lynton dicks supervisor: karen bradshaw co-operative mapping and localization of autonomous robots

8
Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

Upload: nora-price

Post on 18-Jan-2018

218 views

Category:

Documents


0 download

DESCRIPTION

SLAM is a family of algorithms in robotics that is concerned with the creation and maintenance of maps of previously unknown and unchanging environments and then using those maps to determine the robot’s position in the environment Well researched and implemented for use with one robot CSLAM uses a team of robots to create and maintain maps Allow mapping larger areas in parallel, thereby making exploration more efficient SLAM with multiple robots is a relatively new field with many recent research papers INTRODUCTION

TRANSCRIPT

Page 1: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

Lynton Dicks

Supervisor: Karen Bradshaw

CO-OPERATIVE MAPPING ANDLOCALIZATION OF AUTONOMOUS ROBOTS

Page 2: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• Introduction

• Problem Statement and Goals

• History and Background

• Hardware

• Approach

PRESENTATION OVERVIEW

Page 3: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• SLAM is a family of algorithms in robotics that is concerned with the creation and maintenance of maps of previously unknown and unchanging environments and then using those maps to determine the robot’s position in the environment

• Well researched and implemented for use with one robot

• CSLAM uses a team of robots to create and maintain maps

• Allow mapping larger areas in parallel, thereby making exploration more efficient

• SLAM with multiple robots is a relatively new field with many recent research papers

INTRODUCTION

Page 4: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• Extend the SLAM framework that was implemented last year by Shaun Egan to allow multiple robots to map an area

• Implement the framework so that multiple robots are successfully able to implement a solution where their starting positions relative to each other are unknown

• Each robots role

• Centralization

• Aggregation

• Communication methods

PROBLEM STATEMENT AND GOALSProblems

Page 5: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• Generic Framework for both online and offline SLAM

• Implemented SLAM for use with one robot

• Generic Programming Framework to combine standard robotic operations with AI

• Abstracts away the details of interfacing and controlling robots

• Easy to implement new robot hardware classes to allow the framework to work with new hardware

HISTORY AND BACKGROUNDAutonomous Robotic Programming Framework – Leslie Luyt 2009

A Robotic Framework for use in Simultaneous Localization and Mapping Algorithms – Shaun Egan 2010

Page 6: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• Two Encoder Motors

• Two Ultrasonic Sensors

• A Bluetooth Controller – 10m range, ability to keep several connections alive at the same time

HARDWARE – FISCHERTECHNIK ROBOT

Page 7: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

HARDWARE: ADDONS

Motor Encoders Ultrasonic Sensors

Page 8: Lynton Dicks Supervisor: Karen Bradshaw CO-OPERATIVE MAPPING AND LOCALIZATION OF AUTONOMOUS ROBOTS

• By the end of the term relearn python for the online version of the framework and java for the offline version of the framework

• By the end of the term, finish studying both Leslie’s and Shaun’s frameworks and play around with using the framworks

• By the end of second term, complete research and select plan of attack for implementation and complete literature review

• And finally, implementation and testing as well as relevant extensions if time permits

APPROACH