Kinetic energy harvesting: Everyday human activity could power the internet of things

Researchers at Columbia University have conducted the first exhaustive study into kinetic energy harvesting — the harvesting of “free” energy from common human activities, such as walking, writing with a pencil, taking a book off a shelf, or opening a door. Surprisingly, except for those living the most sedentary lifestyles, we all move around enough that a kinetic energy harvester — such as a modified Fitbit or Nike FuelBand — could sustain a wireless network link with other devices, such as a laptop or smartphone.
Energy harvesting is expected to play a very important role in the future of wearable computing and the internet of things, where direct sources of power — such as batteries or solar power — are cumbersome, expensive, and unreliable. At its most basic, a kinetic/inertial energy harvester is a small box with a weight attached to a spring. When the spring moves, the mechanical energy is converted into electrical energy, usually by means of piezoelectrics or MEMS (microelectromechanical systems). If the spring moves with more force, or it bounces back and forth rapidly, more energy is produced.
As you can imagine, some human movements produce more harvestable energy than others, with periodic motions — i.e. repetitive left/right, up/down, back/forth motions — in particular being the key. This is illustrated by the researchers’ finding that writing with a pencil or opening a drawer produces more harvestable energy (10-30 microwatts) than a plane flight at its most turbulant intervals (5 microwatts). For comparison, walking produces somewhere in the region of 100-200 microwatts. The researchers found that intentionally shaking an object, as demonstrated by shake flashlights, creates more than 3,000 microwatts (3 milliwatts).
In the table to the right, the microwatt (µW) figures are from the point of view of the object; as in, the object itself is equipped with an inertial energy harvester. As you can see, except for writing with a pencil (periodic movements), opening a drawer (a strong pull), or shaking an object, most of our interactions with the environment do not produce a lot of energy. According to the researchers this is exacerbated by objects, such as doors and drawers, being dampened for human comfort. If the inertial harvester could be placed in the damper, at the time of production, much more energy could be produced.
The table below shows the amount of energy that humans produce as they go about their everyday lives, relaxing, walking, running, and cycling. Again, you see that the vigorous periodic motion of walking and running produces a lot of energy. The cycling figure, which is very low (10 µW), would be much higher if the harvester was placed lower on the leg.

In the paper, the researchers highlight some other interesting and counterintuitive discoveries: Climbing down stairs generates more energy than going up, due to faster and larger limb movements; push-ups and sit-ups produce less harvestable energy than walking at a normal pace; and that, regardless of placement, each harvester produces similar amounts of energy. An energy harvester in your shirt pocket produces as much power as one in your trouser pocket. Curiously, taller people produce around 20% more harvestable energy than short people — and the weight of the participant also played a role.
All in all, this paper provides invaluable research for the creation and efficient placement of energy harvesting devices. In almost every case, every participant in the study produced enough energy to transmit a steady stream of wireless data to a nearby device — via Bluetooth, say, to a smartphone that manages your body-area sensor network. They also show that some objects, such as drawers, books, and doors might be able to harvest enough energy to wirelessly connect themselves to an internet of things. The first step, which I’m sure some companies are already working on, is to bring kinetic energy harvesting to devices like the Nike FuelBand or Fitbit, where the simple act of being used probably produces enough power to keep these devices charged.
Looking to the future, with advancements to kinetic energy harvesting, and perhaps in concert with RF energy harvesting, a battery-free internet of things begins to sound surprisingly feasible.