The idea is that it reduces administration costs, eliminates poverty traps, and reduces the stigma of unemployment.
I'm no expert in politics, but I think if their benefits were cut they would probably protest, and the protesters would look very sympathetic.
And if we say "OK, we'll have some extra funding for legitimate expenses for the legitimately disabled" you're not going to be able to do away with administration costs.
Why someone that doesn't need a special vehicle would use the subsidy? The normal car fit their needs for the same price! And it's not even accounting the more probable situation in which the subsidy doesn't cover all the cost difference.
You would presumably couple it with a heavily progressive income tax, with the result that people higher up on the scale end up paying way more into the system than the basic income check they receive, and people at the bottom pay no tax at all. So the effective amount of welfare is directly linked to one's income or lack thereof.
Also, UBI can replace most welfare, but not all. In particular, your example with the wheelchair - that sort of thing really ought to be covered by health insurance (which should be managed entirely by the government, or at least price-controlled).
This is one of the primary gripes that I have with UBI. You've just described means testing! "So the effective amount of welfare is directly linked to one's income or lack thereof."
We've moved the means testing around a little bit, but it's still there, and it's still open to politicization. The right demands that the threshold between welfare givers / takers be lowered, the left argues it be raised, and we're right back where we started.
There's no way to avoid means testing - the money has to come from somewhere, and this inherently means that some people pay more money into the system than they receive from it.
The point of UBI is to make this whole setup as transparent and low-overhead as possible.
The nice thing about it is that, in theory, because the redistribution is basically regulated by just two knobs (size of the UBI check, and how progressive the taxes are), adjusting them is politically easy, and adjustment range is pretty much unlimited. I'm hopeful that it would result in fewer situations like the ongoing ACA "debate", where right wing just wants to sink it altogether and replace it with something else entirely, because they can't get where they want by tuning it. If we can get this to work, UBI would always be there, but more left-leaning or more right-leaning governments would simply adjust the knobs more or less gradually depending on how much "freeloading" they're okay with.
And smooth adjustments are easier to adapt to, and cause less social upheaval (think of the shock to the society and the economy if Republicans somehow manage to repeal ACA, say, tomorrow...)
Basic income means everyone gets the money they need to survive without having to go through a humiliating application process, and they only have to care about taxes when they start earning money.
Moreover it means the remaining administration is focused on tax fraud, and can prioritise its resources appropriately on the worst offenders, rather than having an agency dedicated to chasing poor people for relatively trivial amounts of money (likely at greater cost than the fraud itself).
In terms of the politics, I think the peace of mind benefits of welfare will be more obvious to the middle class when it's automatic, rather than it being something they could maybe apply for if they get really desperate.
They'll have much less work anyway now that nobody is starving or homeless by lack of money, might as well focus.
Doesn't that mean that the people who need it the most will just get less than they're getting now?
In other words, are people who are relying on current US welfare systems often using more than (say) 10k/year in total benefits, and giving them 10k in cash instead would actually reduce their total compensation? And if not, then how can the cost of our current welfare systems be comparable to UBI?