Previous literature on the Traveling Salesman Problem (TSP) assumed that the sites to be visited are stationary. Motivated
by practical applications, we introduce a time-dependent generalization of TSP which we call Moving-Target TSP, where a pursuer
must intercept in minimum time a set of targets which move with constant velocities. We propose approximate and exact algorithms
for several natural variants of Moving-Target TSP. Our implementation is available on the Web.
Professor Robins is supported by a Packard Foundation Fellowship and by National Science Foundation Young Investigator Award
MIP-9457412. Additional related papers and our code may be found at http://www.cs.virginia.edu/~robins