基于深度学习的机器人复杂动作控制技术研究  

Research on Robot Complex Motion Control Technology Based on Deep Learning

作　　者：朱煜东 万韬阮[2] 汤汶 薛涛 Zhu Yudong;Wan Taoruan;Tang Wen;Xue Tao(Shaanxi Key Laboratory of Clothing Intelligence,School of Computer Science,Xi'an Polytechnic University,Xi'an 710048,China;University of Bradford;Bournemouth University)

机构地区：[1]西安工程大学计算机科学学院/陕西省服装设计智能化重点实验室,西安710048 [2]布拉德福德大学 [3]伯恩茅斯大学

出　　处：《单片机与嵌入式系统应用》2021年第8期44-47,共4页Microcontrollers & Embedded Systems

摘　　要：本文提出了一种基于深度学习的手势识别来控制机器人做复杂动作的解决方案。本文提出的方案展现出有效的开放性和可扩展性,无论是手势识别还是在机器人复杂动作方面,都可以根据需求做相应调整。同时,本研究工作创建了一个有效的基于人工智能网络的机器人控制实用实验平台。实验设计分别以运用6个基础手势控制机器人在危险环境中清除可疑物品为例,验证了开发的平台可以让机器人实时、远程地完成预设复杂动作,满足现实中自然人机交互的功能。In the paper,a solution based on deep learning gesture recognition to control the robot to do complex actions is proposed.The scheme proposed in this paper shows effective openness and expansibility,and can be adjusted according to the needs in both gesture recognition and complex robot actions.At the same time,this research also created an effective robot control experimental platform based on artificial intelligence network.The experiment design uses six basic gestures to control the robot to clear suspicious objects in dangerous environment.The experiment results show that the developed platform can enable the robot to complete preset complex actions in real time and remotely,and meet the function of natural human-computer interaction in reality.

关 键 词：深度学习 手势识别 机器人复杂动作 人机交互 实时远程控制 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]