Skip to main content

Communicating Branching Plans for Human-Agent Decision Making.

Porteous, J., Lindsay, A. and Charles, F., 2021. Communicating Branching Plans for Human-Agent Decision Making. In: ICAPS 2021 Workshop on Explainable AI Planning, 2-6 August 2021, virtual.

Full text available as:

[img]
Preview
PDF
communicating_branching_plans_.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

3MB

Official URL: https://icaps21.icaps-conference.org/workshops/

Abstract

Recent advances in visualisation technologies have opened up new possibilities for human-agent communication. For systems where agents use automated planning, visualisation of agent planned actions can play an important role in allowing human users to understand agent intent and to help decide when control can be delegated to the agent or when they need to be involved. We are interested in application areas where branched plans are required, due to the typical uncertainty experienced. Our focus is how best to communicate, using visualisation, the key information content of a branched plan. It is important that such visualisations communicate the complexity and variety of the possible executions captured in a branched plan, whilst also connecting to the practitioner's understanding of the problem. Thus we have developed an approach that: generates the complete branched plan, to be able to provide a full picture of its complexity; a mechanism to select a subset of diverse traces that characterise the possible executions; and an interface that uses 3D visualisation to communicate details of these characterising execution traces to practitioners. Using this interface, we conducted a study evaluating the impact of different modes of presentation on user understanding. Our results support our expectation that visualisation of characterising branched plan execution traces increases user understanding of agent intention and range of plan execution possibilities.

Item Type:Conference or Workshop Item (Paper)
Group:Faculty of Science & Technology
ID Code:36040
Deposited By: Symplectic RT2
Deposited On:21 Sep 2021 15:12
Last Modified:14 Mar 2022 14:29

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -