Research On Control System Of Parallel Automatic Tea-picking Machine

China has a long history of tea,huge tea production and a vast market.The picking process is the key to the tea industry,but it is still mainly manual picking,which is not conducive to industrial upgrading.The existing picking equipment is not selective,and can only pick low-quality bulk tea.Therefore,the market urgently needs a selective automatic picking equipment to improve the picking efficiency and reduce costs.There are two major problems to achieve the selective automatic picking of tea.One is the identification and positioning of the buds,and the other is how to pick accurately and efficiently.In response to this,based on the recent progress of tea-picking machinery and agricultural robots,this paper proposes a parallel automatic tea-picking machine that uses machine vision technology to realize bud recognition and positioning,and uses Delta robot to implement tea picking.In-depth research,the main research contents are as follows:First of all,through on-the-spot investigation of tea garden and tea processing plant,combined with the famous tea bud recognition and positioning system based on computer vision that has been designed and developed by the research group and a design scheme of parallel automatic tea harvester that has been put forward,the control system of parallel automatic tea harvester is designed.Based on the picking points obtained by machine vision,combined with the self-designed terminal finger picker on the parallel mechanism,the leaf buds are picked.In view of the tea garden terrain,the crawler car is used as the walking mode,which can achieve stable operation and flexible picking action.Secondly,a delta parallel robot kinematics model is established to lay a theoretical foundation for subsequent tea picking trajectory and path planning.At the same time,the forward and inverse solutions,singularity,and working space of the machine were analyzed based on the kinematics model of the mechanism to ensure that the working range of the equipment can cover the area where the tea buds should be collected,thereby reducing the leakage rate of leaf buds.After that,according to the actual demand of tea picking and the spatial distribution data of leaf buds,the picking trajectory of delta robot is designed andimproved by analyzing the working trajectory of delta robot and the distribution of speed and acceleration on the trajectory.The right angle corner of the transition trajectory is smoothed by Lamé curve,and the position,velocity and acceleration information of the interpolation point of the trajectory are planned by using the modified trapezoid motion law.Further,in view of the problem of the path planning of the suitable tea picking sequence during the tea picking process,by analyzing the actual distribution of the suitable tea picking,this problem is transformed into a mathematical problem of traveling salesman,and the path is planned by using an improved ant colony algorithm.At the same time,in view of the vertical position difference of tea and the operating characteristics of the Delta robot mechanism combined with the inherent problems of the classic ant colony algorithm,the classic ant colony algorithm was adaptively improved,and it was confirmed through simulation picking experiments that the improved algorithm has better optimization results.Finally,according to the needs of the experiment,an indoor prototype of the tea picking system was built,and a set of software for the Delta robot simulation picking control system was designed and developed.Based on the actual picking requirements,the hardware design is selected and selected.Based on the control principle of the servo system,an open control scheme of PC + motion control card is used to build a control platform.At the same time,based on the above hardware platform,using Microsoft's software development environment to develop a software solution for the Delta robot motion control based on the host computer,in order to achieve human-computer interaction,motion control,picking trajectory planning algorithm and picking sequence path optimization algorithm Lay the foundation for test functions such as verification.