In order to address the issues of initial pose estimation and "kidnapping" faced by indoor mobile robots, an innovative relocalization method is proposed. When the robot encounters challenges in initial pose estimation and "kidnapping," a relatively accurate pose is transmitted to the laser localization algorithm through a visual algorithm, achieving the relocalization of the mobile robot. To validate the feasibility and stability of the algorithm, a series of experiments are conducted in real-world scenarios, demonstrating the practicality and robustness of the proposed localization algorithm.
Shuo WangXin LiYu ZhangSonghui MaXiaoxin Ren
Yichi ZhangHeng ZhangZhenhua XiongXinjun Sheng
Yutong WangChaoyang JiangXieyuanli Chen
Nithid MahattansinKanjanapan SukvichaiPished BunnunTsuyoshi Isshiki
Azhar Muhammad HamzaChaoxia ShiYanqing Wang