The command C-o open-line that others have suggested is not quite the same as o in vi, because it splits the current line and lets the cursor remain in the current line.
You get the exact same effect as vi’s o with two strokes: C-e RET, which moves the cursor to the end of the current line and then inserts a new line, which leaves the cursor at the beginning of that line.
You could bind that sequence to a key of its own (perhaps overriding the existing definition of C-o), but I doubt if it’s worth the trouble.
(Incidentally, the symmetric sequence C-a RET gives you the effect of vi’s capital O, inserting a line before the current line.)