Visual control of autonomous mobile robot based on self‐organizing model for pattern learning

Publisher Website
Google Scholar
Abstract
This article proposes a self‐organizing model for pattern learning together with an application to an autonomous mobile robot system. The self‐organizing model consists of a processing rule prescribed and a memory part being blank at the initial stage. To an input signal, the model searches for a similar signal in the memory, and recalls its related information. If the information accompanied with the input signal differs from the recalled information, the model adds the new information to the memory. It influences the subsequent operations. Thus, the model constructs successively a data‐base in a self‐organizing way. This model can universally learn and reproduce any pattern of input‐output response desired. Two principal functions in autonomous movement, i.e., position identification and obstacle avoiding movement were realized based on the self‐organizing model. Furthermore, a camera type autonomous mobile robot system for indoor was made up. The size of the robot is about 0.7 × 0.7 × 0.7 m, and the weight is about 30 kg. The speed of movement is less than 3 km/h. A small computer that has a 16 bit microprocessor and a 1Mbyte RAM controls the motion of the robot with an extendedClanguage.

This publication has 4 references indexed in Scilit: