EEMCS

Full text available as:

PDF 7138 Kb

Abstract

In this thesis, we explore the potential of WSAN technology in dynamic applications requiring a senseand-
react control loop by considering the problem of distributed movement coordination of vehicles on wheels.
The final goal is to obtain a self-organizing group (or swarm) of mobile nodes that maintain a formation by
periodically exchanging sensed movement information. Such a coordinated team of mobile wireless sensor and
actuator nodes can bring numerous benefits for various applications in the field of cooperative surveillance,
mapping unknown areas, disaster management, automated highway and space exploration.
The work we present in this thesis is considers a pair of vehicles that synchronize their movements,
thus trying to maintain a specific formation. We use toy cars as prototype vehicles for this research. One
of the two cars acts as the leader, being remotely controlled by the user, and the other one represents
the follower, which implements a control loop for mimicking the leader’s movements. Each vehicle has a
low-power wireless sensor node attached, featuring a 3D accelerometer and a magnetic compass. Velocity
and heading are computed on both vehicles in real time using inertial navigation techniques. The leader
periodically transmits its measurements to the follower, which implements a lightweight fuzzy logic controller
for imitating the leader’s movement pattern. This solution is not restricted to vehicles on wheels, but
could support any moving entities capable of determining their velocity and heading. We call our initial
implementation FollowMe. We report in detail on all development phases, covering design, simulation,
implementation and testing. We perform the tests on our university’s running track and hockey field.
The simulations and tests show that the system exhibits the desired leader-follower behavior, using
just the compact, low-cost wireless sensors and actuators. Making use of a fuzzy controller facilitates the
implementation on resource constrained sensor nodes and handles robustly the noisy sensor data as well
as the rough mechanical capabilities of the vehicles. The vehicle simulations we devised show adequately
realistic results and provide a means to test, evaluate, tune and debug the controller. However, some issues
are still of concern, such as the effect of inclination changes on the sensor readings, the accumulating errors
seen in the velocity estimate, the sensitivity of the compass sensor to external influence and the stability and
the overall performance of the follower controller. Also, the simulations are open for improvement towards
more realistic physical models of the vehicles.

Export this item as:

• BibTeX
• XML

To correct this item please ask your editor

Repository Staff Only: edit this item