19-08-2014, 02:18 PM
Blind Assistant Navigation System Seminar Report
[attachment=66754]
ABSTRACT
This paper presents the architecture as well as the implementation
of a system that helps blind person navigate independently within
an enclosed environment such as the home. The system uses a
wireless mesh network to provide the first level localization. It
also incorporates additional components to provide more refined
location and orientation information. Optimal path planning is
done by a server that communicates wirelessly with the portable
mobile unit that can be pushed by the blind person. The blind
person issues commands and receives direction responses using
audio signals.
INTRODUCTION
Blindness or visual impairment is a condition that affects many
persons around the world. This condition leads to the loss of the
precious sense of vision to such a degree that makes the concerned
person handicapped due to the need for guidance or assistance and
in some cases special treatment. Guidance by other humans, with
good vision, or specially trained dogs is an obvious solution to
help blind persons navigate their way around both in the house as
well as outside the house. However, dependence on other humans
is highly demanding and constraining in many ways. Trained
dogs are very helpful however they have limitations that include
inability to interpret what the blind persons really wants and
identifying objects. This is in addition to the continuous care cost
for the dog.
Some technological solutions have been introduced recently to
help blind persons navigate independently. Many of those
solutions rely on Global Positioning System (GPS) technology to
identify the position and orientation of the blind person. While
such systems are suitable for outdoor navigation, due to the need
for line of sight access to satellites, they still need additional
components to improve on the resolution and proximity detection
to prevent collision of the blind persons with other objects and
hence subject his/her life to danger. The use of robot-dog is
another technological solution proposed by a number of
researchers. The robot-dog is an attempt to replace the real dog
[1]. It also depends on GPS and incorporates objects avoidance
technologies. These solutions are useful, however they can only
be used outdoor and miss interpretation of the blind person
SYSTEM CONCEPT
The elements required to perform the guidance process include
defining the destination or target, identifying the current position
of the blind person and finally determining the best path to be
taken to reach the desired destination. In order to identify user
position, the guidance system utilizes ZigBee based localization
engine technique that continuously updates the server with the
user location [2]. A digital compass located in the push mobile
cart enables the system to identify the user orientation. The
proximity sensors incorporated in the mobile push cart enable the
detection of obstacles. The user of the guidance system pushes the
cart that houses hardware components in front of him/her while
walking. The cart rolls on passive wheels that support its weight
during regular operation. Also the wheels are equipped with
encoders to determine the relative motion of the user. This
information is used to refine the system localization process
System I/O
Figure 1 shows the top level view of the guidance system. As can
be seen from this diagram, the parameters that are inputted to the
system are User Location, User Request and User Orientation.
The guidance system will update the location of the blind person
with respect to a virtual map in real time. The user can enter
his/her destination verbally using a set of well predefined
locations or objects. After this a series of computational
operations will take place inside the guidance system to generate
the speech commands to direct him to the requested location or
object.
Hardware
The set of components that have been utilized to develop such
system are listed in Figure 5. The major hardware oriented
components are the ZigBee modules and speech recognition and
synthesis modules. A brief description of each component is
provided below.
ZigBee CC2431: This device applies ZigBee localization
technique to estimate the position of the blind node that is
mounted on the Guiding Stick Wheel. The location algorithm
used in the CC2431 location engine is based on Received
Signal Strength Indicator (RSSI) values. The RSSI value will
decrease when the distance increases.
Software
An interface is designed to enable the system administrator to
debug the system, or monitor the movement of the blind person
within the house. In addition, the obtained information can be sent
to E-Health service center in order to diagnose the condition of the
blind person. For example, if there was no sign of activity for a
given period of time, let’s say 15 hours, they would send a
medical team to check the health status of the blind person.
Although the system requires a server; however, the
computational process is done in real-time. Figure 9 shows that
the user icon in the interface continuously follows the location of
the blind node which makes the guidance procedures more
accurate. Also, it can be seen from the interface that it has a drag
and drop menu that enables the system to easily evolve with
respect to the user’s needs. Moreover, one of its features, it
supports the introduction of new desired destinations if the user
wishes to alter the list of objects. Finally, the path-planning
algorithm is also handled by the software, where Re-active path
planning method is used to connect between the user and the
desired target. This algorithm is simply connecting between two
points which are the user and the desired target. However, for
enhancement the system has the ability to adapt any complex path
planning algorithm i.e distance transform method or potential field
algorithm.
CONCLUSIONS
This paper presented the architecture and implementation of a
system that assists a blind person to navigate inside an enclosed
environment such as the home. The system can be considered as a
semi-autonomous device. It provides full autonomy for global
navigation (path-planning & localization), but relies on the skills
of the user for local navigation (Obstacle avoidance).
This device offers innovative solutions in order to replace the
conventional methods of guiding visually impaired person. In
addition, it can be easily applied anywhere where it can handle
places like malls or airports. This system will allow the visually
impaired to wander freely and independently.
