Empowering the Elderly: Implementation of Navigation Assistance

Application for Public Transportation

Samuli Heinonen and Erkki Siira

VTT Technical Research Centre of Finland Ltd, Kaitoväylä 1, 90571, Oulu, Finland

Keywords: Public Transportation, Navigation, Mobile Services, Elderly, Error Detection.

Abstract: When a human ages, the functional abilities could weaken, which can lead to isolation from the society.

Applying technological aid for improving the quality of life of the elderly helps them to stay independent

and active in their community across their life. This article introduces a system, which is empowering the

elderly by aiding them in using public transportation in both rural and urban environments, and helps them

in this way to maintain the mobility and activity. The ASSISTANT system provides a simple, yet effective,

route info for user and shows only relevant information, at the right time and in the appropriate format. The

system is communicating with the user via audio and visual guidance. In addition, it has an error detection

functionality, which helps user in case of common navigation errors. End user evaluation performed in

Helsinki (Finland), Vienna (Austria) and Donostia-San Sebastian area (Spain) confirms that the service

works as intended and helps older people to use public transport by guiding them all the way from the start

to the end of their journeys. The system was found to be particularly helpful on unfamiliar routes.

1 INTRODUCTION

Public transportation is perhaps the most important

mode of transport among the elderly who live in

cities. (Social Exclusion Unit, 2013) Public

transportation becomes a primary mode of transport

when they lose the ability to drive. To learn how to

use the public transportation might be challenging

for elders. Utilizing public transportation in

unfamiliar places might be stressful even for people

with normal cognitive and sensory abilities.

In the ASSISTANT project we have developed a

system, which aids elderly in using public

transportation in urban environments, and provides

simple, yet effective guidance for them.

ASSISTANT system provides only relevant

information, at the right time and in the appropriate

format, and the user interface supports both visual

and audio guidance. In addition, ASSISTANT

supports safety and security by its error trapping and

remediation functionality, which makes it responsive

to the user’s physical, cognitive and mental

capacities and preferences using the ICT already

available.

2 RELATED RESEARCH

In the ASSISTANT project Carmien and Obach

(Carmien and Obach, 2013) have studied the

implications of getting lost and getting back on track

in public transportation. They approach the

happenstance of becoming lost as a form of error,

which anchors it to a long line of error research.

Their philosophy is based on Dekker’s (Dekker,

2006) suggestion in which errors are to be

considered more as a systemic event and not a solely

a human error. This makes errors easier to

understand and react-to from the system design point

of view. ASSISTANT’s theoretical background for

errors is done by Carmien and Obach whose main

research is on error capturing, error classification

and error mitigation in public transportation using

the mobile navigation systems.

Matthews and Wardman (Matthews and

Wardman, 2015) have studied how people with

physical mobility impairments would want to use

mobile application to help them in public transport

and private car setting and how much the users

would be willing to pay for different services the

mobile application could provide. When using

public transport the 59% users would be willing to

Heinonen, S. and Siira, E.

Empowering the Elderly: Implementation of Navigation Assistance Application for Public Transportation.

In Proceedings of the International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2016), pages 35-44

ISBN: 978-989-758-180-9

purchase an app in contrast to 42% for car users. For

the car users, the most important services of which

they are willing to pay were:

 the route directions

 assistance in road breakdown situations, and

 ability to pre-book disabled parking slots

For public transport users the most important

services of which they are willing to pay were:

 the ability to pre-book staff assistance in

transportation hubs

 pre-book a taxi for continuation of the trip

 information on the next station/stop with the

arrival time

 up to date connection information

 accessibility info of the arriving vehicle

The amount of mobile applications which

provides public transport navigation support is

continuously growing. Even though new services are

appearing constantly, most of the newcomers are

aimed at public transport users without taking into

account specific user groups.

Assist-Mi (assist-Mi, 2014) operates in United

Kingdom and is planned for disabled users. It has no

journey planning or travel information, but support

organization people can see the user location and

provide help in difficult situations.

App&Town (App&Town, 2015) is a Spanish

(Barcelona, Madrid and Laval) navigation service

for users that are blind, partially-sighted, deaf or

physically disabled. It has several characteristics that

are shared with ASSISTANT, but it has no error

mitigation functionality.

GeorgiePhone (GeorgiePhone, 2014) operates in

United Kingdom. In the system, the pedestrian

guidance is emphasized. In the error situations, a

“help me” functionality contacts a garegiver.

Transit App (Transit App, 2015) is a free local

transportation planner for mobile phones. It operates

in 100 big cities across the USA. They also support

Canada and many cities in Europe and Mexico.

Transit App does have a support for real-time

information. On the other hand, it is not designed for

special groups and it does not have error mitigation

functionality in it.

Moovit (Moovit, 2015) is also a free, local transit

planner app for mobile phones featuring live arrival

times, updated line schedules, local station maps,

service alerts & changes that affects users’ trip.

Moovit supports hundreds of cities around the world

including many European cities. Moovit presents the

planned route in legs and shows the path in a map. It

also gives textual information how to walk to the

nearest bus stop. While user is in the vehicle, the app

gives a list of the stops during a route. There is no

error mitigation functionality available in the app.

There are even more mobile apps for public

transportation navigation (e.g. TRAFI, Citymapper,

Avenue-Public Transport). Many have a pedestrian

guidance showing a path in a map or by giving a

textual guidance for walking leg, but rarely they do

have error mitigation functionality or other aid for

supporting user in case they get lost.

3 ASSISTANT PROJECT

ASSISTANT (Aiding Sustainable Independent

Senior Travellers to Navigate in Towns) project is a

pan-European joint project under AAL Programme

(Active and Assisted Living Programme, 2015). In

the project there are seven companies from five

European countries – Austria, Finland, France, Great

Britain and Spain. The project contributes to

maintaining the mobility of older people in Europe.

Motivation is in safeguarding their social and

economic participation in aging society by

introducing a navigation assistance service to

support their journeys in urban areas.

The ASSISTANT project was divided into three

phases. First phase in spring 2013 identified which

kind of application would be needed to assist elders

in public transport system. Second phase in spring

2014 had a first version of the system that was tested

in real world environment with the elderly in three

European cities – Vienna, Austria; Helsinki, Finland

and San Sebastian, Spain. Test included both the

creation as well as the execution of a journey from

Point-A to Point-B. To ensure support in case of

operational problems or need for help, a person from

ASSISTANT project was present all the time during

the tests. The third phase in 2015 had a polished

navigation assistance system that was piloted in

same three European cities with the elderly.

3.1 ASSISTANT Use Case

Main goal of the project was to improve the usage of

public transport for older travellers. A user may

navigate independently in public transport network

using a mobile phone application. The application is

connected with back-end system that keeps the

status of the traveller and the status of the public

transport network up to date. The user plans a route

by using a web application and the route is uploaded

to the mobile application for further use.

ASSISTANT project has a clearly defined target

use case that was laid down at the beginning of the

project and consistently pursued. Such use case is

ICT4AWE 2016 - 2nd International Conference on Information and Communication Technologies for Ageing Well and e-Health

split into two parts: route planning and navigation.

Route planning: When the user decided to go to a

theatre. At first, she log in to ASSISTANT website,

enter her preferences and fill in if she has any

caretakers that the system should contact to in case

of an error. On route planner tab, she enters date and

time for the journey, her home as a start address and

the theatre as destination. A map view shows the

route, starting point and the destination. If she is

pleased with the plan, she presses ‘send to mobile’

button and the plan is sent to user’s mobile device.

Navigation: Mobile device will acknowledge that

it has received the information and then waits until

the time for the journey comes up. Application alerts

the user beforehand that she should get ready for the

journey. When the journey starts, the application

will show arrow-based pedestrian guidance to the

public transportation stop. When user gets to the

stop the application guides user to get the right

vehicle and shows how the journey continues while

she is sitting in the vehicle. When destination or

transfer stop gets closer the application will notify

the user. There is an arrow-based guidance from

transfer stop to the stop she needs to go to catch

another vehicle or from the stop to the destination.

4 DESIGN RATIONALES FOR

THE ASSISTANT SYSTEM

The needs of the target group of the elderly users

were reflected on to five rationales that were seen as

essential for the ASSISTANT system to implement.

