Functionalities and Requirements of an Autonomous Shopping Vehicle

for People with Reduced Mobility

Ant

onio Neves

, Daniel Campos

, F

abio Duarte

, In

es Domingues

, Joana Santos

, Jo

ao Le

, Jos

Xavier

, Lu

ıs de Matos

, Manuel Camarneiro

, Marcelo Penas

, Maria Miranda

, Ricardo Silva

and Tiago Esteves

Follow Inspiration, Portugal

IEETA/DETI Universidade de Aveiro, Portugal

Keywords:

Robotics, Retail, Reduced Mobility, Requirements, Functionalities.

Abstract:

This paper concerns a robot to assist people in retail shopping scenarios, called the wGO. The robot’s be-

haviour is based in a vision-guided approach based on user-following. The wGO brings numerous advantages

and a higher level of comfort, since the user does not need to worry about controlling the shopping cart. In

addition, this paper introduces the wGOs functionalities and requirements to enable the robot to successfully

perform personal assistance while the user is shopping in a safe way. A user satisfaction survey is also pre-

sented. Based on the highly encouraging results, some conclusions and guidelines towards the future full

deployment of the wGO in commercial environments are drawn.

1 INTRODUCTION

This paper describes the design concerns and

decisions when the company Follow Inspira-

tion (de Matos, 2012) developing a robot to help and

assist people (giving special emphasis to people with

reduced mobility) in retail environments, through a

user-following scenario.

The wGO

(Fig. 1) is an autonomous and self-

driven shopping cart, designed to follow people with

reduced mobility (elderly, people in wheelchairs,

pregnant women, temporary reduced mobility, etc.)

in commercial environments. With the robot, the user

can control the shopping cart without the need to push

it. This brings numerous advantages and a higher

level of comfort, since the user does not need to worry

about carrying the groceries or pushing the shopping

cart.

Thanks to the sensors (RGBD cameras and LRF),

wGO detects and identiﬁes its user in less than 2 sec-

onds, he just needs to push the “start” button and the

wGO will start following him. Furthermore, distance

sensors, RGBD cameras and LRF allow wGO to iden-

tify and avoid any obstacle along the way.

The robot is currently patent pending.

Figure 1: wGO: Front view; Back view.

The idea for the product came from the fact that

between 8% and 10% of the population has some

form of disability (CRPG and ISCTE, 2007), and that

in Europe alone there are about 50 million people with

disabilities and 134 million people with reduced mo-

bility. Apart from people using wheelchairs, there are

other cases in which people are temporarily or per-

manently disabled, these include: an elderly person

using a cane, or someone with a foot or leg injury

who requires the use of crutches, pregnant ladies and

parents with prams.

Neves, A., Campos, D., Duarte, F., Domingues, I., Santos, J., Leão, J., Xavier, J., Matos, L., Camarneiro, M., Penas, M., Miranda, M., Silva, R. and Esteves, T.

Functionalities and Requirements of an Autonomous Shopping Vehicle for People with Reduced Mobility.

DOI: 10.5220/0006385903730380

In Proceedings of the 3rd International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2017), pages 373-380

ISBN: 978-989-758-242-4

373

In fact, if we add the disabled, the elderly, preg-

nant women, and couples with children, we ﬁnd that

between 30% to 40% of all Europeans could beneﬁt

from improved accessibility. In addition to those peo-

ple with reduced mobility due to disability or injury,

there are many people without mobility issues who

could beneﬁt from assistance in carrying heavy bags.

Shopping environments are highly heterogeneous

and give rise to a high frequency of dynamic interac-

tions that trigger various senses and emotions in hu-

mans. This often causes a high level of stress in peo-

ple, and those with mobility limitations.

Some of the identiﬁed difﬁculties include (Aus-

tralia, 2016):

• For wheelchair users:

– no adequate forward reach at basins, counters

and tables;

– surfaces that do not provide sufﬁcient traction

(e.g. polished surfaces).

• For people who have trouble walking:

– no seating in waiting areas, at counters and

along lengthy walkways;

– access hazards associated with doors, including

the need to manipulate a handle while using a

walking aid;

– surface ﬁnishes that are not slip-resistant or are

unevenly laid.

