Top Banner
Maze Solving with an AIBO Bernard Maassen, Hans Kuipers, Max Waaijers & Andrew Koster 2005
20

Maze Solving with an AIBO

Jan 15, 2016

Download

Documents

Elma

Maze Solving with an AIBO. Bernard Maassen, Hans Kuipers, Max Waaijers & Andrew Koster 2005. Introduction. Problem: Maze Navigation Performed Research: Theseus found his way out of the Labyrinth Using IR for Maze Navigation, CMU www.cs.cmu.edu/~tekkotsu/media/pgss_2004_paper.pdf - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Maze Solving with an AIBO

Maze Solving with an AIBO

Bernard Maassen, Hans Kuipers, Max Waaijers &

Andrew Koster2005

Page 2: Maze Solving with an AIBO

Introduction Problem:

Maze Navigation Performed Research:

Theseus found his way out of the Labyrinth

Using IR for Maze Navigation, CMU www.cs.cmu.edu/~tekkotsu/media/pgss_2004_paper.pd

f

Many competitions Many geometric algorithms for edge

recognition and L-shapes Computational Geometry: Algorithms and Applications

Page 3: Maze Solving with an AIBO

Introduction

Why? Many aspects, like:

Vision World modeling Self localization

Socially relevant: Rescue robots need to navigate maze-like

environments

Page 4: Maze Solving with an AIBO

Problem Description

Maze Navigation 3 Problems:

Landmark detection Map construction AIBO uses map to solve maze

Collision prevention/detection

Page 5: Maze Solving with an AIBO

Landmark detection Edge detection using scanlines Only look below ‘horizon’

Page 6: Maze Solving with an AIBO

Landmark detection

Possible Intersections

Page 7: Maze Solving with an AIBO

Landmark detection

Possible Intersections

Page 8: Maze Solving with an AIBO

Landmark detection

Possible Intersections

Page 9: Maze Solving with an AIBO

Map construction

On each landmark update graph Remember

Type of landmark Current location

Use depth first search to explore Initially simple mazes, later on

more complex ones.

Page 10: Maze Solving with an AIBO

Complications

Maze contains loops Need to use distances as well as type

of intersection Different mazes

Curves 5-way intersections Other

Page 11: Maze Solving with an AIBO

AIBO uses map to solve maze

Random initial location in maze Use Bayesian filters to determine

most likely location Find exit

Page 12: Maze Solving with an AIBO

Possible problems

Missing landmarks Walking into walls Odometry not reliable

Page 13: Maze Solving with an AIBO

Backup plan

Reinforcement learning with joystick Simpler Uses joystick to train Uses odometry data in stead of vision

Page 14: Maze Solving with an AIBO

Milestone 1

Joystick walking Already in Tekkotsu Integrate into DARPA modules Collect odometry data

Page 15: Maze Solving with an AIBO

Milestone 2a

Control point if 2b is feasible If not extend Joystick module

Page 16: Maze Solving with an AIBO

Milestone 2b

Landmark detection Vision module Edge detection + scanlines Distinguish intersections

Page 17: Maze Solving with an AIBO

Milestone 3

Map construction Model world as topological map Self localization Walking through the maze

Page 18: Maze Solving with an AIBO

Milestone 4

Maze Solving Find place in world Use map to find path to exit Exit maze

Page 19: Maze Solving with an AIBO

Milestones

Milestone 1: 30-9 Milestone 2a: 19-10 Milestone 2b: 26-10 Milestone 3: 1-11 Milestone 4: 11-11

Page 20: Maze Solving with an AIBO

Vragen?