Building Large-Scale Occupancy Maps using an Infinite Mixture of Gaussian Process Experts

Kim, SoohwanKim, Jonghyuk2015-12-10December 39780980740431http://hdl.handle.net/1885/65435This paper proposes a novel method of occupancy map building for large-scale applications. Although Gaussian processes have been successfully applied to occupancy map building, it suffers from high computational complexity of O(n3), where n is the numberAuthor/s retain copyrightKeywords: Dirichlet process mixture; Divide-and-conquer approach; Gaussian Processes; Large-scale applications; Map Building; Mixture of Gaussians; Occupancy maps; Training data; Gaussian distribution; Gaussian noise (electronic); Mixtures; RoboticsBuilding Large-Scale Occupancy Maps using an Infinite Mixture of Gaussian Process Experts20122016-02-24