Besides the difﬁculties brought by the shop-

ping environment itself, conventional shopping carts,

which can carry many products and which are pro-

vided with wheels so that the shoppers can push them,

also have serious drawbacks. One of them being their

considerable size. This is simultaneously an impor-

tant asset and a signiﬁcant drawback, as although

shopping carts can hold large and bulky products,

the increased mass complicates manoeuvrability and

handling. Manoeuvrability is particularly compro-

mised when making turns in supermarket aisles or

when avoiding other carts, shelves, and indeed other

shoppers (Rodriguez et al., 2014). Smaller baskets

appeared on the market to overcome the traditional

shopping cart

s drawbacks. These baskets were devel-

oped to hold a set of items while at the same time be-

ing easy to move. They contain wheels or rolling el-

ements incorporated into the bases which allow them

to be moved when parallel to the ﬂoor or when in-

clined. However, even though these baskets improve

manoeuvrability due to their reduced size and capac-

ity, they also have drawbacks typical of their mor-

phology, such as the need for the user to bend down

for placing or removing items, among others. Fur-

thermore, such baskets can have drawbacks typical of

the way they are stored, since stacking them vertically

can entail a problem for elderly shoppers or shoppers

with any type of physical limitation (Rodriguez et al.,

2014).

While the wGOs software architecture is detailed

in (Neves et al., 2017), here a design point of view

is adopted. The paper starts by reviewing the exist-

ing solutions for people with reduced mobility to shop

(Section 2) and some of the relevant legislation related

with technology equipments (Section 3). Design re-

lated requirements are given in Section 4. Next, in

Section 5 the wGO design evolution and justiﬁcations

for the changes made are given. Section 6 identiﬁes

the main risks related with the wGO usage in a real

scenario. Section 7 analyses a user satisfaction study

made in a relevant, unconstrained scenario. Conclu-

sions and another applications of this technology are

given in Section 8 and Section 9, respectively.

2 EXISTING SOLUTIONS

Looking at the commercial market, the most obvious

existing solutions are those provided by shopping cart

producers

. These providers typically have products

targeted for customers in wheelchairs, but not prod-

ucts for other types of users with reduced mobility

(e.g. pregnant women). A different type of solution

is the adapted system. Some examples are the “amigo

mobility” scooter

and adapted wheelchairs

, etc.

These products are, however, not particularly user

friendly. The user needs to ﬁrst move into the mo-

bility auxiliary device and then to learn how to use it

(which may be particularly hard for the scooter case).

In the case of wheelchair users, the user also needs

to leave their own personal chair, which may cause

discomfort and unnecessary stress. Another problem

with these solutions is that the user is visibly distin-

guishable from the other supermarket clients, which

may discourage some people from using it (Iezzoni

et al., 2001).

While this topic of assisted shopping using

robotics has received very little attention in the aca-

demic research community, several systems exist

where robots are used to help people with reduced

mobility. In (Gurgel Pinheiro et al., 2015), an antici-

pative shared control for robotic wheelchairs, targeted

at people with disabilities is presented. The same

idea, of intelligent wheelchairs, is also the focus of the

e.g. wanzl ( www.wanzl.com)

www.myamigo.com

e.g. meyra (www.meyra.de), promoted by Egiro

(www.egiro.pt)

VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems

374

work in (Faria et al., 2014a) where a data analysis sys-

tem which provides an adapted command language

is presented. A smart companion robot for elderly

people, capable of carrying out surveillance and tele-

presence tasks, is described in (Pavon-Pulido et al.,

2015). Also, with the aim of helping elderly peo-

ple through tele-presence, a low-cost platform capa-

ble of providing augmented reality for pill dose man-

agement was developed in (Martin Rico et al., 2014).

In (Faria et al., 2014b) an approach based on the Dy-

namical System Approach for obstacle avoidance of

a Smart Walker device to help navigation of elderly

people is presented.

Perhaps the closest application to the focus of this

paper is presented in (Gmez-Goiri et al., 2011) where

a product locator application is proposed. The appli-

cation runs on heterogeneous personal mobile devices

keeping the user private information safe on them,

and it locates the desired products over each super-

market’s map. We believe that such a system could

be complementary to the wGO and could be used in

combination to further improve customers’ shopping

experiences.

