Decentralised Monte Carlo Tree Search for Active Perception

Best, G; Cliff, O; Patten, T; Mettu, R; Fitch, R

Decentralised Monte Carlo Tree Search for Active Perception

Best, G Cliff, O Patten, T Mettu, R Fitch, R

Permalink

Publisher:: International Workshop on the Algorithmic Foundations of Robotics
Publication Type:: Conference Proceeding
Citation:: Workshop on the Algorithmic Foundations of Robotics (WAFR), 2016
Issue Date:: 2016-12-18

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download Published versionAdobe PDF (2.68 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Best, G	en_US
dc.contributor.author	Cliff, O	en_US
dc.contributor.author	Patten, T	en_US
dc.contributor.author	Mettu, R	en_US
dc.contributor.author	Fitch, R https://orcid.org/0000-0003-2215-5188	en_US
dc.date.issued	2016-12-18	en_US
dc.identifier.citation	Workshop on the Algorithmic Foundations of Robotics (WAFR), 2016	en_US
dc.identifier.uri	http://hdl.handle.net/10453/97186
dc.description.abstract	We propose a decentralised variant of Monte Carlo tree search (MCTS) that is suitable for a variety of tasks in multi-robot active perception. Our algorithm allows each robot to optimise its own individual action space by maintaining a probability distribution over plans in the joint-action space. Robots periodically communicate a compressed form of these search trees, which are used to update the locally-stored joint distributions using an optimisation approach inspired by variational methods. Our method admits any objective function defined over robot actions, assumes intermittent communication, and is anytime. We extend the analysis of the standard MCTS for our algorithm and characterise asymptotic convergence under reasonable assumptions. We evaluate the practical performance of our method for generalised team orienteering and active object recognition using real data, and show that it compares favourably to centralised MCTS even with severely degraded communication. These examples support the relevance of our algorithm for real-world active perception with multi-robot systems.	en_US
dc.publisher	International Workshop on the Algorithmic Foundations of Robotics	en_US
dc.relation.ispartof	Workshop on the Algorithmic Foundations of Robotics (WAFR)	en_US
dc.title	Decentralised Monte Carlo Tree Search for Active Perception	en_US
dc.type	Conference Proceeding
utslib.location	San Fancisco, USA	en_US
utslib.location.activity	San Francisco, USA	en_US
utslib.for	080101 Adaptive Agents and Intelligent Robotics	en_US
utslib.for	091303 Autonomous Vehicles	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - CAS - Centre for Autonomous Systems
utslib.copyright.status	open_access
pubs.consider-herdc	true	en_US
pubs.place-of-publication	San Fancisco, USA	en_US

Abstract:

We propose a decentralised variant of Monte Carlo tree search (MCTS) that is suitable for a variety of tasks in multi-robot active perception. Our algorithm allows each robot to optimise its own individual action space by maintaining a probability distribution over plans in the joint-action space. Robots periodically communicate a compressed form of these search trees, which are used to update the locally-stored joint distributions using an optimisation approach inspired by variational methods. Our method admits any objective function defined over robot actions, assumes intermittent communication, and is anytime. We extend the analysis of the standard MCTS for our algorithm and characterise asymptotic convergence under reasonable assumptions. We evaluate the practical performance of our method for generalised team orienteering and active object recognition using real data, and show that it compares favourably to centralised MCTS even with severely degraded communication. These examples support the relevance of our algorithm for real-world active perception with multi-robot systems.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/97186