摘要
针对传统的移动机器人人机交互通常采用摇杆或按键等操作方式,存在操作繁琐、系统延时大等缺点。基于此提出一种基于姿态传感器的人机交互系统,操作端以Kinect为姿态传感器采集人体姿态信息并上传至云服务中心,机器人端通过云端发出的指令对机器人进行远程控制。实验结果表明,该人机交互系统可以实时高效地操作机器人完成各种运动,具有操作简单、低延时等优点,解决了传统空间机器人操控中的不足,提供了高效的移动机器人遥操作方案。
The traditional human-computer interaction of the mobile robot usually adopts joystick,button,or other opera-tion modes,which has the disadvantages of complicated operation and long system delay. Therefore,a human-computer interac-tive system based on the attitude sensor is proposed. On the operation terminal,the attitude information of the human body is collected and uploaded to the cloud service center by taking the Kinect as the attitude sensor. On the robot terminal,remote con-trol of the robot is conducted by means of the instructions issued from the cloud terminal. The experimental results show that the human-computer interactive system can operate robots to complete various motions in real time and efficiently,and has the ad-vantages of simple operation and short delay,which can make up for the deficiency in operation and control of the traditional space robot and provide an efficient solution for teleoperation of the mobile robot.
引文
[1]董历亚.基于Kinect三维物体的位置检测[D].天津:天津科技大学,2015.DONG Liya.Position detection based on Kinect 3D objects[D].Tianjin:Tianjin University of Science&Technology,2015.
[2]乔文静.基于RGB-D摄像机的人体动作识别的研究[D].西安:西北工业大学,2015.QIAO Wenjing.Research on human motion recognition based on RGB-D camera[D].Xi’an:Northwestern Polytechnical University,2015.
[3]ZHENG Y,YAMANE K.Adapting human motions to humanoid robots through time warping based on a general motion feasibility index[C]//Proceedings of IEEE International Conference on Robotics and Automation.Seattle:IEEE,2015:6281-6288.
[4]HAN J H,KIM J H.Consideration about the application of dynamic time warping to human hands behavior recognition for human-robot interaction[M].Berlin:Springer,2014:269-277.
[5]王松林.基于Kinect的手势识别与机器人控制技术研究[D].北京:北京交通大学,2014.WANG Songlin.Research on Kinect-based gesture recognition and robot control technology[D].Beijing:Beijing Jiaotong University,2014.
[6]刘娇.基于Kinect的骨骼追踪及肢体动作识别研究[D].西安:西安工业大学,2016.LIU Jiao.Kinect-based bone tracking and limb movement recognition[D].Xi’an:Xi’an Technological University,2016.
[7]边俊霞.基于Kinect的人体动作识别系统与实现[D].济南:济南大学,2015.BIAN Junxia.Implementation of human action recognition system based on Kinect[D].Jinan:University of Jinan,2015.
[8]宋腾.面向家庭服务机器人的动态手势识别方法研究[D].秦皇岛:燕山大学,2016.SONG Teng.Research on dynamic gesture recognition for home service robots[D].Qinhuangdao:Yanshan University,2016.
[9]郑中华.基于肢体动作的人体姿态识别研究[D].西安:西安工业大学,2015.182ZHENG Zhonghua.Research on human pose recognition based on limb movements[D].Xi’an:Xi’an Technological University,2015.
[10]蒋亚杰.基于Kinect的人体姿态识别和机器人控制[D].深圳:深圳大学,2017.JIANG Yajie.Kinect-based human pose recognition and robot control[D].Shenzhen:Shenzhen University,2017.
[11]柴琳.基于立体视觉的人体动作识别方法研究[D].青岛:中国海洋大学,2015.CHAI Lin.Research on human motion recognition based on stereo vision[D].Qingdao:Ocean University of China,2015.