A multi-mode switching system control strategy and a parameter design method thereof

Abstract

The application discloses a kind of multi-mode switching system control strategies and its parameter design method.The application respectively establishes the control equation constraint of switching control of admittance control and impedance control, variable parameter impedance control and switching control of impedance control and position control, then give the switching control parameter that meets admittance control and impedance control and impedance control and position control, finally determine the switching control parameter of robot system that meets the stability of robot system switching, to switch robot system switching control;The application establishes the switching control strategy based on state space for the autonomous switching between multi-mode such as admittance control, impedance control and position control, with improve system autonomy, enhance system operation precision, overcome the system time delay and vibration caused by traditional manual switching of operator and other deficiencies.

Description

TECHNICAL FIELD

[0001] The present application relates to a robot-human interaction control method, in particular to a multi-mode switching system control strategy and a parameter design method thereof. BACKGROUND

[0002] Admittance control, impedance control, position control and other control strategies are commonly used robot control strategies, especially the former two are commonly used as force control interaction strategies between robots and humans or between robots and the environment. In general simple robot applications, one of the control strategies is usually used as the main control strategy. However, in complex robot application tasks, multiple control modes are often involved in control. For example, in the orthopedic operation of a surgical robot, the robot end is first guided to the operation area, the second step is to pass through the surgical guide hole, and the third step is to complete the surgical operation according to the planned path.

[0003] The above three steps of operation involve admittance control, impedance control and position control. In the existing operation, the switching between these control modes needs to be paused and switched manually, which brings problems such as unsmooth operation process, increased operation time, vibration interference, and reduced precision of autonomous robot operation due to manual switching.

[0004] Multi-mode switching is a hot research topic. For example, foreign researchers have proposed a concept based on duty cycle, which enables the control system to quickly switch between admittance control and impedance control and achieve system stability. However, this mode switching is usually based on time variables. Researchers have also conducted multi-mode switching based on sliding mode control, but this switching often causes vibration when the robot system switches on the sliding surface.

[0005] The aforementioned complex robot operation task involves switching from admittance control to impedance control and then from impedance control to position control. The switching system is based on state variables, and there is currently no corresponding multi-mode autonomous switching control strategy. SUMMARY

[0006] To solve the problems in the prior art, the present application provides a multi-mode switching system control strategy and a parameter design method thereof, which overcomes the lack of autonomy in switching between admittance control, impedance control and position control, and has the advantages of improving human-robot interaction efficiency and enhancing robot operation precision.

[0007] The multi-mode switching system control strategy and the parameter design method thereof of the present application comprise the following steps:

[0008] 1) System setting:

[0009] a) The multi-mode switching system control strategy and its parameter design method is for a robot system, a terminal tool is installed at the end of the robot, a terminal force sensor is installed between the terminal tool and the joint of the robot end, forming a robot system, and the terminal force sensor can sense the terminal interaction force applied on the robot end by the environment through the terminal tool;

[0010] The terminal tool is guided to reach the guide hole, then passes through the narrow guide hole to reach the target object surface, and finally completes the operation on the target object; The space allowing the movement of the robot end guided by the terminal tool is defined as a Cartesian impedance control motion space S1, and the impedance control operation space where the guide hole is located is defined as a Cartesian impedance control motion space S2, in which space the robot end is autonomously controlled to complete the plug-in force control operation of the terminal tool in the narrow guide hole according to the set operation requirements, and the space where the terminal tool completes the operation on the target object is defined as a Cartesian position control motion space S3; The target object surface is located on the interface between the Cartesian impedance control motion space S2 and the Cartesian position control motion space S3.

[0011] The robot system sequentially experiences the impedance control of the Cartesian impedance control motion space S1, the switching control of the impedance control and the impedance control at the interface between the Cartesian impedance control motion space S1 and the Cartesian impedance control motion space S2, the impedance control of the Cartesian impedance control motion space S2, the switching control of the impedance control and the position control at the interface between the Cartesian impedance control motion space S2 and the Cartesian position control motion space S3, and the position control of the Cartesian position control motion space S3, thereby completing the operation on the target object.

