A safe physical human-computer interaction control method, terminal and medium

Abstract

The application discloses a safe physical human-computer interaction control method, a terminal and a medium. The method comprises the following steps: generating a desired trajectory of a rendered virtual interaction dynamic according to an interaction control equation of a robot and a human and an interaction force; shaping the generated desired trajectory according to a safety limit to obtain a safety reference trajectory; and issuing the safety reference trajectory to a robot controller for execution. The application solves the problem that traditional interaction control (admittance / impedance) cannot generally meet speed and position safety constraints. Speed limiting avoids the robot running too fast to affect interaction safety and comfort, and ensuring that the robot does not exceed the position constraint can avoid the collision between the robot and the environment, and also ensure that the human-computer interaction process does not exceed the set safety area, greatly improving the safety in physical human-computer interaction.

Description

Technical Field

[0001] This application relates to the field of human-computer interaction control technology, and in particular to a safe physical human-computer interaction control method, terminal and medium. Background Technology

[0002] Unlike traditional robots that strictly limit safe areas and prohibit contact with humans, physical human-robot interaction technology enables collaborative robots and rehabilitation robots to coexist and interact with humans. Ensuring operator safety is the fundamental safety consideration during human-robot interaction. Current strategies for ensuring operator safety can be mainly divided into proactive prevention, post-collision detection, and safety-based design. Compared to the passive safety strategy of post-collision detection and safety-based design, proactive prevention strategies utilize multiple sensors to detect the environment and more actively modify the robot's movement to prevent collisions, positioning, and speeds that exceed the operator's tolerance requirements.

[0003] Current robot interaction control is mainly divided into admittance control and impedance control. However, research on ensuring position and velocity safety during interaction remains limited. The paper "Admittance-Based Controller Design for Physical Human-Robot Interaction in the Constrained Task Space" uses a control Lyapunov function to design a robot controller that meets constraints, ensuring the robot's motion is confined within a set space and guaranteeing the semi-globally consistent and eventually bounded state of the system. However, the controller design is complex, and compensation is needed for uncertainties in the robot's dynamics model. Patent CN115256395A discloses a model-uncertain robot safety control method based on a control obstacle function, using Gaussian process regression to compensate for model uncertainties. However, generating the control law requires solving a quadratic optimization problem online, and the effectiveness of this method depends on the accuracy of the learned model. In addition, the artificial potential field method is also a commonly used method for robot safety control. However, the artificial potential field method is more conservative, and the robot may be trapped at a local equilibrium point and unable to reach the target point; furthermore, the control signal may oscillate.

[0004] In conclusion, achieving a practical interactive control method that comprehensively considers safety constraints such as speed and position, does not rely on model accuracy, and has a low application threshold remains a challenging problem in engineering applications. Summary of the Invention

[0005] This application provides a safe physical human-machine interaction control method, terminal, and medium, which solves the problem that traditional interaction control (admittance / impedance) often cannot meet speed and position safety constraints. By limiting the speed, the method prevents the robot from running too fast, which would affect the safety and comfort of the interaction. By ensuring that the robot does not exceed the position constraints, the method can avoid collisions between the robot and the environment, and also ensure that the human-machine interaction process does not exceed the set safety area, thus greatly improving the safety of physical human-machine interaction.

[0006] The technical solution of this application is as follows:

[0007] On the one hand, this application provides a safe physical human-computer interaction control method, including the following steps:

[0008] S1: Generate the desired trajectory for rendering virtual interactive dynamics based on the interaction control equations and interaction forces between the robot and the human.

[0009] S2: Reshape the generated desired trajectory according to safety constraints to obtain a safe reference trajectory;

[0010] S3: Based on the safety reference trajectory obtained in step S2, issue the command to the robot controller for execution.

