zlacker

If you can detect it early enough an absolutely miniscule amount of force can multiply over time to an enormous difference in position later on. Much less force than would be required to stop it from spinning or sending material to act as a deflector.

>>zamada+(OP)
You would want to design and test a vectoring nuclear thermal rocket built to absorb neutrons and reduce thrust on one side, and burn 100% on the other side, with the ability to gimbal the thrust vector depending on the asteroid's orientation. This gives you a much bigger margin of error versus "white paint" or conventional rockets imo.

You probably wouldn't want to nuke it, because you risk buck shotting the earth with a cloud of asteroids. Having a vectored nuclear rocket also allows you to change your trajectory if your initial calculations are off.

>>willma+y72
The other problem is propellent. I think using the material of the asteroid itself as propellent makes the most sense.