This paper presents a self-localization algorithm that allows a recursive state estimation process to be collective in a multi-robot coalition team that is guaranteed connected. In a coalition team, a robot may make mistakes of controlling and measurement for mobility and have restricted ability of identifying the other robots in the team such as ranging and angle of sensing. Our approach allows a leader robot to obtain a temporary estimate from at the current ^_@span style=color:#999999 ^_# ... ^_@/span^_#^_@a href=javascript:; onclick=onClickReadNode('NODE02034074');fn_statistics('Z354','null','null'); style='color:#999999;font-size:14px;text-decoration:underline;' ^_#전체 초록 보기^_@/a^_#