3 EXISTING LEGISLATION

In a joint effort started in 1995, the United Na-

tions Economic Commission for Europe (UNECE)

and IFR, engaged in working out a preliminary ser-

vice robot deﬁnition and classiﬁcation scheme, which

has been absorbed by the current ISO Technical Com-

mittee 184/Subcommittee 2 resulting in a novel ISO-

Standard 8373 which became effective in 2012 (for

Standardization, 2 03).

There, a robot is an actuated mechanism pro-

grammable in two or more axes with a degree of au-

tonomy, moving within its environment, to perform

intended tasks. Autonomy in this context means the

ability to perform intended tasks based on current

state and sensing, without human intervention.

A service robot is a robot that performs useful

tasks for humans or equipment excluding industrial

automation application.

A robot system is a system comprising robot(s),

end-effector(s) and any machinery, equipment, de-

vices, or sensors supporting the robot performing its

task (of Robotics, 2017).

Being an autonomous and self-driven shopping

cart, designed to follow people with or without re-

duced mobility in commercial surfaces, wGO follows

under the service robot category. Thanks to the sen-

sors (RGBD cameras and LRF) wGO detects and

identiﬁes its user in less than 2 seconds, he just needs

to push the “start” button and the wGO will start fol-

lowing him. Furthermore, distance sensors, RGBD

cameras and LRF allow wGO to identify and avoid

any obstacle along the way.

The international standard IEC 60950-1 gives the

general requirements for the safety of Information

technology equipments (Commission, 2005). As

stated, it is essential that designers understand the

underlying principles of safety requirements in or-

der that they can engineer safe equipment. Designers

should take into account not only normal operating

conditions but also likely fault conditions, consequen-

tial faults, foreseeable misuse and external inﬂuences.

The standard also assumes that users will not in-

tentionally create a hazardous situation. The priorities

depicted in Figure 2 should be observed in determin-

ing what design measures to adopt.

Figure 2: Priorities for design measures to adopt (Commis-

sion, 2005). inp stands for “if not possible”.

The application of a safety standard is intended to

reduce the risk of injury or damage due to: electric

shock, energy related hazards, ﬁre, heat related haz-

ards, mechanical hazards, radiation or chemical haz-

ards.

Here, we are mostly interested in the mechanical

hazards, whose injuries may result from: sharp edges

and corners, moving parts, equipment instability, ﬂy-

ing particles.

Suggestions of measures to reduce risks include:

rounding of sharp edges and corners, guarding, provi-

sion of safety interlocks, providing sufﬁcient stability

to free standing equipment. May also comprise se-

lecting cathode ray tubes and high pressure lamps that

are resistant to implosion and explosion respectively,

and provision of markings to warn users where access

is unavoidable.

Concerning Stability, the standard (Commission,

2005) states that under conditions of normal use, units

and equipment shall not become physically unstable

to the degree that they could become a hazard to

an operator or to a service person. Compliance is

checked by the following tests, where relevant. Each

test is conducted separately. During the tests, con-

Functionalities and Requirements of an Autonomous Shopping Vehicle for People with Reduced Mobility

375

tainers are to hold the amount of substance within

their rated capacity producing the most disadvanta-

geous condition.

• A unit having a mass of 7 kg or more shall not fall

over when tilted to an angle of 10

◦

from its normal

upright position.

• A ﬂoor-standing unit having a mass of 25 kg or

more shall not fall over when a force equal to 20%

of the weight of the unit, but not more than 250 N,

is applied in any direction except upwards, at a

height not exceeding 2 m from the ﬂoor.

• A ﬂoor-standing unit shall not fall over when a

constant downward force of 800 N is applied at

the point of maximum moment to any horizontal

surface of at least 125 mm by at least 200 mm, at

a height up to 1 m from the ﬂoor. The 800 N force

is applied by means of a suitable test tool having a

ﬂat surface of approximately 125 mm by 200 mm.

