrust along each axis of the robot body coordinate system, both of which are unknown quantities;

[0135] f x , f y , f z Respectively represent the components of the friction force along each axis of the robot body coordinate system. In this embodiment f z is 0; f x , f y is an unknown quantity;

[0136] N x , N y , N z Respectively represent the support force N The components along each axis of the robot body coordinate system are N x , N y All are 0; N z is an unknown quantity;

[0137] G x , G y , G z Represents the total gravity G The components along each axis of the robot body coordinate system are all known quantities;

[0138] a x , a y , a z Respectively represent acceleration a The components along each axis of the robot body coordinate system are all known quantities; in this embodiment, a z is 0;

[0139] m represents the mass of the robot, which is a known quantity;

[0140] μ represents the friction coefficient, which is a known quantity;

[0141] F pis the thrust size, is an unknown quantity;

[0142] It represents the angle of the first degree of freedom, which is an unknown quantity;

[0143] Represents the angle of the second degree of freedom, which is an unknown quantity.

[0144] In this embodiment, the supporting force N The magnitude is equal to its component along the Zc axis of the robot body coordinate system N z .

[0145] We get 8 linearly independent equations, including 9 unknown parameters, which are:

[0146] Support N Component along the Zc axis of the robot body coordinate system N z , the friction force components along the Xc and Yc axes f x , f y , thrust size F p , the three components of thrust in the robot body coordinate system F xc , F yc , F zc , the first degree of freedom angle and the second degree of freedom angle .

[0147] Combine the above 8 equations to calculate the first degree of freedom angle and the second degree of freedom angle And thrust size F p The conversion relationship of the other unknown parameters can be obtained by calculation and deduction. and The calculation formula represented by . Finally, the solution model is obtained, which is specifically expressed as:

[0148]

[0149] In order to simplify the representation, the solution model express , express , express .

[0150] Furthermore, the optimization algorithm is a particle swarm algorithm.

[0151] Get the robot's real-time acceleration, real-time first posture angle, total gravity and friction coefficient at the current moment, and substitute them into the solution model.

[0152] Then, the step of optimizing the thrust value and the second attitude angle by using an optimization algorithm to obtain a target thrust value and a target second attitude angle comprises:

[0153] S41: constructing a target optimization function according to the thrust;

[0154] Specifically, the objective optimization function is expressed by Formula 1:

[0155] Formula 1;

[0156] in, f 0 means the target optimization function, F p Indicates the magnitude of thrust.

[0157] In the optimization process, the smaller the target optimization function is, the closer it is to the optimal solution. In this embodiment, it is only necessary to find the second attitude angle corresponding to the minimum thrust to obtain the optimal solution. The final target optimization function is set as shown in Formula 1, and the thrust size is directly used as the optimization target.

[0158] Specifically, according to the solution model, using a particle swarm optimization algorithm, optimizing the magnitude of the thrust and the second attitude angle, and obtaining a target thrust value and a target second attitude angle comprises:

[0159] S42: setting a search space for the second posture angle; setting the number of particles of the particle swarm algorithm, the initial position and initial velocity of each particle in the search space, and the maximum number of iterations; setting a termination condition according to the target optimization function;

[0160] In this embodiment, the second posture angle includes: a first degree of freedom angle and a second degree of freedom angle.

[0161] The search space is set as follows: the first degree of freedom angle is in the interval of (-180°, 180°]; the second degree of freedom angle is in the interval of (-180°, 180°];

[0162] The number of particles is set to 100, the initial position is set to follow the normal distribution in the search space, the initial speed is set randomly, the maximum number of iterations is 100, and the termination condition is: the maximum number of iterations is reached or the historical minimum value of the target optimization function does not decrease after multiple consecutive iterations. In this embodiment, it is set to no longer decrease after 10 consecutive iterations.

