A Software Suite for the Control and the Monitoring of Adaptive Robotic Ecologies

Adaptive robotic ecologies are networks of hetero- geneous robotic devices (sensors, actuators, automated appli- ances) pervasively embedded in everyday environments, where they learn to cooperate towards the achievement of complex tasks. While their flexibility makes them an increasingly popu- lar way to improve a system's reliability, scalability, robustness and autonomy, their effective realisation demands integrated control and software solutions for the specification, integration and management of their highly heterogeneous and computa- tional constrained components. In this extended abstract we briefly illustrate the characteristic requirements dictated by robotic ecologies, discuss our experience in developing adaptive robotic ecologies, and provide an overview of the specific solutions developed as part of the EU FP7 RUBICON Project. I. INTRODUCTION Robotic ecologies are an emerging paradigm, which crosses the border between the fields of robotics, sensor net- works, and ambient intelligence (AmI). Central to the robotic ecology concept is that complex tasks are not performed by a single, very capable robot (e.g., a humanoid robot butler), instead they are performed through the collaboration and cooperation of many networked robotic devices performing several steps in a coordinated and goal oriented fashion while also exchanging sensor data and other useful information in the process. Building smart spaces in this way reduces application complexity and costs, and enhances the individ- ual values of the devices involved, by enabling new services that cannot be performed by any device by itself. Consider for instance the case of an ecology-supported robot vacuum cleaner that avoids cleaning when any of the inhabitants are home after receiving information from the home alarm system, or of a robot informing an elderly person living alone that she has forgotten to switch off the stove, after receiving a signal from a wireless sensor installed in the kitchen. One of the key strengths of such an approach is the possibility of using alternative means to accomplish application goals when multiple courses of action are available. For instance, a robot may be able to localise itself with the help of an environmental camera or through the use of an on-board laser sensor, if the network connection to the camera is disrupted or if the light is not sufficient for the camera to track the location of the robot.