Analysis of simultaneous use of extended Kalman filter and Q-learning algorithm in autonomous navigation inside dynamic environments

• Grant number: 18/05228-3
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): June 01, 2018
• Effective date (End): May 31, 2019
• Field of knowledge: Engineering - Electrical Engineering
• Principal researcher: Marcelo Nicoletti Franchin
• Grantee: Hugo Ferraz Sangaletti
• Home Institution: Faculdade de Engenharia (FE). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil

Abstract

Path planning and localization problems in autonomous navigation of mobile robots are often independently approached. However, both are essential to develop systems on autonomous robots field. This project intends to analyze the simultaneous approach of extended Kalman filter to solve not localization problem and q-learning algorithm to address path planning with unknown obstacle avoidance problem. The expected results are those that indicate performance improvement on both methods compared to each of them individually. (AU)