Double-end virtual fixture auxiliary training method for skill training

Abstract

The invention discloses a double-end virtual fixture auxiliary training method for skill training. The method comprises the following steps: establishing a dynamic model based on a scene in which an expert terminal and a student terminal respectively control two hand controllers to control the same slave terminal; designing an expert-end virtual fixture, wherein the expert-end virtual fixture comprises a real-time position of an expert-end movement process, a variable radius and a real-time position of a student-end movement process; and based on the real-time position of the expert end in the motion process, the variable radius and the real-time position of the student end in the motion process, determining the feedback force of the expert end virtual clamp acting on the expert end. The technical problems that in the prior art, when double users control a robot, tactile feedback on an in-loop expert end is not considered, and the guidance on a student end only takes the real-time position of an expert as a center target, so that the guidance of the expert on the student is unstable when the student receives training are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of teleoperation skill training, and in particular to a double-ended virtual fixture auxiliary training method for skill training. Background Art

[0002] The skill level of the operator in teleoperation determines the success or failure of the task implementation. For situations where the skill level of teleoperation is required to be high, the operator needs to be trained to master the skills of operating the robot and integrate professional knowledge and advanced technology. Therefore, the present invention designs and implements a method for skill training of operators assisted by a virtual fixture. The method of using software-generated motion constraints is called a "virtual fixture". The main purpose is to guide or restrict the actions of human operators. It is used in applications that require the precision and accuracy of the robot system and the decision-making ability and intelligence of the human operator. The concept of virtual fixture is used in skill training to use the guiding force on the trainee side as a guiding clue to improve his skill level. On the expert side, the feedback force it generates to react to the trainee and the end position state and learning state helps the expert to make guiding actions accurately and quickly.

[0003] Previous research has not considered tactile feedback for the in-loop expert side, and the guidance for the trainee side is only centered on the real-time position of the expert. For example, a simple spherical virtual fixture is used to record the instantaneous expert's guidance position, which is unstable. The present invention proposes a double-end virtual fixture assisted training method for skill training. A virtual fixture is designed on the expert side to provide tactile feedback of the real-time student's motion state. At the same time, an innovative closed double-layer virtual pipe fixture is designed based on the expert's operation trajectory. At the same time, a skill level parameter is introduced according to the student's real-time skill level, and the virtual fixture is adjusted in real time according to the parameter. This method can not only be adjusted according to the student's real-time skill level, but also provide differentiated auxiliary operation strategies to ensure the safety of the trainee during training and the flexibility of training. Summary of the invention

[0004] An embodiment of the present invention provides a dual-end virtual fixture assisted training method for skill training, so as to at least solve the technical problem that when two users control the robot in the prior art, tactile feedback to the expert end in the loop is not considered, and the guidance to the trainee end is only centered on the expert's real-time position, resulting in unstable guidance of the expert to the trainee when the trainee is receiving training.

[0005] According to one aspect of an embodiment of the present invention, a dual-end virtual fixture assisted training method for skill training is provided. The method may include: establishing a dynamic model based on a scenario in which an expert end and a trainee end respectively control two hand controllers to control the same slave end;

[0006] The expression of the kinetic model is:

[0007]

[0008] Among them, M 1 、M 2 、M s Represent the inertia matrix of the expert side, the student side and the slave side respectively, C 1 , C 2 , C s Represent the Coriolis centrifugal force matrices of the expert side, the student side, and the slave side, G 1 , G 2 , G s Represent the gravity matrices of the expert side, the student side, and the slave side, respectively, and F h1 、F h2 are the forces exerted by the expert and the trainee on hand controller 1 and hand controller 2 respectively, F s1 、F s2 They represent the feedback force from the end to the expert end and the student end, respectively, vf1 、F vf2 They represent the feedback force of the virtual fixture on the expert side and the guiding force of the virtual fixture on the trainee side, respectively. sd represents the control power of the expert and the student over the slave, x 1 、x 2 、x s Represent the real-time positions of the expert, student, and slave during the movement process, Represent the speed of the expert end, the student end and the slave end during the movement, Represent the acceleration of the expert end, the student end and the slave end during the movement process respectively; design a virtual fixture on the expert end, which includes: the real-time position of the expert end during the movement process, the variable radius and the real-time position of the student end during the movement process; determine the feedback force of the expert end virtual fixture on the expert end based on the real-time position of the expert end during the movement process, the variable radius and the real-time position of the student end during the movement process; design a virtual fixture on the student end, which includes: the target trajectory of the expert end, the real-time positions of the expert end and the student end during the movement process, the first radius, the second radius, the closest point of the student end to the expert end target trajectory and the speed of the expert end when the expert end moves to the closest point; determine the guiding force of the student end virtual fixture on the student end based on the real-time positions of the expert end and the student end during the movement process, the first radius, the second radius, the closest point of the student end to the expert end target trajectory and the speed of the expert end when the expert end moves to the closest point; determine the motion state of the student end and adjust the training strategy for the student end based on the feedback force of the expert end virtual fixture on the expert end, and guide the student end to move toward the expert end target trajectory based on the guiding force of the student end virtual fixture on the student end.

