CN107097231A - A kind of concentric tube robot precise motion control method of view-based access control model servo - Google Patents

Abstract

The invention discloses a precise motion control method for a concentric tube robot based on visual servoing. The method establishes an eye-in-hand visual servoing system based on images, and aims at extracting image features without the need for camera calibration, and through the image Jacobian The mapping of the matrix completes the extraction of the control quantity; the Kalman filter is used to complete the online estimation of the image Jacobian matrix, the image feature change is used as the observation vector, and the image Jacobian matrix is used as the state vector; and then the estimated value of the image Jacobian matrix is used As the observation matrix of the Kalman filter, the joint angle error is used as the state vector, and the image error is used as the observation vector, so that the filter estimation can obtain the control amount of the joint angle; when the image error tends to zero, the positioning of the target point will be completed. The present invention relies on relatively reliable image information to feed back motion control, and finally achieves precise positioning of tiny trembling target points during the operation, thereby completing the operation.

Description

A precise motion control method for concentric tube robot based on visual servoing

technical field

The invention relates to a precise motion control method for a concentric tube medical robot, in particular to a precise motion control method for a concentric tube robot based on visual servoing.

Background technique

At present, the mainstream of the motion control of the concentric tube robot is the teleoperation algorithm, which is mainly based on the establishment of its kinematic model, which has high requirements for the accuracy of the kinematic model and the consideration of factors such as friction and torsion. It is difficult to accurately locate the surgical target point, and it can only be roughly reached. In recent years, related technologies such as X-ray, ultrasound, MRI, and stereoscopic imaging have been applied to feedback the moving shape of concentric tubes, but the image-based visual servoing method does not require much modification of the hardware and can provide accurate image information as a moving image. ring feedback. At the same time, the image-based method is based on fewer kinematic models, and it does not require calibration between the robot body and the camera, and can accurately locate the target point so that it is always kept at the center of the image.

Contents of the invention

The object of the present invention is to solve the shortcomings in the prior art, and propose a precise motion control method for a concentric tube robot based on visual servoing.

In order to achieve the above object, the present invention adopts the following technical solutions:

A precise motion control method for concentric tube robots based on visual servoing. This method establishes an eye-in-hand visual servoing system based on images. It does not require camera calibration and aims to extract image features, and completes it through the mapping of the image Jacobian matrix. The extraction of the control quantity; use the Kalman filter to complete the online estimation of the image Jacobian matrix, take the image feature change as the observation vector, and use the image Jacobian matrix as the state vector; then use the estimated value of the image Jacobian matrix as the Kalman filter The observation matrix uses the joint angle error as the state vector, and the image error as the observation vector, so that the filter estimation can obtain the control amount of the joint angle; when the image error tends to zero, the positioning of the target point will be completed.

The purpose of the present invention is to solve the defect that the teleoperation algorithm fails to accurately locate the target point for the robot, and rely on more reliable image information to feed back the motion control, and finally realize the precise positioning of the micro-vibrating target point during the operation , to complete the operation. The present invention aims at the defect that the prior art fails to guarantee the precise positioning of the surgical position by the robot, and introduces visual feedback by using the image information collected by the camera at the end of the robot. The application of this method can also complete the tracking of the moving point to achieve the purpose of its accurate positioning.

Description of drawings

FIG. 1 is a schematic structural view of a concentric tube medical robot applied to minimally invasive surgery proposed by the present invention.

Fig. 2 is a flowchart of a precise motion control method for a concentric tube robot based on visual servoing proposed by the present invention.

Fig. 3 is the experimental platform of the concentric tube robot based on visual servoing proposed by the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention.

