I share your skepticism.
However, as a practical matter, once we've made the basic assumptions that are required to do anything (there is a real objective world, I can infer something about that world through sensory perception, etc.), we find that some approaches to gaining knowledge work.
Objects seem to behave in a consistent way, and as we measure their behavior more and more accurately, we're able to predict what they'll do next more, and more accurately. Maybe tomorrow objects will change their habits, or start behaving in an erratic fashion, but we don't have any reason to think they will. All of our experience tells us that the laws of physics don't just decide to change, so it would be unreasonable for us to expect them to. When we observe the structure of the universe, it looks the way we'd expect it to, if it had always worked the way that it works now.
Maybe all of our experiences, and observations are illusory, but this is the best we can do.