[0163] S43: according to the current position of each particle in the search space, obtain the corresponding i-th second posture angle; the initial value of i is 1;

[0164] S44: Substituting the i-th second attitude angle into the solution model to calculate the i-th thrust;

[0165] Since the second attitude angle is substituted into the solution model as a known quantity, the specific values ​​of the remaining unknown parameters, including the magnitude of the thrust, can be calculated.

[0166] S45: Substituting the i-th thrust into the target optimization function to calculate and obtain the i-th target optimization value;

[0167] S46: Update the position of each particle; update the speed of each particle;

[0168] Furthermore, the method of updating the position of each particle includes:

[0169] Arrange the i-th target optimization values ​​corresponding to the current positions of all particles from small to large to obtain the i-th arrangement order;

[0170] Move the particles corresponding to the i-th target optimization value of the first set proportion at the front in the i-th arrangement order to the position of the particle with the largest i-th target optimization value corresponding to the current position;

[0171] Use the particle swarm position update formula to update the positions of all particles.

[0172] Use the particle group velocity relationship formula to update the velocity of all particles.

[0173] In this embodiment, by default, the values ​​at the front of the sorting order are smaller than the values ​​at the back, and the values ​​at the back of the sorting order are larger than the values ​​at the front. For example, the first i-th target optimization value at the front of the sorting order is the minimum value, and the last one is the maximum value.

[0174] Specifically, this method of updating the position can move the particles to the vicinity of the current optimal solution for searching more quickly, so that the particle swarm can converge to the global optimal solution at a faster rate to obtain the target second attitude angle.

[0175] Since in the application scenario of this embodiment, the components of gravity, thrust, friction and support force need to be equal to zero or the components of the resultant force, after adjusting the direction of the thrust, the magnitude of the thrust needs to be adjusted synchronously.

[0176] To satisfy the above relationship, the thrust component in one direction must always be equal, and the thrust magnitude and direction can be changed by changing the component in the other direction. In the above situation, the thrust magnitude and direction combination is the optimal solution only when the thrust component in the other direction is the smallest. It can be seen that the thrust magnitude has only one minimum value.

[0177] Therefore, in the case of this embodiment, there is no local optimum. An algorithm solution with faster convergence as shown above can be used. Other algorithm design solutions must consider the problem of falling into the local optimum, so the convergence speed is definitely lower than that in this embodiment.

[0178] By applying the above method, in the actual optimization process, the number of iterations can be greatly reduced, the response time of the robot can be significantly shortened, and the stability of the robot in dealing with bumpy curved walls can be improved.

[0179] Furthermore, the first set ratio is greater than or equal to zero and less than or equal to one, and increases monotonically with the increase of the number of iterations.

[0180] Furthermore, the first set ratio increases monotonically with the increase of the number of iterations in the following manner:

[0181] When the number of iterations is less than or equal to half of the maximum number of iterations, the first set ratio is equal to 50%;

[0182] When the number of iterations is greater than half of the maximum number of iterations, the first set ratio is equal to the ratio of the current number of iterations to the maximum number of iterations.

[0183] In some embodiments, half of the particles are moved directly to the current optimal solution position during the first 50 iterations. If the termination condition is still not met after 50 iterations, a more radical approach is taken, that is, after 50 iterations, as the number of iterations increases, more and more particles are moved directly to the current optimal solution position.

[0184] This method allows the algorithm to search more finely near the global optimal solution, making the final result closer to the global optimal solution.

[0185] S47: Determination: When the termination condition or the maximum number of iterations is reached, the historical optimal solution is used as the target second attitude angle and the target thrust value; otherwise, the value i is increased by one, and steps S43 to S47 are repeated.

[0186] In this embodiment, the historical optimal solution is specifically: the jth second attitude angle and the jth thrust corresponding to the minimum value from the 1st target optimization value to the i-th target optimization value; j is an integer value in the interval [1, i].

[0187] Through the above-mentioned method, the adaptability of the particle swarm algorithm applied to the present embodiment can be greatly improved, making the algorithm more targeted and realizing fast and accurate optimization of the second posture angle.

