Research On Fiber Optic Tactile Soft-bodied Manipulators

The development of robots has played an important role in the mechanization and automation of industrial production.As application objects become more complex,tra-ditional rigid robots have some drawbacks in terms of safety and flexibility.Due to the flexibility of the body material,the robot software hand provides a new idea and direc-tion for solving the problem of intricate adaptability and inflexibility of the rigid oper-ator's complex environment.However,the nonlinear,complex geometry of the soft body material makes software hand modeling and control a problem.To this end,this paper takes the advantages of high sensitivity,light weight and softness of the fiber and implanted FBG into the interior of the soft body,so the bending and contact force are detected without affecting the movement of the soft hand to realize perception function.This paper analyzes the development status of robot software hand and its tactile sensing,and introduces the classification of robot software hand and common flexible sensor.After pointing out the shortcomings and shortcomings of the current flexible sensors,a detection method using a fiber optic implanted flexible soft body is proposed.Firstly,the theoretical analysis of the pneumatic soft finger structure is carried out based on the Yeoh model,and the structure is further optimized by finite element sim-ulation to determine the internal dimensions of the soft finger.The FBG sensing prin-ciple and demodulation technology are introduced.The sensing system scheme is de-termined according to the finger size and the fiber wavelength division multiplexing technology.Then,the soft finger and the optical fiber were fabricated by using the lost wax casting process,and the corresponding tactile sensing system was built.The effect of FBG-based soft finger bending and contact force detection was verified,indicating the effectiveness and rationality of the sensor.The FBG for bending and pressure testing was then calibrated.The bending calibration uses image fitting and the pressure cali-bration is based on a standard pressure tester.The results show that the sensitivity of the curvature is 35.23 pm/m-1 and the pressure sensitivity is 29.53 pm/N.At the same time,a kind of slip sensor is designed.The sensor structure makes it sensitive to tangential force and insensitive to positive pressure,and uses the second derivative al?gorithm to realize the object sliding recognition.Finally,the methods and effects of the tactile sensing software hand for the surface shape and roughness recognition of the object are studied based on the software hand touch sensing system of the constructed FBG.With the passive compliance of the finger,the surface profile of the object is scanned and the recognition effect is satisfied.At the same time,the identification method of the softness and hardness of the object is studied using the pressure and bending perception information of the finger by increasing the experimental conditions of the contact force.The results show that the method has good feasibility.In the end,the intelligent tactile perception function of the robot software hand is realized to a certain extent.At the end of the article,the paper summarizes the full text,and points out the short-comings of this research and the areas that need improvement,and looks forward to the next work.