[0011] Furthermore, step S1 specifically includes the following steps:

[0012] S11: Obtaining External Forces for Robot-Human Interaction ;

[0013] S12: Based on the robot-human interaction equation

[0014] (1)

[0015] The expected trajectory is calculated. ,in These are the expected acceleration, velocity, and position in Cartesian space. Represents the robot's actual acceleration, velocity, and position in Cartesian space;

[0016] S13: Transform the Cartesian space trajectory to joint space, specifically calculated as follows:

[0017] (2)

[0018] (3)

[0019] (4)

[0020] in These represent the desired acceleration, velocity, and position in joint space, respectively. Indicates the inverse kinematics of the robot. The Jacobian matrix representing the robot. Describe the generalized inverse of the Jacobian matrix. Then it represents the derivative of the Jacobian matrix with respect to time.

[0021] Furthermore, step S2 specifically includes the following steps:

[0022] S21: Establish the control barrier function , define when When the robot is in a safe state, it is considered to be in a safe state; and when At that time, it is considered that the robot in interaction is unsafe. Used to indicate the safe distance between the robot and the environment, the robot itself, and the operator, or to indicate the safety limits for the position of the robot's joints;

[0023] S22: Calculate the partial derivatives of the control barrier function. ;

[0024] S23: Calculate the safety reference trajectory and safety reference velocity. Must meet

[0025] (5)

[0026] in Represents extended K ∞ Function-like.

[0027] Furthermore, the aforementioned The function is selected as And parameters , This indicates the convergence speed of the robot's speed tracking error. For parameters related to the initial velocity tracking error, This represents the Lipschitz constant of the control barrier function.

[0028] Furthermore, in step S23, considering position-related safety, the desired velocity in equation (1) is shaped to obtain the safe reference velocity at the current time k.

[0029] (6)

[0030] (7)

[0031] in This represents the operation of finding the minimum value;

[0032] From safety reference speed Obtain safety reference acceleration

[0033] (8)

[0034] and safety reference location

[0035] (9)

[0036] China T s Indicates the control cycle.

[0037] Furthermore, in step S23, considering position and velocity-related safety, the velocity in Cartesian space must satisfy constraints.

[0038] (10)

[0039] Joint space velocity satisfies constraints

[0040] (11)

[0041] , Represents the minimum and maximum allowable velocities in Cartesian space. , Indicates the minimum and maximum allowable velocities in the joint space;

[0042] When the Cartesian space robot interacts, it must satisfy the requirement of equation (10), then we have

[0043] (12)

[0044] in , This represents the absolute value operation. Represents symbolic operations, Obtained from the interaction equation;

[0045] The desired velocity in the joint space is obtained from equation (12).

[0046] (13)

[0047] Since the joint space velocity constraint equation (11) needs to be satisfied, the desired joint space velocity is obtained.

[0048] (14)

[0049] Equation (14) yields the control barrier function that satisfies positional safety. Joint safety reference speed

[0050] (15)

[0051] Same as formula (7);

[0052] The safe reference acceleration and position that satisfy the position and velocity constraints are obtained from equation (15).

[0053] Through the above scheme, the method of the present invention explicitly provides a method for calculating the safe reference velocity that meets the position safety requirements, without the need to solve a quadratic problem in general safety control, thus improving the solution efficiency.

[0054] This application limits the robot's speed to prevent it from running too fast and affecting the safety and comfort of the interaction. Ensuring that the robot does not exceed the positional constraints can prevent collisions between the robot and the environment, and also ensure that the human-computer interaction process does not exceed the set safety area, which greatly improves the safety of physical human-computer interaction.

[0055] Furthermore, in step S3, the safety physical human-machine interaction control includes either safety admittance control or safety impedance control:

[0056] Safety admittance control: The safety reference speed obtained in step S2 Or safety reference location Send out robot speed or position trackers;

[0057] Safety impedance control: based on the desired reference acceleration obtained in step S2 Calculate robot control torque

[0058] (16)

[0059] in Represents the inertia matrix. Represents Coriolis force and centripetal force. Indicates the effect of gravity. Indicates friction. This indicates the equivalent effect of external force acting on the joint space.

