*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Abstract: We describe a novel approach to mapping an unknown environment based on sparse sampling of paths with minimalist sensor data. The problem is inspired by mapping the geometry of a building floorplan via “uncooperative sensing” — using data from such as camera feeds and other tracking-capable sensors. Unique challenges include having no knowledge of sensor placement, coverage or parameters. The methods used are, at first, topological, to build a combinatorial model with the appropriate topology. This model is then augmented to include a weak form of geometric information, and optimization techniques are used to approximate domain geometry. The dual problem is also discussed, where we leverage a topological map of the environment, built from “uncooperative” sensors, to
coarsely localize an agent in a complex environment.
Bio: Alberto Speranzon is currently a Technical Fellow at Honeywell Aerospace, Advanced Technology Group, based in Minneapolis, MN, USA. Alberto received the “Laurea” degree in computer engineering from University of Padova, Italy in 2000, and a Ph.D. in automatic control from the School of Electrical Engineering, Royal Institute of Technology (KTH), Stockholm, Sweden in 2006. At Honeywell, Alberto is working on autonomous systems, machine learning and systems of systems, leading such research areas as program manager and principal investigator. He is currently also Co-PI of a DARPA sponsored project on new optimization methods based on homological algebra and sheaf theory. Alberto received the Best Poster Award at the “Hybrid Systems: Computation and Control (HSCC) conference” in 2017 on novel abstraction and compositional methods based on sheaf and category theory, in collaboration with MIT. Before joining Honeywell, he was a research scientist at United Technologies Research Center (UTRC), in East Hartford, CT, USA where, since 2008, he covered various roles of increasing responsibility. At UTRC, Alberto served as project manager and principal investigator of various DARPA sponsored projects on novel methods for navigation in GPS degraded and denied and new mathematics for swarm intelligence where methods, cutting across robust adaptive filtering, artificial intelligence, graph theory and algebraic topology were combined. His research interests are mainly in the area of autonomy/robotics, machine learning and cyber–physical systems, and in particular on distributed control and optimization, analysis and design of complex systems and abstraction methods for intelligent and learning systems. At UTRC he received the Outstanding Achievement Award in 2009 from UTRC, highest award given by UTRC. He is a Senior Member of the IEEE, an associate editor of the IEEE Transactions on Control Systems Technology Journal and part of the Technical Program Committee of various conferences in the area of cyber–physical systems, robotics and networked control systems.
