CN-121989260-A - Force and position cooperative control method, system, equipment and medium for robot

Abstract

The application discloses a force and position cooperative control method, a system, equipment and a medium of a robot, which are applied to the field of robot control, wherein the method comprises the steps of acquiring contact force data, resistance data, vibration data and pose data; generating a coupling sensing tensor based on the contact force data, the resistance data and the vibration data, determining a force-bit coupling adjustment factor based on the coupling sensing tensor, generating a force control signal and a position control signal based on the force-bit coupling adjustment factor, the contact force data and the position data, respectively calculating a force control channel output and a position control channel output based on the control coefficient, the force control target value, the contact force data and the position control target value and the position data, generating a force-bit mixed adjustment instruction based on the force control channel output, the position control channel output, the force control signal and the position control signal, and driving the robot to adjust the contact force and the contact position based on the force-bit mixed adjustment instruction. The application can improve the operation precision of the robot to the force and the position under multi-mode sensing.

Inventors

• LI DUANJIAO
• Yao Juanwen
• NING XUEFENG
• LU HUAQING
• ZHANG MIN
• MAI WEIHUA
• Zhang Peirun
• ZHANG JIANGUO
• XU WENTAI
• SUN WENXING
• YUE XIAO
• LIU JIANMING
• JIA ZIRAN
• LIANG YONGCHAO
• QIN GUANGSHENG
• CHEN BIN
• WU XIAOLIANG

Assignees

• 广东电网有限责任公司

Dates

Publication Date

20260508

Application Date

20260408

Claims (10)

