Principle and implementation steps of robot dragging teaching based on admittance control

Drag-to-drag teaching based on admittance control is a commonly used robot control strategy that allows direct interaction between humans and robots for natural and intuitive control. In admittance control, the robot is modeled as a force, and its dynamic characteristics are described by an equivalent admittance model. The admittance model is similar to a combination of a spring and a damper, and is used to describe the robot's response to external forces and motions.

1. Principle

Admittance model: Robot admittance control is a robot control strategy used to achieve force interaction and force control between the robot and the external environment. Its basic principle is to model the robot as a force control system, similar to a spring-damper system, which can respond to external forces and torques. The core idea of ​​robot admittance control is to respond to external forces applied to the robot instead of predetermined trajectories or positions. This enables the robot to adapt to different force interaction scenarios, such as interacting with human operators or working in uncertain environments.

Drag teaching principle: The user manually operates the end of the robot or the connection with the robot, and the robot follows the user's movement in real time according to the response of the admittance model, and has a certain response to the external force. This method allows the user to intuitively guide the robot to complete the task, such as guiding the robot to drag parts during the assembly process.

2. Implementation steps

1. Configuration: Install a force sensor or force/torque sensor at the end of the robot to measure the force and torque applied to the robot in real time.

2. Admittance model design: Design an appropriate admittance model, including stiffness and damping parameters, based on task requirements and user operating characteristics.

3. Force Design: Develop a force controller based on the admittance model that calculates the robot's control commands based on the measured external forces and the desired admittance model response, so that the force and motion at the end of the robot respond to the user's operation.

4. Teaching process: The user manually operates the robot end or joystick, and the robot follows the user's movement in real time according to the instructions of the force controller. In this process, the movement of the robot is affected by the force applied by the user.

5. Performance implementation and optimization: In practical applications, performance optimization may be required, including adjustment of the admittance model, gain adjustment of the controller, and filtering and calibration of sensor data to improve the control accuracy and stability of the robot.

Admittance control is often used to control robots that physically interact with their environment or objects. The goal is to achieve a desired admittance characteristic between the interacting lever and the robot's motion.

Drag teaching is done by directly applying a force in a certain direction on the end of the robot or the connecting rod. The controller force system estimates the external drag torque, guides the robot to make corresponding follow-up movements, and then records the position of the dragging process to complete the teaching work.

During the teaching process, it is necessary to overcome the gravity torque, friction torque and inertia torque of the robot connecting rod.

The dragging teaching module based on the dynamic model is mainly divided into four functional parts: dynamic parameter identification, dynamic model, torque pre-processing and dragging control.

The admittance control method is suitable for the motion control of robot dragging teaching. The admittance control takes the external drag torque as input and calculates the position command of the joint dragging following according to the motion law of the spring-damper-oscillator second-order system. The admittance control simulates the robot link motion into a spring damping system!!!

Get joint torque through or joint torque sensor!!!

The ideal zero-force mode is that the arm can be dragged to any position, but when the admittance/impedance control drags the robotic arm, if the external force remains unchanged, there will be a limit to the dragging position.

Application scenario: For mobile lower limb rehabilitation, the admittance controller is used to allow the cart to move with the patient with minimal effort!!!

The key difference between admittance control and impedance control is that the former controls motion after measuring force, whereas the latter controls force after measuring motion or deviation from a set point.

Admittance control, similar to impedance control, aims to impose a desired dynamic behavior on the robot subject to external contact forces based on the available admittance parameters (i.e., inertia, stiffness, and damping).

Unlike the impedance control law that calculates the reference joint torque, the output of the admittance controller is based on the reference motion of the inner motion control loop that measures (or estimates) the contact forces.

The evaluation of admittance control performance usually involves the following considerations:

1. Response Speed: The response speed of a robot under admittance control to external forces is a key indicator. A fast and stable response can improve the robot's operating efficiency, especially in tasks that require rapid adaptation to changes in external forces.

2. Stability: The stability of the control system is very important. The admittance control system should be able to maintain a stable force balance and avoid instability or oscillation caused by external interference.

3. Coordination of force and motion (Fce-Moon Coordination): A robot under admittance control needs to be able to coordinate the externally applied force and its own motion to ensure the balance between force and motion so that the robot can perform task operations stably.

4. Adaptability: One advantage of admittance control is its adaptability, that is, the robot can adapt to external forces of different sizes and directions. A good admittance control system should be able to show good adaptability in different working environments and task requirements.

5. Disturbance Rejection: The admittance control system should be able to resist external disturbances, including sudden applied forces or changes in disturbing forces. A stable admittance control system should be able to adjust quickly to offset these disturbances.

6. User Experience: If admittance control is used for human-robot interaction, user experience is very important. The control system should be intuitive, allowing users to interact with the robot naturally, and flexible enough to adapt to the user's actions and needs.

7. Energy Efficiency: The energy consumption of the robot is also an important performance indicator. An efficient admittance control system should be able to minimize energy consumption while maintaining stability and performance.

The connotation of admittance control:

1. Force control: The core idea of ​​admittance control is a force-dominated control strategy. Different from traditional position control, admittance control focuses on the robot's response to external forces.

2. Modeling: In admittance control, the robot is modeled as an equivalent force control system. This model usually includes an admittance model, which describes the stiffness and damping characteristics of the robot and is used to simulate the robot's response to external forces.

3. Balance of force and motion: Admittance control seeks the balance between force and motion. The robot can adjust its own motion according to the external force to maintain the coordination between force and motion.

4. Adaptability: Admittance control systems are usually highly adaptable and can adapt to external forces of different sizes, directions and speeds to suit different task requirements and working environments.

5. User interaction: Admittance control is often used in human-machine interaction. Users can guide the robot's movement by manually operating the robot's end or joystick, achieving intuitive control and collaborative work.

6. Stability: The stability of the control system is a key factor. The admittance control system should be able to remain stable in the presence of external disturbances and force changes, avoiding instability or oscillation.

7. Force balance: Admittance control seeks to maintain force balance. The robot's response to external forces should be balanced with the external forces to achieve the control goal.

8. Real-time response: Admittance control requires the robot to respond to external forces and user inputs in real time to maintain the coordination of forces and motions.

Admittance Control is a robot control strategy that focuses on the force interaction and force control between the robot and the external environment. Its core idea is to model the robot as a force control system, similar to a combination of springs and dampers. This control strategy allows the robot to interact with the external environment, enabling the robot to adapt to different force interaction scenarios, such as interacting with human operators or working in uncertain environments.

1. Admittance model: In admittance control, the robot is usually modeled as an equivalent admittance model, which includes two main parameters:

Stiffness: The stiffness parameter determines how quickly the robot responds to external forces. Higher stiffness causes the robot to respond faster to changes in external forces, while lower stiffness makes the response slower. Stiffness is usually expressed as a force/displacement ratio.

Damping: The damping parameter determines the smoothness of the response. Higher damping will slow down the robot's response and reduce oscillations, while lower damping may result in a more unstable response. Damping is usually expressed as a force/velocity ratio.