The five characteristics are:

 Personalization and customization

 Research-based interface for the elderly

 Route planning metadata with accessibility

information

 Help for the last kilometre

 Design for failure

These characteristics themselves are not new or

unique, but the use of the combination of them

makes the system unique for elderly and disabled

users. Our main contribution to the field of

navigational services is ‘design for failure’ and it is

being discussed in Chapter 5 in more detail.

Personalization and customization was the first

characteristic that was chosen. The elderly people

have very different kind of disabilities and the

ASSISTANT system was designed to be helpful for

the people with different needs. When users have

disabilities the normal look-and-feel might not be

functional enough and the service is thus not usable.

To create functional service for disabled users, the

system needs to take into account the users’ special

needs. Here are two possible approaches. System

can be split up and design services individually to

address different disabilities or it may be designed to

be flexible and customizable. ASSISTANT took the

latter approach. ASSISTANT’s customization and

personalization tries to provide only the relevant

information, at the right time and using visual,

haptic and audio cues to communicate.

The second characteristic is research-based

interface for the elderly. There is a large body of

research about how user interfaces should be

constructed for elderly users. In addition to

academic research, we have performed two

developmental incremental phases with the target

user group, where they have been able to affect

which kind of user interfaces they prefer and they

have been able to confirm or unconfirm the choices

made by the design team. From research we derived

several user interface guidelines for ASSISTANT

(Carmien and Garzo, 2014). In addition, TELSCAN

project provided lots of guidelines how to implement

for elderly and disabled. (TELSCAN, 1997)

The third characteristic is using accessibility

information in route planning. This characteristic is

ASSISTANT’s way to create possibility for

continuous accessible path of travel. Normal route

planning services do not utilize this information

when routing people from place to another. For the

elderly this might be a deal breaker, because of

disabilities they do not want to end up on a train

platform where they can’t exit for example because

of long and steep stairs. This characteristic has

proved the most difficult to implement in pilot cities,

because getting the accessibility information is

difficult as the information is usually not up to date

and might not even reflect the real world conditions.

The fourth characteristic is help for the last (and

first) kilometre. This means that system will help the

pedestrian part of the journey in addition to the use

of public transportation system. The grounding

philosophical approach in ASSISTANT was that the

user will be guided from the doorstep to the

destination. ASSISTANT last-kilometer guidance

includes both a compass-based arrow guidance and a

map-based guidance. Route planners and

navigational guides typically only do routing

through the public transportation system and maybe

add a rough map for the pedestrian part of the

journey.

Empowering the Elderly: Implementation of Navigation Assistance Application for Public Transportation

5 CATCHING ERRORS

ASSISTANT system is designed to be able to catch

and mitigate the errors during the navigation. An

ability to catch and recover from errors is one of the

main features of ASSISTANT system and it is the

main difference with other similar navigation

systems.

5.1 Design for Failure

As mentioned in design rationales for the system in

previous chapter (Chapter 4), one of our design

principles in ASSISTANT project was the so-called

“Design for Failure”. Design for Failure in the

context of ASSISTANT means, that at some point of

use something will go wrong. It might be a human

error, lack of network coverage or fault in the public

transportation system, but inevitably everything does

not go as planned. Because of this, the navigational

system needs to be able to react for errors somehow,

and be reliable enough so that the user can trust that

he or she is not left alone and lost. ASSISTANT has

a process to enable recovery from failure.

5.2 Error Detection

In ASSISTANT system there are several ways to

detect an error. Our approach to error detection is

that the route navigation is based on a state machine.

In every state there might be different errors.

5.3 State Machine

In order to be able to give precise and compact

guidance to the user, the application needs to have

reliable identification of the user state (e.g. walking,

in the vehicle or at a stop). When the system detects

an error, it categorizes it based on the user’s state

and then it will try to remediate the situation.

The ASSISTANT mobile application has four

main context states: idle, walking, at stop and in

vehicle. (See Figure 1). Transitions between the

states are identified by different sensors: compass,

speed, bearing and accelerometer. State transitions

are described here:

 T1 idle to walking

 T2 walking to stop

 T3 stop to vehicle

 T4 vehicle to stop

 T5 stop to walking

 T6 walking to idle

A successful route starts from the home (or other

origin) and goes through all of these states. If route

includes the vehicle changes, it is possible that state

machine cycles between WALKING - AT STOP -

IN VEHICLE – states as long as route lasts.

