Mobile Robot Navigation on Partially Known Maps using a Fast A* Algorithm Version

Mobile robot navigation in total or partially un-known environments is still an open problem. The path planningalgorithms lack completeness and/or performance. Thus, thereis the need for complete (i.e., the algorithm determines in finitetime either a solution or correctly reports that there is none) andperformance (i.e., with low computational complexity) orientedalgorithms which need to perform efficiently in real scenarios.In this paper, we evaluate the efficiency of two versionsof the A∗algorithm for mobile robot navigation inside indoorenvironments with the help of two software applications and thePioneer 2DX robot. We demonstrate that an improved versionof the A∗algorithm which we call thefastA∗algorithm canbe successfully used for indoor mobile robot navigation. Weevaluated the A∗algorithm first, by implementing the algorithmsin source code and by testing them on a simulator and second,by comparing two operation modes of thefastA∗algorithm w.r.t.path planning efficiency (i.e., completness) and performance (i.e.,time need to complete the path traversing) for indoor navigationwith the Pioneer 2DX robot. The results obtained with thefastA∗algorithm are promising and we think that this results canbe further improved by tweaking the algorithm and by usingan advanced sensor fusion approach (i.e., combine the inputs ofmultiple robot sensors) for better dealing with partially knownenvironments.