[0009] Optionally, the feedback force from the slave end to the expert end and the trainee end and the control force from the expert end and the trainee end to the slave end are expressed as:

[0010]

[0011]

[0012] Among them, K p1 , K p2 , K ps They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end on the slave end, respectively. K d1 , K d2 , K ds They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end to the slave end, respectively. sd is the expected position of the proportional derivative control of the slave end by the expert end and the student end, is the expected speed of the expert and student end proportional-derivative control over the slave end.

[0013] Optionally, the expressions of the expected position and expected speed of the proportional differential control of the slave end by the expert end and the trainee end are:

[0014] x sd =γx 1 +(1-γ)x 2

[0015]

[0016] Among them, γ∈[0,1] is the participation factor set in skill training.

[0017] Optionally, based on the real-time position of the expert end during the motion process, the variable radius, and the real-time position of the trainee end during the motion process, the expression for the feedback force of the virtual fixture of the expert end on the expert end is determined as follows:

[0018]

[0019] Among them, μ (0 < μ < 1) represents a multiplicative reduction gain, F vf1|t-1 represents the feedback force generated by the virtual fixture at the expert end at the last moment, K vf1 represents the proportional gain of the virtual fixture on the expert side, r 1 The variable radius.

[0020] Optionally, based on the real-time positions of the expert end and the trainee end during the motion process, the first radius, the second radius, the nearest point of the trainee end to the target trajectory of the expert end and the speed of the expert end when moving to the nearest point, the guiding force of the virtual clamp on the trainee end is determined, including: determining the guiding virtual clamp force of the virtual clamp on the trainee end based on the real-time positions of the expert end and the trainee end during the motion process, the speed of the expert end when moving to the nearest point and the second radius; determining the restricted area type virtual clamp force of the virtual clamp on the trainee end based on the real-time positions of the expert end and the trainee end during the motion process, the first radius, the second radius and the nearest point of the trainee end to the target trajectory of the expert end; determining the guiding force of the virtual clamp on the trainee end based on the guiding virtual clamp force and the restricted area type virtual clamp force.

[0021] Optionally, based on the real-time positions of the expert end and the trainee end during the movement process, the speed of the expert end when the expert end moves to the closest point, and the second radius, determining the guided virtual fixture force of the trainee end virtual fixture on the trainee end includes:

[0022]

[0023] Among them, F gvf is the guiding virtual fixture force acting on the trainee end by the virtual fixture at the trainee end, R out is the second radius, v 1 (x near ) is the speed of the expert end when it moves to the closest point, K gvf F gvf The calculation coefficient, F gvfmax F gvf The maximum value of .

[0024] Optionally, based on the real-time positions of the expert end and the student end during the motion process, the first radius, the second radius, and the closest point of the student end to the target trajectory of the expert end, determining the restricted area type virtual clamp force of the student end virtual clamp acting on the student end includes:

[0025]

[0026] Among them, F frvf is the restricted area virtual fixture force acting on the student end by the student end virtual fixture, F frvfmax is the maximum value of the virtual fixture force of the restricted area on the student side, R in is the first radius, x near is the closest point of the target trajectory between the student end and the expert end, and η is the potential field constant.

[0027] Optionally, based on the guiding virtual fixture force and the restricted area virtual fixture force, the guiding force of the virtual fixture at the trainee end acting on the trainee end is determined as follows:

[0028] F vf2 =F gvf +F frvf

[0029] Among them, F vf2 It is the guiding force exerted by the virtual fixture on the trainee.