IDLE WALKING ATSTOP INVEHICLE

T1 T2 T3

T6 T5 T4

Figure 1: Mobile app state machine.

Errors in the journey itself are easier to catch and

recover, because the system itself can work to fix the

problem. More difficult problems are those where

the system itself does not function as planned. For

example loss of network, loss of battery power or

weak GPS signal are problems that affect the

performance of the system severely. To react these

types of problems, the system has a caretaker

caretakers are notified that the user is missing and

the system cannot help enough. The back-end

system will then help the caretaker to its best ability

to regain contact with the user. Thus, the user can

trust that if everything else fails there still is a person

that knows where she was going to and what was the

error that happened during the journey.

5.4 Error Remediation

Next we take a look what are the most common

errors and how the system recovers from these errors

during the route

User is Walking to Wrong Direction

While the user is guided on the First/Last kilometre,

she is travelling by walk. Normally she may use the

compass to see the direction where to go or use the

map to see her route to the next waypoint. However,

it might be possible that she put the mobile phone in

her pocket, when she may forget the right direction.

Now, if she has been walking to the totally wrong

direction (the distance to the waypoint is continually

extended) for more than 3 seconds, the Assistant app

warns (as seen in Figure 2) her by the dialog and

also by saying aloud: “You are walking to the wrong

direction!”

Missed Stop while Travelling by Vehicle

While user is travelling e.g. by the bus and she is

approaching the destination stop, she is informed

about this on the second-last stop. When user arrive

ICT4AWE 2016 - 2nd International Conference on Information and Communication Technologies for Ageing Well and e-Health

Figure 2: Warning - dialogs.

the destination stop, the user is advised to get off. If

the user travels past the destination stop, assistant

app informs user (see Figure 2) by a dialog saying:

“You have missed your bus stop. Please get off at

the next stop.” When user has step out of the bus, the

system calculates the new route to destination, and

the navigation continues as normally.

Missed Vehicle while Waiting the Transport on a

Stop

Sometimes user may be a bit late from the stop

where she was planned to get on the transport, and

her vehicle may have already left the stop. If there is

real-time information available, ASSISTANT

system could provide information about the next

vehicle departure time. If real-time information is

not available, system needs to utilize the schedule

information for further guidance. Then, the user is

asked to refresh the route to see when the next

appropriate vehicle will arrive to the stop. Figure 4

shows the dialog to the user.

Figure 3: Missed vehicle dialog on the stop.

User Gets Lost during the Route

If user gets lost during the route in any point, he can

ask help from the Assistant app HELP-screen which

is seen in Figure 5. There are 5 different alternatives:

1. Get a route to destination. This action creates a

new route to the planned destination.

2. Get a route to starting point. This action creates

a new route to the starting point.

3. Get a route back home. This action creates a

new route to the user’s home.

4. Connect to the contact person by Phone. This

action makes a phone call to the user’s contact

person for asking more help.

5. Connect to the contact person by SMS. This

action send’s a SMS message to the user’s

contact person for informing her about troubles

on the route.

Figure 4: HELP-screen.

GPS Signal Lost

Navigating in urban environment with the help of

GPS is sometimes challenging. High buildings form

urban canyons, where GPS signal may be weak

temporarily. The navigation module in the mobile

application tolerates the short and temporary GPS

signal lost during navigation without disturbing user

about that too sensitively. The user is informed only

if the GPS signal has been lost more than 15 seconds

by saying: “GPS signal not available!” In addition,

Assistant app tells user the next waypoint where she

was travelling, so she can try to continue the

navigation on her own.

Figure 5: Guidance during the Metro leg.

Metro Leg, no GPS Signal Available

If user travels by metro, it is mandatory to go below

the ground, where GPS signal is not available. In the

Figure 6 it can be seen how the application informs

user about the lost GPS signal, and request to follow

the signs in the station to get to the metro stop.

