论文部分内容阅读
为了解决中型组比赛环境下的足球机器人自定位、绑架和跟踪问题,提出一种基于遗传算法的机器人连续自定位方法.首先,从全景图像中提取出场上黑线与机器人辐射线的交点;然后,把该交点与环境地图对应黑线间距离之和的最小值做为适应度函数;最后,以机器人中心坐标做为个体,进行选择、交叉和变异操作,收敛到机器人的自定位坐标.另外,针对绑架和连续定位,在遗传算法全局自定位基础上利用梯度优化算法局部修正主位姿,提高了自定位精度和算法的鲁棒性.仿真结果和场地环境机器人自定位试验说明了该定位方法的有效性.
In order to solve the problem of self-localization, abduction and tracking of soccer robots under medium-sized competition, a continuous robot self-localization method based on genetic algorithm is proposed.First, the intersection of the black line and the robot’s radiation is extracted from the panoramic image; , The minimum value of the sum of the distance between the intersection and the corresponding black line of the environment map is taken as the fitness function. Finally, the robot center coordinates are used as the individual, and the selection, crossing and mutation operations converge to the self-positioning coordinates of the robot. , Aiming at the abduction and continuous positioning, based on the global self-localization of genetic algorithm, a gradient optimization algorithm is used to locally correct the pose and pose, which improves the self-localization accuracy and the robustness of the algorithm. The simulation results and the self-positioning experiment of the site environment robot illustrate the positioning The effectiveness of the method.