Safe path planning and replanning with unmapped objects detection

This paper deals with the definition of safe tasks for displacement missions of a nonholonomic mobile robot in an outdoor environment. The goal of this study is to plan actions (paths following) taking into account new exteroceptive sensor measurements when they appear. There are two kinds of exteroceptive sensor measurements: telemetric measurements and absolute localization measurements. Absolute localization measurements (in combination with odometric measurements) are managed by an extended Kalman filtering. Telemetric measurements are given to a credibilist Kalman filter algorithm which manages appearance, disappearance, reassociation and tracks objects. Thus, a path planning method dealing with uncertainties is proposed, where both uncertainties in mobile robot localization and new objects localization are taken into account. The safeness of the proposed method is due to the mixing between the planning phase and the navigation phase.