Runtime self-adaptation in a component-based robotic framework

Abstract

The development and maintenance of software for robotic systems is a hard task due to the complexity inherent in these systems. Besides, the resulting applications have to deal with limited resources and variable execution conditions that must be considered in order to keep an acceptable system performance. To address both problems we have integrated a set of dynamic adaptation policies inside CoolBOT, a component oriented framework for programming robotic systems. CoolBOT contributes to reduce the programming effort, promoting robustness and code reuse, while the adaptation scheme provides a dynamic modulation of system performance to meet available computational resources at runtime. In this paper we also present two demonstrators that outline the benefits of using the proposed approach in the development of real robotic applications.