First you need to estimate the ground state level of the nucleus. That is the kinetic energy, or p^2/2m. The ground state is stationary, so <p>=0. So using \Delta p=<(p-<p>)^2>^{1/2}, \Delta p=<p^2>^{1/2}=(2m<H>)^{1/2}.
So from the uncertainty relation \Delta p=(2m<H>)^{1/2}>\frac{\hbar}{2 \Delta x}
or <H> > \frac{\hbar^2}{8m (\Delta x)^2}
So what you know is that the potential has a value at least equal to this ground state energy, in order to be confined. So to escape the potential, the particle must have at least this energy. When it is free of the nucleus, it'll still have this energy relative to the nucleus. So at the very least, the kinetic energy is equal to \frac{\hbar^2}{8m (\Delta x)^2}, but it is actually more relative to infinity where it is free instead of relative to the nucleus. So set p^2/2m equal to \frac{\hbar^2}{8m (\Delta x)^2}, and solve for p:
p=\frac{\hbar}{2 (\Delta x)}
if the size of the nucleus is 10^-14, then the momentum is 3.3*10^(-20). If the alpha particle has mass 6.6*10^(-27), then the velocity is 1/2*10^7 m/s. I think this is close enough to the speed of light (if you had used 10^-15 meters for the nucleus instead of 10^-14 then you'd get there) where you might have to use relativity to get the speed, which means you'd set (c^2p^2+m^2c^4)^(1/2) equal to the energy, and solve for p, and then use the relativistic relationship between p and v to solve for v.
Anyways, if I didn't make a mistake, this is the long explanation of how to use the uncertainty principle.