Font Size: a A A

Research On Grasping Force Control Of Robot Underactuated Flexible Multi-fingered Hand

Posted on:2024-03-31Degree:MasterType:Thesis
Country:ChinaCandidate:Y L XuFull Text:PDF
GTID:2568307100459394Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
The grippers commonly used in current industrial robots have few degrees of freedom,poor flexibility and lack of flexible operation capability,and cannot meet the multi-functional grasping requirements.As a generalized end-effector,the underactuated flexible multi-fingered hand has the dexterity and grasping force closure characteristics of a multi-fingered hand,and can flexibly operate objects of different shapes and complete complex operation tasks,such as grasping,assembly and maintenance operations.Therefore,the development of underactuated multi-fingered hand with stable grasping performance,multiple grasping functions and easy operation has become an important research direction.Based on the mechanical body of an eightdegree-of-freedom underactuated three-fingered hand,the control board and driver board are designed,integrating various sensors such as force,tactile,angle,etc.The motors and drivers are installed inside the multi-fingered hand,with an overall compact structure and high flexibility.Standard serial communication is adopted to realize flexible grasping through data parsing of control commands.The main contents are as follows:(1)The development of the prototype control system,sensing system,communication system,control algorithm and control software is completed,and the underactuated flexible multi-fingered hand grasping test platform is established.A control board based on STM32F407 as the main chip and a power driver board are designed and installed inside the palm of the hand to collect a variety of sensor data such as force tactile and angle in real time.The underlying control program is deployed in the control board for implementation,and the upper computer interaction interface is developed based on the Qt platform,using serial communication to send control commands for instant connection.(2)The kinematic model of the underactuated finger is established,and its workspace is drawn using MATLAB.Based on the robot grasping point contact constraint model and force closed grasping theory,the multi-fingered grasping sphere and rectangular object are analyzed in detail with its grasping mapping matrix as an example.Based on the multi-fingered force-closed grasping equation and the external force on the object,the multi-fingered grasping force is solved iteratively to provide a basis for the grasping force control test.(3)To improve the end trajectory control accuracy in the underactuated multifingered grasping process,an interference observer-based sliding mode control method for underactuated fingers is proposed.By introducing an interference observer and a saturation function sliding mode convergence law,which are used to estimate the magnitude of the external interference value and suppress the jitter phenomenon in real time,respectively,the interference estimate acts on the control input and gives the corresponding compensation at the control side to improve the robustness of finger control.(4)In the developed underactuated multi-fingered grasping experimental platform,the grasping control scheme of single-finger force control and two-finger position control is adopted.The middle fixed finger is controlled by a force control method,which is realized by PI force controller and incremental PID controller in series.The symmetrical fingers on both sides adopt position control method,which is realized by incremental PID control.All the three fingers of underactuated multi-fingered hand work together to complete the grasping force control experiments on some target objects and realize the flexible grasping.(5)To address the interference problem of system process noise and sensor observation noise in the underactuated robot system.A LQG/LTR control method is proposed for underactuated two-degree-of-freedom robotic arm.The control method consists of two parts,the optimal state feedback gain regulator and the Kalman filter,and further uses the loop transfer recovery technique to improve the stability margin of the control system,and the simulation test proves its effectiveness.
Keywords/Search Tags:Multi-fingered hand, Grasping force, Underactuated
PDF Full Text Request
Related items