Analysis And Experimental Research On Robot Picking Action

Fruit harvesting is the most time-consuming and energy-consuming process in fruit production.Thanks to the complexity of picking operation,the degree of automation of picking is still very low.At present,fruit picking works are basically carried out manually in China.With the aging of population and the reduction of agricultural labor force,the labor cost increases year by year.Robot picking will become the main trend of fruit picking.The working environment of the picking robot is greatly influenced by the seasons and weather.the complexity of the working environment,the randomness of the fruit growth position,the overlap of fruit and fruit,the occlusion of fruit and branches and leaves,and the uncertainty of illumination conditions,which lead to the existence of noise and various interference information in the image and reduce the accuracy of identifying and locating the target fruit.It limits the wide application of picking robots in practice.Studying the robot picking action and exploring the fuzzy picking method suitable for the robot has important application significance.Based on the analysis of the manual picking action,this study proposes a fuzzy picking method based on the characteristics of robot picking.The apple picking can be completed without relying on the posture information of apple and the position information of the fruit stem.The dynamic model of apple picking process was established.The process of breaking,twisting,pulling,compound action,assistant restraint action and picking effect were evaluated by simulation.The effects of branch stiffness,restraint position,movement speed and fruit posture on fruit picking performance were analyzed.The fuzzy picking end effector is designed.By designing the auxiliary support,the porous suction cup increases the tolerance of the end effector to the size,shape and posture of the apple,and realizes fuzzy picking.The feasibility of the proposed method is proved by experiments.Research and experimental results show that the proposed robotic fuzzy picking method can pick apples of different postures and sizes,reduce the dependence of visual system,and improve the picking efficiency.The fuzzy picking end effector has good applicability for apple picking with the attitude angle in the range of [-45 °,45 ° ] and the diameter of75mm-95 mm,which provides an effective technical approach for robot picking.