An adaptive control method for humanoid robot hand joints

Abstract

This invention discloses an adaptive control method for the hand joints of a humanoid robot. The method first uses pressure sensors mounted on the fingertips to determine the position vector of the geometric center point of the grasping surface. The method then calculates the expected pressure and actual pressure vectors at each fingertips. Finally, the contact elasticity matrix is ​​used to calculate the position correction vectors for the fingertips, enabling adaptive correction of the fingertips' positions. The advantages of this method are that it allows for adaptive grasping of objects based on the set contact elasticity coefficients using only the pressure sensors mounted on the fingertips. It offers flexibility, adjustability, a simple algorithm, and is applicable to various humanoid robots with varying numbers of fingers.

Description

Technical Field

[0001] The present invention relates to the field of robotics and automation, and in particular to an adaptive control method for hand joints of a humanoid robot. Background Art

[0002] Humanoid robots are a class of robotic systems with highly complex kinematic models and strong interactions with the environment. As the number of fingers in their hands increases, position control becomes increasingly complex, and it becomes difficult to adjust the contact force with the environment. Therefore, an adaptive control method for humanoid robot hand joints is urgently needed to achieve dual control of finger position and force. Summary of the Invention

[0003] The problem to be solved by the present invention is to provide an adaptive control method for the hand joints of a humanoid robot.

[0004] To solve the above problems, the present invention provides an adaptive control method for the hand joints of a humanoid robot. To achieve the above objectives, the technical solution adopted by the present invention to solve the technical problems is:

[0005] An adaptive control method for a humanoid robot hand joint comprises: step S1: closing the palm; step S2: calculating the position vector of the geometric center point of the grasping surface; step S3: calculating the expected pressure and the actual pressure vector of each finger end in the palm; step S4: calculating the correction vector of the finger end point position; step S5: compensating the correction vector to the control system;

[0006] The center point of the rotation joint connecting the i-th finger and the palm of the humanoid robot hand in step S1 is A i (i=1,2,…,n), the end point of the i-th finger is B i , the palm center defines the coordinate system O-xyz, n represents the number of fingers; given the end point B of the i-th finger i Expected position instruction The expected rotation angle of each finger driving joint motor is calculated by position inverse solution, and closed-loop control is implemented to make the palm perform a closing movement;

[0007] Step S2 includes installing at each finger end point B during the execution of step S1. i The pressure sensor collects pressure signals in real time And use the following formula to calculate the judgment factor λ:

[0008]

[0009] As a further improvement of the present invention, step S3 is to determine the gravity f of the grasped object. gThe expected pressure vector at the end point of the i-th finger is given by the friction coefficient μ with the surface of the grasped object. With the actual pressure vector for:

[0010]

[0011] As a further improvement of the present invention, step S4 sets the contact elastic coefficient matrix K of the finger end point according to actual needs, and then calculates the contact elastic coefficient matrix of the i-th finger end point B using the following formula: i Correction vector for position:

[0012]

[0013] As a further improvement of the present invention, step S5 uses the following formula to calculate the end point B of the i-th finger: i Corrected position vector:

[0014]

[0015] The position calculation result of the above formula is inversely solved to calculate the expected rotation angle of each finger driving joint motor, and closed-loop control is implemented to enable the palm to adaptively grasp the object.

[0016] As a further improvement of the present invention, in step S2, when λ=0 is satisfied, step S1 is continued to be executed; when λ≠0 is satisfied, the position solution is used to calculate the end point of the i-th finger as B i The position vector is r Bi , and then use the following formula to calculate the position vector of the geometric center C of the grasping surface:

[0017]

[0018] As a further improvement of the present invention, in step S4, k x 、k y 、k z Represents the contact elastic coefficient along the three directions of the coordinate system O-xyz.

[0019] The beneficial effects of adopting the above-mentioned technical solution of the present application include: only using the pressure sensor installed at the end of the finger to adaptively grasp the object according to the set contact elasticity coefficient, it is flexible and adjustable, the algorithm is simple, and it is suitable for various types of humanoid robots with different numbers of fingers. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is a schematic diagram of the robot hand joints and grasping positions according to one embodiment of the present invention;

[0022] Figure 2 This is an adaptive control block diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The present invention will be described in further detail below with reference to specific embodiments:

[0024] The present invention provides an adaptive control method for a humanoid robot hand joint, comprising the following steps:

[0025] Step S1: Implement palm closure.

