The ultimate limit of any engineering pursuit is the strength of materials. Right now, it's very likely that the maximum speed of a traction vehicle on the earth's surface has been reached. To drive a vehicle at the speed of sound, the tread of the wheels has to turn slightly faster than the speed of sound. This becomes a problem because a sonic boom at ground level is very violent.
It's possible to turn a fighter jet upside down and just let the wheels roll, but the sonic boom is still a problem, because even if the wheels are just rolling, they are still covering the ground at the speed of sound. Despite this, it has been done, but 763.035 mph was the best they could manage. This is slower than the text book speed of sound, but SoS depends on air density. This might not be such a problem on a planet with very little atmosphere.
It's possible some similar natural phenomena could occur at speeds approaching the speed of light.
The real question is how to generate the energy. The first practical rocket was powered by black powder, and explosive with so little thrust, it can barely lift the unburned fuel. A couple thousand years of intense research has led us to rockets which go very fast and move enormous loads.
Current rocket technology is focused on escaping Earth gravity. Once free of gravity, very little force is needed and most of that is used for direction control. The obvious problem of propelling an object to near the speed of light is that whatever is pushing it has to be going faster than the object. That maybe the key. All we have to do is find a way to push something to the speed of light, in the opposite direction of where we want to go.
Materials Science is definitely the key to the next (possibly every 'next') giant leap in technology... and will be the next multi-billion dollar industry.
As for the sonic boom issue... I completely disagree that any physical limit exists, much less be reached, by the issue of managing (or failing to manage) a sonic boom. Test pilots said that it was impossible to fly faster than the speed of sound due to unmanageable flight characteristics at the trans-sonic level. Fly-by-wire technology and redesign of the control surfaces solved the problem.
perhaps the solution may be in the elimination of 'the wheel', but perhaps not. But, BA, you sound like Bill Gates when he said that no one will ever need more than 640K of addressable memory in a computer.