[0030] Beneficial effects of the present invention:

[0031] The present invention proposes a double-end virtual fixture assisted training method for skill training. When an expert and a trainee respectively control two hand controllers to control the same robot, a dynamic model is established, and a virtual fixture on the expert end and a virtual fixture on the trainee end are designed. According to the designed virtual fixture on the expert end, the feedback force of the virtual fixture on the expert end is obtained. According to the designed virtual fixture on the trainee end, the guiding force of the virtual fixture on the trainee end is obtained. According to the feedback force of the virtual fixture on the expert end, the motion state of the trainee end is obtained and the training strategy for the trainee end is adjusted. According to the guiding force of the virtual fixture on the trainee end, the trainee end is guided to move toward the target trajectory of the expert end, which solves the problem of double users controlling the robot in the prior art. When controlling, tactile feedback to the expert end in the loop was not taken into consideration, and the guidance to the trainee end was only centered on the expert's real-time position, which resulted in the technical problem of unstable guidance of the expert to the trainee during training. The technical effect of stable guidance of the expert to the trainee was achieved by designing an adaptive spherical virtual clamp on the expert end and a double-layer closed pipe virtual clamp on the trainee end. The feedback force of the virtual clamp on the expert end was obtained, and the guiding force of the virtual clamp on the trainee end was obtained. According to the feedback force of the virtual clamp on the expert end, the motion state of the trainee end was determined and the training strategy for the trainee end was adjusted. According to the guiding force of the virtual clamp on the trainee end, the trainee end was guided to move toward the target trajectory of the expert end, achieving the technical effect of stable guidance of the expert to the trainee. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0033] Figure 1 is a flow chart of a double-ended virtual fixture assisted training method for skill training according to an embodiment of the present invention;

[0034] Figure 2 is a principle block diagram based on double-ended virtual fixture assisted training according to an embodiment of the present invention;

[0035] Figure 3 is a schematic diagram of an expert-side virtual fixture according to an embodiment of the present invention;

[0036] Figure 4 is a schematic diagram of a prohibited feedback force of a virtual pipeline clamp at a trainee end according to an embodiment of the present invention;

[0037] Figure 5 is a schematic diagram of a guided feedback force of a virtual pipe clamp at a trainee end according to an embodiment of the present invention;

[0038] Figure 6 is a schematic diagram of generating a composite spherical shell according to an embodiment of the present invention;

[0039] Figure 7 It is a schematic diagram of a skill training trajectory tracking simulation effect according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Example 1

[0043] According to an embodiment of the present invention, a double-ended virtual fixture-assisted training method for skill training is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system comprising at least one set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0044] Figure 1 is a flow chart of a double-ended virtual fixture assisted training method for skill training according to an embodiment of the present invention. Figure 1As shown, the method may include the following steps:

[0045] Step S101, establishing a dynamic model based on a scenario in which an expert terminal and a trainee terminal respectively operate two hand controllers to control the same slave terminal;

[0046] The expression of the kinetic model is:

[0047]

[0048] Among them, M 1 、M 2 、M s Represent the inertia matrix of the expert side, the student side and the slave side respectively, C 1 , C 2 , C s Represent the Coriolis centrifugal force matrices of the expert side, the student side, and the slave side, G 1 , G 2 , G s Represent the gravity matrices of the expert side, the student side, and the slave side, respectively, and F h1 、F h2 are the forces exerted by the expert and the trainee on hand controller 1 and hand controller 2 respectively, F s1 、F s2 They represent the feedback force from the end to the expert end and the student end, respectively, vf1 、F vf2 They represent the feedback force of the virtual fixture on the expert side and the guiding force of the virtual fixture on the trainee side, respectively. sd represents the control power of the expert and the student over the slave, x 1 、x 2 、x s Represent the real-time positions of the expert, student, and slave during the movement process, Represent the speed of the expert end, the student end and the slave end during the movement, Represent the acceleration of the expert side, the student side and the slave side during the movement respectively.

[0049] In the technical solution provided in the above step S101 of the present invention, when the expert and the trainee respectively operate two hand controllers to control the same robot, a dynamic model of the expert end, the trainee end and the slave end is established, such as Figure 2 As shown, Figure 2 It is a principle block diagram of double-ended virtual fixture-assisted training according to an embodiment of the present invention.