[0026] The center point of the rotation joint connecting the i-th finger and the palm of the humanoid robot hand is A i (i=1,2,…,n), the end point of the i-th finger is B i , the palm center defines the coordinate system O-xyz, n represents the number of fingers. Given the end point B of the i-th finger i Expected position instruction The expected rotation angles of the motors driving each finger joint are calculated through position inverse solution, and closed-loop control is implemented to make the palm perform a closing movement.

[0027] Step S2: Calculate the position vector of the geometric center point C of the grasping surface.

[0028] During the execution of step S1, the finger end point B is installed i The pressure sensor collects pressure signals in real time And use the following formula to calculate the judgment factor λ:

[0029]

[0030] When λ=0, continue to step S1; when λ≠0, use the position solution to calculate the end point of the i-th finger as B i The position vector is r Bi , and then use the following formula to calculate the position vector of the geometric center C of the grasping surface:

[0031]

[0032] Step S3: Calculate the expected pressure and actual pressure vector at each finger end.

[0033] When the gravity f on the object is known g The expected pressure vector at the end point of the i-th finger is given by the friction coefficient μ with the surface of the grasped object. With the actual pressure vector for:

[0034]

[0035] Step S4: Calculate the correction vector of the finger end point position.

[0036] Set the contact elastic coefficient matrix K of the finger end point according to actual needs, and then calculate the i-th finger end point B using the following formula i Correction vector for position:

[0037]

[0038] Where, k x 、k y 、k z Represents the contact elastic coefficient along the three directions of the coordinate system O-xyz.

[0039] Step S5: Compensate the correction vector to the control system.

[0040] The end point B of the i-th finger is calculated using the following formula: i Corrected position vector:

[0041]

[0042] The position calculation result of the above formula is inversely solved to calculate the expected rotation angle of each finger driving joint motor, and closed-loop control is implemented to enable the palm to adaptively grasp the object.

[0043] like Figure 1 As shown, the palm includes five fingers, each of which is a double-link mechanism hinged to each other. One end of each finger is hinged to the palm surface.

[0044] like Figure 2 As shown in the figure, the position inverse algorithm drives the joint motor to the desired rotation angle, and then uses the PID feedback controller, servo drive, motor, and finally drives the joint of the manipulator.

[0045] The above embodiments are only for illustrating the technical concept and features of the present invention. Their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it. They are not intended to limit the scope of protection of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adaptive control of hand joints of a humanoid robot, characterized in that: include: Step S1; Implement palm closure; Step S2; Calculate the position vector of the geometric center point of the grasping surface; Step S3: Calculate the expected pressure and actual pressure vector at the end of each finger in the palm; Step S4: Calculate the correction vector of the finger end point position; Step S5: Compensating the correction vector to the control system; Wherein, in step S1, the humanoid robot hand is The center point of the rotation joint connecting the fingers and the palm is ( ), No. The end points of the fingers are , the palm center defines the coordinate system , Indicates the number of fingers; given the Finger tip points Expected position instruction , the expected rotation angle of each finger driving joint motor is calculated by position inverse solution, and closed-loop control is implemented to make the palm perform a closing movement; The step S2 includes installing at each finger end point during the execution of step S1. The pressure sensor collects pressure signals in real time , and use the following formula to calculate the judgment factor : ， In step S3, the gravity of the grasped object is known. Coefficient of friction with the surface of the object being grasped In the case of The expected pressure vector at the finger tip With the actual pressure vector for: ， ， The step S4 sets the contact elastic coefficient matrix of the finger end point according to actual needs , and then use the following formula to calculate the Finger tip points Correction vector for position: ， In step S2, when When , continue to execute step S1; when When the position is correct, the first The end points of the fingers are The position vector is , and then use the following formula to calculate the position vector of the geometric center C of the grasping surface: 。 2. The adaptive control method for the hand joints of a humanoid robot according to claim 1, characterized in that: The step S5 uses the following formula to calculate the Finger tip points Corrected position vector: ， The position calculation result of the above formula is inversely solved to calculate the expected rotation angle of each finger driving joint motor, and closed-loop control is implemented to enable the palm to adaptively grasp the object.

3. The adaptive control method for hand joints of a humanoid robot according to claim 1, wherein: In the step S4, , 、 、 Indicates along the coordinate system Contact elastic coefficients in three directions.

Images