[0012] b) The dynamic equation of the robot system in the Cartesian space is:

[0013]

[0014] In the above formula, M(X) is the inertia matrix of the robot in the Cartesian space, is the centrifugal force and Coriolis force matrix of the robot in the Cartesian space, and G(X) is the gravity matrix of the robot in the Cartesian space; X is the actual position of the robot end in the Cartesian space, including three translational variables and three rotational variables; F is the output control total force of the robot end in the Cartesian space; F ext is the terminal interaction force between the robot end in the Cartesian space and the environment, which is obtained by the terminal force sensor;

[0015] 2) Switching control of impedance control and impedance control is established:

[0016] a) The switching control of impedance control and impedance control satisfies:

[0017]

[0018] In the above formula, F a is the admittance control output of the Cartesian space, F i is the impedance control output of the Cartesian space;

[0019] b) The admittance control output that meets the switching control requirement is:

[0020]

[0021] In the above formula, K p and K d are the admittance control position proportional coefficient and the admittance control velocity proportional coefficient respectively, and both are positive definite diagonal matrices; a X d is the admittance control calculation position command, which is the key to realize the switching between the admittance control and the impedance control;

[0022] c) The admittance control calculation position command a X d also needs to meet the impedance control equation:

[0023]

[0024] In the above formula, M θ , D θ and K θ are the impedance control inertia matrix, the impedance control damping matrix and the impedance control stiffness matrix respectively, and all are positive definite diagonal matrices; a X0 is the expected position value of the expected position X0 on the interface between the Cartesian admittance control motion space and the Cartesian impedance control motion space, and meets and

[0025] 3) Establish variable parameter impedance control:

[0026] After the robot system completes the first switching on the interface between the Cartesian admittance control motion space S1 and the Cartesian impedance control motion space S2, it enters the Cartesian impedance control motion space S2 and switches to impedance control. When the robot end is in collision or contact with the target object surface, it is hoped that the collision is minimized during this process, so the variable parameter control setting is performed in the impedance control.

[0027] a) In the impedance control, the actual position of the robot end should meet the impedance control equation:

[0028]

[0029] b) The impedance control output that meets the switching control is:

[0030]

[0031] In the above formula, I is a unit matrix, X0 is the desired position;

[0032] c) Cartesian impedance control motion space division:

[0033] By setting the impedance control inertia matrix M θ and the impedance control damping matrix D θ , the end tool can be guided to complete the plug-in force control operation in the narrow guide hole, and the end tool contacts the target object surface at this stage;

[0034] Since the modeling of the narrow guide hole cannot be completely accurate, and there is an error in the position feedback in the narrow guide hole, it is necessary to establish a force control to guide the end tool to safely contact the target object surface based on the end interaction force;

[0035] The Cartesian impedance control motion space between the Cartesian guide control motion space and the Cartesian position control motion space is divided into J parts, which are respectively the Cartesian impedance control first motion space S 21 to the Cartesian impedance control Jth motion space S 2J , J is a natural number ≥2;

[0036] d) Set the inertia matrix and damping matrix of impedance control:

[0037] In the Cartesian impedance control first motion space, set the inertia matrix and damping matrix of impedance control:

[0038] M θ = 1 M θ , D θ = 1 D θ 1 M θ and 1 D θ are respectively the inertia matrix and damping matrix of impedance control in the Cartesian impedance control first motion space;

[0039] In the Cartesian impedance control jth motion space, set the inertia matrix and damping matrix of impedance control:

[0040] M θ = j M θ = γ j · j-1 M θ , D θ = j D θ = γ j · j-1 D θ

[0041] j M θ and j D θ are inertia matrix and damping matrix of impedance control in the j-th Cartesian impedance control motion space, respectively, j = 2, …, J, γ j is the amplification factor of the j-th Cartesian impedance control motion space relative to the (j-1)-th Cartesian impedance control motion space;

[0042] The J parts of the Cartesian impedance control motion space in the above impedance control correspondingly adopt different impedance control inertia matrix and damping matrix variable settings, which can quickly and smoothly adjust the plug-in direction when the end tool enters the guide hole, and can also ensure that the impact force is minimized when the end tool contacts the target surface and the robot end speed is reduced to zero in the shortest time;

[0043] e) For impedance control, simplify its control strategy, set K θ to 0;

[0044] 4) Switching control between impedance control and position control is established:

[0045] a) The switching control between impedance control and position control satisfies:

[0046]

[0047] In the above formula, F p is the position control output of the Cartesian space, and the robot system can control the end tool to complete the operation on the target object based on the position control output;

[0048] b) The position control output that meets the switching control is:

[0049]

[0050] In the above formula, N p and N d are the position control position proportional coefficient and the position control speed proportional coefficient, respectively, and are both positive definite diagonal matrices;

