Immuno-inspired heterogeneous mobile robotic systems

Abstract

Artificial immune systems are primarily inspired by human immune functions that exhibit self organization, learning, adaptation, recognition, robustness and scalability. This makes such systems a suitable choice to be experimented in robotic applications in unstructured environments. Out of many models of immune system, immune network model is applied in this research on mobile robotic systems with heterogeneous capabilities. The environment is translated into antigen according to the sensory arrangement of each robot. Antibody stimulation and suppression is implemented to navigate robots along with the idiotypic connections to establish inter-robot communications. These immune metaphors have been translated and simulated to exhibit a search and rescue operation in an obstacle filled arena. Experimental results show that search-robot can efficiently arbitrate the behaviors of wander, obstacle avoidance, target search and inter-robot communication. The rescue robot(s) can then navigate through the obstacles to effectively rescue the target(s).