I-Corps: RoboSPRITE: The Social, Swarming, Educational Robot

Georgia Tech Research Corporation, Atlanta GA

Investigators

Abstract

This project addresses robotics and robotic swarm collaboration. This I-Corps team will explore commercialization of systems of multi-robots that will be transitioned onto swarming, social, educational, programmable, robotic toys. The resulting product, the RoboSPRITE, will form and evolve relationships with other robots, which in turn manifest themselves through different types of coordinated interaction maneuvers. The envisioned project, the RoboSPRITE, has never existed before outside the research lab. As a toy, the robots remember each other, form 'friendships' over time, and adaptively change their behaviors and interaction rules based on the history of their interactions. These interaction rules are moreover programmable, turning the toys into vehicles for learning robotics, coding, mathematics, and even biology since different types of biological herding, hunting, and foraging strategies can be encoded. As such, broader impact can be found in the educational domain, where children will be exposed to STEM concepts in a playful setting. The intellectual driver behind the proposed educational, robotic toy is research that focuses on the design of algorithms for large swarms of cooperating robots, with particular focus on mobility algorithms. Through this research, biologically inspired coordination strategies have been developed for making robot teams solve problems based solely on local interaction rules. These interaction rules are designed in such a way that global properties emerge in the team in a provable manner. This combination of robotics, biological principles, and control theory, has previously allowed for the development of particular types of algorithms, such as formation control and coverage algorithms. Through the proposed work, they will be taken even further to allow for the development of a novel set of algorithms for the creation of programmable, customizable interaction behaviors, and consequently, the development of robots that can interact with each other through such behaviors.