[0188] Further, the step of controlling the thrust assembly to move to the target second attitude angle includes:

[0189] Obtain the current posture of the thrust assembly in the robot body coordinate system to obtain the third posture angle;

[0190] According to the second attitude angle and the third attitude angle, calculating the angle of rotation of the thrust assembly from the third attitude angle to the second attitude angle along the positive direction of each degree of freedom, and obtaining the rotation angle of each degree of freedom;

[0191] When the rotation angle is greater than 180 degrees, a compensation angle is rotated in the opposite direction of the degree of freedom; the compensation angle is equal to the difference between 360 degrees and the rotation angle.

[0192] Specifically, the third attitude angle includes: the third degree of freedom angle and the fourth degree of freedom angle of the thrust assembly in the current state, which are detected by encoders on the second drive device 7 and the third drive device 9 respectively.

[0193] The rotation angle of each degree of freedom is calculated by subtracting the third degree of freedom angle from the first degree of freedom angle and subtracting the fourth degree of freedom angle from the second degree of freedom angle.

[0194] In some embodiments, the actual orientation of the fan 6 of the thrust assembly may need to rotate more than 180 degrees in the positive direction to the target second posture angle. Therefore, when encountering such an accidental situation, directly rotating the compensation angle in the reverse direction can achieve faster rotation of the fan 6 to the second posture angle, thereby improving the response speed of the robot.

[0195] Furthermore, the method of providing thrust according to the target thrust value includes:

[0196] When the acceleration of the robot is equal to zero, the target thrust value is rounded up to obtain a thrust adjustment value;

[0197] The thrust assembly is controlled to provide a thrust having the magnitude of the thrust adjustment value.

[0198] Specifically, since the particle swarm algorithm is used in this embodiment to optimize the second posture angle, there is generally an error between the final target second posture angle and the actual target second posture angle. In order to avoid the error causing adverse effects on the normal operation of the robot, causing the robot to fall off due to insufficient thrust. When the robot remains stationary or moves at a constant speed, the actual thrust provided is made slightly greater than the target thrust value to offset the effect of the error.

[0199] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) to form a technical solution.

Claims

1. A robot control method, characterized in that: The robot comprises a first platform and a moving wheel group and a thrust assembly mounted on the first platform; the moving wheel group is used to drive the first platform to move along a curved wall; the thrust assembly is used to provide thrust to the first platform so that the moving wheel group abuts against the curved wall; the thrust provided by the thrust assembly is adjustable in size and direction; Methods include: Obtaining a solution model; the solution model is used to represent the relationship between the robot's acceleration, the first attitude angle, the robot's total gravity, the friction coefficient of the curved wall, the magnitude of the thrust provided by the thrust assembly, and the second attitude angle; the first attitude angle is the attitude angle of the robot in the world coordinate system; the second attitude angle is the attitude angle of the thrust provided by the thrust assembly in the robot body coordinate system; Get the robot's real-time acceleration, real-time first attitude angle, total gravity and friction coefficient at the current moment; Substituting the real-time acceleration, the real-time first attitude angle, the total gravity and the friction coefficient into the solution model, and using the optimization algorithm to optimize the magnitude of the thrust and the second attitude angle, to obtain a target thrust value and a target second attitude angle; the target thrust value is the minimum value to ensure that the robot's moving wheel group abuts against the curved wall; the target second attitude angle is the second attitude angle of the thrust when the magnitude of the thrust is equal to the target thrust value; The thrust assembly is controlled to move to the target second attitude angle, and thrust is provided according to the target thrust value.

2. A robot control method according to claim 1, characterized in that: The steps to obtain the solution model include: Determine, according to the first posture angle, the total gravity and the acceleration, the components of the resultant force acting on the robot along each axis of the robot body coordinate system to obtain a plurality of resultant force components; Determine the components of the external force applied to the robot along each axis of the robot body coordinate system according to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust, and the friction coefficient, and obtain multiple force components; For each axis of the robot body coordinate system, the resultant force component is respectively set equal to the force component to obtain the solution model.

