You would have to be fairly close to the black hole in question to do this. A distance of r=3 MG/c² according to Stuckey '93 [1], which is referred to in the paper referenced [2] in the article. Skimming the articles, I can't see any reference to how sensitive to error in aiming this would be.

[1] The Schwarzschild black hole as a gravitational mirror https://sci-hub.se/https://doi.org/10.1119/1.17434 [2] http://coolworlds.astro.columbia.edu/halodrive_preprint.pdf