[0051] c) p X0 is the expected position value of the desired position X0 on the interface between the Cartesian impedance control motion space and the Cartesian position control motion space, and is the key to realize the switching between impedance control and position control, therefore, the desired position

[0052] X0 is the expected position value of the desired position X0 on the interface between the Cartesian impedance control motion space and the Cartesian position control motion space p X0 needs to satisfy the impedance control equation:

[0053]

[0054] In the above formula, simultaneously satisfy

[0055] 5) Determine the switching control parameters of the robot system that satisfy the switching stability of the robot system:

[0056] a) Define the switching control state space variable Φ:

[0057]

[0058] b) Give the switching control parameter design requirements that satisfy admittance control and impedance control:

[0059] The switching control state space description equation that satisfies admittance control is:

[0060]

[0061]

[0062] In the above formula, the positive definite diagonal matrix K e is the environmental stiffness parameter;

[0063] The switching control state space description equation that satisfies impedance control is:

[0064]

[0065]

[0066] The switching control state space description equation of admittance control and impedance control is:

[0067]

[0068] In order to satisfy the switching control, the admittance control position proportional coefficient K p and the admittance control velocity proportional coefficient K d , the impedance control inertia matrix M θ , the impedance control damping matrix D θ , the impedance control stiffness matrix K θ , and the environmental stiffness parameter K e should be such that for any real number α belonging to 0≤α≤1, the admittance impedance switching confusion first matrix T ia and the admittance impedance switching confusion second matrix T ai both satisfy the Hurwitz matrix, where the admittance impedance

[0069] The switching confusion first matrix T ia and the admittance impedance switching confusion second matrix T ai are respectively:

[0070] T ia = a · T a + (1 - a) · T i

[0071]

[0072] If admittance control position proportional coefficient K p and admittance control velocity proportional coefficient K d , impedance control inertia matrix M θ , impedance control damping matrix D θ and impedance control stiffness matrix K θ and environment stiffness parameter K e , so that for any real number a belonging to 0≤a≤1, matrix T ia and T ai satisfy Hurwitz matrix, then the set of parameters can be used as stable admittance control and impedance control switching control parameters to control the switching of admittance control and impedance control; if it does not satisfy Hurwitz matrix, the parameters are reselected until the parameters satisfying Hurwitz matrix are obtained;

[0073] c) Give the switching control parameter design requirements of the robot system satisfying impedance control and position control:

[0074] The switching control state space description equation satisfying position control is:

[0075]

[0076]

[0077] The switching control state space description equation of impedance control and position control is:

[0078]

[0079] In order to satisfy the switching control, impedance control inertia matrix M θ , impedance control damping matrix D θ and impedance control stiffness matrix K θ , environment stiffness parameter K e , position control position proportional coefficient N p and position control velocity proportional coefficient N d , should be such that for any real number a belonging to 0≤a≤1, impedance position switching confusion first matrix T ip and impedance position switching confusion second matrix T pi satisfy Hurwitz matrix, wherein the impedance position switching confusion first matrix T ipImpedance position switching confusion second matrix T pi Respectively:

[0080] T ip = alpha * T i + (1-alpha) * T p

[0081]

[0082] If impedance control inertia matrix M θ , impedance control damping matrix D θ And impedance control stiffness matrix K θ , environmental stiffness parameter K e , position control position proportional coefficient N p And position control velocity proportional coefficient N d , so that for any real number alpha belonging to 0 <= alpha <= 1, impedance position switching confusion first matrix T ip And impedance position switching confusion second matrix T pi Both satisfy Hurwitz matrix, then the set of parameters can be used as stable impedance control and position control switching control parameters to control the switching of impedance control and position control; if not satisfy Hurwitz matrix, then reselect parameters until the parameters satisfying Hurwitz matrix are obtained;

[0083] Switching system state space description is:

[0084]

[0085] Wherein, in step 3) d), J is a natural number satisfying 2 <= J <= 5; gamma j Is 1-6.

[0086] Switch to position control to complete the operation on the target object includes cutting, drilling, grinding and the like. The aperture of the narrow guide hole is 2-10mm.

