Energy Harvesting Using Motion

1
Energy Harvesting Using Motion
Center for Embedded Networked Sensing
UCLA UCR Caltech USC CSU
Mohammad Rahimi, Hardik Shah, Sandeep Babel,
Gaurav Sukhatme, John Heidemann, Deborah
Estrin www-robotics.usc.edu/robomote
- Introduction Increase the Sensor Network
Longetivity.

• Novel Approach for Energy Hunting
• There are some nodes in the network who have
  ability to charge themselves and move.
• Using motion to carry energy from low energy
  zones to high.
• Acts as an Energy distribution equalizer.

• Motivation
• Network Energy is limited.
• Environmental Energy distribution may vary.
• Current Methods are trying to seek balance
  between energy producers and consumer.
• .

Low Energy
High Energy
Service Network
Charge
Consumer -Sense -Communicate
Producer Hunt Energy
Problem Obtain Self Contained Network with
Serving Mobile Nodes.

• Network Energy
• En(t) ?tt0 (?x?y Pp(x,y,t)-Pc(x,y,t) )
• Self-contained network
• En(t) gt 0 ? tgt0
• Energy Cell Area (ECA) is the moving robot
  territory. it shows the zone of service of a
  mobile node with diagonal distance
• (Emax-Epacket) /2Emov
• Where Emax is max energy carried by node,
  Epacket max energy any node can reserve and Emov
  is energy required for movement per unit of
  distance.

• Number of Serving Robot (AN / ECA) (Pc/Pp)
  (ECA ?s)
• AN / ECA is the number of territories in the
  network
• ECA ?s Number of clients in each territory
• Pc/Pp average rate of energy consumption vs.
  Production
• Benefit Function Pc / Enav

Pc
Enav

• Robot calculates the benefit function to select
  energy source to go.

 
Solution Energy Harvesting Using Mobility in
Mobile Sensor Networks
Experiments
Results

• Experiments showed in case of maximum traffic,
  we need about 6 robots to guarantee network
  longevity.
• 0.40 robots to static nodes ratio for
  guaranteeing network lifetime.

Test Bed
Robomote

• Future Work
• Improved Robot Architecture for Mobile Sensor
  Platform
• More Number of static nodes
• More Number of robots to serve static nodes.
• Multiple charger docking stations.

• 15 static Berkeley Motes working as beacons.
• 3 Robotmotes equipped with solar panels.
• Overhead Vision based system for Ground Truth.

• Implementing a self contained ecology of sensor
  network nodes
• Studying the migration pattern of the mobile nodes

Conclusion Motion can extend the longevity of the
network or convert a limited lifetime network to
an ever lasting network.
Robomote Moving Across The Testbed
Simulation of Robots Movement.