[0060] Through the above scheme, the method of this application can be easily and conveniently implemented in admittance / impedance control based on the generated safe speed, acceleration, and position. It is easy to implement and suitable for use in interactive scenarios with safety restrictions such as speed and position.

[0061] Furthermore, it also includes step S4: during the human-computer interaction process, detecting whether the interaction force exceeds the limit; if... or If the limit is exceeded, the robot will abandon the current task and switch to zero-force mode to wait for safe operation to resume.

[0062] In another aspect, this application provides a robot control terminal, including a processor and a memory, wherein the memory stores computer instructions, and the processor is configured to: invoke and execute the computer instructions to implement the safe physical human-machine interaction control method as described above.

[0063] In another aspect, this application provides a computer-readable medium storing computer instructions, which, when invoked and executed by a computer, implement the secure physical human-computer interaction control method described above.

[0064] In summary, the beneficial effects of this application are as follows:

[0065] 1. This invention provides a safe physical human-machine interaction control method that solves the problem that traditional interaction control (admittance / impedance) often fails to meet speed and position safety constraints. By limiting the speed, the method prevents the robot from running too fast, which would affect the safety and comfort of the interaction. By ensuring that the robot does not exceed the position constraints, it can avoid collisions between the robot and the environment, and also ensure that the human-machine interaction process does not exceed the set safety area, thus greatly improving the safety of physical human-machine interaction.

[0066] 2. The safe physical human-machine interaction control method provided by this invention is compatible with existing major interaction control methods. Based on the generated safe speed, acceleration, and position, it can be easily and conveniently implemented in admittance / impedance control. It is easy to implement and suitable for use in interaction scenarios with safety restrictions such as speed and position.

[0067] 3. The present invention provides a safe physical human-machine interaction control method that, when generating safe reference acceleration, velocity, and position, eliminates the need for a precise robot dynamics model, which is generally required by the control obstacle function method. This greatly reduces the difficulty of achieving safety-critical control using the control obstacle function method and realizes a simple and feasible safety control method.

[0068] 4. This invention also provides a method for adjusting key parameters of safe human-machine interaction control, adjusting parameters according to the robot's speed tracking performance: the better the robot's speed tracking performance, the higher the parameter... The larger the value, the better; conversely, the smaller the value, the worse the robot's speed tracking performance. It is advisable to choose a smaller value.

[0069] 5. This invention explicitly provides a method for calculating the safe reference velocity that meets position safety requirements, eliminating the need to solve quadratic problems in general safety control and improving solution efficiency.

[0070] 6. In addition to interactive control, the safety acceleration, velocity, and position shaping methods provided by this invention are also applicable to robot control applications with position and velocity limitations. Attached Figure Description

[0071] Figure 1 This is a control block diagram of a secure physical human-computer interaction control method provided in a specific embodiment of this application;

[0072] Figure 2 This is a schematic diagram illustrating the implementation of robot safety impedance control in a specific embodiment of the safety physical human-machine interaction control method provided in this application;

[0073] Figure 3 This is a schematic diagram illustrating the implementation of robot safety admittance control in a specific embodiment of the safety physical human-machine interaction control method provided in this application;

[0074] Figure 4 This is a curve of joint movement and interaction force in a specific embodiment of the safe physical human-computer interaction of this application. Detailed Implementation

[0075] The specific embodiments of this application are described in detail below with reference to the accompanying drawings.

[0076] Example: A specific embodiment of this application provides a safe physical human-computer interaction control method, referencing... Figure 1 This includes the following steps:

[0077] S1: Based on the interaction control equations and interaction forces between the robot and the human, generate the desired trajectory for rendering the virtual interactive dynamics. Step S1 specifically includes the following steps:

[0078] S11: Obtaining External Forces for Robot-Human Interaction ;

[0079] S12: Based on the robot-human interaction equation

[0080] (1)

[0081] The expected trajectory is calculated. ,in These are the expected acceleration, velocity, and position in Cartesian space. Represents the robot's actual acceleration, velocity, and position in Cartesian space; The specific form is determined by the specific interactive task.

