Well the basic idea is when a touch begins you mark the coordinates on a variable(StartTouchX,StartTouchY) and the starting screen position(ScrollStartX,ScrollStartY) on another variable, then you take the current coordinates that the touch is at and the difference between the two are used in a system scrollto action to place the camera at a different position.
If you already have a scrollto behavior on an object or anything else conflicting with this camera position they will be battling each other.