During the operation of the manipulator,it is prone to collision among the manipulators and manipulator with people and objects in the external environment,resulting in property damage or injury or even death.In view of this,after comparing and analyzing the current mainstream path planning algorithms,the convolutional neural network and the RRT Connect algorithm were fused to make up for the shortcomings of the original algorithm,such as strong randomness and slow algorithm execution efficiency,and the digital twin was introduced into the collision detection process,the data interaction between the virtual and real systems was established to realize the real-time monitoring of the grasping system.Finally,based on CoppeliaSim software,a manipulator twin simulation platform was built and the simulation experiments were carried out.The results show that the proposed detection method has the advantages of high grasping accuracy,strong stability and high reliability.