The study and proposition of methods for the general Chance-Constrained qualitative state planning problem

Abstract

The research project aims to study and propose methods of resolution for General Chance-Constrained State Planning Qualitative Problem with flexible scheduling and existence of obstacles. The problem has been established and is being studied by the team of Prof. Brian c. Williams in the Computer Science and Artificial Intelligence Laboratory (CSAIL) at the Massachusetts Institute of Technology (MIT), motivated by the concept of a personal transport system model (autonomous vehicle) of the Boeing Company. The solution of this problem seeks to raise the level of interaction between human operators and the vehicle as treating automatically the uncertainties inherent in the planning of missions in a real environment. Mathematical models based on mixed integer linear programming (PLIM) were proposed in papers published by Prof. Brian c. Williams. The project will be executed by Prof. Claudio Fabian Motta Toledo during the period from 8/4/2014 to 8/3/2015 in CSAIL, as acceptance letter issued by MIT. The solutions can be validated in simulators and in real situations, considering the applications in autonomous vehicles developed at CSAIL/MIT and unmanned aerial vehicle (UAV) developed by Embedded and Evolutionary Systems group (EES) of the ICMC/USP. (AU)