PATH PLANNING ALGORITHM, BASED ON USER-DEFINED MAXIMAL LOCALIZATION ERROR

Abstract

A model-based path planning algorithm will be presented in this paper. The whole model, like the algorithm, is divided into 2 parts. The 1st part begins with a map-building process over an unknown environment, which is based on the construction of the so-called Potential Field (PF) of the environment. In this part, the above mentioned PF will be created by three autonomous robots, equipped with US sensors, which will cooperate and update the global potential map on the remote host. The 2nd part begins with the calculation of the environment´s 2D mathematical model. The calculation is realized through the ythresholding of the global potential map. Furthermore, a landmark arrangement will be defined on this model. The yArtificial Error Field (AEF), which covers the entire workspace, will be calculated and the result will depend on the sensory system of the mobile robot/robots and on the landmark arrangement. Actually, the three-dimensional yAEF contains the ylocalization errors corresponding to each `x, y´ position of the work space´s free space. In respect of the yuser-defined maximal localization error (ϵ_\max), some ynavigation paths (NP) can be generated. These paths serve as the base for the calculation of the possible routes in form of directed and yweighted graph-map. The route with yminimal complexity between the start and the ydocking positions on this graph-map will be selected. Each point of the mobile robot´s exact trajectory must fit in the selected navigation path (NP). This maintains the allowed position error below the defined limit. The shape of the trajectory is calculated by the use of cubic B-splines.