I'm in the UK, so I don't know how it works in other countries. Over here though, we have various sources of government funding which give an incentive for academic institutions to work with industry. Funding is particularly geared to research into things which aren't currently widespread in industry.
The Industrial Strategy Challenge Fund invests in a few key areas. Civil engineering isn't really applicable to them, but other areas are.
More generally, Innovate UK has a raft of projects it's looking for applicants for, or you can apply for funding for your research if you think you fit their criteria.
And for European funding (which British academics can't get now - deep joy) there are Research Councils which fund research from EU budgets.
As an academic though, one of your problems is going to be showing that you know how to apply your theory to a practical problem. Industry doesn't care about theory, it cares about building things, and your "respected" colleagues have earned respect by understanding this. I'm sorry to say though that industry generally regards academics as ivory-tower amateurs - and that opinion is generally justified.
A personal example. One piece of opthalmic test kit my company produces was closely tied to research from two academics on "deskilling" the screening test they were taking. We worked closely with them, and assumed the tests they had designed were valid (we're software engineers after all, not opthalmologists). It was only when customers around the world brought up some problems that we had to go back to first principles. We found their entire methodology was so fatally flawed that any conclusions they wanted to draw from the data were completely invalid. Put simply, these senior professors who'd been important academics in their field for 40 years had no understanding of how to run an experiment.
In another case, we bought technology for improving the response of a nanopositioning system. The theory was (and is) compelling, and the academic concerned had data which backed up the theory. It was only when we integrated this technology into our systems and tried to reproduce his methods that we found none of his reports used the same circuit diagram twice, and the circuits he'd used didn't match the transfer functions in his reports, and even then we couldn't reproduce his results. We've been working with another university to try to figure out how to apply this technology properly, because it does look to have potential if it's done properly. So far we've had three postgraduate students look at it, under the guidance of a professor of control theory. None of them has had any kind of clue about how to run an experiment either, and we've had very limited useful data from them.
Can you do better than that? If you can, you have a chance. If you can't, then honestly you should quit your job and get some experience, because you're a hazard to the careers of the people you're supposed to be teaching.