A robot safety control method and system for safety operating space expansion and obstacle avoidance direction guidance of aircraft engine blade detection

Abstract

This invention belongs to the technical field of robot safety control and discloses a robot safety control method and system for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection. The method includes the following steps: enveloping the point cloud data of the engine blade using a hyperellipsoidal surface to obtain multiple hyperellipsoidal surfaces; moving the robot to the blade according to a preset reference trajectory and determining whether the robot needs to bypass the surface boundary of the blade; when bypassing the surface boundary is required, calculating the robot's safety control force, converting this safety control force into control torques for the robot joints, and controlling the robot to bypass the blade surface according to these control torques; otherwise, the robot continues to inspect the blade surface according to the preset reference trajectory. This invention solves the control problem of robots being unable to bypass surfaces along the desired direction with a shorter path when the operating space is limited and the obstacle avoidance path lengths vary.

Description

Technical Field

[0001] This invention belongs to the field of robot safety control technology, and more specifically, relates to a robot safety control method and system for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection. Background Technology

[0002] In the process of inspecting aircraft engine blades, the operating space between adjacent blades is small and the path lengths for different obstacle avoidance directions vary due to the tortuous characteristics of the blades. Therefore, it is crucial to accurately characterize the curved surface boundary to obtain a larger operating space and to consider the obstacle avoidance direction to obtain a shorter path for the curved surface inspection task.

[0003] For representing the envelope of curved surfaces, the traditional spherical envelope, while simple in form, easily leads to a significant waste of operational space. The minimum volume hyperellipsoidal envelope method requires continuous adjustment of the hyperellipsoid's center position, orientation, and semi-axis length, making parameter optimization complex. In addition, the method of obtaining combined hyperellipsoids based on neural network training has high training costs, limiting its application in the field of robot obstacle avoidance.

[0004] For robot obstacle avoidance, traditional position-constrained high-order obstacle control functions lack effective guidance for obstacle avoidance direction, leading to directional uncertainty. Methods such as constructing enhanced obstacle control functions by adding direction terms, and constructing separate obstacle control functions to ensure safe distances and feasible obstacle avoidance directions, often result in infeasible solutions for obstacle avoidance on complex curved surfaces in confined environments. Therefore, there is an urgent need for a robot control method that can solve the aforementioned technical problems in the process of aircraft engine blade inspection. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a robot safety control method and system for expanding the safe operating space and guiding the obstacle avoidance direction for aircraft engine blade inspection, which solves the control problem that the robot cannot bypass the curved surface along the desired direction with a shorter path when the operating space is small and the obstacle avoidance direction path is of varying length.

[0006] To achieve the above objectives, according to one aspect of the present invention, a robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection is provided, the method comprising the following steps: The point cloud data of aircraft engine blades is partitioned, with each blade corresponding to a region. The point cloud in each region is clustered, and a hyperellipsoidal surface is used to enclose each cluster, thereby obtaining multiple hyperellipsoidal surfaces in each region. The robot moves to the blade according to a preset reference trajectory, calculates the intersection point of the hyperellipsoidal surface on the blade with the reference trajectory, and determines whether the robot needs to go around the surface boundary of the blade based on the angle between the intersection point and the normal of the blade surface. When it is necessary to bypass a curved surface boundary, calculate the obstacle avoidance direction and the required guiding force for the robot to bypass the curved surface boundary; combine the constraints of the robot bypassing the curved surface boundary, calculate the safe control force of the robot under the guiding force, convert the safe control force into the control torque of the robot joint, and control the robot to bypass the blade surface according to the control torque; Otherwise, the robot continues to inspect the blade surface according to the preset reference trajectory; thus, the robot is controlled during blade inspection.

[0007] More preferably, the clustering is performed according to the following steps: (11) Randomly select multiple points as initial cluster centers, calculate the distance between the points in the point cloud and the cluster centers, and take the point with the shortest distance from the center as a class, thereby dividing the points in the region into multiple classes; (12) Calculate the center of each class and return to step (11) with the calculated center of the class as the initial cluster center. (13) Repeat step (12) until the class obtained now is no different from the class obtained previously.