1. 1. The cooperative control method for the force position of the robot is characterized by comprising the following steps of: Acquiring contact force data, resistance data, vibration data and pose data of a robot end effector generated in the process of interacting with a contact environment by the robot end effector; Generating a coupling perception tensor for representing physical properties of a contact environment based on the contact force data, the resistance data and the vibration data, determining a force position coupling adjustment factor based on the coupling perception tensor, generating a force control signal and a position control signal based on the force position coupling adjustment factor, a preset force control target value and a preset position control target value and the contact force data and the position control data, respectively calculating a force control channel output quantity and a position control channel output quantity based on a preset control coefficient, the force control target value, the contact force data and the position control target value and the position control data, and generating a force position mixing adjustment instruction for control based on the force control channel output quantity, the position control channel output quantity, the force control signal and the position control signal; And driving the robot end effector to adjust the contact force and the contact position based on the force-position mixing adjustment instruction so as to realize the cooperative contact control of the force position between the robot end effector and the contact environment.
2. 2. The method of cooperative control of force bits of a robot of claim 1, further comprising, prior to the generating a coupled perceived tensor characterizing physical properties of a touch environment based on the touch force data, the resistance data, and the vibration data: Normalizing the acquired initial contact force data according to a plurality of axial directions to obtain the contact force data; And respectively carrying out standardization processing on the acquired initial resistance data and the acquired initial vibration data to respectively obtain the resistance data and the vibration data.
3. 3. The method of cooperative control of force bits of a robot of claim 1, wherein the generating a coupled sensing tensor for characterizing physical properties of a contact environment based on the contact force data, the resistance data, and the vibration data comprises: Determining a multi-dimensional tensor structure indexed by contact spatial position and time based on the contact force data, the resistance data, and the vibration data; Determining mapping relations among the contact force data, the resistance data and the vibration data and tensors based on the multi-dimensional tensor structure so as to map the contact force data, the resistance data and the vibration data to force attribute dimensions, resistance attribute dimensions and vibration attribute dimensions corresponding to the multi-dimensional tensor structure respectively; combining the force attribute dimension, the resistance attribute dimension and the vibration attribute dimension under the same tensor coordinate to obtain an initial coupling perception tensor; inputting the historical contact force data, the historical resistance data and the historical vibration data into a preset weak supervision learning model, taking the error between the minimized predicted tensor and the target tensor as an optimization target, and carrying out iterative updating on the mapping weight and the mapping bias in the mapping relation to obtain a target mapping weight and a target mapping bias; based on the target mapping weights, the target mapping bias, and the initial coupled perceptual tensor, a coupled perceptual tensor is determined that characterizes physical properties of the touch environment.
4. 4. The method of cooperative control of force and position of a robot according to claim 1, wherein the determining a force and position coupling adjustment factor based on the coupling-aware tensor comprises: Analyzing the coupled sensing tensor based on a tensor decomposition algorithm, and extracting to obtain hardness characteristic distribution, friction characteristic distribution and flatness characteristic distribution for representing physical properties of the contact environment; respectively determining hardness weight parameters, friction weight parameters and flatness weight parameters corresponding to the hardness characteristic distribution, the friction characteristic distribution and the flatness characteristic distribution; And carrying out weighted fusion operation on the basis of the hardness weight parameter, the friction weight parameter, the flatness weight parameter, the corresponding hardness characteristic distribution, the friction characteristic distribution and the flatness characteristic distribution, and determining a force-position coupling adjusting factor for representing the influence proportion of the contact environment on force control and position control.
5. 5. The method of claim 1, wherein generating the force control signal and the position control signal based on the force position coupling adjustment factor, the preset force control target value and the preset position control target value, and the contact force data and the pose data comprises: respectively determining a corresponding force control target value and a corresponding position control target value based on a preset force control target range and a preset position control target range; performing weighted fusion processing on the force control target value and the contact force data based on the force bit coupling adjustment factor to generate the force control signal; and carrying out weighted fusion processing on the position control target value and the pose data based on the force-position coupling adjustment factor to generate the position control signal.
6. 6. The method for cooperative control of force and position of a robot according to claim 1, wherein the calculating the force control channel output and the position control channel output based on the preset control coefficient, the force control target value, the contact force data, and the position control target value and the pose data, respectively, comprises: calculating a force control error signal based on the force control target value and the contact force data; Based on a preset force control proportional coefficient, a preset force control integral coefficient and a preset force control differential coefficient, proportional operation, integral operation and differential operation are respectively carried out on the force control error signal to respectively obtain a corresponding first proportional operation result, a first integral operation result and a first differential operation result, and the first proportional operation result, the first integral operation result and the first differential operation result are overlapped to obtain a force control channel output quantity; calculating a position control error signal based on the position control target value and the position and posture data; And respectively performing proportional operation, integral operation and differential operation on the bit control error signal based on a preset bit control proportional coefficient, a preset bit control integral coefficient and a preset bit control differential coefficient to respectively obtain a corresponding second proportional operation result, a second integral operation result and a second differential operation result, and superposing the second proportional operation result, the second integral operation result and the second differential operation result to obtain a bit control channel output quantity.
7. 7. The method of cooperative control of force and position of a robot according to claim 1, wherein the generating a force and position mixture adjustment command for control based on the force channel output, the position channel output, the force control signal, and the position control signal comprises: determining a force control adjustment output based on the force control channel output and the force control signal; determining a position control adjusting output quantity based on the position control channel output quantity and the position control signal; And fusing the force control adjusting output quantity and the position control adjusting output quantity to obtain a force-position mixed adjusting instruction for simultaneously controlling the contact force and the contact position.
8. 8. The force and position cooperative control system of the robot is characterized by comprising an acquisition module, a calculation module and a control module; the acquisition module is used for acquiring contact force data, resistance data, vibration data and pose data of the robot end effector, which are generated in the process of interacting with a contact environment by the robot end effector; The computing module is used for generating a coupling perception tensor for representing physical properties of a contact environment based on the contact force data, the resistance data and the vibration data, determining a force-bit coupling adjustment factor based on the coupling perception tensor, generating a force control signal and a position control signal based on the force-bit coupling adjustment factor, a preset force control target value and a preset position control target value and the contact force data and the position control data, respectively calculating a force control channel output quantity and a position control channel based on a preset control coefficient, the force control target value, the contact force data and the position control target value and the position control data, and generating a force-bit mixed adjustment instruction for control based on the force control channel output quantity, the position control channel output quantity, the force control signal and the position control signal; and the control module is used for driving the robot end effector to adjust the contact force and the contact position based on the force-position mixing adjustment instruction so as to realize the force-position cooperative contact control between the robot end effector and the contact environment.
9. 9. A terminal device comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing a method of cooperative control of force bits of a robot according to any of claims 1-7 when the computer program is executed by the processor.
10. 10. A computer readable storage medium comprising a stored computer program, wherein the computer program, when run, controls a device in which the computer readable storage medium is located to perform a method of cooperative control of force bits of a robot according to any of claims 1-7.

