In this paper the problem of the autonomous navigation of a cooperative robot formation is studied. Robots navigate in an unknown environment, encumbered by obstacles which have a circular shape or which can be modele...
详细信息
ISBN:
(纸本)9783642108167
In this paper the problem of the autonomous navigation of a cooperative robot formation is studied. Robots navigate in an unknown environment, encumbered by obstacles which have a circular shape or which can be modeled by polygons. Robots must move between an initial known position and a final known position. While going ahead robots discover the obstacles which block their passage and must avoid the collision and go to their final destination. This paper tackles two problems: 1. path planning, which is described as an optimal control problem under constraints (robot physical limits, collision avoidance). 2. localization, in order to plan its trajectory, the robot has to know its localization in the environment : a mono-landmarkalgorithm of localization in 2D is proposed. This mono-landmarkalgorithm tracks a landmark during the navigation and uses it to localize the robot in an algebraic framework.
暂无评论