Development of evolutionary and self-assembling robot-organisms

Abstract

Symbiotic robotics is a discipline within collective robotics that is concerned with artificial multi-cellular robot-organisms that define their morphological structure by aggregation through self-assembling and they are also able to disaggregate afterwards. This contribution is concerned to the description of evolutionary and cognitive principles that governs such a symbiotic cycle to build artificial organisms of different forms and operate with them. The evolutionary approach starts with a artificial genome, will be continued by the insertion of different types of regulative cycles, and ends up in an embryogenetic formed body. Hereby there is differentiation between the genetic based learning and the fitness based learning. Further there are dominant differences between multi-cellular organism and structured cooperative aggregations of swarm members. The cognitive approach is focused on cognitive maps, on cognitive sensor data fusion and finally to the definition of information that governs the the process of organism formation and body survival in a given environment. This more engineering oriented approach is used to build all HW-components and all kinds of embedded “operating systems” to control and to operate symbiotic robot organisms.