A learning algorithmic scheme for collision free navigation

A learning algorithmic scheme for collision-free navigation is presented. This scheme determines the minimum collision-free navigation path of an autonomous platform by using a trial and error process and past navigation knowledge and by extracting the current information from the surrounding environment. This scheme is related to the Odysseus autonomous navigation platform. However, it could be made into a general-purpose scheme with simple modifications. Simulated results for an illustrative example in a graphical representation are provided.<
>