[0008] More preferably, the center position and attitude matrix of the hyperellipsoidal surface are determined according to the following steps: (21) Obtain the point closest to the center of each class, and calculate the position of the center of the hyperellipsoid surface according to the following formula: ,in, For the thickness of the curved surface, It is the identity matrix. The location of the point in each class that is closest to the center of the class. The normal vector of the point in each class that is closest to the center of the class; (22) Calculate the lengths of each type of point cloud along the X and Y axes of the world coordinate system; When the length along the X-axis is greater than the length along the Y-axis, the hyperellipsoid... shaft and The shafts are calculated according to the following formulas: , ,in, The X-axis of the world coordinate system. The Z-axis attitude vector of the hyperellipsoid; When the length along the Y-axis is greater than the length along the X-axis, the hyperellipsoid... shaft and The shafts are calculated according to the following formulas: , ,in, The Y-axis of the world coordinate system The Z-axis attitude vector of the hyperellipsoid; (23) Normal of the center of the class As a hyperellipsoid Axis, attitude matrix of the hyperellipsoid .

[0009] More preferably, the semi-axial length of the hyperellipsoid is determined according to the following steps: (31) Set each type of point cloud along the world coordinate system shaft and The length of the axis is taken as the hyperellipsoid along shaft and Initial half-axis length of the shaft and , Shaft half shaft length To directly set it as a multiple of the blade surface thickness, thus forming a hyperellipsoidal surface; (32) Obtain the outermost point of the hyperellipsoid. Calculate the point To the center of the superellipsoid vector In respectively shaft and Projection on axis and ; when ,along The half-axis length of the shaft is updated to Otherwise, along Half shaft length of the shaft Updated to , and For the scaling factor, return to step (31) until the point. Enclosed by a hyperellipsoid.

[0010] More preferably, the determination of whether the robot needs to bypass the curved boundary of the blade is performed in the following manner: Calculate the normal of the intersection point and the blade surface. The included angle , ; When the included angle When using this method, it is necessary to bypass the curved boundary of the blade; otherwise, it is not necessary.

[0011] More preferably, the obstacle avoidance direction is L in the following formula i The direction corresponding to the maximum value of i:

[0012] Where i = 1, 2, 3, the values ​​of i 1, 2, 3 correspond to the hyperellipsoids. Axial direction, As a weighting factor, The first of the hyperellipsoidal attitude matrix List, For the first One half-axis length, yes The first in the hyperellipsoidal coordinate system One component; The formula for calculating the guiding force is as follows:

[0013] in, This is the maximum value of the guiding force. This determines the growth rate of guiding force. The duration of the guiding force.

[0014] More preferably, the formula for calculating the safety control force is as follows:

[0015] in, It is a new nominal control. For safety control, For security certificates based on hyperellipsoidal functions, For the derivative of the security certificate, For input irrelevant Lie derivative components, To input the relevant Lie derivative components, and linear Function-like.

[0016] More preferably, the formulas for calculating the nominal tracking force and the new nominal tracking force are as follows:

[0017]

[0018] in, and The gain is controlled by the scaling-differential method in Cartesian space. and For the position and velocity of the reference trajectory, and This refers to the robot's end-effector position and velocity.

[0019] According to another aspect of the present invention, a robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, as described above, is provided, wherein the calculation formula for the control torque of the robot joint is as follows:

[0020] in, For the Jacobian matrix of the robot, For safety control.

[0021] According to another aspect of the present invention, a robot safety control system for expanding the safe operating space and guiding the obstacle avoidance direction for aircraft engine blade inspection is provided. The system includes an actuator for the robot safety control method for expanding the safe operating space and guiding the obstacle avoidance direction for aircraft engine blade inspection described above.

[0022] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art: 1. This invention first uses a hyperellipsoid to enclose clustered points on each blade, then determines the angle between the velocity of the intersection point of the reference trajectory and the hyperellipsoid and the surface normal, and decides whether to detour based on the result. This method enables obstacle avoidance guidance for the robot at the blade boundary. In situations where the operating space is small and the path lengths for different obstacle avoidance directions vary due to the tortuous characteristics of the blade, it achieves the task of detecting the blade surface with a shorter path.