The downward force is applied with the complete

ﬂat surface of the test tool in contact with the

Equipment Under Test (EUT); the test tool need

not be in full contact with uneven surfaces (for

example, corrugated or curved surfaces).

Moreover, materials and components used in the

construction of equipment should be selected and ar-

ranged so that they can be expected to perform in a

reliable manner for the anticipated life of the equip-

ment.

Risk assessment as given in ISO14121-1 (for

Standardization, 2007) is depicted in Fig. 3. The ﬁrst

level concerns the severity of damage of injury:

• S1: Reversible, e.g. medical treatment or ﬁrst aid

required

• S2: Irreversible, e.g. loss or breaking of limbs

The second level describes the frequency and/or du-

ration of exposure to hazard:

• F1: 1 day up to 2 weeks; 2 weeks up to 1 year

• F2: Less than 1 hour; 1 hour up to 1 day

The last level is the possibility of avoiding the hazard

• A1: Possible, probable

• A2: Impossible

4 wGO: DESIGN

REQUIREMENTS

Several of the robot desired functionalities and re-

quirements were taken into consideration when de-

signing the wGO. These include:

Figure 3: Risk assessment as per (for Standardization,

2007).

• The robot should not have sharp edges

• The bag should be accessible to every type of user

• Product placement in the bag should be easy

• The robot should be able to carry at least 20 Kg

• The Start/Stop button should be well identiﬁed

and accessible

• The Emergency button should be well identiﬁed

and accessible

• There should be redundancy in the sensors

• The robot should detect and follow people with

1.3 m height or more

• The robot should be easy to clean

Since not every type of user would be able to reach

a touch screen (e.g. users in wheel chairs), a further

requirement was speciﬁed that teh wGOs operation

should not depend on a touch interface.

5 wGO: DESIGN EVOLUTION

The wGO is designed to have an ergonomic shape,

friendly both to the target users (people with reduced

mobility) and the environment (commercial retail en-

vironment).

It has, however, undergone an evolution, as can be

seen in Fig. 4.

Figure 4: wGO evolution.

VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems

376

The design on the left of Fig. 4 was the initial pro-

posal (v0.1). From that version to the next (v0.2) sev-

eral alterations were implemented, namely:

• LRF position was changed: It was noted that the

different height would make visible more of the

obstacles typically present in a commercial site.

Moreover, stability of the support assembly easi-

ness was increased.

• Smaller base: A smaller increases maneuverabil-

ity. It cannot, however, be too small due to stabil-

ity issues which might jeopardize safety.

• Emergency button was repositioned: The emer-

gency button needs to be visible and of easy ac-

cess. It cannot, however, be positioned in a place

where it can be accidentally triggered. A new lo-

cation was found that complies with these require-

ments.

• Changes to the electronic box: Some changes to

the electronic box were performed in order to in-

crease safety. For example, the box was electri-

cally isolated.

• Redesign of the bag holder support: Angular ver-

tices were smoothed in order to make the product

less prone to accidents.

From that version to the next (v0.3) most of the

changes were related to the sensors:

Change of the LRF. Initial tests revealed that the

LRF (RP Lidar) had difﬁculties in seeing black. A

better LRF was thus chosen to minimize this problem.

Change of the Active Camera. Kinect 2 was found

to be too resource demanding. A different active cam-

era, with lower speciﬁcation was tested and found to

be enough for our application, at the gain of being less

resources consuming.

Inclusion of a Pan and Tilt System. A pan and tilt

system was included so that the active camera respon-

sible for performing tracking would better follow the

user.

Inclusion of Two Additional Active Cameras.

Two active cameras were added in order to decrease

the dead area and have a better obstacle avoidance

performance.

Changes in the Structure. The middle main struc-

ture was redesigned in order to be thinner (a sufﬁ-

ciently wide diameter to allow the required cables to

pass through it) and less heavy.

Reorganization of the Interior of the Base. The

base was redesigned in order to accommodate two

batteries and thus increase the energetic efﬁciency of

the robot.

Head Redesign. The inclusion of the additional ac-

tive cameras forced a complete redesign of the head

of the robot in order to accommodate them.