Empowering the Elderly: Implementation of Navigation Assistance Application for Public Transportation

When user has is on the metro, the name of the

destination stop is shown. Finally, when user has

reached the destination stop by metro and she comes

on the ground, the GPS signal comes available and

user is guided to her destination as normally.

Mobile Data Connection Lost

While user is travelling, Assistant system keeps

track on her progress by sending so called ‘I-am-

alive’-message to the server every five minutes

automatically. ‘I-am-alive’-messages include

information about mobile phone battery level,

current state of the route (walking-stop-in vehicle),

and current location. If server does not receive

messages from the mobile phone in every five

minutes, a SMS message is sent to the user’s contact

person. The reason for system fail may be a mobile

data connection lost or some unpredictable and

serious application error. The example of the SMS

message is here: “Mobile phone is unreachable for

Mr. West. Phone battery level was 50% and Mr.

West was travelling by tram. Mr. West has most

recently been located at Mannerheimintie 2,

Helsinki.” When SMS is sent to the contact person, a

responsibility to help user is moving to the contact

person.

6 SYSTEM ARCHITECTURE

ASSISTANT system follows a traditional client-

server architecture, where mobile phone client

communicates with back-end server-based

application. Server retrieves updated maps,

schedules and telematics data from external sources.

The system offers a web site with the set of features

such as trip planning and caretaker support. The

Figure 6 gives a high-level architecture of the

system and its relations to the external systems.

The main components of the ASSISTANT

system consist of a mobile application running on

the Android device carried by the user, and a

backend system interacting with the mobile

application asynchronously over the Internet. The

main role of the backend system is to provide public

transport route planning functionality, and profile

management services. The server is also used to

monitor the systems interactions for storing error

states in order that the system can provide error

mitigation functions for user.

Figure 6: High-level system design.

6.1 Components

There are five main components in the ASSISTANT

system: Web Client, Mobile Application, Server,

External Routing APIs and External Map Providers.

Web client is used for user registration, user

management and trip planning. With the web client,

user can create user accounts and adjust user

preferences, like walk speed, home addresses,

telephone numbers etc. More importantly, user can

plan public transport routes by using trip planner.

The Server hosts the web site, where the web client

is installed. There are two ways to get routes for the

user: the trip guide engine may request route plans

from external routing APIs and modify the route to

ASSISTANT-specific format or it can use

OpenTripPlanner (OpenTripPlanner, 2014) module

(inside of Trip Guide Engine) to create routes from

the raw schedule information. When route is created,

it is sent to the mobile application, which starts the

guidance when the planned route is about to start.

External Map providers are used to clarify the

overall view of route for user. ASSISTANT server

can utilize maps from Google and OpenStreetMap.

6.2 Communication

Communication between server and client is handled

using REST API. It means that server is stateless, it

has uniform interface, communication is based on

client-server architecture and system follows the

standard https-based security mechanisms. If the

server needs to contact the mobile client, it sends

GCM (Google Cloud Messaging) message to the

client with appropriate parameters. For example, if

user has created a new route by using the web client,

the server sends a GCM-message to the mobile

client to inform about the new route. The mobile

ICT4AWE 2016 - 2nd International Conference on Information and Communication Technologies for Ageing Well and e-Health

application’s GCM-listener notices the message and

starts to download route information from the server

in pre-determined XML format.

6.3 Real-time Information

The ASSISTANT system is able to utilize real-time

information during the navigation. Real-time

information in public transportation can be utilized

two ways:

1. The local route planner’s telematics server

provides the real-time information through the

web interface. The ASSISTANT server requests

the data from the telematics server and delivers

it to the mobile application. Real-time

information is used to make public

transportation guidance more accurate. For

example, while user is waiting a vehicle on the

stop, she can receive real-time information of

the vehicle arrival time.

2. Gathering real-time data in vehicle stops and in

vehicles over the on-board vehicle

WLAN/Bluetooth. Every vehicle could have

their own WLAN/Bluetooth base station, which

is used to identify the right vehicle in heavy

traffics stops. In addition, the vehicle itself can

provide route information, disruption data, etc.

for user through the specified API. In metros,

user could get exact information about time

schedules and upcoming stops.

