To investigate an effective physical pest control method for tea trees, we designed and manufactured a suction-based pest-capture machine (hereafter labeled the “pest vacuum”) and conducted a test and field experiment to evaluate its effectiveness in the control of Empoasca vitis Göthe (E. vitis). Based on the proposed model, the minimum practical air velocity of the pest vacuum was 5.85 m s⁻¹. The field experiment included two treatments and a blank control. In treatment 1, we used the pest vacuum along tea trees only once, while in treatment 2, the pest vacuum was used along trees twice, separately, in 2 consecutive periods, and with an interval of more than 30 min. The results show that a highly significant difference arose among the dropping rates of the two treatments and the blank control instantly after suction, a significant difference among the dropping rates 7 days later, and no significant difference 14 days later. The dropping rate and the effective rate of prevention of treatment 2 reached 81.87% and 80.60% instantly after suction. Moreover, the effective rate of prevention of nymphs was slightly higher than that of adults. Overall, the pest vacuum had a prominent, albeit short-term, effect on the control of E. vitis. Seven days after the suction, it is better to repeat the suction five days after the first suction. The pest vacuum provides a valid physical method for pest control, so more attention should be paid in future investigations to reducing the weight and working noise of the pest vacuum.

Neural Network has been used in approximation of dead zone nonlinearity when modeling the manipulator control systems. However the existed method fail to minimize the possible input saturation effect and the NN mapping accuracy also be degraded, which leads to degrading in tracking precise and accuracy. To establish an accurate control model of the tea picking robot, an adaptive compensator of m-RBF (modified radial basis function neural network) and adaptive law were designed aimed at the nonlinearity units and dead zone. The closed-loop tracking error of the proposed control system is eventually going to be stable and bounded. A simulation was carried out with Simulink, which show that m-RBF provides a good approximation to characteristic of Dead zone nonlinearity, and that the control scheme based on m-RBF had a excellent and stable tracking accuracy. The tea picking experiment with six-axis manipulator verifies the effectiveness of the proposed algorithm, in which the proposed method got a higher score of 95.3, near two times that of traditional PID control methods. It can be drown that the m-RBF has a faster learning rate and can avoid local minima; the control scheme based on m-RBF has a excellent performance on control accuracy, robustness and self-adaption, which is especially appropriate for real-time control, like tea picking robot.