CSCW12-SeriousMRG-1_mdf_kxp_jog_jef-chAcc - FischerEtAl

Serious Mixed Reality Games

Abstract

We argue for serious

mixed reality

games as an

approach to study and design for chal

lenging real

word

scenarios, such as disaster response, for which

empirical data is difficult to obtain and efficacy of

purely computational simulations is questionable. We

outline our approach and give an example of a serious

mixed reality

game, which all

ows the study and

analysis of human

agent interaction in a disaster

response scenario on the ground.

Author Keywords

pervasive games,

mixed reality

games, agents

ACM Classification Keywords

H.5.m. Information interfaces and presentation:

HCI

General Te

rms

Design, Human Factors

Introduction

We suggest serious mixed reality games as a platform

to explore scenarios in the real world that are typically

hard to study in realistic settings, such as disaster

response. Such scenarios typically include groups of

human, computational and embodied agents (such as

UAVs) coordinating a response to a disaster such as an

earthquake, a flooding, a terrorist attack or an

epidemic outbreak. Responders may have to coordinate

and perform their operations potentially under c

ritical

CSCW’12

February 11

–

15, 2012, Seattle, Washington, USA

ACM 978

4503

1051

/12/02.

Joel E. Fischer

jef@cs.nott.ac.uk

Martin Flintham

mdf@cs.nott.ac.uk

Dominic Price

Dominic.Price@nottingham.ac.uk

James Goulding

James.Goulding@nottingham.ac.uk

Nadi

a Pantidi

kxp@cs.nott.ac.uk

Tom Rodden

1,2

Tom.Rodden@nottingham.ac.uk

The

Mixed Reality Laboratory

Scho

ol of Computer Science

University of Nottingham

Wollaton Road

NG8 1BB, Nottingham, UK

Horizon Research Institute

University of Nottingham

Triumph Road

Nottingham, UK, NG72TU

temporal and spatial constraints, with limited resources

and personnel, where failure may costs human lives.

The ORCHID project investigates the potential of

human

agent collectives in a disaster response scenario

[4]. In particular, the project f

ocuses on how autonomy

can be flexibly negotiated between agents, how agile

teaming can assist arrangement of human

agent

response teams, how incentives to respond might be

provided and sustained, and how the provenance and

trustworthiness of the informati

on gathered by large

numbers of humans, sensors, and agents may be

visualized and evaluated.

In this position paper, we argue that serious mixed

reality games provide currently underexplored

opportunities for the pervasive/

mixed reality

gaming

community t

o contribute to research in complex real

world scenarios,

such as natural disasters

By situating

both agents and participants in real world

environments, and presenting them with compelling

game

scenarios

analogous to disasters

, it is possible to

study co

ordination, interaction and patterns of

communication amongst actors

while

also having

greater confidence in the efficacy of behavioural

observations.

In so doing, we hope to fertilize cross

disciplinary research in Human

Computer Interaction,

Machine Lear

ning, Multi

Agent Systems and Robotics.

After reviewing some related work and motivating our

approach, we outline

iZombie!

, an exemplary mixed

reality game (MRG), and show how the research aims

stated above may be tackled from various disciplines.

Related w

ork and approach

Disaster and emergency situations epitomize the non

linearity of human events. Not only are computational

simulations of such scenarios extremely difficult to

construct, but the veracity of their results is almost

impossible to verify [

]

. Particularly, these do not

allow us to study human

agent interaction in a realistic

disaster setting. The impact of emotional response

likely in a disaster situation, such as stress, fear, or

panic remains understudied

or misconceived [11

]

approaches

relying on simulation

However, inspiration can be drawn from robotics and

adjacent fields that focus on computational intelligence,

which have pushed the state

the

art forward by

relying on competitions set out to solve problems in the

real world. Not

able examples include how the team that

won the DARPA Red Balloon Challenge drew on a

crowd

sourcing approach [9], and how the RoboCup

Rescue initiative evaluates its contributions in real

world test settings

that integrate

intelligent agents into

disaster

management situations

[6]. Furthermore, the

growing popularity in crowd

sourcing information in

crisis situations and sharing ‘hard’ data (from sensors)

and ‘soft’ data (from humans) on portals such as

Ushahidi or Pachube open up ways to fuse and analyze

that data in novel ways, for example to infer intent [5].

We propose that in addition to such real

world problem

solving competitions and crowd

sourcing initiatives,

mixed reality

gaming provides an approach to study the

outlined research aims includin

g emotional issues of

human

agent interaction in disaster settings.

particular,

MRGs

(or pervasive games) provide a setting

under which people have been shown to suspend

disbelief [8], and share some

further

key

characteristics

with disaster

response

enarios

, of which we highlight

three

by drawing on related work.

First

, MRGs and disaster response scenarios

bridge the

physical and the digital [1].

Over the past decade we

have created a series of MRGs that provide such hybrid

spaces to enable playful a

nd artistic performances and

publicly accessible exp

eriences.

For example,

Can You

See Me Now? [

, 1

]

pitched online players in a virtual

city against runners in the streets of a real city in a

fast

paced chase game.

Secondly, both MRG and disaster response are highly

chestrated activities. For example, the online

components of our

mixed reality

games are somewhat

comparable to the control room, with the games

striving to provide online players with a rich, accurate

and real

time view of the physical world through the

mixed reality

city in order to support meaningful game

play [2]. All of these games also require considerable

behind the scenes support,

characterized by activities

such as

monitor

ing

for failures,

managing access to

physical resources via

dedicated inte

rfaces.