[0087] The advantages of the present application are:

[0088] The robot system switching control of the present application establishes a switching control strategy based on state space for the autonomous switching between admittance control, impedance control and position control and the like, has the advantages of improving system autonomy, enhancing system operation precision, overcoming the system time delay and vibration caused by manual switching of the operator in the traditional way and the like. BRIEF DESCRIPTION OF DRAWINGS

[0089] Figure 1 The robot system switching task operation schematic diagram of the present application;

[0090] Figure 2Schematic diagram of autonomous switching of admittance control / impedance control / position control for a robot system. DETAILED DESCRIPTION

[0091] The application will be further described below with reference to the drawings and specific embodiments.

[0092] The multi-mode switching system control strategy and its parameter design method of the embodiment include the following steps:

[0093] 1) System setting:

[0094] a) The multi-mode switching system control strategy and its parameter design method are directed to a robot system, as shown in the figure, a terminal tool is installed at the end of the robot, a terminal force sensor 3 is installed between the terminal tool 2 at the end of the robot 1 and the end joint of the robot, forming a robot system, and the terminal force sensor can sense the terminal interaction force applied on the end of the robot by the environment through the terminal tool; Figure 1 The terminal tool needs to reach the guide hole under guidance, and after the terminal tool is guided to reach the guide hole, it passes through the narrow guide hole 4 to reach the surface of the target object 5, and finally completes the operation on the target object 5; the space allowing the movement of the robot end guided by the terminal tool is defined as a Cartesian admittance control movement space S1, the impedance control operation space where the guide hole is located is a Cartesian impedance control movement space S2, in this space, the robot end is autonomously controlled to achieve the plug-in force control operation of the terminal tool in the narrow guide hole according to the set operation requirements, and the space where the terminal tool completes the operation on the target object is defined as a Cartesian position control movement space S3; the surface of the target object is located on the interface of the Cartesian impedance control movement space S2 and the Cartesian position control movement space S3.

[0095] i S = S1, S2, S3; Figure 2

[0096] The robot system sequentially experiences admittance control in the Cartesian admittance control movement space S1, switching control of admittance control and impedance control at the interface of the Cartesian admittance control movement space S1 and the Cartesian impedance control movement space S2, impedance control in the Cartesian impedance control movement space S2, switching control of impedance control and position control at the interface of the Cartesian impedance control movement space S2 and the Cartesian position control movement space S3, and position control in the Cartesian position control movement space S3, thereby completing the operation on the target object;

[0097] b) The dynamics equation of the robot system in the Cartesian space is:

[0098]

[0099] ​​In the formula, M(X) is the inertia matrix of the Cartesian space robot, is the centrifugal force and Coriolis moment matrix of the Cartesian space robot, G(X) is the gravity matrix of the Cartesian space robot; X is the actual position of the robot end in the Cartesian space, which contains three translational variables and three rotational variables; F is the output control force of the Cartesian space robot end; F ext is the end interaction force between the Cartesian space robot end and the environment, which is obtained by the end force sensor;

[0100] 2) Switching control of admittance control and impedance control is established:

[0101] a) The switching control of admittance control and impedance control satisfies:

[0102]

[0103] In the formula, F a is the admittance control output in the Cartesian space, F i is the impedance control output in the Cartesian space; S1 is the Cartesian admittance control motion space, in which the robot end is guided by the end tool to move; S2 is the Cartesian impedance control motion space, in which the robot end is autonomously controlled, and the plug-in force control operation of the end tool is completed in the narrow guide hole space;

[0104] b) The admittance control output meeting the switching control requirement is:

[0105]

[0106] In the formula, K p and K d are the position proportional coefficient and the velocity proportional coefficient of the admittance control respectively, and are positive definite diagonal matrices; a X d is the admittance control calculation position command, which is the key to realize the switching of admittance control and impedance control;

[0107] c) a X d Also needs to satisfy the impedance control equation:

[0108]

[0109] In the formula, M θ , D θ and K θ are the inertia matrix, the damping matrix and the stiffness matrix of the impedance control respectively, and are positive definite diagonal matrices; a X0 is the expected position value of the expected position X0 on the interface between the Cartesian admittance control motion space and the Cartesian impedance control motion space, which satisfies and

[0110] 3) Establish variable parameter impedance control:

[0111] The robot system enters the Cartesian impedance control motion space S2 after the first switch on the interface of the Cartesian guide control motion space S1 and the Cartesian impedance control motion space S2, and switches to impedance control; When the robot end collides or contacts with the target surface, that is, the impact force quickly stabilizes to zero, and it is hoped that the collision is minimized during this process, so the variable parameter control is set in the impedance control, and the impact force falling to zero also helps to complete the switch from impedance control to position control;