[0050] Step S102, designing a virtual fixture on the expert side, the virtual fixture on the expert side includes: the real-time position and variable radius of the expert side during the movement process and the real-time position of the trainee side during the movement process.

[0051] In the technical solution provided in the above step S102 of the present invention, Figure 3 is a schematic diagram of an expert-side virtual fixture according to an embodiment of the present invention, such as Figure 3 As shown, x 1 is the real-time position of the expert end during the movement process, x 2 is the real-time position of the student during the movement, x 1 The ball edge to the virtual fixture at the expert end has a radius, which is a variable radius.

[0052] Step S103, based on the real-time position of the expert end during the movement process, the variable radius and the real-time position of the trainee end during the movement process, determine the feedback force of the virtual fixture of the expert end on the expert end.

[0053] In the technical solution provided in the above step S103 of the present invention, the real-time position of the expert end during the movement, the variable radius and the real-time position of the trainee end during the movement are calculated to obtain the feedback force of the expert end virtual fixture acting on the expert end.

[0054] Step S104, design a virtual fixture on the student side, which includes: the target trajectory of the expert side, the real-time positions of the expert side and the student side during the movement process, the first radius, the second radius, the closest point of the student side to the target trajectory of the expert side, and the speed of the expert side when it moves to the closest point.

[0055] In the technical solution provided in the above step S104 of the present invention, Figure 4 is a schematic diagram of a prohibited feedback force of a virtual pipe clamp at a trainee end according to an embodiment of the present invention, Figure 5 is a schematic diagram of a virtual pipe clamp guided feedback force at a trainee end according to an embodiment of the present invention, such as Figure 4 and Figure 5 As shown, the dotted line in the middle of the fixture is the expert target trajectory, x near It is the speed of the expert end when it moves to the closest point.

[0056] The calculation process of the first radius, second radius and variable radius is as follows:

[0057] The sizes of the inner radii of the virtual fixture on the expert side and the virtual pipe on the student side are adaptively determined based on the student's real-time skill level evaluated by considering the student's motion state relative to the expert.

[0058] The first is to establish the skill level evaluation indicators: the relative path distance between the trainee and the expert, the smoothness of the trainee's movement trajectory, the similarity between the trainee's and the expert's trajectory, and the trainee's movement pause reaction together represent the trainee's movement status relative to the expert:

[0059] The relative path distance between the trainee and the expert σ(t): x near is the closest point of the student to the expert trajectory, and x neat =x1 (t mear ) Known t near is the time value of the nearest point on the expert trajectory, t real is the time value of the expert's real-time position on the expert trajectory, and σ(t) represents the length between the closest point and the expert's real-time position on the expert trajectory, which is used to measure the trainee's real-time skill level to adjust the size of the virtual fixture radius:

[0060]

[0061] Among them, x 1 (t), y 1 (t), z 1 (t) is Figure 3 x in 1 Corresponding x-axis, y-axis and z-axis.

[0062] Motion smoothness ν(t): Instantaneous jitter is the rate at which acceleration changes over time. The smoothness of the student's motion v(t) is used as a measure of the student's real-time skill level to adjust the size of the virtual fixture radius:

[0063]

[0064] Among them, x 2 (t), y 2 (t), z 2 (t) is Figure 4 and Figure 5 x in 2 Corresponding x-axis, y-axis and z-axis.

[0065] The similarity between the trainee and the expert trajectory ρ(t): The improved windowed DTW algorithm is used to sample the operation trajectory sequence x of the expert and the trainee. 1 and x 2 To match, only a limited number of the latest recorded actual trajectory points of the trainee and their matching points are selected on the operation path, and the average distance d between them is calculated as the real-time similarity between the trainee and the expert trajectory. As a measure of the trainee's real-time skill level to adjust the radius of the virtual fixture:

[0066]

[0067] Among them, x 1 (i) and x 2 (j) represents the optimal matching point obtained by the windowed DTW algorithm

[0068]

[0069] Where w represents the window length of the windowing dtw algorithm (a fixed value)

[0070]

[0071] Student movement pause time ω(t): The matching points between the student trajectory and the expert trajectory within the real-time windowing range are obtained by the above-mentioned DTW algorithm. The number of times each student matching point is matched can be calculated, thereby obtaining the student trajectory point with the maximum number of repeated matching times. The maximum number of matching times can represent the student movement pause time ω(t).

