47 haptic maps ntu

Evaluation of Haptic RIA Maps
David J. Brown and Lindsay J. Evett,
ISRG, Nottingham Trent University

AEGIS Workshop and International Conference, Brussels

Introduction
People who are blind use sequential, route based strategies
for navigating round the real world, rather than external,
or allocentric, frames of references
More map-based strategies better for navigational tasks for
people who are blind (Hill, et al, 1993); training in such
strategies greatly improves performance (Cummins and
Rieser, 2008; Simonnet et al, 2006)
Allocentric mapping involves identifying location relative to
perceptible landmarks (external frames of reference) and
encoding vectors between landmarks to provide a ﬂexible
system to determine location as the person moves
around the environment (Feigenbaum and Morris (2004))


Spatial mental models
External frames of reference and map based strategies are
more efficient and flexible – easier to remember,
alternative routes can be taken, shortcuts made and
destinations changed because they encompass a more
complete spatial representation (Martinsen et al, 2007)
Oliver and Burnett (2008) – route guidance systems
suppress cognitive map development
Active exploration of virtual worlds and maps to support
development of map-based strategies and spatial mental
models (Tversky, 1993) to support independent
navigational skills


Spatial representations

We are investigating a range of games based and
assistive technologies that can support
development of allocentric navigational
strategies in people who are blind, and
continually assessing their efficacy
In some circumstances these technologies offer
advantages over real world route learning as
they may help generate a fuller spatial cognitive
representation, involve active learning positions,
and be available for use on a daily basis (unlike
real world training support)


Haptic RIA Maps

Haptic RIA maps is an application whereby
visually impaired users can explore a web-
based representation of a street map using a
force feedback/haptic device
As well as haptic feedback there are auditory
cues, such as street names, and a sonification
mechanism which provides distance
information
Can this system provide information equivalent
to that provided to sighted users by
conventional 2D maps?


Spatial representation and Haptic
Maps
Can people who are blind read Haptic
Maps and get useful information out of
them? (spatial information, contextual
information…)
What is the level of improved spatial
information? e.g., find out more about
an unknown space, used it to extend a
known route, or take alternative routes,
or create a new route, take different
perspectives…..?


Nature of Information generated by
Haptic Maps
Tests proposed to assess whether active
exploration of Haptic Maps can support
development of spatial mental models, and
the complexity of those representations
The efficacy of the Haptic RIA Maps can be
compared to the use of Touch Over Maps,
and other navigational and Way Finding
support systems, such as the Virtual Cane,
Route Mate and Point Nav


AEGIS Haptic RIA Maps Test Tasks
1. Search for a location; map generated through
OpenStreetMap; after a map has loaded, generate
3D representations (i.e., create haptic map)
2. Explore the map, feel the haptic feedback, hear
the auditory feedback: While moving on the streets,
different sounds generated e.g., when standing at an
intersection; pitch of sound indicates approximately
distance to next intersection; press LCtrl for street; ~
for POI information
3. Search for a specific street name
4. Move/relocate the map: move around to see more of
the map using the arrow keys. Press the spacebar to
restore haptics
5. Zoom: Zoom the map in and out one step; wait until
street info resumes. Space bar restores haptics.


Participants
LUND tested with 9 people: 1 test leader who also
performed a heuristic evaluation, 2 pilot testers and 6
users. Both the pilot testers and 2 end users were fully
sighted but used simulated cataract glasses. 1 user is
blind, 3 have low vision
FONCE/UPM tested with 5 blind users, and 3 experts
formed the focus group
EPR tested with 8 blind/partially sighted users, 10 experts
NTU tested with 2 blind users, 1 tutor and 3 experts (all
sighted). All took part in the focus group discussion.
LUND used the PHANToM OMNI; all others used the Novint
Falcon


Participants - experience

LUND users all had varying degrees of experience
with force feedback haptics
NTU blind users had some limited experience of
tactile maps; both use GPS and had some very
limited experience with the Falcon
FONCE/UPM users were experienced with Braille
maps and GPS but had no experience of haptic
devices
EPR users had experience with swell paper and
relief usage, but none had ever worked with a
Falcon device before


Degree of Visual Impairment

Partial sightedness: someone who has serious loss
of vision even when corrected
Blind: severe sight loss even when corrected; may
be total
Low vision: moderate sight loss (NHS choices,
2011; Wikipedia, 2011)
Aegis:
Low vision users (users with a sight impairment and blindness with
useful residual vision) rate: 1. mild; 2. moderate; 3. severe
Blind users (without useful residual vision) rate: 4. total