[0082] S13: Transform the Cartesian space trajectory to joint space, specifically calculated as follows:

[0083] (2)

[0084] (3)

[0085] (4)

[0086] in These represent the desired acceleration, velocity, and position in joint space, respectively. Indicates the inverse kinematics of the robot. The Jacobian matrix representing the robot. Describe the generalized inverse of the Jacobian matrix. Then it represents the derivative of the Jacobian matrix with respect to time.

[0087] S2: Shape the generated desired trajectory according to safety constraints to obtain a safe reference trajectory. Step S2 specifically includes the following steps:

[0088] S21: Establish the control barrier function To ensure robot safety during human-computer interaction, the following definition is provided: When the robot is in a safe state, it is considered to be in a safe state; and when At that time, it is considered that the robot in interaction is unsafe. Used to indicate the safe distance between the robot and the environment, the robot itself, and the operator, or to indicate the safety limits for the position of the robot's joints;

[0089] S22: Calculate the partial derivatives of the control barrier function. Partial derivatives can be obtained using analytical methods or numerical differentiation.

[0090] S23: Calculate the safety reference trajectory and safety reference velocity. Must meet

[0091] (5)

[0092] in Represents extended K ∞ Function class. A feasible one. The function is selected as And parameters , This indicates the convergence speed of the robot's speed tracking error. For parameters related to the initial velocity tracking error, This represents the Lipschitz constant of the control barrier function.

[0093] In step S23, considering position-related safety, the desired velocity in equation (1) is shaped to obtain the safe reference velocity at the current time k.

[0094] (6)

[0095] (7)

[0096] in This represents the operation of finding the minimum value;

[0097] From safety reference speed Obtain safety reference acceleration

[0098] (8)

[0099] and safety reference location

[0100] (9)

[0101] China T s Indicates the control cycle.

[0102] In some other specific embodiments, step S23 differs from the above-mentioned consideration of only position-related safety; instead, it considers both position-related and velocity-related safety, requiring the velocity in Cartesian space to satisfy constraints.

[0103] (10)

[0104] Joint space velocity satisfies constraints

[0105] (11)

[0106] , Represents the minimum and maximum allowable velocities in Cartesian space. , This indicates the minimum and maximum allowable velocities in the joint space; generally speaking... , .

[0107] When the Cartesian space robot interacts, it must satisfy the requirement of equation (10), then we have

[0108] (12)

[0109] in , This represents the absolute value operation. Represents symbolic operations, Obtained from the interaction equation;

[0110] The desired velocity in the joint space is obtained from equation (12).

[0111] (13)

[0112] Since the joint space velocity constraint equation (11) needs to be satisfied, the desired joint space velocity is obtained.

[0113] (14)

[0114] Equation (14) yields the control barrier function that satisfies positional safety. Joint safety reference speed

[0115] (15)

[0116] Same as formula (7);

[0117] The safe reference acceleration and position that satisfy the position and velocity constraints are obtained from equation (15), and the calculation method is the same as that of equations (8) and (9).

[0118] S3: Based on the safety reference trajectory obtained in step S2, issue the command to the robot controller for execution. Depending on the implementation method, safe physical human-machine interaction control can be divided into safety admittance control and safety impedance control, the specific implementation methods of which are as follows:

[0119] Safety admittance control: The safety reference speed obtained in step S2 Or safety reference location Send out robot speed or position trackers;

[0120] Safety impedance control: based on the desired reference acceleration obtained in step S2 Calculate robot control torque

[0121] (16)

[0122] in Represents the inertia matrix. Represents Coriolis force and centripetal force. Indicates the effect of gravity. Indicates friction. This represents the equivalent effect of external forces acting on the joint space. The control law in equation (16) can be combined with methods such as neural networks and observers to compensate for the uncertainty of the robot model.