Description

Force and position cooperative control method, system, equipment and medium for robot Technical Field The invention relates to the field of robot control, in particular to a method, a system, equipment and a medium for cooperative control of force and position of a robot. Background With the development of robotics, automation systems and intelligent manufacturing, robots are increasingly interacting with external environments in the context of industrial production, service work, medical rehabilitation, etc. When the robot finishes grabbing, assembling, carrying or contacting operations, not only the contact force needs to be accurately controlled, but also the contact position, the contact attitude and the physical characteristics of the contact surface, such as hardness, friction coefficient, surface flatness, vibration state and the like, need to be considered. In order to achieve stable and accurate interaction, the robot must have real-time sensing and adaptive adjustment capabilities for various physical properties. In the prior art, conventional force control methods or position control methods have typically focused on a single control target, such as controlling only the contact force or controlling only the position. Although force control and position control combined methods have appeared in recent years, most of the methods are based on data of a single sensing channel, and cannot fully utilize heterogeneous information from a multi-mode sensor, and also lack dynamic sensing and comprehensive processing capabilities of complex physical properties of a contact surface. In addition, the existing method is difficult to realize self-adaptive adjustment under the environment of real-time change, and the hardness, friction and surface state of the contact surface are not responsive enough, so that the operation precision, stability and adaptability of the robot in a complex environment are affected. Disclosure of Invention The invention provides a force and position cooperative control method, a system, equipment and a medium for a robot, which can improve the operation precision of the robot on force and position under multi-mode sensing. In a first aspect, an embodiment of the present invention provides a method for cooperatively controlling force and position of a robot, including: Acquiring contact force data, resistance data, vibration data and pose data of a robot end effector generated in the process of interacting with a contact environment by the robot end effector; Generating a coupling perception tensor for representing physical properties of a contact environment based on the contact force data, the resistance data and the vibration data, determining a force position coupling adjustment factor based on the coupling perception tensor, generating a force control signal and a position control signal based on the force position coupling adjustment factor, a preset force control target value and a preset position control target value and the contact force data and the position control data, respectively calculating a force control channel output quantity and a position control channel output quantity based on a preset control coefficient, the force control target value, the contact force data and the position control target value and the position control data, and generating a force position mixing adjustment instruction for control based on the force control channel output quantity, the position control channel output quantity, the force control signal and the position control signal; And driving the robot end effector to adjust the contact force and the contact position based on the force-position mixing adjustment instruction so as to realize the cooperative contact control of the force position between the robot end effector and the contact environment. The embodiment of the invention provides accurate and rich input information for subsequent control by collecting multi-mode sensing data in real time and reflecting mechanical and physical interaction characteristics in the contact process, improves the sensing capability of a system to a contact environment, establishes a coupling sensing tensor through multi-mode data fusion, realizes comprehensive characterization of heterogeneous physical properties such as hardness, friction coefficient, flatness and the like of a contact surface, generates a force position coupling adjustment factor, enables a system to dynamically adjust a force position distribution proportion according to the environmental properties so as to enhance the adaptability and the accuracy of a control strategy, realizes dynamic generation of force control and position control signals by coupling the environment sensing information with a task target value, enables the robot to adjust the output of force and position under different contact conditions, better meets the task requirements, improves the operation stability and the accuracy, calculates the channel output q