受限视觉下RoboCup3D中机器人的定位  

Self localization under restricted vision in RoboCup3D

作　　者：赵业锦[1] 

机构地区：[1]南京邮电大学自动化学院,南京210003

出　　处：《电子测试》2010年第4期81-85,共5页Electronic Test

摘　　要：针对RoboCup3D中机器人视觉模型的重大改变,即由原来的完美视觉模型变为受限视觉模型,在综合完美视觉下机器人的定位方法,即陀螺仪定位、1个标志杆定位,3个标志杆定位,提出一种受限视觉下机器人自身定位策略,并应用卡尔曼滤波器来提高定位的精度,在Rcssserver3D仿真环境中进行了定位数据的采样和对比,实验表明此定位方法具有很好的竞赛适应性与健壮性。There is a great change of robot’s visual model in RoboCup3D,which changes the original perfect visual model to the restricted one.Summary the robot localization under perfect visual,for example： gyroscope positioning,a sign pole position,three sign pole position.In accordance with it,we put forward a strategy that robots can position by themselves under the restricted visual.We also use Kalman filter to improve the positioning accuracy.Besides,we they sample and analysis the data under the simulation environment of Rcssserver3D.Many experiments have shown this positioning strategy is characterized by high adaptive competition and robustness.

关 键 词：受限视觉 卡尔曼滤波器 自身定位 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]