And that's just how evolution works.
I don't think so. The role of mutation in evolution is that individuals (preferably pairs of individuals) have numerous offspring, which retain a large percentage of the genome of the parent(s) except for maybe one minor mutation. That mutation may increase or decrease each offspring's fitness (carefully avoiding the idea of Darwinian fitness, which relates to how many offspring it has). It's really hard for me to apply this idea to what we're talking about here. I can't see any of these wheeled designs as the offspring of its predecessor. In most cases I can't see any evolutionary inheritance at all. Without that, it's impossible to identify whatever mutation may have occurred, and so it's equally impossible to determine whether that mutation increased or decreased the vehicle's fitness.