[0123] S4: During human-computer interaction, in addition to ensuring that position and speed meet safety constraints, it is necessary to detect whether the interaction force exceeds the limit. or If the limit is exceeded, the robot will abandon the current task and switch to zero-force mode to wait for safe operation to resume.

[0124] The following section uses a rope-driven robot to illustrate how to achieve safe physical human-robot interaction control. The robot makes direct physical contact with the patient according to different rehabilitation training tasks. During rehabilitation training, the robot's speed and position need to be limited. Controlling the speed to within limits ensures the operator's safety during rehabilitation training, while keeping the position within limits prevents the robot from moving beyond the patient's safety limits and also prevents the robot from colliding with mechanical limits. The robot's safety impedance controller is constructed as follows... Figure 2As shown, the robot controller first estimates the external force applied by the operator using the external force observer module M1. The external force observer is implemented using robot dynamics and a generalized momentum observer, with the generalized momentum observer gain set to 500. Using the external force estimation result as input, the interactive control module M2 outputs the desired velocity. The safety control module M3 outputs the desired velocity and determines whether it meets the velocity and position constraints. If not, it calculates the safety reference acceleration, velocity, and position according to the method of this invention. The impedance control module M4 receives the safety reference acceleration and calculates the torque, sending it to the motor control module. The motor control module M5 receives the torque command and sends the command to the motor amplifier component using FOC control. In the aforementioned safety module M3, the M31 submodule monitors the interaction force between the robot and the operator. If the external force exceeds a set threshold, the robot switches to a zero-force mode to ensure that the force between the robot and the operator is not excessive. The robot safety admittance controller is as follows: Figure 3 As shown, its structure is similar to a safety impedance controller, with the impedance control module M4 replaced by a servo drive module M4.

[0125] To further explain how to ensure speed and position safety during the rehabilitation process, a safety admittance control scheme is adopted. First, an external force estimation algorithm is executed. Based on the robot's motion state, joint torques, and dynamic model, an external force observer is used to estimate the external force currently applied to the robot by the user. The estimated external force is as follows: Figure 4 As shown. Then, the desired acceleration is calculated based on the interaction equation of the rehabilitation model. and speed The desired speed and acceleration may not meet safety requirements. If the robot executes this desired command, it is highly likely to exceed the speed limit, causing discomfort to the user. Simultaneously, the robot may also exceed the set training range and the safe rehabilitation area set by the therapist. Therefore, the safety module performs safety shaping on the desired speed and acceleration to obtain a safe reference acceleration and speed. Finally, admittance control is used to generate a safe speed and send it to the robot servo controller, achieving safe physical human-machine interaction control. During training, the speed limit is 0.733 rad / s, and the user's rehabilitation training range is -1.31 rad to -0.125 rad. Figure 4 As shown, during user training, the joint speed generally does not exceed the set value. When the user approaches the boundary of the set range, even if the user still exerts force, the robot cannot break through the boundary, thus ensuring that the positional constraints are not exceeded during rehabilitation training. Therefore, the safe physical human-machine interaction control algorithm proposed in this invention can effectively limit the speed and position of the machine during human-machine interaction. The proposed method is compatible with both admittance control and impedance control implementation schemes, greatly enhancing operator safety during physical human-machine interaction and improving the rehabilitation training experience.

[0126] In another embodiment of this application, a robot control terminal is provided, including a processor and a memory. The memory stores computer instructions, and the processor is configured to call and execute the computer instructions to implement the safe physical human-machine interaction control method as described above.

[0127] In another embodiment of this application, a computer-readable medium is provided, which stores computer instructions. When the computer instructions are invoked and executed by a computer, the secure physical human-computer interaction control method described above is implemented.

[0128] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the inventive concept of this application, and these all fall within the protection scope of this application.

