RRL-MR: Mobile robot navigation based on relational reinforcement learning

• Grant number: 09/04489-9
• Support Opportunities: Scholarships in Brazil - Master
• Start date: August 01, 2009
• End date: May 31, 2011
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Anna Helena Reali Costa
• Grantee: Tiago Matos
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 08/03995-5 - Logprob: probabilistic logic --- foundations and computational applications, AP.TEM

Abstract

Solutions for autonomous navigation of mobile robots typically do not consider solutions previously given for similar problems, preventing the robot to improve its performance from past experiences in similar tasks. The objective of this research project is to propose, develop and evaluate a method based on relational reinforcement learning that generates navigation abstract policies representations from places visited by the robot during the execution of its tasks. The relational abstraction generated will allow the generalization of navigation plans already executed, as well as the transfer of policies to other similar tasks.

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

MATOS, Tiago. Using stochastic abstract policies in robotic navigation.. 2011. Master's Dissertation - Universidade de São Paulo (USP). Escola Politécnica (EP/BC) São Paulo.