[0112] a) In impedance control, the actual position of the robot end should satisfy the impedance control equation:

[0113]

[0114] b) The impedance control law that meets the switching control is:

[0115]

[0116] In the above formula, I is the unit matrix;

[0117] c) Division of Cartesian impedance control motion space:

[0118] By setting the impedance control inertia matrix M θ and the impedance control damping matrix D θ , the end tool can be guided to complete the plug-in force control operation in the narrow guide hole, and the end tool contacts the target surface at this stage;

[0119] Because the modeling of the narrow guide hole cannot be completely accurate, and there is also an error in the position feedback in the narrow guide hole, it is necessary to establish force control based on the end interaction to guide the end tool to safely contact the target surface;

[0120] Divide the Cartesian impedance control motion space between the Cartesian guide control motion space and the Cartesian position control motion space into three parts, the first Cartesian impedance control motion space S 21 near the Cartesian guide control motion space, the third Cartesian impedance control motion space S 23 near the Cartesian position control motion space, and the second Cartesian impedance control motion space S 22 ;

[0121] d) Set the inertia matrix and damping matrix of impedance control:

[0122] In the Cartesian impedance control first movement space, set the inertia matrix and damping matrix of the impedance control:

[0123] M θ = 1 M θ , D θ = 1 D θ

[0124] 1 M θ and 1 D θ are respectively the inertia matrix and damping matrix of the impedance control in the Cartesian impedance control first movement space;

[0125] In the Cartesian impedance control second movement space, set the inertia matrix and damping matrix of the impedance control:

[0126] M θ = 2 M θ = 3· 1 M θ , D θ = 2 D θ = 3· 1 D θ

[0127] 2 M θ and 2 D θ are respectively the inertia matrix and damping matrix of the impedance control in the Cartesian impedance control second movement space;

[0128] In the Cartesian impedance control third movement space, set the inertia matrix and damping matrix of the impedance control:

[0129] M θ = 3 M θ = 3· 2 M θ , D θ = 3 D θ = 3· 2 D θ

[0130] 3 M θ and 3 D θ are respectively the inertia matrix and damping matrix of the impedance control in the Cartesian impedance control third movement space;

[0131] The inertia matrix and damping matrix of the impedance control are set differently in the three parts of the Cartesian impedance control motion space, which ensures that the end tool can quickly and smoothly adjust the insertion direction when entering the guide hole, and also ensures that the end tool can form a small impact force when contacting the target surface and quickly reduce the robot end speed to zero;

[0132] e) For impedance control, simplify the control strategy, i.e. set K θ to 0;

[0133] 6) Switching control between impedance control and position control is established:

[0134] a) The switching control between impedance control and position control satisfies:

[0135]

[0136] In the above formula, F p is the position control output in the Cartesian space, and the robot system can control the end tool to complete the operation on the target object based on the position control output;

[0137] b) The position control output that meets the switching control is:

[0138]

[0139] In the above formula, N p and N d are the position control position proportional coefficient and the position control speed proportional coefficient respectively, and both are positive definite diagonal matrices;

[0140] c) p X0 is the expected position value of the desired position X0 on the interface between the Cartesian impedance control motion space and the Cartesian position control motion space, which is the key to realizing the switching between impedance control and position control, therefore, the desired position

[0141] X0 is the expected position value of the desired position X0 on the interface between the Cartesian impedance control motion space and the Cartesian position control motion space p X0 needs to satisfy the impedance control equation:

[0142]

[0143] 4) In the above formula, since the end interaction force F ext tends to 0, therefore satisfies

[0144] 5) Determine the switching control parameters of the robot system that meet the switching stability of the robot system:

[0145] a) Define the switching control state space variable Φ:

[0146]

[0147] b) give the switching control parameter design requirements that satisfy the admittance control and impedance control:

[0148] The switching control state space description equation that satisfies the admittance control is:

[0149]

[0150]

[0151] In the above formula, the positive definite diagonal matrix K e is an environmental stiffness parameter;

[0152] The switching control state space description equation that satisfies the impedance control is:

[0153]

[0154]

[0155] The switching control state space description equation of admittance control and impedance control is:

[0156]

[0157] In order to satisfy the switching control, the admittance control position proportional coefficient K p and the admittance control speed proportional coefficient K d , the impedance control inertia matrix M θ , the impedance control damping matrix D θ and the impedance control stiffness matrix K θ and the environmental stiffness parameter K e should be such that for any real number α belonging to 0≤α≤1, the admittance impedance switching confusion first matrix T ia and the admittance impedance switching confusion second matrix T ai both satisfy the Hurwitz matrix, where the admittance impedance switching confusion first matrix T ia and the admittance impedance switching confusion second matrix T ai are respectively:

