在RoboCup3D仿真比赛中,机器人自定位非常重要,定位不准确会对仿真比赛产生严重的影响.为了模拟真实环境,比赛中加入了视觉噪声,这使机器人定位变得更加困难.本文针对RoboCup3D仿真中的机器人视觉特征,提出一种观测值加权融合的卡尔曼滤波方法来实现机器人自定位,采用此方法能得到更精确的观测值.仿真实验结果表明,此定位方法大大提高了机器人自定位的精度.%In RoboCup3D simulation game, the robot self-location is very important, inaccurate location will have serious affect on simulation game. In order to simulate the real environment, it adds visual noise in the game, which makes it more difficult to lo-cate the robot. In this paper, considering the robot visual features of RoboCup3D simulation, it proposes an observations weighted fusion Kalman filter approach to realize robot self-location, using this method to get more accurate observations. The simulation ex-perience result shows that the method greatly improves the accuracy of robot self-location.
展开▼