Results – tasks (b/ps users only)
1. Search: Search stage not accessible by screenreader
2. Explore Map: All groups found the Haptic feedback to be unstable.
Those using the Falcon found it very difficult to use, haptic feedback
was erratic and inconsistent, the device was prone to violently
lurching and sometimes no haptic feedback could be felt at all
3. Search for specific street:
LUND – all users able to do tasks at least partly without help, but hard
to find street with no guidance
FONCE/UPM – needed to know the area, or have explored Braille map
to do this
EPR – mostly failed
NTU – one could, one couldn‟t


Results - tasks (contd.)
4. Move/relocate the map
LUND – all users able to do tasks at least partly without help
FONCE/UPM – not mentioned
EPR – couldn‟t do it
NTU – if any key press etc. while map relocating system crashed, so
didn‟t do this
5. ZOOM
LUND – all users able to do tasks at least partly without help
FONCE/UPM – easier with fewer streets
EPR - couldn‟t do it
NTU – didn‟t work when in map


Results – auditory feedback
NTU: position of tilde key on English keyboards made it difficult to use.
One of the blind users was left handed, found all the key commands
awkward
All Falcon users had to go very slowly to be able to use it at all, to try
and keep on the streets. Very easy to lift off the streets, audio
feedback stopped.
No cues to leaving the map, so difficult to know if left the boundary of
the map, or lifted the device off the map
Spanish users found verbal output difficult – Street names in Spanish
BUT the rest in English, and pronunciations were difficult to
understand. NTU found the voice difficult to understand. EPR
reported the auditory feedback overall as good, but sometimes the
CTRL key did not generate any feedback
Lag on TTS and sonification could cause problems; sonification could be
difficult to understand, unpleasant


Results – Summary
Verbal feedback difficult to understand; sonification OK but could be
improved (some lag on both)
Relocating and zooming unreliable; only LUND able to do these, but
users didn‟t like losing their reference points, and having to turn
haptics back on
All wanted indication of edge of map
The PHANToM appears to be easier to use than the Falcon, but still
haptic feedback is unstable
Both interaction devices involve a complicated relationship with the
map representation (actions don‟t have direct/simple relationship
with the map); PHANToM works better but too expensive
All groups liked the idea of the application; there were limitations
with cues and feedback, but overriding difficulty was with the
haptic feedback, especially for the Falcon


Results – NTU comments

One blind user could do it all, but found it
difficult to “keep on the map”; found the
voice difficult to understand and the
sonification a bit difficult, slightly
unpleasant
NONE of the other NTU users could use it
easily (1 blind, 4 sighted), although the
sighted users did manage to move to
the specified street


What to do?

Better device? MS haptic mouse? Better haptics?
Consider aims of app – to help blind users build a
useful, spatial cognitive representation of the
map area
Consider relationship between cognitive
representation, actions, map representation
Need more direct and reliable relationship between
them
Need reference points (implicit in a spatial mental
model, multiple perspectives)


Touch Over maps

HaptiMap demonstrator (HaptiMap, 2011)
Has all the desirable attributes:
user actions have a direct, reliable, relationship with maps
Simple reference points (may need more on tablet)
easy to use (2 blind users)
can reproduce the map (2 blind users)
BUT 1 blind user with resid. vision found areas with patchy
feedback, frustrating and poor info.

MS haptic mouse could have similar attributes
To have both would give desktop and mobile apps


Further work

Investigate MS haptic mouse; improve haptics?
Evaluate the information obtained from exploring
the map; can users create spatial mental
models?
Can they use this information:
to know about the layout and content of an area?

to find out about an area in which they have known routes (to
overcome obstacles, changes)?
to extend a known route?
to create a new route?