[0158] T ia =α·T a +(1-α)· T i

[0159]

[0160] A set of admittance control position proportional coefficients K p and admittance control speed proportional coefficients Kd , impedance control inertia matrix M θ , impedance control damping matrix D θ , and impedance control stiffness matrix K θ , and environmental stiffness parameter K e Then, Hurwitz matrix is used for verification. If the set of parameters makes the matrix T ia and T ai satisfy Hurwitz matrix for any real number α belonging to 0≤α≤1, the set of parameters can be used as the switching control parameter of admittance control and impedance control for controlling the switching of admittance control and impedance control. If Hurwitz matrix is not satisfied, the parameters are reselected within the range of experience value until the parameters satisfying Hurwitz matrix are obtained.

[0161] c) The switching control parameter design requirement of the robot system satisfying impedance control and position control is given:

[0162] The switching control state space description equation satisfying position control is:

[0163]

[0164]

[0165] The switching control state space description equation of impedance control and position control is:

[0166]

[0167] In order to satisfy the switching control, impedance control inertia matrix M θ , impedance control damping matrix D θ , and impedance control stiffness matrix K θ , environmental stiffness parameter K e , position control position proportional coefficient N p , and position control velocity proportional coefficient N d should make the impedance position switching confusion first matrix T ip and the impedance position switching confusion second matrix T pi satisfy Hurwitz matrix for any real number α belonging to 0≤α≤1, wherein the impedance position switching confusion first matrix T ip and the impedance position switching confusion second matrix T pi are respectively:

[0168] T pi = α·T i +(1-α)·T p

[0169]

[0170] a set of impedance control inertia matrix M θ , impedance control damping matrix D θ and impedance control stiffness matrix K θ , environmental stiffness parameter K e , position control position proportional coefficient N p and position control velocity proportional coefficient N d , then using Hurwitz matrix verification, if the set of parameters makes any real number α belonging to 0≤α≤1, impedance position switching confusion first matrix T ip and impedance position switching confusion second matrix T pi satisfy Hurwitz matrix, the set of parameters can be used as stable impedance control and position control switching control parameters to control the switching of impedance control and position control; if it does not satisfy Hurwitz matrix, the parameters are reselected in the range of empirical values until the parameters satisfying Hurwitz matrix are obtained;

[0171] The switching system state space description is:

[0172]

[0173] The range of empirical values of the parameters: admittance control position proportional coefficient K p is: 0.1·I~4·I, admittance control velocity proportional coefficient K d is: 0.01·I~0.3·I, impedance control inertia matrix M θ is: 0.8·I~10·I, impedance control damping matrix D θ is: 9·I~140·I, impedance control stiffness matrix K θ is: 100·I~1000·I, environmental stiffness parameter K e is: 500·I~2000·I, position control position proportional coefficient N p is: 0.1·I~4·I, position control velocity proportional coefficient N d is: 0.01·I~0.3·I.

[0174] Finally, it should be noted that the purpose of the disclosed embodiments is to help further understand the present application, but those skilled in the art can understand that various substitutions and modifications are possible without departing from the spirit and scope of the present application and the appended claims. Therefore, the present application should not be limited to the disclosed content, and the scope of the present application claimed is the scope defined by the claims.