[0072] The second is the calculation of the real-time skill level. According to the relative path distance σ(t) between the trainee and the expert, the motion smoothness ν(t), the similarity ρ(t) between the trainee and the expert trajectory, and the trainee's motion pause time ω(t), the trainee's real-time skill level parameter α(t) is obtained, with a value range of 0 to 1:

[0073]

[0074] Finally, according to the trainee's real-time skill level evaluation index α(t), the size of the virtual fixture radius R is obtained:

[0075] R=K r (1-α(t))+R 0

[0076] Among them, R 0 is the minimum radius of the pipe (set), K r For gain.

[0077] The outer radius of the virtual pipeline in step 4 is generated by considering whether collision detection occurs:

[0078] The hierarchical bounding box algorithm is used to perform collision detection between the unit shell of the generated pipeline and obstacles. If no collision occurs, the radius of the unit shell remains unchanged; if a collision occurs, the greedy idea is used to reduce the radius of the unit shell by half, and the generated new hemisphere and the other half of the original shell form a composite unit shell (such as Figure 5 ):

[0079]

[0080] The composite unit spherical shells are still connected by the triangular patch method using the maximum tangent circle of the adjacent unit spherical shells perpendicular to the center line, thereby generating a smooth pipe.

[0081] Step S105, based on the real-time positions of the expert end and the student end during the movement process, the first radius, the second radius, the closest point of the student end to the expert end's target trajectory, and the speed of the expert end when moving to the closest point, determine the guiding force of the student end virtual clamp on the student end.

[0082] In the technical solution provided in the above step S105 of the present invention, the real-time positions of the expert end and the student end during the movement process, the first radius, the second radius, the closest point of the student end to the target trajectory of the expert end and the speed of the expert end when moving to the closest point are calculated to obtain the guiding force of the virtual clamp on the student end.

[0083] Step S106, based on the feedback force of the virtual clamp on the expert side, determine the movement state of the student side and adjust the training strategy for the student side, and based on the guiding force of the virtual clamp on the student side, guide the student side to move toward the target trajectory of the expert side.

[0084] In the technical solution provided in the above step S106 of the present invention, during the operation training, the expert receives tactile feedback from the virtual fixture force on the expert side and the feedback force from the slave side, and understands the motion state of the trainee and the slave side in real time, thereby adjusting the training strategy in real time. At the same time, the trainee is also guided by the virtual fixture force on the trainee side and the feedback force from the slave side, and is guided by adaptive changes according to the trainee's motion state and skill level, thereby achieving safe and flexible training support for the trainee.

[0085] The above method of this embodiment is further introduced below.

[0086] As an optional embodiment, in step S101, the feedback force from the slave end to the expert end and the trainee end and the control force from the expert end and the trainee end to the slave end are expressed as:

[0087]

[0088] Among them, K p1 , K p2 , K ps They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end on the slave end, respectively. K d1 , K d2 , K ds They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end to the slave end, respectively. sd is the expected position of the proportional derivative control of the slave end by the expert end and the student end, is the expected speed of the expert and student end proportional-derivative control over the slave end.

[0089] As an optional embodiment, in step S102, the expressions of the expected position and expected speed of the slave proportional differential control by the expert end and the trainee end are respectively:

[0090] xsd =γx 1 +(1-γ)x 2

[0091]

[0092] Among them, γ∈[0,1] is the participation factor set in skill training.

[0093] As an optional embodiment, in step S103, based on the real-time position of the expert end during the movement process, the variable radius and the real-time position of the trainee end during the movement process, the expression of the feedback force of the virtual fixture of the expert end on the expert end is determined as follows:

[0094]

[0095] Among them, μ (0 < μ < 1) represents a multiplicative reduction gain, F vf1|t-1 represents the feedback force generated by the virtual fixture at the expert end at the last moment, K vf1 represents the proportional gain of the virtual fixture on the expert side, r 1 The variable radius.

[0096] In this embodiment, when ||x 2 -x 1 || <r 1 When F vf1 is a form of expression where ||x 2 -x 1 ||≥r 1 When F vf1 It is a form of expression.

