Search

For the current problems of poor grasping force control performance and complex control systems in robot hands, a forward-looking predictive intelligent force control method for complaisant grasping of thin-skinned fruits based on the structural performance advantages of flexible hands was proposed. Four grasping planning methods suitable for flexible hand structure configuration were proposed with the optimal configuration and force closure as constraints, and grasping stability evaluation mechanism was established. The nonlinear regression prediction model based on radial basis function neural network was constructed, and the prospective intelligent force control algorithm based on minimum grasping force was put forward emphatically. The structure of the grasping system was analyzed, a layered modular flexible hand control system was designed, and finally a flexible hand grasping experimental platform was built to conduct grasping performance testing. The experimental results show that our proposed method reduces the grasping force by 10.64% and 6.91% compared to the other two methods, respectively. At the same time, the success rate of grasping is increased by 2.67% and 5.00%, respectively, with an average success rate of 98.00%. The study provides a reference for further research on flexible hands. [ABSTRACT FROM AUTHOR]