摘要
针对三峡等大型电站水轮机叶片坑内修焊机器人测控过程中的定位问题,设计了一种移动机器人沿复杂空间曲面运行时的定位算法。该算法以二维平面中的测程法为基础,利用机器人车体驱动轮编码器的测量信息和三维叶片曲面的离散点空间坐标信息,通过等时采样、瞬时平面近似及空间坐标变换等,计算任意采样时刻机器人在叶片表面的位置及姿态,实现机器人的空间定位。以水轮机叶片典型形貌的圆柱面为例,进行了机器人不同运行方式的仿真实验,结果表明该算法的定位误差小于1.5%。
A localization algorithm was developed to improve control of wheeled mobile robots for on-site repair of hydraulic turbine blades in large hydropower stations like the Three Gorges Dam. The algorithm was the odometry method in a 2-D plane with equal interval sampling, and instant plane approximations to calculate the robot＇s position and stance on the blade surface at any sampling time, using signals from encoders on the vehicle＇s wheels and 3-D coordinate information for discrete points on the blade. The algorithm was validated by localization simulations of various mobile robot operations on a cylindrical surface represented a typical blade shape. Results show that the total localization error of the algorithm is less than 1.5%.
出处
《清华大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2008年第2期171-175,共5页
Journal of Tsinghua University(Science and Technology)
基金
国家“八六三”高技术项目(2007AA04Z258)
国家自然科学基金资助项目(50275083)
高等学校博士学科点专项科研基金资助项目(20060003069)
关键词
轮式移动机器人
定位
复杂曲面
水轮机叶片
测程法
wheeled mobile robot
localization
complicated curved surface
hydraulic turbine blade
odometry