[0097] As an optional implementation method, step S105 determines the guiding force of the virtual clamp on the trainee end based on the real-time positions of the expert end and the trainee end during the movement process, the first radius, the second radius, the nearest point of the trainee end to the target trajectory of the expert end, and the speed of the expert end when the expert end moves to the nearest point, including: determining the guiding virtual clamp force of the virtual clamp on the trainee end based on the real-time positions of the expert end and the trainee end during the movement process, the speed of the expert end when the expert end moves to the nearest point, and the second radius; determining the restricted area type virtual clamp force of the virtual clamp on the trainee end based on the real-time positions of the expert end and the trainee end during the movement process, the first radius, the second radius, and the nearest point of the trainee end to the target trajectory of the expert end; determining the guiding force of the virtual clamp on the trainee end based on the guiding virtual clamp force and the restricted area type virtual clamp force.

[0098] In this embodiment, the real-time positions of the expert end and the trainee end during the movement process, the speed of the expert end when it moves to the nearest point, and the second radius are calculated to determine the guiding virtual clamp force of the trainee end virtual clamp on the trainee end, the real-time positions of the expert end and the trainee end during the movement process, the first radius, the second radius, and the nearest point of the trainee end to the target trajectory of the expert end are calculated to obtain the restricted area type virtual clamp force of the trainee end virtual clamp on the trainee end, the guiding virtual clamp force and the restricted area type virtual clamp force are calculated to obtain the guiding force of the trainee end virtual clamp on the trainee end.

[0099] As an optional embodiment, based on the real-time positions of the expert end and the trainee end during the movement process, the speed of the expert end when the expert end moves to the closest point, and the second radius, the guided virtual clamp force of the trainee end virtual clamp acting on the trainee end is determined, including:

[0100]

[0101] Among them, F gvf is the guiding virtual fixture force acting on the trainee end by the virtual fixture at the trainee end, R out is the second radius, v 1 (x near ) is the speed of the expert end when it moves to the closest point, K gvf F gvf The calculation coefficient, F gvfmax F gvf The maximum value of .

[0102] In this embodiment, when ||x 2 -x 1 ||>R out When F gvf is an expression, ||x 2 -x 1 ||≤R out When F gvf is an expression.

[0103] As an optional embodiment, based on the real-time positions of the expert end and the student end during the motion process, the first radius, the second radius, and the closest point of the student end to the target trajectory of the expert end, determining the restricted area type virtual clamp force of the student end virtual clamp acting on the student end includes:

[0104]

[0105] Among them, F frvf is the restricted area virtual fixture force acting on the student end by the student end virtual fixture, F frvfmax is the maximum value of the virtual fixture force of the restricted area on the student side, R in is the first radius, xnear is the closest point of the target trajectory between the student end and the expert end, and η is the potential field constant.

[0106] In this embodiment, if ||x near -x 2 ||<R in , F frvf is an expression, if||x near -x 2 ||>R out , F frvf is an expression, x near The KD tree algorithm in the nearest neighbor search method is used to obtain it. The specific steps are as follows:

[0107] Step a. Construct a KD tree.

[0108] First, for the point set on the curve, a KD tree is constructed based on their coordinates. The construction process includes selecting a split axis (usually the data dimension axis), dividing the point set into two subsets according to the median value of the axis, and recursively proceeding until all points are included in the nodes of the KD tree.

[0109] Step b. Dynamic update.

[0110] Since the curve is dynamic, as points are added in one direction and removed in another direction, the KD tree needs to be updated dynamically. This includes two operations: adding new points and deleting old points. In addition, the tree needs to be balanced.

[0111] Step c. Find the nearest point.

[0112] Starting from the root node of the KD tree, recursively search downward and backtrack upward to search the tree, and finally determine the distance x from the student position by continuously updating and comparing distances. 2 The nearest point x near .

[0113] As an optional embodiment, based on the guiding virtual fixture force and the restricted area virtual fixture force, the guiding force of the virtual fixture at the trainee end on the trainee end is determined as follows:

[0114] F vf2 =F gvf +F frvf

[0115] Among them, F vf2 It is the guiding force exerted by the virtual fixture on the trainee.