3. A robot control method according to claim 2, characterized in that: According to the total gravity, the first posture angle, the second posture angle, the magnitude of the thrust and the friction coefficient, the components of the external force applied to the robot along the axes of the robot body coordinate system are determined, and the step of obtaining multiple force components includes: Determine, according to the first posture angle and the total gravity, the components of the total gravity along each axis of the robot body coordinate system to obtain a plurality of total gravity components; According to the second posture angle and the magnitude of the thrust, components of the thrust along each axis of the robot body coordinate system are determined to obtain a plurality of thrust components; According to the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the supporting force of the curved wall on the robot along each axis of the robot body coordinate system to obtain a plurality of supporting force components; According to the friction coefficient, the first posture angle, the second posture angle, the total gravity and the magnitude of the thrust, determining the components of the friction force between the curved wall and the robot along each axis of the robot body coordinate system to obtain a plurality of friction force components; The vector sums of the total gravity component, thrust component, support force component and friction force component along each axis of the robot body coordinate system are calculated to obtain a plurality of the force components.

4. A robot control method according to claim 1, characterized in that: The optimization algorithm is a particle swarm algorithm; The step of optimizing the thrust value and the second attitude angle by using an optimization algorithm to obtain a target thrust value and a target second attitude angle comprises: S41: constructing a target optimization function according to the thrust; S42: setting a search space for the second posture angle; setting the number of particles of the particle swarm algorithm, the initial position and initial velocity of each particle in the search space, and the maximum number of iterations; setting a termination condition according to the target optimization function; S43: according to the current position of each particle in the search space, obtain the corresponding i-th second posture angle; the initial value of i is 1; S44: Substituting the i-th second attitude angle into the solution model to calculate the i-th thrust; S45: Substituting the i-th thrust into the target optimization function to calculate and obtain the i-th target optimization value; S46: Update the position of each particle; update the speed of each particle; S47: Determination: When the termination condition or the maximum number of iterations is reached, the historical optimal solution is used as the target second attitude angle and the target thrust value; otherwise, the value i is increased by one, and steps S43 to S47 are repeated.

5. A robot control method according to claim 4, characterized in that: Ways to update the position of each particle include: Arrange the i-th target optimization values ​​corresponding to the current positions of all particles from small to large to obtain the i-th arrangement order; Move the particles corresponding to the i-th target optimization value of the first set proportion at the front in the i-th arrangement order to the position of the particle with the largest i-th target optimization value corresponding to the current position; Use the particle swarm position update formula to update the positions of all particles.

6. A robot control method according to claim 5, characterized in that: The first set ratio is greater than or equal to zero and less than or equal to one, and increases monotonically with the increase of the number of iterations.

7. A robot control method according to claim 6, characterized in that: The first set ratio increases monotonically with the increase of the number of iterations as follows: When the number of iterations is less than or equal to half of the maximum number of iterations, the first set ratio is equal to 50%; When the number of iterations is greater than half of the maximum number of iterations, the first set ratio is equal to the ratio of the current number of iterations to the maximum number of iterations.

8. A robot control method according to claim 1, characterized in that: The step of controlling the thrust assembly to move to the target second attitude angle comprises: Obtain the current posture of the thrust assembly in the robot body coordinate system to obtain the third posture angle; According to the second attitude angle and the third attitude angle, calculating the angle of rotation of the thrust assembly from the third attitude angle to the second attitude angle along the positive direction of each degree of freedom, and obtaining the rotation angle of each degree of freedom; When the rotation angle is greater than 180 degrees, a compensation angle is rotated in the opposite direction of the degree of freedom; the compensation angle is equal to the difference between 360 degrees and the rotation angle.

9. A robot control method according to claim 1, characterized in that: The method of providing thrust according to the target thrust value includes: When the acceleration of the robot is equal to zero, the target thrust value is rounded up to obtain a thrust adjustment value; The thrust assembly is controlled to provide a thrust having the magnitude of the thrust adjustment value.