Development Of Data Collection Gloves And Research On Manual Harvesting Behavior Of Tomatoes

With the development of robotics,picking robots have great potential for development in agriculture.However,it is still very challenging to complete non-destructive picking for the current picking robots,and the human hand is very dexterous,and the five fingers of the human hand are coordinated with each other.It can complete complex grasping tasks.Therefore,in order to further improve the picking performance of the picking robot,it is necessary to deeply understand the grasping behavior of human hands when picking fruits.This article mainly expands from the following three aspects:(1)In order to obtain the distribution of the contact force between the human hand and the fruit when picking tomato fruit by hand,this paper designed a wearable 16-channel pressure data collection glove,which is mainly composed of a ring unit,a palm strap unit and a hand ring unit.Each unit is independent and adjustable to adapt to different sizes of hands,and realizes the function of collecting pressure value information during the interaction between different hands and the tomato.(2)On this basis,the pressure data collection glove was used in combination with a twochannel surface electromyography to study the grasping behavior of human hands in the three stages of picking-moving-release during the tomato picking process,and using multiple linear regression analysis to explore the tomato.The influence of fruit growth parameters on the total grasping stress,the number of contact areas and the total picking time;the overall statistical analysis of the contact sequence,number of contact areas,and contact frequency when picking tomatoes by hand;focusing on contact force and muscle potential.The relationship between activities,a multiple nonlinear regression analysis model is established to predict muscle potential activities.The results show that during the separation stage of fruit stems,the number of contact areas increases with the increase of the geometric mean diameter of the fruit.The angle of fruit stems has a significant effect on the total separation time;the end regions of the thumb,middle finger and index finger are always for each fruit.The picking produces the greatest pressure effect.The entry and exit sequence of the 16 hand regions is different during the picking process;the pressure exerted by the thumb and the other four fingers on the surface of the fruit is the same as that of the flexor pollicis longus and superficial digita flexor muscles,respectively.The potential activity level is moderately correlated.(3)Lastly,the key parts of the pressure data collection gloves were optimized to further improve the comfort and stability of the pressure data gloves.In addition,a Force-s EMG data collection system was developed in the visual studio development environment.The serial port is connected with the hardware equipment to realize the synchronous work of the pressure data collection glove and the electromyography instrument.In summary,these research systems reflect the coordination of human grasping and some efficient and non-destructive picking strategies,which provide an important technical basis for the design of fruit picking robots and intelligent picking control.