Tel: ++ 213 4 85 54 54. 213 4 85 83 75 Fax: +t AbstractAlthough path planning in known, partially unknown or completely unknown environment has been-extensively studied by scientists in general and researchers in particular, the problem is far from being completely solved. In this paper a new method of global path planning for an autonomous automated guided vehicle is presented. The strategy of the developed method is based on two parts: partition of free working space according to obstacles models, and calculation of a set of points within the free working space according to the dimensions of the mobile robot. This set of points constitutes the eventual passage points for the AGV during its navigation between starting and goal points. Once starting and goal points are given, a visibility graph is constructed and an optimal path is searched. The new approach in this method is that the mobile robot is not punctual and the obstacles are not expanded. The validity and the efficiency of the developed algorithm are confirmed by the results obtained by simulation.

No abstract

No abstract

No abstract

This paper deals with the design and implementation of an improved localization method. The study presented in this paper focuses on the displacement and heading errors. The technique proposed improves considerably on dead reckoning methods by reducing the systematic and non‐systematic errors. Detection of the floor irregularities is the main feature of the designed method in the correction of displacement errors. However, for internal errors correction, a trailing wheel is used. Following that, a predefined trajectory is provided by the control of the trailing wheel deviation angle.