摘要
本文使用Leap Motion手势识别设备获取手势轨迹路径,使用中值滤波和均值滤波对原路径进行去噪,获取较平滑的原手势轨迹。在需要工业机器人进行多次重复高效的轨迹运行时,将手势轨迹相对于关节空间进行时间-冲击最优轨迹规划,根据需要选取一定数量关键点。本文使用粒子群算法对所给关键点进行轨迹优化,根据结果对比表明优化后粒子群具有收敛速度快,寻优效果好的优点,并使用粒子群算法对手势轨迹进行机器人6轴关节空间轨迹优化,获得时间-冲击最优轨迹。
In order to realize the robot trajectory with gesture trajectory,this paper uses the Leap Motion gesture recognition device to acquire the gesture trajectory path,and uses median filtering and averaging filtering to denoise the original path for obtaining a smoother original gesture trajectory.When the industrial robot is required to perform repeated and efficient trajectory operation,the time-impact optimal trajectory planning of the gesture trajectory relative to the joint space is performed,and a certain number of key points are selected according to the needs.In this paper,the particle group algorithm is used to track the given key points.According to the result comparison,the optimized particle swarm has the advantages of fast convergence speed and good optimization effect.The particle swarm algorithm is used to optimize the six-axis joint space trajectory of the gesture trajectory,and the time-impact optimal trajectory is obtained.
作者
吕亚辉
严雨灵
Lü Yahui;YAN Yuling(School of Air Transportation,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《智能计算机与应用》
2019年第1期122-126,共5页
Intelligent Computer and Applications
基金
上海工程技术大学研究生科研创新项目(E 3-0903-18-01176)
关键词
手势识别
轨迹规划
粒子群优化
gesture recognition
trajectory planning
particle swarm optimization