[0116] In this embodiment, the sum of the guiding virtual clamp force and the restricted area virtual clamp force is determined as the guiding force of the student end virtual clamp on the student end. Figure 6is a schematic diagram of generating a composite spherical shell according to an embodiment of the present invention; Figure 7 It is a schematic diagram of a skill training trajectory tracking simulation effect according to an embodiment of the present invention.

[0117] In an embodiment of the present invention, a dynamic model is established by respectively controlling two hand controllers based on an expert end and a trainee end to control a scene of the same slave end; a virtual fixture on the expert end is designed, and the virtual fixture on the expert end includes: a real-time position of the expert end during the motion process, a variable radius, and a real-time position of the trainee end during the motion process; based on the real-time position of the expert end during the motion process, the variable radius, and the real-time position of the trainee end during the motion process, the feedback force of the virtual fixture on the expert end is determined; a virtual fixture on the trainee end is designed, and the virtual fixture on the trainee end includes: a target trajectory of the expert end, the real-time positions of the expert end and the trainee end during the motion process, a first radius, a second radius, the closest point of the trainee end to the target trajectory of the expert end, and the speed of the expert end when the end moves to the closest point; based on the real-time positions of the expert end and the trainee end during the motion process, the first radius, the second radius, the closest point of the trainee end to the target trajectory of the expert end, and the speed of the expert end when the end moves to the closest point, the guiding force of the virtual fixture on the trainee end is determined; based on the virtual fixture on the expert end The feedback force acting on the expert end is used to determine the motion state of the trainee end and adjust the training strategy for the trainee end. Based on the guiding force of the virtual fixture on the trainee end, the trainee end is guided to move toward the target trajectory of the expert end. This solves the problem that when two users control the robot in the prior art, tactile feedback to the expert end in the loop is not considered, and the guidance to the trainee end only takes the real-time position of the expert as the central goal, which makes the technical problem that the expert's guidance to the trainee is unstable when the trainee is receiving training. The feedback force of the virtual fixture on the expert end and the guiding force of the virtual fixture on the trainee end are obtained by designing an adaptive spherical virtual fixture on the expert end and a double-layer closed pipe virtual fixture on the trainee end. According to the feedback force of the virtual fixture on the expert end, the motion state of the trainee end is determined and the training strategy for the trainee end is adjusted. According to the guiding force of the virtual fixture on the trainee end, the trainee end is guided to move toward the target trajectory of the expert end, and the technical effect of stable guidance of the expert to the trainee is achieved.

[0118] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0119] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0122] In addition, each functional unit in each embodiment of the present invention may be integrated into a first processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0123] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A double-ended virtual fixture assisted training method for skill training, characterized in that: include: A dynamic model is established based on the scenario where the expert and the trainee respectively control two hand controllers to control the same slave. The expression of the kinetic model is: Among them, M1, M2, M s Represent the inertia matrices of the expert side, the student side, and the slave side, C1, C2, C s Represent the Coriolis centrifugal force matrices of the expert side, the student side, and the slave side, G1, G2, G s Represent the gravity matrices of the expert side, the student side, and the slave side, respectively, and F h1 、F h2 are the forces exerted by the expert and the trainee on hand controller 1 and hand controller 2 respectively, F s1 、F s2 They represent the feedback force from the end to the expert end and the student end, respectively, vf1 、F vf2 They represent the feedback force of the virtual fixture on the expert side and the guiding force of the virtual fixture on the trainee side, respectively. sd represents the control power of the expert and the student over the slave, x1, x2, x s Represent the real-time positions of the expert, student, and slave during the movement process, Represent the speed of the expert end, the student end and the slave end during the movement, Represent the acceleration of the expert end, the student end, and the slave end during the movement, respectively; Design the virtual fixture on the expert side, which includes: the real-time position and variable radius of the expert side during the movement process and the real-time position of the trainee side during the movement process; Based on the real-time position and variable radius of the expert end during the motion process and the real-time position of the trainee end during the motion process, the feedback force of the virtual fixture of the expert end on the expert end is determined; Design a virtual fixture for the student end, which includes: the target trajectory of the expert end, the real-time positions of the expert end and the student end during the movement process, the first radius, the second radius, the closest point of the student end to the target trajectory of the expert end, and the speed of the expert end when it moves to the closest point; Determine the guiding force of the virtual fixture on the trainee end based on the real-time positions of the expert end and the trainee end during the movement process, the first radius, the second radius, the closest point of the trainee end to the target trajectory of the expert end, and the speed of the expert end when the expert end moves to the closest point; Based on the feedback force of the virtual fixture on the expert side, the movement state of the trainee side is determined and the training strategy for the trainee side is adjusted. Based on the guiding force of the virtual fixture on the trainee side, the trainee side is guided to move toward the target trajectory of the expert side.