Claims

1. A control strategy and parameter design method for a multi-mode switching system, characterized in that, The control strategy and parameter design method for the multi-mode switching system include the following steps: 1) System Settings: a) Multi-mode switching system control strategy and parameter design method are oriented towards robot systems. An end tool is installed at the end of the robot, and an end force sensor is installed between the end tool and the end joint of the robot to form a robot system. The end force sensor can sense the end interaction force applied to the end of the robot by the environment through the end tool. After the end-effector is guided to the guide hole, it passes through the narrow guide hole to reach the surface of the target object, and finally completes the operation on the target object. The space that allows the robot end to move by guiding the end of the robot through the end-effector is defined as the Cartesian admittance control motion space S1, and the impedance control operation space where the guide hole is located is defined as the Cartesian impedance control motion space S2. In this space, the robot end is autonomously controlled to realize the insertion and extraction force control operation of the end-effector within the narrow guide hole according to the set operation requirements. The space in which the end-effector completes the operation on the target object is defined as the Cartesian position control motion space S3. The surface of the target object is located at the interface between the Cartesian impedance control motion space S2 and the Cartesian position control motion space S3. The robot system sequentially undergoes admittance control in Cartesian admittance control motion space S1, switching between admittance and impedance control at the interface between Cartesian admittance control motion space S1 and Cartesian impedance control motion space S2, impedance control in Cartesian impedance control motion space S2, switching between impedance and position control at the interface between Cartesian impedance control motion space S2 and Cartesian position control motion space S3, and finally position control in Cartesian position control motion space S3, thereby completing the operation on the target object. b) The dynamic equations of the robot system in Cartesian space are: In the above formula, M(X) is the Cartesian inertial matrix of the space robot. Let G(X) be the centrifugal and Coriolis force matrices of the Cartesian space robot, and G(X) be the gravity matrix of the Cartesian space robot; X is the actual position of the robot's end effector in Cartesian space, which includes three translational variables and three rotational variables; F is the resultant output control force of the Cartesian space robot's end effector; F ext The force of interaction between the end effector of the Cartesian space robot and its environment is obtained from the end effector force sensor. 2) Establish a switching control between admittance control and impedance control: a) The switching control between admittance control and impedance control satisfies: In the above formula, F a For the admittance control output in Cartesian space, F i For impedance control output in Cartesian space; b) The admittance control output that meets the switching control requirements is: In the above formula, K p and K d These are the admittance control position proportional coefficient and the admittance control velocity proportional coefficient, respectively, and both are positive definite diagonal matrices; a X d The position command for admittance control calculation is the key to switching between admittance control and impedance control; c) Admittance control calculation position command a X d At the same time, the impedance control equation must also be satisfied: In the above formula, M θ D θ and K θ These are the impedance-controlled inertia matrix, impedance-controlled damping matrix, and impedance-controlled stiffness matrix, respectively, and all are positive definite diagonal matrices; a X0 is the desired position value of X0 at the interface between the Cartesian admittance-controlled motion space and the Cartesian impedance-controlled motion space, while satisfying... and 3) Establish variable parameter impedance control: After the robot system completes the first switch at the interface between the Cartesian admittance control motion space S1 and the Cartesian impedance control motion space S2, it enters the Cartesian impedance control motion space S2 and switches to impedance control. When the robot end effector collides or comes into contact with the surface of the target object, it is desirable to minimize the collision during this process. Therefore, variable parameter control is set in the impedance control. a) During impedance control, the actual position of the robot's end effector should satisfy the impedance control equation: b) The impedance control output conforming to the switching control is: In the above formula, I is the identity matrix, and X0 is the desired position; c) Cartesian impedance control of motion space partitioning: By setting the impedance control inertia matrix M θ and impedance control damping matrix D θ It can guide the end tool to complete the insertion and extraction force control operation in a narrow guide hole, and make the end tool contact the surface of the target object during this stage; Since modeling a narrow guide hole cannot be completely accurate, and there are also errors in position feedback within a narrow guide hole, it is necessary to establish force control based on end-effector interaction force to guide the end-effector tool to safely contact the target surface. The Cartesian impedance control motion space, which lies between the Cartesian admittance control motion space and the Cartesian position control motion space, is divided into J parts, namely the first Cartesian impedance control motion space S. 21 To Cartesian impedance control of the Jth motion space S 2J J is a natural number ≥ 2; d) Set the inertia matrix and damping matrix for impedance control: In the first motion space controlled by Cartesian impedance, the inertia matrix and damping matrix for impedance control are set: M θ = 1 M θ ,D θ = 1 D θ 1 M θ and 1 D θ These are the inertial matrix and damping matrix for impedance control in the first motion space under Cartesian impedance control, respectively. In the Cartesian impedance-controlled j-th motion space, set the inertia matrix and damping matrix for impedance control: