Year

2020

Season

Spring

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Computer and Information Sciences (MS)

Department

Computing

NACO controlled Corporate Body

University of North Florida. School of Computing

First Advisor

Dr. Ayan Dutta

Second Advisor

Dr. O Patrick Kreidl

Third Advisor

Dr. Anirban Ghosh

Department Chair

Dr. Sherif Elfayoumy

College Dean

Dr. William F Klostermeyer

Abstract

This research activity is primarily focused to obtain information from an environment with the help of a group of coordinated robots. Each robot is responsible to plan its path independently but the robots, as an overall system, have a common goal of maximum information collection. This domain of research is known as Multi-Robot Informative Path Planning (MIPP). MIPP is very motivating due to its challenging nature and numerous real-world applications. It has shown its presence from semiautomatic applications like robotic search and rescue to fully automatic applications like interplanetary missions.

We consider the NP-Hard problem of MIPP in an unknown environment having communication constraints and budget limitations in robots. We propose a novel approach that uses continuous region partitioning to efficiently divide an initially unknown environment among the robots based on the discovered obstacles. The research objective is to collect a higher amount of information and target a uniform work distribution among robots. Simulation results show that our proposed approach is successful in reducing the initial imbalance in the work distribution of the robots while ensuring close-to-reality spatial modeling within a reasonable amount of time.

Download

Included in

Robotics Commons

Share

COinS