6 wGO: THE RISKS

Safety is a critical characteristic for robots designed

to operate in human environments. Looking back at

the risk assessment as given in ISO14121-1 (for Stan-

dardization, 2007) and depicted in Fig. 3, we observe

that:

• the severity of damage of injury is irreversible

(e.g. breaking of limbs) - S2

• the frequency and duration of exposure to hazard

is in the order of 30 minutes, twice a week (typical

duration and frequency of a shopping experience)

- F1

• it is impossible to avoid the hazard - A2

In this way, wGO is categorized in the d level.

The two main risks associated with the wGO are:

• To turn due to someone hanging on it

• Collision with people

Besides the user manual and the instructions on

how to use the robot properly, several of the design

options had in mind the minimization of the identiﬁed

risks.

Changes in the structure included in v0.3 made the

robot less heavy on its top part, giving it a lower centre

of gravity. The addition of one battery, besides the

obvious increase in the robot autonomy, also had the

side effect of adding weight to the base making the

robot even more stable and harder to turn.

Collision avoidance is assured mostly by the use

of redundant sensors (Ponz et al., 2016). Their

existence, type and location has underwent several

changes. The change in the LRF position in v0.2, for

instance, made visible more of the obstacles typically

present in a commercial site. The change in the LRF

itself also allowed the detection of darker obstacles

(people wearing a black suit for instance). The addi-

tion of two active cameras in v0.3 is another example

of redundancy of sensors to detect obstacles.

Finally, there is an emergency button, clearly vis-

ible and positioned in a place with easy access for ev-

ery type of user in case an unpredicted situation hap-

pens. It is important to note that this button has not yet

Functionalities and Requirements of an Autonomous Shopping Vehicle for People with Reduced Mobility

377

been used in any of the internal, external, controlled

or uncontrolled tests.

7 RESULTS

This section summarizes the demonstration of wGO

in a relevant, unconstrained scenario (Figure 5). Two

wGOs were available for the tests and only users with

reduced mobility were asked to participate.

Figure 5: wGO in a relevant scenario.

Transportation of the wGOs from their production

site to the destination where the demonstration (more

than 1800 km) was made by truck with the robots ac-

commodated in disposable plywood boxes.

Concerning the location description, the site had

enough space for circulation, and there were no areas

where the wGO did not ﬁt. Its use was, however, re-

stricted to days where the store was too crowded. The

hypermarket had locations with several different lev-

els of brightness and the main corridor’s ceiling had

big skylights.

143 clients tested the wGO during two weeks of

demonstration. An average of 14 users a day used

the wGO with the exception, as already mentioned,

of days where the store was too crowded.

An average of 35 minutes was spent per trip, with

most users being female (Fig. 6).

Figure 6: Population gender.

Most of the users had ages between 25 and 36,

with the second most represented class people with

55 or more years (Fig. 7).

Figure 7: Population age.

The categorization of users is shown in Fig. 8. As

can be seen most of the volunteers presented some

type of reduced mobility. Among those, parents with

a baby stroller are the most represented followed by

people in wheelchair and the elderly.

Figure 8: User’s categorization.

A small questionnaire was made to the partici-

pants with the following questions:

• Do you ﬁnd the wGO more agreeable than the al-

ternatives?

• Would the wGO be a reason for you to come to

this store?

• Would you reuse the wGO?

• Would you recommend the wGO?

• From a scale of 0 to 4 (being 0 not al all and 4

completely), how satisﬁed are you with the wGO?

Results are presented in Figs 9, 10, 11, and 12, respec-

tively. It can be seen that the vast majority of the users

ﬁnd wGO better than existing alternatives. More than

64% would ﬁnd it a reason to return to this particular

store. More than 90% would reuse the wGO and more

than 97% would recommend it. Average satisfaction

was 3.5 out of 4.

7 out of the 143 have returned (within the short pe-

riod of the demonstration) to use the wGO. Of these,

5 were male and 2 female.

Comments from the users included:

Detection and Identiﬁcation: at times, wGO follows

someone else; the wGO should have the ability to fol-

low the customer in crowded environments; a passive

bracelet or plotter should be provided so that wGO

will follow one person; and the wGO should function

in exterior light.

VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems

378

Figure 9: Do you ﬁnd wGO

more agreeable than the al-

ternatives?

