Not a bad choice, but, as digimike suggests, approach it in stages.
Learn how to connect and control the motors that move the robot BEFORE putting them in the chassis.
Learn how to move the servo that holds the sensor, before attaching the sensor or mounting the servo.
Learn how to use the sensor, and what to make of the data you get back, before attaching the sensor to the servo.
Then, learn how to make the robot simply roll forward 5 feet and stop. Then, make it roll forward 5 feet, stop, and back up 5 feet.
Then, forward, turn left, and stop.
Add capabilities one at a time. Get each one working, then put the working code in a function.
You should be able to make the robot go forward until it sees an obstruction with a single function call.
When that function ends, you know that the robot is stopped because there is something in its way.
Then, you can scan to determine if turning left or right is appropriate. Create a scan function that returns -1, 0, or +1 (for clear(er) to the left, straight, or right), and call the appropriate turn function.
Doing this will make it much easier to debug a function (my robot does not turn left correctly) than if you have an 11,000 line loop method.