October 28, 2010
Schools of Engineering, Coimbatore & Amritapuri
Prof. Stefano Panzieri is the Director of the Automatic Laboratory of the Department of Information & Automation, at the University of “Roma-Tre” in Italy.
Committed to the development and continued exploration of robotic technology, and having published several papers on mobile and industrial robots, Prof. Panzieri recently spoke at Amrita Vishwa Vidyapeetham, elaborating on current issues challenging Mobile Robotics.
The lecture was titled Spatially Structured Genetic Algorithm over Complex Networks for Mobile Robot Localization.
“One of the most important problems in Mobile Robotics is to realize the complete robot’s autonomy,” he shared. “In order to achieve this goal, several tasks have to be accomplished. Among them, the robot’s ability to localize itself turns out to be critical.”
Finding an effective solution for the localization problem has long been sought by the research community. Previously, mobile robot localization problems were solved by applying evolutionary computing techniques like genetic algorithms (GA), particle swarm optimization (PSO) and ants system (AS).
Introducing a new approach to the localization problem, a specialization of the genetic algorithm, Prof. Panzieri remarked, “The novelty of this approach is to take advantage of the complex networks theory for the spatial deployment of population to more quickly find optimal solutions.”
“In fact, modeling the search space with complex networks and exploiting their typical connectivity properties, results in a more effective exploration of space,” he explained.
Panzieri went on to suggest the possibility of using either a Space Structured GA or a Predictor/Corrector Bayesian Filter GA to locate a mobile robot when there is no prior knowledge of its pose or to find a kidnapped robot.
“Evaluating data provided by the sensors determines which GA should be used in the location process,” he elaborated.
Other topics addressed by Dr. Panzieri included the advantages of applying the Evolutionary Cellular Automata Theory to a genetic population in the form of a complex network system and how the GA is used in a multi-robot environment.
Exposing students to cutting- edge advancements in robotic technology, Panzieri’s lecture on localization and Mobile Robotics was an inspiration to Amrita M. Tech. students, complementing their endeavors, as they strive to harness the power of modern technology for the betterment of society.