2. The method according to claim 1, characterized in that The expressions of the feedback force from the slave end to the expert end and the student end and the control force from the expert end and the student end to the slave end are: Among them, K p1 , K p2 , K ps They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end on the slave end, respectively. K d1 , K d2 , K ds They represent the proportional differential control coefficient of the feedback force generated by the slave end to the expert end, the proportional differential control coefficient of the feedback force generated by the slave end to the trainee end, and the proportional differential control coefficient of the feedback force generated by the expert end and the trainee end to the slave end, respectively. sd is the expected position of the proportional derivative control of the slave end by the expert end and the student end, is the expected speed of the expert and student end proportional-derivative control over the slave end.

3. The method according to claim 2, characterized in that The expressions of the expected position and expected speed of the proportional differential control of the slave end by the expert end and the student end are: x sd =γx1+(1-γ)x2 Among them, γ∈[0,1] is the participation factor set in skill training.

4. The method according to claim 1, characterized in that: Based on the real-time position and variable radius of the expert end during the motion process and the real-time position of the trainee end during the motion process, the expression of the feedback force of the virtual fixture on the expert end is determined as follows: Among them, μ (0 < μ < 1) represents a multiplicative reduction gain, F vf1|t-1 represents the feedback force generated by the virtual fixture at the expert end at the last moment, K vf1 It represents the proportional gain of the virtual fixture on the expert side, and r1 is the variable radius.

5. The method according to claim 1, characterized in that Based on the real-time positions of the expert end and the student end during the motion process, the first radius, the second radius, the closest point of the student end to the target trajectory of the expert end, and the speed of the expert end when moving to the closest point, the guiding force of the student end virtual fixture on the student end is determined, including: Based on the real-time positions of the expert end and the trainee end during the movement process, the speed of the expert end when it moves to the closest point, and the second radius, the guiding virtual fixture force of the trainee end virtual fixture on the trainee end is determined; Based on the real-time positions of the expert end and the student end during the motion process, the first radius, the second radius, and the closest point of the student end to the target trajectory of the expert end, determine the restricted area type virtual fixture force of the student end virtual fixture on the student end; Based on the guiding virtual fixture force and the restricted area virtual fixture force, the guiding force of the virtual fixture on the trainee end is determined.

6. The method according to claim 5, characterized in that Based on the real-time positions of the expert end and the trainee end during the motion process, the speed of the expert end when it moves to the closest point, and the second radius, the guided virtual fixture force of the trainee end virtual fixture on the trainee end is determined, including: Among them, F gvf is the guiding virtual fixture force acting on the trainee end by the virtual fixture at the trainee end, R out is the second radius, v1(x near ) is the speed of the expert end when it moves to the closest point, K gvf F gvf The calculation coefficient, F gvfmax F gvf The maximum value of .

7. The method according to claim 5, characterized in that Based on the real-time positions of the expert end and the student end during the motion process, the first radius, the second radius, and the closest point of the student end to the target trajectory of the expert end, the restricted area type virtual fixture force of the student end virtual fixture on the student end is determined, including: Among them, F frvf is the restricted area virtual fixture force acting on the student end by the student end virtual fixture, F frv f max is the maximum value of the virtual fixture force of the restricted area on the student end, R in is the first radius, x near is the closest point of the target trajectory between the student end and the expert end, and η is the potential field constant.

8. The method according to claim 5, characterized in that Based on the guiding virtual fixture force and the restricted area virtual fixture force, the formula for determining the guiding force of the virtual fixture at the trainee end on the trainee end is: Fvf22=Fgvf+'Ffrvf Among them, F vf2 It is the guiding force exerted by the virtual fixture on the trainee.

9. A processor, characterized in that: The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 8 when running.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 8.

Images