Figure 10: Would the wGO

be a reason for you to come

to this store?

Figure 11: Would you

reuse the wGO?

Figure 12: Would you rec-

ommend the wGO?

Visibility: to be recognizable by the customer, and

more visible to others (e.g. beacon, light signal); to

show when it is on or off; and increase the volume.

Capacity: larger bag; support for heavy products,

such as water bottles; and hook for personal items

(e.g. handbag).

Usability: to go faster and to shorten the distance be-

tween the wGO and the user.

Movement: more ﬂuidity, especially in narrow pas-

sages and angles and reactivity in crowded environ-

ments.

Usage Restrictions: to go outside and to go to the

car’s trunk.

Other Features: to call a collaborator; to be able to

scan products, know the price and remaining bag ca-

pacity; system for fetching products; speech recogni-

tion (dialogue with wGO); guide the customer in the

store.

8 CONCLUSION

In this paper the wGOs evolution is presented. A high

focus was given to security concerns and its implica-

tion the design of the robot.

A user study made with 143 volunteers was also

analysed.

Several positive aspects were pointed out, such as

the high interest from the customers (some came ex-

pressly to test the wGO); positive general opinion; ex-

istence of loyal customers; technical satisfaction of

87% (versus 73% on previous tests).

There were also, some points to be improved:

• During the process, the wGO should not change

the user

• Difﬁcult to move in crowded environments

• The wGO does not work outside

To attack these problems, several measures will be

taken. Improvements on the identiﬁcation algorithm

will be made. This will both help with the ﬁrst and

second points to be improved. When identifying bet-

ter the user, the wGO will not start following a differ-

ent person and will behave better in crowded environ-

ments by, again, always following the same user.

Concerning the exterior lightning, new sensors

that are able to acquire depth information in the ex-

terior will be analysed. At the same time, new algo-

rithms that do not make use of depth information and

only the RGB will be studied.

9 OTHER APPLICATIONS

A great part of Follow Inspiration’s R&D strategy

is based on the wGO technology that can be ap-

plied to several scenarios, including industry, for ma-

terial handling and transportation for the automa-

tive industry. As consequence, a new project were

started in 2015 in order to develop a robot ca-

pable of performing tasks in an autonomous way

through an industrial environment (NORTE-01-0247-

FEDER-011109). New developments have been

made in mapping, localization (Andry Maykol Pinto

and Moreira, 2014), navigation (Costa et al., 2016)

and multi-robot cooperation algorithms (Santos et al.,

2015). The application of autonomous vehicles in

these scenarios has several advantages in the LEAN

process offering ﬂexibility, reducing times and there-

fore the optimization of the operational costs.

Figure 13: wGO Technology applied in an industrial sce-

nario.

Functionalities and Requirements of an Autonomous Shopping Vehicle for People with Reduced Mobility

379

ACKNOWLEDGEMENTS

The authors would like to thank all the institu-

tional supporters, including TIC RISCO, Portugal

Capital Ventures, SGPS, S.A, CEiiA, Fund

ao City

Hall. We would also like to thank to R&D entities

including, Tente, CEiiA, Centimfe and Alma De-

sign. These developments were undertaken under

several EU cofunding programs, namely: QREN

(projects: CENTRO-07-0201-FEDER-023962 and

CENTRO-07-0202-FEDER-024692); Portugal 2020

(projects: PFOCI-01-0247-FEDER006398 and

NORTE-01-0247-FEDER-011109).

REFERENCES

Andry Maykol Pinto, P. G. C. and Moreira, A. P.

(2014). Architecture for visual motion perception of a

surveillance-based autonomous robot. In IEEE Inter-

national Conference on Autonomous Robot Systems

and Competitions (ICARSC).

Australia, G. o. W. (2016). Section ﬁve access and inclu-

sion. Disability access and inclusion plan - training

package, Disability Services Comission. Accessed:

2016-11-18.

Commission, I. E. (2005). Part 1: General requirements.

Information technology equipment Safety.