Using WLAN/Bluetooth hotspot in vehicles will

solve the problem of user mistakenly boarding a

wrong vehicle. In Helsinki there is a new vehicle on-

board WLAN/BT system under development, but

other pilot cities (Vienna and San Sebastian) do not

offer WLAN hotspots in their vehicles, so the

ASSISTANT system did not utilize on-board

WLAN for real-time data delivery.

7 GRAPHICAL USER

INTERFACE

In this section a Graphical User Interface (GUI) of

the ASSISTANT System is introduced. First we go

through Web Route Planner GUI, and after that, a

GUI of a Mobile Application is introduced.

7.1 Web Route Planner

The use case of ASSISTANT System is was

described in section 3.1. If the user is using the

system first time, the user should first create and

define a user profile in ‘My Information’- and

‘Preferences’-tabs. The user should give a name, a

phone number, an email address and a home street

address. In addition, in user can define attributes,

like personal walking speed and alert sounds to

customize a system to better fit on the user’s needs.

Figure 7: Route Planner-tab on ASSISTANT web site.

When the user has finished creating a user profile, it

is possible to plan a route by using ‘Route Planner’-

tab. The figure (Figure 7) is a screenshot from

‘Route Planner’-tab. In this screen, the user gives

start and end addresses, and desired time for the

route. This makes the system to calculate a new

route and which can be seen in a map. The route can

be send to the mobile client by “Send to mobile”-

button.

7.2 Mobile Application

Currently Loaded Routes – Screen

The main screen of the mobile application can be

seen in Figure 8. Screen lists all the routes which are

loaded to the mobile client. In this screen user can

delete old routes, or see the details of the route. In

addition, user has an option to get help in any

moment by opening the HELP-screen.

Navigation – Screen

This screen shows all the textual information on the

top of the screen. There are three kind of

information available: 1. TO-field telling user the

next goal of the route; 2. INFO-field giving user a

textual advice how to reach the next goal, and; 3.

NEXT LEG-field which prepares user for the next

leg of the route. In the bottom of the screen there are

three buttons, which gives different kind of

assistance for user during the navigation: Help-

button opens the HELP-screen, Replay-button

reads aloud the INFO-field text again, and

Empowering the Elderly: Implementation of Navigation Assistance Application for Public Transportation

Figure 8: Currently loader routes - screen.

Map/Compass-button is used for switching between

map/compass views. Navigation screen can be seen

in Figure 9

Figure 9: Navigation-Screen (Compass- and Map-view).

Help – Screen

Figure 5 is showing the screenshot from the HELP –

screen. Mobile application utilizes the GPS signal to

locate user, and by taking advantage of the

geocoding, the app tells user the current location as

street address. In addition, there are buttons which

helps user to create routes to three different

destinations: 1) To destination, 2) To starting point

and 3) Back home. In the bottom of the screen there

are two buttons, which makes it possible to connect

with the user’s contact person by making a phone

call or sending a SMS message.

8 END USER PILOTS

In ASSISTANT project we had two evaluation

phases: the first one was arranged in spring 2014 and

the second one was arranged in spring 2015. The

system was improved based on the analysis of the

first pilot’s results. The general objective of the

second evaluation phase was to gather feedback

from the representatives of the target user group

(mobile older adults who are over 65 years old) not

only to evaluate the usability and the usefulness of

the system but also to track system bugs and errors

for future development.

8.1 Organization

End user testing in spring 2015 was organized in

three cities simultaneously: Helsinki (Finland),

Vienna (Austria) and San Sebastian (Spain). The test

users were outfitted with smartphones for two

weeks, and it was ensured that everybody have a

personal computer with internet connection at their

home. The test users were asked to use the

ASSISTANT system in their everyday life and not

to change their mobility habits for the testing period.

The project partners have a common

methodology for the testing sessions ensuring that

the results from the three testing sites are

comparable and that all the test participants are

treated in an equal manner. At each testing site, one

project partner was responsible for organizing the

testing sessions, recruitment of the testers,

acquisition of the testing material and carrying out

the tests by applying the testing methodology.

The test users participated in an initial workshop,

which ensured that they understood their role in the

evaluation and learned how the system functions.

