Kalman filtering for non-linear problems

Hi,

I am doing a sonar tracking problem in which my state equations are
non-linear. I am aware of two Kalman filtering techniques to tackle
non-linear problems : EKF and the unscented filter. I was wondering
what are the respective  advantages and disadvantages of using an
extended Kalman filter and unscented filter? Are there any other
methods in adapting the Kalman filter to non-linear problems? Are their
any articles/literature surveyscomparing the different ways to adapt a
Kalman filter to a non-linear problems. Any information would be great,
thanks.