M θ =jM θ =γ j · j-1 M θ ,D θ = j D θ =γ j · j-1 D θ j M θ and j D θ Let γ be the inertia matrix and damping matrix of the impedance control in the j-th motion space under Cartesian impedance control, j = 2, ..., J. j The magnification factor of the j-th motion space controlled by Cartesian impedance relative to the (j-1)-th motion space controlled by Cartesian impedance. In the above impedance control, different impedance control inertial matrix and damping matrix variable parameter settings are adopted for the J part of the Cartesian impedance control motion space to ensure that the end tool can quickly and smoothly adjust the insertion and extraction direction when it enters the guide hole, while ensuring that the impact force is minimized when the end tool contacts the surface of the target object and the robot end speed is reduced to zero in the shortest time. e) For impedance control, simplify its control strategy by setting K... θ Set to 0; 4) Establish a switching control between impedance control and position control: a) The switching control between impedance control and position control satisfies: In the above formula, F p The position control output is in Cartesian space. Based on the position control output, the robot system can control the end tool to perform operations on the target object. b) The position control output that conforms to the switching control is: In the above formula, N p and N d These are the position control proportional coefficient and the position control speed proportional coefficient, respectively, and both are positive definite diagonal matrices; c) p X0 represents the desired position value of X0 at the interface between the Cartesian impedance control motion space and the Cartesian position control motion space. It is crucial for switching between impedance control and position control. Therefore, the desired position value of X0 at the interface between the Cartesian impedance control motion space and the Cartesian position control motion space... p X0 needs to satisfy the impedance control equation: In the above formula, both conditions must be met. 5) Determine the switching control parameters of the robot system that satisfy the switching stability of the robot system: a) Define the switching control state space variables: b) Provide the design requirements for switching control parameters that satisfy both admittance control and impedance control: The state-space description equation for switching control that satisfies admittance control is: In the above formula, the positive definite diagonal matrix K e For environmental stiffness parameters; The state-space description equation for the switching control that satisfies impedance control is: Therefore, the state-space description equation for switching between admittance control and impedance control is: To meet the switching control requirements, the admittance control position proportional coefficient K p And admittance control speed proportional coefficient K d Impedance control inertia matrix M θ Impedance control damping matrix D θ and impedance control rigid matrix K θ and environmental stiffness parameter K e The admittance-impedance switching confusion first matrix T should be such that for any real number α belonging to 0≤α≤1. ia The second matrix T is confused with admittance impedance switching. ai All satisfy the Hurwitz matrix, where the admittance-impedance switching confusion first matrix T ia The second matrix T is confused with admittance impedance switching. ai They are respectively: T ia =α·T a +(1-a)·T i If the admittance control position proportional coefficient K p And admittance control speed proportional coefficient K d Impedance control inertia matrix M θ Impedance control damping matrix D θ and impedance control rigid matrix K θ and environmental stiffness parameter K e Such that for any real number α belonging to 0≤α≤1, matrix T ia and T ai If all parameters satisfy the Hurwitz matrix, then this set of parameters can be used as stable switching control parameters between admittance control and impedance control to control the switching between admittance control and impedance control; if they do not satisfy the Hurwitz matrix, then reselect the parameters until parameters that satisfy the Hurwitz matrix are obtained. c) Provide the design requirements for the switching control parameters of the robot system that satisfy both impedance control and position control: The state-space description equation for switching control that satisfies position control is: Therefore, the state-space description equation for the switching between impedance control and position control is: To meet the switching control requirements, the impedance control inertia matrix M θ Impedance control damping matrix D θ and impedance control rigid matrix K θ Environmental stiffness parameter K e Position control position proportional coefficient N p And position control speed proportional coefficient N d It should be such that for any real number α belonging to 0≤α≤1, the impedance position switching confusion first matrix T ip The second matrix T is confused with impedance position switching. pi All satisfy the Hurwitz matrix, where the impedance position switching confusion first matrix T ip The second matrix T is confused with impedance position switching. pi They are respectively: T ip =α·T i +(1-a)·T p If the impedance control inertia matrix M θ Impedance control damping matrix D θ and impedance control rigid matrix K θ Environmental stiffness parameter K e Position control position proportional coefficient N p And position control speed proportional coefficient N d Such that for any real number α belonging to 0≤α≤1, the impedance position switching confusion first matrix T ip The second matrix T is confused with impedance position switching. pi If all parameters satisfy the Hurwitz matrix, then this set of parameters can be used as stable switching control parameters for impedance control and position control; if they do not satisfy the Hurwitz matrix, then the parameters are reselected until parameters that satisfy the Hurwitz matrix are obtained. The description of switching system state space is as follows:

2. The control strategy and parameter design method for the multi-mode switching system as described in claim 1, characterized in that, In step 3), d), J is a natural number that satisfies 2 ≤ J ≤ 5.

3. The control strategy and parameter design method for the multi-mode switching system as described in claim 1, characterized in that, In step 3), d), γ j The range is 1 to 6.

Images