The only errors are in the form of misreading numbers or misidentifying what you're looking at - with more than 500,000 known objects, there's bound to be a few 'discovered' twice or more. That's why it's important to get as many observations over time as possible. The more you see an asteroid drift through space, the more you know about its orbit, and can use mathematics that are older than the telescope in order to find out where they're heading.

Have NASA ever been wrong about the direction of asteroids.

They've refined predictions over the course of time. But wrong? The laws of physics don't change for asteroids, so no.

Toutatis has been known about since 1934. NASA and all the other entities and individuals capable of watching it are not wrong.

Meaning have we ever encountered way before the day they predicted.

No.

Can the asteroid suddenly make a turn back for.

No.

I dont anything about this stuff, these might be crazy questions to you

There are lots of things humans haven't figured out. The universe is huge and mysterious, and there will likely never be a shortage of new things to learn. One thing we do know a lot about and are very good at, however, is making observations and calculating orbits to a high degree of precision.

Totanious. I think thats how u say it

Er, that would be Toutatis, and it was discovered in 1934, closest recent approach was Dec. 12, 2012 at a distance of 18 LD (that's 18 times the distance to the Moon, or 6,919,218 km). It's mentioned on the Asteroids page.

Scientists can figure out where objects will go using precise observations and math. It works.

Ok, I'm not even going to answer this. Google it, seriously, one little google search, hit Wikipedia and read. Where in Gods name are you getting your info?