Neural-Based Navigation Approach for a Bi-Steerable Mobile Robot
Ouahiba Azouaoui, Ouadah Noureddine, Aouana Salem, Chabi Djeffer · 2008
In the implemented neural-based navigation, the two intelligent behaviors necessary to the navigation, are acquired by learning using GBP algorithm. They enable Robucar to be more autonomous and intelligent in partially structured environments. Nevertheless, there are a number of issues that need to be further investigated. At first, the Robucar must be endowed with one or several actions to come back to eliminate a stop in a dead zone situation. Another interesting alternative is the use of a better localization not only based on odometry but by fusing data of other sensors such as laser scanner.