Here's how I made myself stop obsessing about step-count accuracy: If you think of the steps tracked as just a number that you will compare to every day, then it doesn't matter so much if it counts a few steps while driving because (assuming most days are the same: you drive to and from work) that will be a "constant" in your overall day. So if this is included in your average day, and your more-or-less average day is, say 4,000 steps - before you intentionally decided to up your game - then it will also be a part of your new average, INTENTIONAL day of say 6,500 steps to start. See what I mean? As long as it counts in the same way every day, you get the good data you're looking for. Sure, it won't be quite as comparable on the weekends, depending upon how much driving you do, but in the end you have the data you need to chose to get up a move a little more in order to reach your goal, and that's the bottom line, for me.
...my $.02.