Thirdly, in both MRGs and in disaster response people

on the ground work with people online to solve a

common problem

[3]

. For exam

ple, Uncle Roy All

Around You [1

] involved online and physical players

collaborating in order to achieve a common goa

–

finding the mysterious Uncle Roy in the back streets of

London

–

but also questioning when and how much to

trust the often contrary instructions given by

anonymous players on the Internet, or fictitious

instructions from the narrative voice of the game

ur approach

rries

the prefix

‘

serious

’

despite

that

he popu

larity of so

called

‘

serious games

’

has perhaps

waned

owever

we believe that a

MRG

that is

not

inherently or specifically

‘

serious

’

could provide

sufficient analog

s with a truly serious re

world

scenario as to provide a useful test bed for

understanding situational awareness, trust and

interaction with embodied agents, and

mixed reality

collaboration in an adrenaline fueled situation.

iZombie!

–

A serious mixed reality game

iZombie!

gamif

ies a potential viral epidemic outbreak

by means of a MRG based on th

e Zombie theme.

Players in the ‘

game’

have

to survive an outbreak of a

Zombie ‘virus’,

while

volunteers

act as

infected

Zombies whose motivation is to attack and kill non

infected players

. The Zombie ‘virus’ is spread through

sustained exposure to an airborne strain present in a

locality

and

through direct physical contact with a

Zombie where the player manages to ‘fend off’ a

Zombie attack but sustains an ‘infection’. Whilst the

scenario

of a Zombie epidemic is clearly fictional, we

believe that it forms the basis for a compelling game to

play whilst maintaining an analogue to real world

scenarios such as a virus outbreak.

Crucially, players in the ‘game’ have to collaborate in

‘agile team

s’ to achieve situational awareness to reach

their objectives, by instructing agents and by being

instructed by agents. To escape the Zombies and reach

safety players must coordinate with each other and

teams orchestrating rescue attempts (remote players

ho themselves are not in danger of infection but

whose goal is to rescue the players ‘on the ground’, as

the Army and other emergency rescue teams would in

a real

life situation). We also propose the introduction

of embodied agents such as UAVs or ground b

ased

robots that would be able to coordinate rescue

attempts through such tasks as aerial mappings or

leading players away from danger zones to safety.

iZombie!

seeks to leverage the viral popularity of other

big urban games, including 2.8 Hours Later, and

scavenger hunts such as Encounter which attract

hundreds

[7]

of publi

c players, in order to create a

mixed reality

game that both engages the public but

that also provides a meaningful research environment.

Conclusions and outlook

MRGs let us explore im

portant interactional issues of

human agent

collectives under stress

inducing realistic

scenarios. How do humans interact with agents in a

way that flexible autonomy can be achieved, monitored

and amended in a context sensitive manner? Secondly,

how can hu

mans and agents be supported in their need

for agile teaming as and when needed to achieve a

particular task? Thirdly, how can they be motivated to

complete their missions in spite of conflicting interest

and self

interest and so on? Finally, how is

trustw

orthiness of the information of various agents

(computational and human) evaluated by players in the

real

world, how can effective sharing of it be supported

and how can it be presented in an understandable way?

In the future, we aim to draw on our extens

ive

experience in

mixed reality

game

and experience

development

and draw on existing relationships with

artists and designers to refine our game design and

work towards a public performance of our first serious

mixed reality

game by mid 2012. Collaboratio

ns and

workshops within the Orchid project will further shape

the design in terms of the research questions and

development challenges to investigate future human

agent collectives for disaster response scenarios.

Acknowledgements

This work is supported by

EPSRC grant

EP/I011587/1.

References

[1]

Benford,

S.,

Magerkurth,

C. and

Ljungstrand

, P

Bridging the physical and digital in pervasive gaming.

Commun. ACM

48 (3), ACM Press (2005), 54

57.

[2]

Crabtree, A., Benford, S., Rodden, T., Greenhalgh,

C. Flintham, M. Ana

stasi, R., Drozd, A, Adams, M. Row

Farr, J., Tandavanitj, N. and Steed, A. Orchestrating a

mixed reality game 'on the ground'.

Proc.

CHI '0

. ACM

Press (2004), 391

398.

[3]

Flintham,

Benford,

Anastasi,

R.,

Hemmings,

T.,

Crabtree,

A.,

Greenhalgh,

Tanda

vanitj,

N.,

Adams,

and

Row

Farr, J.

Where on

line meets on

the streets: experiences with mobile mixed reality

games. In

Proc.

CHI '03

. ACM Press (2003), 569

576.

[4]

ORCHID. Human

Agent Collectives: from

Foundations to Applications. EPSRC Grant EP/I011587/.

http://www.orchid.ac.uk

[5]

Roberts,

S.,

Reece,

S.,

Nicholson

and

Llloyd, C.

Determining intent using hard/soft data and Gaussian

process classifiers.

Proc. Fusion 2011.

IEEE (2011).

[6]

RoboCup Rescue

http://www.robocuprescue.org

[7]

Olivia Solon. Zombies invade

Bristol for Igfest. In

Wired.co.uk.

http://bit.ly/b6GKQl

[8]

Stenros,

J., Montola, M. and

Waern

Pervasive

Game Design Strategies. In Pervasive games: theories

and design. Elsevier (2009).

[9]

Tang,

J.C.,

Cebrian,

M.,

Glacobe,

Kim,

W.,

Kim, T.

and Wicke

, D

. Reflecting on the DARPA Red

Balloon Challenge.

Commun. ACM

54 (4), ACM Press

(2011), 78

[10]

S. P. Simonovic, Systems Approach to Management

of Disa

sters: Methods and Applications

, Wiley, 2009, in

Disaster Prevention and Management, Emerald, 20 (4).

[11]

Drury,

J.,

Cocking

and

Reicher,

Everyone for

themselves? A comparative study of crowd solidarity

among emergency survivors

, Brit. Journ. of Soc.

Psych

2009

(

)

, 487

–

50.