Then, during the two weeks, they used ASSISTANT

system to guide their routes in the local public

transports. The testers recorded their user

experiences in a specific diary. In the end of the test

phase, the participants were gathered for a final

discussion where they were able to exchange their

user experiences with the test administrator.

8.2 Results and Analysis

As a result of end user testing, some testers used the

system more often than the others. Furthermore,

some of the users were more comfortable with the

system than others, and in consequence some

requested help from the test administrator and the

others did not. The Table 1 below demonstrates

statistical information from the end user pilots 2015.

Even though some system bugs and some

usability difficulties occurred during the two-week

autonomous testing, the general attitude of the users

were very positive and they consider that once they

trust ASSISTANT system, it would encourage them

to be more mobile and to take the public transports

more often. Furthermore, several participants said

ICT4AWE 2016 - 2nd International Conference on Information and Communication Technologies for Ageing Well and e-Health

Table 1: Statistics from the second end user pilots.

CITY

A number

of created

routes

A number of

routes sent to

the Mobile

client

A number

finished

routes

HELSINKI 315 162 23

VIENNA 271 150 14

SAN

SEBASTIAN

465 - 22

that this service is helpful when taking unknown

route or when travelling in previously unknown city.

The essential recommendation that the testers

gave is that ASSISTANT system should interact

more with the user’s needs; all functionalities should

be optional so that the user could adapt them

according to their ICT-skills and functionality needs.

Another important feedback from the testers was

that the users need the possibility to change and

create routes on the mobile phone too. People want

to be flexible: they cannot plan their entire journey

ahead and they don’t carry a computer with them. It

is nice to have the possibility to send routes from a

computer but when people get used to handling a

smartphone, they can use the service with a

smartphone more efficiently. There was quite much

discussion of enabling route planning in mobile

application in early design stages of the project, but

in the end we limited it to the three choices of ‘to

original goal’, ‘to home’ and ‘to route start’. The

design rationale for these was twofold:

1. We decided early on to eliminate the need to

use on-screen keyboards, because elderly may

have difficulty of using tiny keyboards on

mobile phone, and

2. When the screen is divided into data display and

keyboard / text input areas the data must be

displayed in much smaller font making the use

of the system difficult for elders.

This evaluation confirmed that the system

provides a helpful service for the end-user when

using public transportation and notably when taking

new unknown routes where the user needs to trust

the guidance provided by the system. The general

reaction was that the system is relatively easy to use

in the context of both devices. As a conclusion, it

was noticed that based on the evaluation the

ASSISTANT project fulfilled 85 % of the initial

user requirements (14/17 items achieved).

9 FUTURE WORK

Even though the current version of the ASSISTANT

system is a well-functioning service, it is not

completely ready yet. There are a couple of missing

features which needs some development work

before the system is ready for markets.

One of the features missing in the current system

is its inability to recognize the right vehicle while

user was waiting it on the stop. As mentioned on

section 6, on-board WLAN/Bluetooth hotspots could

help with this challenge. Furthermore, there has been

recently published new devices on markets called

Bluetooth Beacons. In essence, Bluetooth Beacon

technology allows mobile applications to understand

their position on a micro-local scale, and deliver

hyper-contextual content to users based on location.

In that sense, Bluetooth Beacons could be utilized in

ASSISTANT as follows: Beacons could be mounted

in public transport vehicles. While the user is

waiting a vehicle on a stop, the application can

detect the arriving vehicles by listening the vehicle

Beacon signals. When the signal is strong enough,

the user can be guided to get on to the right vehicle.

Another missing feature in ASSISTANT is its

inability to plan new routes on-the-fly during the

navigation. This feature is easy to implement, and a

test version of that is existing already.

In order that the system could be taken into use

in other cities, a specific server module and a couple

of configuration files should be updated and tailored

to support the local transportation operator’s route

format. Finally, all the GUI text labels on Mobile

Application and Web Route Planner must be

translated to the host country language.

10 CONCLUSIONS

In this paper we introduced The ASSISTANT

system, which is developed in ASSISTANT project.