Evaluate Spatial Mental Models

Does model contain spatial relationships,
allow multiple perspectives, sufficient for
actual route finding? Possible tasks:
1. Recreate map
2. Describe routes from A to B and from C to B
3. Study area where known route:
• demonstrate ability to deal with obstacles
• extend known route
• create new route


Conclusions

Both Falcon and Phantom have significant usability
issues and are expensive
Alternative device – (Microsoft Haptic Mouse, £30).
This gives much more direct correspondence
between real movement and virtual response
Improve haptic information in app
Touch Over Maps – direct correspondence between
actions and map; reference points
Can blind/partially sighted obtain useful
information from these apps? Usable spatial
cognitive representations – research to evaluate

Comparison to other technologies

Virtual Cane – to support development of
both egocentric and allocentric strategies
(Evett at al, 2009). Similarities to the aims
of Haptic Maps, but levels of scale and
details are different
Point Nav (HaptiMap, 2011)
Talking GPS (Trekker; Mobile Accessibility)
RouteMate (Brown et al, 2011)
Bluetooth/wireless indoor way points (Evett
at al, 2011)

References
Aegis (2011). DOWNLOADS: WAI ARIA enabled plugins; Haptic RIA Maps,
http://www.aegis-project.eu/ accessed 23/11/11
Brown, D. J., D. McHugh, P. Standen, L. Evett, N. Shopland, S. Battersby
(2011), Designing location-based learning experiences for people with
intellectual disabilities and additional sensory impairments, Computers and
Education, vol. 56, pp. 11–20.
Code Factory (2011b) Mobile Accessibility for Android,
http://www.codefactory.es/en/products.asp?id=415 accessed 29/3/11
Craik, K. J. W. (1943) The Nature of Explanation, CUP
Cummins P A & Rieser J J (2008), Strategies of maintaining dynamic spatial
orientation when walking without vision, In Blindness and Brain Plasticity in
Navigation and Object Perception (J J Rieser, D H Ashmead, F F Ebner and A L
Corn, Eds), Lawrence Erlbaum Associates, New York, pp. 227-238
Evett, L., T. Allen, M. Javad Akhlaghinia, N. Shopland (2011). I need
assistance: Smart phones as assistive devices, Proceedings Interactive
Technology and Games, Nottingham UK


References (contd.)
Evett L, Battersby S, Ridley A, and Brown DJ. (2009). An interface to virtual
environments for people who are blind using Wii technology – mental models
and navigation. Journal of Assistive Technologies, 3 (2), pp.30-39
Feigenbaum, J. D. and Morris, R. G. (2004). “Allocentric versus egocentric
spatial memory after unilateral temporal lobectomy in humans”.
Neuropsychology, 18, 462-472
HaptiMap (2011) HaptiMap project outline,
http://www.haptimap.org/home/about-haptimap.html accessed 14/11/11
Hill E W, Rieser J J, Hill M, Halpin J & Halpin R (1993), How persons with visual
impairments explore novel spaces: strategies of good and poor performers, J.
Vis. Imp. and Blindness, 87, 8, pp. 295-301
Humanware (2011a) Trekker Breeze, http://www.humanware.com/en-
united_kingdom/products/blindness/talking_gps/trekker_breeze/_details/id_10
1/trekker_breeze_handheld_talking_gps.html accessed 12/10/11
Martinsen, H., J. M. Tellevik, B. Elmerskog, M. Storlilokken (2007). Mental
effort in mobility route learning, J. of Vis. Imp. and Blindness, 101, pp1-18.


References (contd.)
NHS choices (2011) Visual impairment, http://www.nhs.uk/conditions/visual-
impairment/Pages/Introduction.aspx accessed 14/11/11
Oliver, K. J., Burnett, G. E. (2008). Learning-oriented vehicle navigation
systems: a preliminary investigation in a driving simulator. In Proceedings of
the 10th International Conference on Human–Computer Interaction with Mobile
Devices and Services. pp. 119–126
Simonnet M, Guinard J-Y & Tisseau J (2006), Preliminary work for vocal and
haptic navigation software for blind sailors, Proc. 6th Intl Conf. Disability,
Virtual Reality & Assoc. Tech. (ICDVRAT), Esbjerg, Denmark, 2006, pp. 255-
262
Tversky, B (1993) Cognitive maps, cognitive collages and spatial mental
models, in Frank, A U and Campari, I (Eds.) Spatial Information Theory: A
Theoretical Basis for GIS, Proceedings COSIT „93, Lecture Notes in Computer
Science, 716, pp. 14-24, Springer, Berlin
Wikipedia contributors (2011) Low vision. Wikipedia, The Free Encyclopaedia.
November 1, 2011, 18:42 UTC. Available at:
http://en.wikipedia.org/w/index.php?title=Low_vision&oldid=458503177
Accessed November 15, 2011.

47 haptic maps ntu

More Related Content

Viewers also liked (10)

Similar to 47 haptic maps ntu (20)

More from AEGIS-ACCESSIBLE Projects (20)

Recently uploaded (20)

47 haptic maps ntu