In reading the article about the antibodies being traceable within urine, that brings me to one question… Is it just the anti-bodies that can be detected in the urine, or is there also the presence of the virus itself? If it's just the anti-bodies, then there isn't really a risk... If however the virus itself does somehow transfer to the urine, then that's a whole other ball of wax. It's definitely something worth taking a closer look at.
Hematuria could be one possible example of HIV in urine. Blood in urine is not always visible if it's just in small amounts.
EDIT: Even with the "small amounts" of blood in the urine, when they're only in such levels that you cannot see them and you have an undetectable viral load, it's fairly safe to say that the chances are not very likely.