[0023] 2. This invention employs an improved K-means clustering algorithm to partition the surface point cloud data, and then uses the partitioned point cloud with a rotated and scaled hyperellipsoidal envelope to achieve an accurate analytical representation of the surface boundary, effectively expanding the operable space.

[0024] 3. This invention generates a guiding force for the desired obstacle avoidance direction when the robot approaches the curved surface. The magnitude of the guiding force gradually increases and is combined with the nominal tracking force to form a new nominal control force. The higher-order obstacle control function converts the nominal control force into a safe control force, thereby enabling the robot to smoothly bypass the curved surface boundary along the desired direction with a shorter path and complete the detection task of multiple blade curved surfaces. Attached Figure Description

[0025] Figure 1 This is a robot safety control framework diagram for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, constructed according to a preferred embodiment of the present invention.

[0026] Figure 2 This is a flowchart of the improved K-means clustering algorithm according to a preferred embodiment of the present invention.

[0027] Figure 3This is a flowchart of the semi-major axis optimization algorithm according to a preferred embodiment of the present invention.

[0028] Figure 4 This is a schematic diagram of the obstacle avoidance direction guidance principle according to a preferred embodiment of the present invention.

[0029] Figure 5 The result is the envelope of the hyperellipsoidal envelope of the rotating and scaling according to the preferred embodiment of the present invention, ranging from simple to complex surfaces.

[0030] Figure 6 The simulation results of obstacle avoidance in the desired direction on three curved surfaces according to a preferred embodiment of the present invention are shown, wherein (a) is a schematic diagram of the three-dimensional trajectory and (b) is a partial enlarged view of the third curved surface for obstacle avoidance.

[0031] Figure 7 This is an experimental platform for aircraft engine blade model inspection according to a preferred embodiment of the present invention, wherein (a) is the experimental equipment and (b) is a magnified view and an endoscopic view.

[0032] Figure 8 The experimental trajectory of the aircraft engine blade model detection task according to a preferred embodiment of the present invention is shown in (a) as a schematic diagram of the three-dimensional trajectory and (b) as a magnified view of a part. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0034] like Figure 1 The diagram shows a robot safety control framework for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, which specifically includes the following steps: S1 utilizes point clouds partitioned from multiple rotated and scaled hyperellipsoidal envelope surfaces to achieve accurate analytical representation of surface boundaries, thereby obtaining a larger operable space.

[0035] S11 improves the K-means clustering algorithm to partition curved surface point clouds. The algorithm flowchart is as follows: Figure 2 As shown: Random selection Using 10 points as initial cluster centers, calculate the distance between the sample and the center, and select the center with the shortest distance as the first cluster:

[0036]

[0037] in For the first From point 1 to point 2 Improved distance to class center and They are the first The position vector and normal vector of the class center, The cosine similarity weighting factor is used. and They are the first The position vector and normal vector of each point It is the number of point clouds. It is the number of clusters. It is the first The number of point clouds in the class, For the first Point cloud in the class.

[0038] Recalculate the center of each class and determine whether the clustering result has changed compared with the previous clustering. If it has changed, recalculate the distance between the sample and the center and cluster again; otherwise, end the point cloud partitioning.

[0039] S12 uses the projection formula to calculate the center position and orientation of the hyperellipsoid required for the envelope partitioning point cloud: (1) The general equation of the hyperellipsoid can be expressed as: The matrix form can be represented as ,in The center of the hyperellipsoid. Let be any point on the surface of the hyperellipsoid. Let be the semi-axis length of the hyperellipsoid. A diagonal matrix describing the length of the semi-axis. Let be the attitude matrix of the hyperellipsoid. For shape parameters, when When it approaches infinity, the shape of the hyperellipsoid approximates a cube.