Claims

1. A safe physical human-computer interaction control method, characterized in that, Includes the following steps: S1: Generate the desired trajectory for rendering virtual interactive dynamics based on the interaction control equations and interaction forces between the robot and the human. S11: Obtaining External Forces for Robot-Human Interaction ; S12: Based on the robot-human interaction equation (1) The expected trajectory is calculated. ,in These are the expected acceleration, velocity, and position in Cartesian space. Represents the robot's actual acceleration, velocity, and position in Cartesian space; S13: Transform the Cartesian space trajectory to joint space, specifically calculated as follows: (2) (3) (4) in These represent the desired acceleration, velocity, and position in joint space, respectively. Indicates the inverse kinematics of the robot. The Jacobian matrix represents the robot. Describe the generalized inverse of the Jacobian matrix. This represents the derivative of the Jacobian matrix with respect to time; S2: Reshape the generated desired trajectory according to safety constraints to obtain a safe reference trajectory; Step S2 specifically includes the following steps: S21: Establish the control barrier function , define when When the robot is in a safe state, it is considered to be in a safe state; and when At that time, it is considered that the robot in interaction is unsafe. Used to indicate the safe distance between the robot and the environment, the robot itself, and the operator, or to indicate the safety limits for the position of the robot's joints; S22: Calculate the partial derivatives of the control barrier function. ; S23: Calculate the safety reference trajectory and safety reference velocity. Must meet (5) in Represents extended K ∞ Class function; In step S23, considering position-related safety, the desired velocity in equation (1) is shaped to obtain the safe reference velocity at the current time k. (6) (7) in This represents the operation of finding the minimum value; From safety reference speed Obtain safety reference acceleration (8) and safety reference location (9) China T s Indicates the control cycle; Alternatively, in step S23, considering position and velocity-related safety, the velocity in Cartesian space must satisfy constraints. (10) Joint space velocity satisfies constraints (11) , Represents the minimum and maximum allowable velocities in Cartesian space. , Indicates the minimum and maximum allowable velocities in the joint space; When the Cartesian space robot interacts, it must satisfy the requirement of equation (10), then we have (12) in , This represents the absolute value operation. Represents symbolic operations, Obtained from the interaction equation; The desired velocity in the joint space is obtained from equation (12). (13) Since the joint space velocity constraint equation (11) needs to be satisfied, the desired joint space velocity is obtained. (14) Equation (14) yields the control barrier function that satisfies positional safety. Joint safety reference speed (15) Equation (7); The safe reference acceleration and position that satisfy the position and velocity constraints are obtained from equation (15); S3: Based on the safety reference trajectory obtained in step S2, issue the command to the robot controller for execution.

2. The safe physical human-computer interaction control method according to claim 1, characterized in that, The The function is selected as And parameters , This indicates the convergence speed of the robot's speed tracking error. For parameters related to the initial velocity tracking error, This represents the Lipschitz constant of the control barrier function.

3. The safe physical human-computer interaction control method according to claim 1, characterized in that, In step S3, the safety physical human-machine interaction control includes either safety admittance control or safety impedance control: Safety admittance control: The safety reference speed obtained in step S2 Or safety reference location Send out robot speed or position trackers; Safety impedance control: based on the desired reference acceleration obtained in step S2 Calculate robot control torque (16) in Represents the inertia matrix. Represents Coriolis force and centripetal force. Indicates the effect of gravity. Indicates friction. This indicates the equivalent effect of external force acting on the joint space.

4. The safe physical human-computer interaction control method according to claim 1, characterized in that, It also includes step S4: during the human-computer interaction process, detecting whether the interaction force exceeds the limit; if... or If the limit is exceeded, the robot will abandon the current task and switch to zero-force mode to wait for safe operation to resume.

5. A robot control terminal, characterized in that, It includes a processor and a memory, the memory storing computer instructions, and the processor being configured to: invoke and execute the computer instructions to implement the secure physical human-computer interaction control method as described in any one of claims 1-4.

6. A computer-readable medium, characterized in that, The computer-readable medium stores computer instructions, which, when invoked and executed by a computer, implement the secure physical human-computer interaction control method as described in any one of claims 1-4.

Images