论文部分内容阅读
针对机器人在没有任何初始位姿先验知识的情况下,通过传感器感知信息确定位姿的全局定位问题,将智能空间技术与ROS(robot operating system)服务机器人相结合,设计了一种智能空间技术支持下的基于WIFI指纹定位和蒙特卡洛粒子滤波定位的复合服务机器人全局定位系统.在该复合定位方法中,首先利用智能空间中的基于BP(backpropagation)神经网络的WIFI指纹定位对机器人进行粗定位,并将估计位置与估计误差发送给ROS服务机器人;在粗定位的基础上使用蒙特卡洛粒子滤波算法方法最终获得服务机器人的精确位置.实验结果表明,本文设计的系统实现了ROS机器人与智能空间之间的零配置与松耦合,可有效地提高服务机器人全局定位精度,缩短计算迭代时间.
Aiming at the problem that the robot can determine the position and orientation of the robot by sensing the information without any priori knowledge of the pose, a smart space technology is designed by combining the intelligent space technology with robot operating system (ROS) Based on WIFI fingerprinting and Monte-Carlo particle filter localization, a hybrid service robot global positioning system is proposed.In this method, the robots are first coarsened by WIFI fingerprinting based on BP (backpropagation) neural network in smart space And then send the estimated position and estimation error to the ROS service robot.On the basis of rough positioning, the Monte Carlo particle filter algorithm is used to get the exact position of the service robot.Experimental results show that the system designed in this paper realizes the optimization of ROS robot and Zero configuration and loose coupling between smart spaces can effectively improve the global positioning accuracy of service robots and shorten the computation iteration time.