System can be used by the home PC and smartphone

and it helps older travellers plan public transport

journeys and then provides guidance during their

journeys. ASSISTANT includes:

 A web browser based on-line means of

planning a trip

 Guidance on making multi-step journeys

 Alerts which tells the user when it is time to

leave or board the vehicle

 Error detection and mitigation during the trip

ASSISTANT system helps older people to travel

safely and independently by public transport. The

system offers a seamless support for the entire

length of a journey in urban environments.

An ability to catch and recover from errors is one

Empowering the Elderly: Implementation of Navigation Assistance Application for Public Transportation

of the main features of ASSISTANT system and it

makes the difference with other navigation systems.

End user evaluation confirmed that the

ASSISTANT system provides a helpful service for

the end-user when using public transportation and

notably when taking unknown routes where the user

needs to trust the guidance provided by the system.

ACKNOWLEDGEMENTS

The authors would like to thank ASSISTANT

project partners for the good cooperation during the

project. In addition, we would like to thank our

financier Tekes (the Finnish Funding Agency for

Innovation) and AAL Program for their support for

this research work.

REFERENCES

Active and Assisted Living Programme Referenced from

http://www.aal-europe.eu/ 4th of September 2015.

App&Town - website 2015 Referenced from http:/

/www.appandtown.com/ 11th of September 2015.

assist-Mi -website 2014 Referenced from www.assist-

mi.com 15th of October 2015.

Avenue - Public Transport, Google Play Market website

2015 Referenced from https://play.google.com/store/

apps/details?id=com.pocketavenue&hl=en 15th of

October 2015.

Carmien, S. and Garzo, A., Elders using smartphones: a

set of research based heuristic guidelines for designers,

in 16th International Conference on Human-Computer

Interaction (HCII 09) in the parallel session

"Cognitive Accessibility and Cognitive Support" in the

8th international Conference on Universal Access in

Human-Computer Interaction (UAHCI),2014,

Springer LNCS, Lecture Notes in Computer Science

Heraklion, Crete, Greece. p. 26-37.

Carmien, S. and Obach, M. 2013 Back on Track: Lost and

Found on Public Transportation. 15

International

Conference on Human-Computer Interaction (HCII

09) in the parallel session "Cognitive Accessibility and

Cognitive Support" in the 7th International

Conference on Universal Access in Human-Computer

Interaction (UAHCI), 21-26 July, 2013, Las Vegas,

Nevada, USA, Springer LNCS, Proceedings Part II

Part V pp 575-584.

Citymapper - website 2015 Referenced from

https://citymapper.com/berlin 11th of September 2015.

Dekker, S. 2006. The field Guide to Understanding

Human Error, p. 236. Ashgate, Burlington, Vermont,

USA.

GeorgiePhone – website 2015. Referenced from

www.georgiephone.com 15th of October 2015.

HaptiMap-project – website 2015 Referenced from

http://www.haptimap.org/ 15th of October 2015.

Matthews and Wardman 2015, Travel apps for disabled

people: Evaluating their contribution toward inclusive

transport, Proceedings of Transed2015 conference

Jun/2015 Lisbon, Portugal.

Moovit - website 2015 Referenced from http://

moovitapp.com/ 10th of September 2015.

OnTheBus 2014. Referenced from www.onthebus-

project.com/index.html 15th of October 2015.

OpenTripPlanner 2014 The open trip planner organisation

- Multimodal trip planning & analysis. 201 7/5/2014];

Available from: http://www.opentripplanner.org.

Social Exclusion Unit. 2003. Making the connections:

Final report on transport and social exclusion. London:

Social Exclusion Unit, Office of the Deputy Prime

Minister.

TELSCAN 1997, Inventory of ATT System Requirements

for Elderly and Disabled Drivers and Travellers.

Report from Transport Telematics Project No. TR

1108. Deliverable 3.1 Referenced from

ftp://ftp.cordis.europa.eu/pub/telematics/docs/tap_tran

sport/telscan_d3.1.pdf 26th of February 2015.

Trafi App - website 2015 Referenced from http://

www.trafi.com/ 10th of September 2015.

Transit App - website 2015 Referenced from http://

transitapp.com/ 8th of October 2015.

ICT4AWE 2016 - 2nd International Conference on Information and Communication Technologies for Ageing Well and e-Health