Costa, P., Moreira, N., and Campos, D. (2016). Local-

ization and navigation of an omnidirectional mobile

robot: The robot@factory case study. IEEE Re-

vista Iberoamericana de Tecnologias del Aprendizaje,

pages 1–9.

CRPG and ISCTE (2007). Elementos de caracterizacc

das pessoas com deﬁci

encias e incapacidades em

portugal. In Programa Operacional de Assist

encia

ecnica ao QCA III: eixo FSE.

de Matos, L. (2012).

Faria, B., Reis, L., and Lau, N. (2014a). User modeling

and command language adapted for driving an intelli-

gent wheelchair. In IEEE International Conference on

Autonomous Robot Systems and Competitions, pages

158–163.

Faria, V., Silva, J., Martins, M., and Santos, C. (2014b).

Dynamical system approach for obstacle avoidance in

a smart walker device. In IEEE International Con-

ference on Autonomous Robot Systems and Competi-

tions, pages 261–266.

for Standardization, I. O. (2007). Safety of machinery – risk

assessment – part 1: Principles. Standard, ISO.

for Standardization, I. O. (2012-03). Robots and robotic

devices – vocabulary. Standard, ISO.

Gmez-Goiri, A., Castillejo, E., Ordua, P., Laiseca, X., Lpez-

de-Ipia, D., and Fnez, S. (2011). Easing the mobil-

ity of disabled people in supermarkets using a dis-

tributed solution. In Bravo, J., Hervs, R., and Vil-

larreal, V., editors, Ambient Assisted Living, number

6693 in Lecture Notes in Computer Science, pages

41–48. Springer Berlin Heidelberg.

Gurgel Pinheiro, P., Cardozo, E., and Gurgel Pinheiro,

C. (2015). Anticipative shared control for robotic

wheelchairs used by people with disabilities. In IEEE

International Conference on Autonomous Robot Sys-

tems and Competitions, pages 91–96.

Iezzoni, L. I., McCarthy, E. P., Davis, R. B., and Siebens,

H. (2001). Mobility difﬁculties are not only a prob-

lem of old age. Journal of General Internal Medicine,

16(4):235–243.

Martin Rico, F., Rodriguez Lera, F., and Matellan Olivera,

V. (2014). Myrabot+: A feasible robotic system for

interaction challenges. In IEEE International Con-

ference on Autonomous Robot Systems and Competi-

tions, pages 273–278.

Neves, A., Campos, D., Domingues, I., Santos, J., Leo, J.,

Xavier, J., de Matos, L., Camarneiro, M., Penas, M.,

Miranda, M., Morais, R., Silva, R., and Esteves, T.

(SUBMITTED 2017). A personal robot as an im-

provement to the customers in-store experience. In-

Tech.

of Robotics, I. F. (2017). Deﬁnition of service robots.

http://www.ifr.org/service-robots/. Accessed: 2017-

01-23.

Pavon-Pulido, N., Lopez-Riquelme, J., Pinuaga-Cascales,

J., Ferruz-Melero, J., and Morais Dos Santos, R.

(2015). Cybi: A smart companion robot for el-

derly people: Improving teleoperation and telepres-

ence skills by combining cloud computing technolo-

gies and fuzzy logic. In IEEE International Con-

ference on Autonomous Robot Systems and Competi-

tions, pages 198–203.

Ponz, A., Rodrguez-Garavito, C. H., Garca, F., Lenz, P.,

Stiller, C., and Armingol, J. M. (2016). Automatic ob-

stacle classiﬁcation using laser and camera fusion. In

International Conference on Vehicle Technology and

Intelligent Transport Systems.

Rodriguez, V. J. L., Fernandez, L. M., Hansen, M. M., Mod-

rego, C. E., Peligero, F. J. A., Torijano, J. L., Guallar,

E. E., Cornudella, B. S., and Verdejo, I. G. (2014).

Shopping cart. Justia Patent.

Santos, J., Costa, P., Rocha, L., Moreira, A., and Veiga,

G. (2015). Time enhanced a*: Towards the develop-

ment of a new approach for multi-robot coordination.

IEEE International Conference on Industrial Technol-

ogy (ICIT).

VEHITS 2017 - 3rd International Conference on Vehicle Technology and Intelligent Transport Systems

380