[0040] (2) Calculate the center location: Find the point closest to the class center in the point cloud of each class. By the nearest point normal to that point The center position of the hyperellipsoid is obtained by translating it in the opposite direction by a distance equal to half the thickness of the surface. , ,in For the thickness of the curved surface, It is an identity matrix.

[0041] (3) Calculate the attitude matrix: Definition The coordinate axes of the world coordinate system. These are the coordinate axes of the hyperellipsoid's own coordinate system. The normal to the center of the class is... As a hyperellipsoid The axis is used to calculate the point cloud along the world coordinate system. shaft and The length of the axis, along The length of the axis refers to the X in the class. max -X min The length along the Y-axis refers to the Y-axis length of the class. max -Y min .

[0042] If along The length of the shaft is greater than The length of the axis, then the hyperellipsoid Shaft may have get, Shaft may have get; If along The length of the shaft is greater than The length of the axis, then the hyperellipsoid Shaft may have get, Shaft may have get; Finally, the attitude matrix of the hyperellipsoid is obtained. , .

[0043] S13 uses a semi-major axis optimization algorithm to obtain the semi-axis length of the hyperellipsoid. The algorithm flowchart is as follows Figure 3 As shown: Based on each type of point cloud along the world coordinate system shaft and The length of the axis is obtained along the hyperellipsoid. shaft and Initial half-shaft length and , Shaft half shaft length Set directly to blade surface thickness The value after multiplying by the magnification factor. Find the outermost point that is not enveloped by the hyperellipsoid. Calculate the distance from this point to the center of the hyperellipsoid. vector In respectively shaft and Projection on axis and ,if So along The half-axis length of the shaft is updated to ; Otherwise, along Half shaft length of the shaft Updated to , and Use the scaling factor, increasing in steps of 0.05, until the outermost point is reached. Enclosed by a hyperellipsoid.

[0044] S2 utilizes a proportional-derivative controller to obtain the nominal tracking force for the robot to track the reference trajectory. When the robot approaches the surface boundary, it performs a surface normal approach judgment and generates a guiding force for the desired obstacle avoidance direction. This guiding force, combined with the nominal tracking force, yields a new nominal control force. Using a higher-order control obstacle function as a constraint, the new nominal control force is converted into a safe control force. This safe control force is used to control the robot to follow the desired obstacle avoidance direction and traverse the surface, ultimately completing the detection task of multiple surfaces. The principle diagram of the desired direction obstacle avoidance is shown below. Figure 4 As shown.

[0045] S21 calculates the nominal tracking force of the robot following the preset reference trajectory using a proportional-derivative controller as follows:

[0046] in and The gain is controlled by the scaling-differential method in Cartesian space. and For the position and velocity of the reference trajectory, and This refers to the robot's end-effector position and velocity.

[0047] S22 When the robot moves along the preset reference trajectory to the corresponding blade, calculate the intersection point of the hyperellipsoid in that blade with the reference trajectory. Then determine whether the reference velocity direction at that point is close to the surface normal: Based on the matrix form of the hyperellipsoid function Considering a certain safety margin, the shortest semi-axis length of the enclosed hyperellipsoid is enlarged by a factor of 1.5, and the lengths of the other semi-axis lengths are enlarged by a factor of 1.2. At that time, the reference trajectory intersects with the hyperellipsoid of the envelope surface. .

[0048] calculate Point speed With the unit normal of the surface The included angle , , where the surface normal It can be approximated by the direction of the shortest axis of the hyperellipsoid.

[0049] When the included angle When the reference trajectory approaches the blade surface roughly along the normal direction of the blade surface, it is necessary to consider that the robot can bypass the surface boundary with a shorter detour along the desired obstacle avoidance direction.

[0050] Otherwise, it will not avoid obstacles and will continue to move according to the predicted reference trajectory.

