Flexible mechanical hand based on machine learning

Abstract

An embodiment of the invention provides a flexible mechanical hand based on machine learning. Each flexible mechanical finger comprises a hollow mechanical finger and a hollow flexible finger stall. Alight reflecting material coats the inner wall of each finger stall, an opening is formed in the upper end of each finger stall, and each finger stall sleeves the tail end of the corresponding mechanical finger through the corresponding opening. Optical fibers are arranged inside a hollow structure of each mechanical finger, extend to the interior of the corresponding flexible finger stall and sense the deformation generated by the corresponding light reflecting material on the inner wall of the corresponding finger stall. The transmitting ends of the optical fibers transmit images to the receiving ends of the optical fibers through an optical fiber array, the receiving ends of the optical fibers are connected with a control system, the control system receives the images, a grabbed biological sample is judged through machine learning, and an actuator is controlled by a driving mechanism to conduct the grabbing work. According to the flexible mechanical hand based on machine learning,the mechanical hand can judge the kind and the weight of the grabbed biological sample by itself through the mode of machine learning, and therefore the appropriate force is given to grab the sample,and the integrity of the biological sample is guaranteed.

Description

technical field [0001] The present application relates to the technical field of manipulators, in particular to a machine learning-based software manipulator. Background technique [0002] The function and form of traditional rigid manipulators have been widely used in the production of human society. These rigid manipulators are mainly made of metal materials and are often responsible for structural and repetitive tasks in industrial sites. However, modern manipulators are developing towards high positioning accuracy, high flexibility and high human-computer interaction. In recent years, soft manipulators designed and manufactured using soft materials have attracted widespread attention from scholars and institutions at home and abroad, and have received continuous research, which provides solutions for essentially solving the problems of rigid manipulators such as poor environmental interaction and poor adaptability to complex environments. and directions. Researchers f...

AI Technical Summary

Problems solved by technology

Rigid manipulators can easily cause damage when collecting marine biological samples and cannot obtain the integrity of biological samples

Images