The system described in the paper is at the prototype stage and
hence there are many options for improvements. Some of the
improvements are
[attachment=66754]
ABSTRACT
This paper presents the architecture as well as the implementation
of a system that helps blind person navigate independently within
an enclosed environment such as the home. The system uses a
wireless mesh network to provide the first level localization. It
also incorporates additional components to provide more refined
location and orientation information. Optimal path planning is
done by a server that communicates wirelessly with the portable
mobile unit that can be pushed by the blind person. The blind
person issues commands and receives direction responses using
audio signals.
INTRODUCTION
Blindness or visual impairment is a condition that affects many
persons around the world. This condition leads to the loss of the
precious sense of vision to such a degree that makes the concerned
person handicapped due to the need for guidance or assistance and
in some cases special treatment. Guidance by other humans, with
good vision, or specially trained dogs is an obvious solution to
help blind persons navigate their way around both in the house as
well as outside the house. However, dependence on other humans
is highly demanding and constraining in many ways. Trained
dogs are very helpful however they have limitations that include
inability to interpret what the blind persons really wants and
identifying objects. This is in addition to the continuous care cost
for the dog.
Some technological solutions have been introduced recently to
help blind persons navigate independently. Many of those
solutions rely on Global Positioning System (GPS) technology to
identify the position and orientation of the blind person. While
such systems are suitable for outdoor navigation, due to the need
for line of sight access to satellites, they still need additional
components to improve on the resolution and proximity detection
to prevent collision of the blind persons with other objects and
hence subject his/her life to danger. The use of robot-dog is
another technological solution proposed by a number of
researchers. The robot-dog is an attempt to replace the real dog
[1]. It also depends on GPS and incorporates objects avoidance
technologies. These solutions are useful, however they can only
be used outdoor and miss interpretation of the blind person
SYSTEM CONCEPT
The elements required to perform the guidance process include
defining the destination or target, identifying the current position
of the blind person and finally determining the best path to be
taken to reach the desired destination. In order to identify user
position, the guidance system utilizes ZigBee based localization
engine technique that continuously updates the server with the
user location [2]. A digital compass located in the push mobile
cart enables the system to identify the user orientation. The
proximity sensors incorporated in the mobile push cart enable the
detection of obstacles. The user of the guidance system pushes the
cart that houses hardware components in front of him/her while
walking. The cart rolls on passive wheels that support its weight
during regular operation. Also the wheels are equipped with
encoders to determine the relative motion of the user. This
information is used to refine the system localization process
System I/O
Figure 1 shows the top level view of the guidance system. As can
be seen from this diagram, the parameters that are inputted to the
system are User Location, User Request and User Orientation.
The guidance system will update the location of the blind person
with respect to a virtual map in real time. The user can enter
his/her destination verbally using a set of well predefined
locations or objects. After this a series of computational
operations will take place inside the guidance system to generate
the speech commands to direct him to the requested location or
object.
Hardware
The set of components that have been utilized to develop such
system are listed in Figure 5. The major hardware oriented
components are the ZigBee modules and speech recognition and
synthesis modules. A brief description of each component is
provided below.
ZigBee CC2431: This device applies ZigBee localization
technique to estimate the position of the blind node that is
mounted on the Guiding Stick Wheel. The location algorithm
used in the CC2431 location engine is based on Received
Signal Strength Indicator (RSSI) values. The RSSI value will
decrease when the distance increases.
Software
An interface is designed to enable the system administrator to
debug the system, or monitor the movement of the blind person
within the house. In addition, the obtained information can be sent
to E-Health service center in order to diagnose the condition of the
blind person. For example, if there was no sign of activity for a
given period of time, let’s say 15 hours, they would send a
medical team to check the health status of the blind person.
Although the system requires a server; however, the
computational process is done in real-time. Figure 9 shows that
the user icon in the interface continuously follows the location of
the blind node which makes the guidance procedures more
accurate. Also, it can be seen from the interface that it has a drag
and drop menu that enables the system to easily evolve with
respect to the user’s needs. Moreover, one of its features, it
supports the introduction of new desired destinations if the user
wishes to alter the list of objects. Finally, the path-planning
algorithm is also handled by the software, where Re-active path
planning method is used to connect between the user and the
desired target. This algorithm is simply connecting between two
points which are the user and the desired target. However, for
enhancement the system has the ability to adapt any complex path
planning algorithm i.e distance transform method or potential field
algorithm.
CONCLUSIONS
This paper presented the architecture and implementation of a
system that assists a blind person to navigate inside an enclosed
environment such as the home. The system can be considered as a
semi-autonomous device. It provides full autonomy for global
navigation (path-planning & localization), but relies on the skills
of the user for local navigation (Obstacle avoidance).
This device offers innovative solutions in order to replace the
conventional methods of guiding visually impaired person. In
addition, it can be easily applied anywhere where it can handle
places like malls or airports. This system will allow the visually
impaired to wander freely and independently.
The system described in the paper is at the prototype stage and
hence there are many options for improvements. Some of the
improvements are