[0051] S23 The generation of the desired obstacle avoidance direction guiding force includes determining the desired obstacle avoidance direction and calculating the magnitude of the guiding force: pass The values ​​of i, 1, 2, and 3, correspond to the X, Y, and Z directions, respectively. The desired obstacle avoidance direction is determined, and the calculation is performed. The maximum value, at which point the index The direction of the semi-axis of the corresponding hyperellipsoid is the desired obstacle avoidance direction. As a weighting factor, The first of the hyperellipsoidal attitude matrix List, For the first One half-axis length, yes The first in the hyperellipsoidal coordinate system Each component.

[0052] pass The magnitude of the guiding force is calculated and gradually increased to prevent sudden movement of the robot's end effector. This is the maximum value of the guiding force. This determines the growth rate of guiding force. For the duration of the guiding force, at the intersection Guiding force begins to be generated at a certain point, and when the motion reaches a point on the hyperellipsoid... Make The guiding force disappears.

[0053] S24 Constructs higher-order control barrier function constraints, and uses a quadratic programming algorithm to calculate the safe control force that meets the safety constraints and is closest to the nominal control force:

[0054] in The sum of guiding force and nominal tracking force constitutes the new nominal control force. For safety control, For security certificates based on hyperellipsoidal functions, For the derivative of the security certificate, For input irrelevant Lie derivative components, To input the relevant Lie derivative components, and for Function-like.

[0055] S25 obtains the joint torque for controlling the robot by transposing the calculated safety control force using the Jacobian transpose: ,in Let be the Jacobian matrix of the robot. The method of rotating and scaling the hyperellipsoidal envelope surface described above was used to complete the envelope of surfaces with shapes ranging from simple to complex, such as... Figure 5 As shown, the number of hyperellipsoids is 2, 2, 4, and 6 respectively.

[0056] The obstacle avoidance results for the three curved surfaces using the above-described desired direction-guided obstacle avoidance method are as follows: Figure 6 As shown, the reference trajectory sequentially verifies that the X-axis negative direction, Z-axis positive direction, and X-axis negative direction pass through the curved surface. After adopting the obstacle avoidance method guided by the desired direction, the actual trajectory sequentially verifies that the X-axis negative direction, Z-axis positive direction, and X-axis negative direction bypass the curved surface.

[0057] To verify the effectiveness of the proposed hyperellipsoidal envelope and desired direction-guided obstacle avoidance method, accuracy verification was conducted using an embodiment. In this embodiment, an aircraft engine blade model inspection task platform was built for experimental verification, such as... Figure 7 As shown, the experiment includes the Franka robot, an industrial endoscope, and a model of an aircraft engine blade. In experimental testing, given a reference trajectory traversing the blade surface, the robot tracks the reference trajectory to detect the blade, and when approaching the blade, it bypasses it in the desired direction.

[0058] When using the method proposed in this invention to perform aircraft engine blade model inspection, the method first accurately encloses the blade boundary using a hyperellipsoid to obtain a larger operable space, and then guides the robot to bypass the blade boundary along the desired direction by using the desired obstacle avoidance direction. Figure 8 As shown, when the robot is far from the blade boundary, it tracks the reference trajectory, while when it is close to the blade boundary, it bypasses the blade boundary along the positive Z-axis in a shorter path.

[0059] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, characterized in that, The method includes the following steps: The point cloud data of aircraft engine blades is partitioned, with each blade corresponding to a region. The point cloud in each region is clustered, and a hyperellipsoidal surface is used to enclose each cluster, thereby obtaining multiple hyperellipsoidal surfaces in each region. The robot moves to the blade according to a preset reference trajectory, calculates the intersection point of the hyperellipsoidal surface on the blade with the reference trajectory, and determines whether the robot needs to go around the surface boundary of the blade based on the angle between the intersection point and the normal of the blade surface. When it is necessary to bypass a curved surface boundary, calculate the obstacle avoidance direction and the required guiding force for the robot to bypass the curved surface boundary; combine the constraints of the robot bypassing the curved surface boundary, calculate the safe control force of the robot under the guiding force, convert the safe control force into the control torque of the robot joint, and control the robot to bypass the blade surface according to the control torque; Otherwise, the robot continues to inspect the blade surface according to the preset reference trajectory; thus, the robot is controlled during blade inspection.

