Solving 3D mazes with machine learning and humanoid robots

Vishnu K. Nath, Stephen E. Levinson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present a system that integrates computer vision with machine learning to enable a humanoid robot to accurately solve any 3 dimensional maze that has not been previously given to it. The robot can construct the optimum path policy based on previous iterations and does not require any specialized programming. The experimental setup includes a constructed 3D maze with a start and end point. The robot solves the maze using a red-colored ball. The robot can physically tilt the base of the maze with its hand so that the ball can roll into the desired region. The robot would begin tilting the maze only if a path exists between the start and the end point. If none exists, the robot would remain idle. This work is important and novel for a couple of reasons. The first is to determine if constant repetition of a task leads to gradually increasing performance and eventual mastery of a skill. If yes, can that skill be adapted to a generic ability ( Fleishman, 1972)? Also, can a robot's performance match or exceed that of an average human in the acquired ability?

Original languageEnglish (US)
Title of host publicationMachine Learning for Interactive Systems
Subtitle of host publicationBridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report
PublisherAI Access Foundation
Pages20-26
Number of pages7
ISBN (Electronic)9781577356684
StatePublished - Jan 1 2014
Event28th AAAI Conference on Artificial Intelligence, AAAI 2014 - Quebec City, Canada
Duration: Jul 28 2014 → …

Publication series

NameAAAI Workshop - Technical Report
VolumeWS-14-07

Other

Other28th AAAI Conference on Artificial Intelligence, AAAI 2014
CountryCanada
CityQuebec City
Period7/28/14 → …

Fingerprint

Learning systems
Robots
Computer vision

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Nath, V. K., & Levinson, S. E. (2014). Solving 3D mazes with machine learning and humanoid robots. In Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report (pp. 20-26). (AAAI Workshop - Technical Report; Vol. WS-14-07). AI Access Foundation.

Solving 3D mazes with machine learning and humanoid robots. / Nath, Vishnu K.; Levinson, Stephen E.

Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report. AI Access Foundation, 2014. p. 20-26 (AAAI Workshop - Technical Report; Vol. WS-14-07).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nath, VK & Levinson, SE 2014, Solving 3D mazes with machine learning and humanoid robots. in Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report. AAAI Workshop - Technical Report, vol. WS-14-07, AI Access Foundation, pp. 20-26, 28th AAAI Conference on Artificial Intelligence, AAAI 2014, Quebec City, Canada, 7/28/14.
Nath VK, Levinson SE. Solving 3D mazes with machine learning and humanoid robots. In Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report. AI Access Foundation. 2014. p. 20-26. (AAAI Workshop - Technical Report).
Nath, Vishnu K. ; Levinson, Stephen E. / Solving 3D mazes with machine learning and humanoid robots. Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report. AI Access Foundation, 2014. pp. 20-26 (AAAI Workshop - Technical Report).
@inproceedings{cb96aa1fbb484085aa693351c85b44c6,
title = "Solving 3D mazes with machine learning and humanoid robots",
abstract = "In this paper, we present a system that integrates computer vision with machine learning to enable a humanoid robot to accurately solve any 3 dimensional maze that has not been previously given to it. The robot can construct the optimum path policy based on previous iterations and does not require any specialized programming. The experimental setup includes a constructed 3D maze with a start and end point. The robot solves the maze using a red-colored ball. The robot can physically tilt the base of the maze with its hand so that the ball can roll into the desired region. The robot would begin tilting the maze only if a path exists between the start and the end point. If none exists, the robot would remain idle. This work is important and novel for a couple of reasons. The first is to determine if constant repetition of a task leads to gradually increasing performance and eventual mastery of a skill. If yes, can that skill be adapted to a generic ability ( Fleishman, 1972)? Also, can a robot's performance match or exceed that of an average human in the acquired ability?",
author = "Nath, {Vishnu K.} and Levinson, {Stephen E.}",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
series = "AAAI Workshop - Technical Report",
publisher = "AI Access Foundation",
pages = "20--26",
booktitle = "Machine Learning for Interactive Systems",

}

TY - GEN

T1 - Solving 3D mazes with machine learning and humanoid robots

AU - Nath, Vishnu K.

AU - Levinson, Stephen E.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this paper, we present a system that integrates computer vision with machine learning to enable a humanoid robot to accurately solve any 3 dimensional maze that has not been previously given to it. The robot can construct the optimum path policy based on previous iterations and does not require any specialized programming. The experimental setup includes a constructed 3D maze with a start and end point. The robot solves the maze using a red-colored ball. The robot can physically tilt the base of the maze with its hand so that the ball can roll into the desired region. The robot would begin tilting the maze only if a path exists between the start and the end point. If none exists, the robot would remain idle. This work is important and novel for a couple of reasons. The first is to determine if constant repetition of a task leads to gradually increasing performance and eventual mastery of a skill. If yes, can that skill be adapted to a generic ability ( Fleishman, 1972)? Also, can a robot's performance match or exceed that of an average human in the acquired ability?

AB - In this paper, we present a system that integrates computer vision with machine learning to enable a humanoid robot to accurately solve any 3 dimensional maze that has not been previously given to it. The robot can construct the optimum path policy based on previous iterations and does not require any specialized programming. The experimental setup includes a constructed 3D maze with a start and end point. The robot solves the maze using a red-colored ball. The robot can physically tilt the base of the maze with its hand so that the ball can roll into the desired region. The robot would begin tilting the maze only if a path exists between the start and the end point. If none exists, the robot would remain idle. This work is important and novel for a couple of reasons. The first is to determine if constant repetition of a task leads to gradually increasing performance and eventual mastery of a skill. If yes, can that skill be adapted to a generic ability ( Fleishman, 1972)? Also, can a robot's performance match or exceed that of an average human in the acquired ability?

UR - http://www.scopus.com/inward/record.url?scp=84974824342&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84974824342&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84974824342

T3 - AAAI Workshop - Technical Report

SP - 20

EP - 26

BT - Machine Learning for Interactive Systems

PB - AI Access Foundation

ER -