"Network Localization From Relative Measurements With Application to Robotics"

Luca Carlone, Politecnico di Torino, Italy

April 8th (Friday), 10:30am
ENGR II Room 2217

Several fields of application ranging from sensor networks to mobile robotics require the agents to have accurate knowledge of their positions in a common reference frame. When no direct method is viable (e.g., GPS) the network configuration has to be estimated from relative measurements among nodes. In this talk we discuss two case studies in which, depending on the available relative measurements, the network localization problem is rephrased in terms of nonlinear optimization and solved with suitable mathematical tools. The first part of the talk deals with network localization from relative distance measurements and presents two distributed techniques for solving the problem in a decentralized fashion. The second part discusses the case of relative pose measurements and proposes an approximate linear estimation framework for network localization, which is amenable for distributed implementation. As a conclusion, experimental results in mobile robotics applications are presented, underlining open issues and future research lines.

About Luca Carlone:

Luca Carlone received his bachelor in Mechatronic Engineering in 2006 from Politecnico di Torino. In 2008 he achieved his MS degree in Mechatronic Engineering from the same university and MS degree in Automation Engineering from Politecnico di Milano. At present he is a PhD student at Politecnico di Torino. His current research interests involve localization, mapping and SLAM algorithms for mobile robots navigation.

Hosted by: Professor Francesco Bullo