2. The robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 1, characterized in that, The clustering is performed according to the following steps: (11) Randomly select multiple points as initial cluster centers, calculate the distance between the points in the point cloud and the cluster centers, and take the point with the shortest distance from the center as a class, thereby dividing the points in the region into multiple classes; (12) Calculate the center of each class and return to step (11) with the calculated center of the class as the initial cluster center. (13) Repeat step (12) until the class obtained now is no different from the class obtained previously.

3. A robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 1 or 2, characterized in that, The center position and attitude matrix of the hyperellipsoidal surface are determined according to the following steps: (21) Obtain the point closest to the center of each class, and calculate the position of the center of the hyperellipsoid surface according to the following formula: ,in, For the thickness of the curved surface, It is the identity matrix. The location of the point in each class that is closest to the center of the class. The normal vector of the point in each class that is closest to the center of the class; (22) Calculate the lengths of each type of point cloud along the X and Y axes of the world coordinate system; When the length along the X-axis is greater than the length along the Y-axis, the hyperellipsoid... shaft and The shafts are calculated according to the following formulas: , ,in, The X-axis of the world coordinate system. The Z-axis attitude vector of the hyperellipsoid; When the length along the Y-axis is greater than the length along the X-axis, the hyperellipsoid... shaft and The shafts are calculated according to the following formulas: , ,in, The Y-axis of the world coordinate system The Z-axis attitude vector of the hyperellipsoid; (23) Normal of the center of the class As a hyperellipsoid Axis, attitude matrix of the hyperellipsoid .

4. The robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 3, characterized in that, The semi-axial length of the hyperellipsoid is determined according to the following steps: (31) Set each type of point cloud along the world coordinate system shaft and The length of the axis is taken as the hyperellipsoid along shaft and Initial half-axis length of the shaft and , Shaft half shaft length To directly set it as a multiple of the blade surface thickness, thus forming a hyperellipsoidal surface; (32) Obtain the outermost point of the hyperellipsoid. Calculate the point To the center of the superellipsoid vector In respectively shaft and Projection on axis and ; when ,along The half-axis length of the shaft is updated to Otherwise, along Half shaft length of the shaft Updated to , and For the scaling factor, return to step (31) until the point. Enclosed by a hyperellipsoid.

5. A robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 1 or 4, characterized in that, The determination of whether the robot needs to bypass the curved boundary of the blade is performed as follows: Calculate the normal of the intersection point and the blade surface. The included angle , ; When the included angle When using this method, it is necessary to bypass the curved boundary of the blade; otherwise, it is not necessary.

6. The robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 1, characterized in that, The obstacle avoidance direction is L in the following formula. i The direction corresponding to the maximum value of i: Where i = 1, 2, 3, the values ​​of i 1, 2, 3 correspond to the hyperellipsoids. Axial direction, As a weighting factor, The first of the hyperellipsoidal attitude matrix List, For the first One half-axis length, yes The first in the hyperellipsoidal coordinate system One component; The formula for calculating the guiding force is as follows: in, This is the maximum value of the guiding force. This determines the growth rate of guiding force. The duration of the guiding force.

7. The robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 1, characterized in that, The formula for calculating the safety control force is as follows: in, It is a new nominal control. For safety control, For security certificates based on hyperellipsoidal functions, For the derivative of the security certificate, For input irrelevant Lie derivative components, To input the relevant Lie derivative components, and linear Function-like.

8. The robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 7, characterized in that, The formula for calculating the new nominal control force is as follows: in, and The gain is controlled by the scaling-differential method in Cartesian space. and For the position and velocity of the reference trajectory, and This refers to the robot's end-effector position and velocity.

9. A robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection as described in claim 8, characterized in that, The formula for calculating the control torque of the robot joint is as follows: in, For the Jacobian matrix of the robot, For safety control.

10. A robot safety control system for expanding safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, characterized in that, The system includes an actuator for performing a robot safety control method for expanding the safe operating space and guiding obstacle avoidance direction for aircraft engine blade inspection, as described in any one of claims 1-9.

Images