Referring to Figure 1, a concentric tube medical robot for minimally invasive surgery, including four servo motors 1, motor end cover 2, coupling 3, rear end cover 4, mobile carrier 6, miniature bearing 7, concentric tube mechanism , a square cylinder 11, a ball screw 16 and a transmission gear set 17, four servo motors 1 are arranged on the motor end cover 2, and the output shafts of the two oppositely arranged servo motors 1 are connected to the rear end cover through a coupling 3 4 connection, the rear end cover 4 is provided with two ball screws 16, the two ball screws 16 are connected with the output shafts of the above two servo motors 1 through the coupling 3, and the mobile carrier 6 is set on the two ball screws In the middle of 16, the ball screw 16 is fixed on the mobile carrier 6 through the flange 15, the limit switch 5 is arranged on the mobile carrier 6, and two square cylinders 11 are arranged on the outer sides of the two ball screws 16, and the remaining two are opposite to each other. The servo motor 1 drives the square cylinder 11 through the transmission gear set 17 on the motor end cover 2; the concentric tube mechanism includes an outer tube 8 and an inner tube 9 inserted into the outer tube 8, and the concentric tube mechanism is fixed on the base through a fixed tube The inner tube 9 is fixed on the mobile carrier 6 through the gear 14, and the square cylinder 11 is provided with a gear transmission module 12 meshed with the gear 14; two servo motors drive the ball screw 16 to rotate through the coupling 3, thereby The mobile carrier 6 on the ball screw 16 moves axially, and the concentric tube mechanism fixed on the mobile carrier 6 also expands and contracts axially; the remaining two servo motors drive the square cylinder 11 to rotate through gear transmission, and then The concentric pipe mechanism fixed on the mobile carrier 6 is driven to rotate by the gear transmission modules 12 and 14 on the mobile carrier 6 .

In the present invention, a micro-endoscopic camera 10 is arranged at the end of the inner tube 9, and the process of the whole method is realized by using its identification and positioning of the target point. The experimental platform is shown in Figure 3.

Referring to Fig. 2, the present invention proposes a method for precise motion control of concentric tube robots based on visual servoing. This method establishes an eye-in-hand visual servoing system based on images, and aims at extracting image features without camera calibration, and The extraction of the control quantity is completed through the mapping of the image Jacobian matrix; the Kalman filter is used to complete the online estimation of the image Jacobian matrix, the image feature change is used as the observation vector, and the image Jacobian matrix is used as the state vector; and then the image Jacobian is used The estimated value of the matrix is used as the observation matrix of the Kalman filter, the joint angle error is used as the state vector, and the image error is used as the observation vector, so that the filter estimation can obtain the control amount of the joint angle; when the image error tends to zero, the target point will be completed positioning.

In this method, the camera placed at the end of the inner tube of the robot is used to collect images in real time and extract image features, and formula (1) is used to obtain the coordinates of the center of the target.

(1)

By individually controlling a certain amount of control of each servo motor, the position of the target point on the image plane changes sequentially. Therefore, the image Jacobian matrix The initial value of . Variations in image features is the state as the observation vector, with is estimated for the state vector, so the state equation is,

(2)

The iterative process is divided into state prediction and prediction update, as follows,

Finally, the estimated image Jacobian matrix is used as the observation matrix of the Kalman filter, the joint angle error is used as the state vector, and the image error is used as the observation vector, and the above filter iteration process is re-passed, so the control amount of the joint angle can be estimated.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto, any person familiar with the technical field within the technical scope disclosed in the present invention, according to the technical solution of the present invention Any equivalent replacement or change of the inventive concepts thereof shall fall within the protection scope of the present invention.

Claims (1)

1. A concentric tube robot precise motion control method based on visual servo, it is characterized in that, this method establishes the visual servo system of eye-in-hand based on image, does not need the demarcation of camera and aims at extracting the feature of image, and through image The mapping of the Jacobian matrix completes the extraction of the control quantity; the Kalman filter is used to complete the online estimation of the image Jacobian matrix, taking the image feature change as the observation vector, and the image Jacobian matrix as the state vector; and then using the image Jacobian matrix to estimate The value is used as the observation matrix of the Kalman filter, the joint angle error is used as the state vector, and the image error is used as the observation vector, so that the filter estimation can obtain the control amount of the joint angle; when the image error tends to zero, the positioning of the target point will be completed.