Human-computer cooperation human body behavior intention discrimination method based on recurrent neural network

Abstract

The invention discloses a human-computer cooperation human body behavior intention discrimination method based on an improved recurrent neural network. The human-computer cooperation human body behavior intention discrimination method comprises the steps of establishing a deep learning network model for a human body behavior intention estimation problem in a process of completing a disassembly task through man-machine cooperation, and analyzing an intention advanced perception problem in combination with characteristics of video data; collecting corresponding video data in combination with thedisassembly task; solving the intention category by adopting an improved LSTM recurrent neural network, and training a deep learning network model through the acquired video data to obtain an optimalparameter; and adjusting a loss function of the deep learning network model according to the optimal parameters, testing discrimination results of different data lengths of a single video, and searching an optimal early pre-judgment effect. According to the human-computer cooperation human body behavior intention discrimination method, a real man-machine cooperation dismounting scene is combined,and an effective solution is provided for the robot to predict human body behaviors in advance in man-machine cooperation.

Description

technical field [0001] The invention is suitable for solving the problem of early prediction and discrimination of human behavior intention in the field of human-computer cooperation, and relates to a human-computer cooperation human behavior intention discrimination method based on a cyclic neural network. Background technique [0002] In recent years, human-machine collaboration has become a hot spot in intelligent manufacturing. In traditional manufacturing scenarios, due to safety reasons, human operators and robots are separated in different work areas, and each independently completes its assigned tasks. In the human-machine collaborative system, robots can assist humans to perform complex tasks, thereby improving production efficiency and reducing human load. In order to achieve human-robot collaboration, robots need to track human actions and estimate the behavioral intentions of human workers, which is crucial for robots to intelligently assist humans to complete c...

AI Technical Summary

Problems solved by technology

Since the state of each waste product is different, the disassembly of waste products cannot be done in a fixed process, and human-machine collaborative disassembly has great uncertainty, which makes the estimation of human behavior intention in human-machine collaborative disassembly very complicated.

Images