Kinetic energy harvesting has been the subject of a significant research effort in the last twenty years. Unfortunately most of the energy available at the microscales comes in the form of random vibrations with a wide spectrum of frequencies while standard harvesting methods are based on linear oscillators that are resonantly tuned in narrow frequency ranges. In this paper we present a novel approach based on the exploitation of nonlinear stochastic dynamics and show that, under
proper conditions nonlinear oscillators can beat the standard linear approaches with significant increase in the harvesting efficency. For the sake of demonstration we present experimental results from a toy-model bistable oscillator made by a piezoelectric inverted pendulum