(Abridged) Out-of-the-ecliptic trajectories that are beneficial to space observatories such as the Terrestrial Planet Finder and other potential mid-IR missions are introduced. These novel trajectories result in significantly reduced background noise from the zodiacal dust radiation and therefore allow a reduction in telescope collecting area and hence cost. Two types of optimal trajectories that are energetically feasible were derived using genetic algorithms. These are highly inclined non-Keplerian heliocentric orbits. The first optimal trajectory can use existing launch technology and the zodiacal brightness for this trajectory is reduced by at least 50% for 60% of the mission lifetime. The second optimal trajectory requires planned improvement in launch technology but it renders a zodiacal cloud brightness reduction by at least 70% for 82% of the mission lifetime. Heliocentric orbits at 5 AU have been discussed because the zodiacal dust concentration is extremely low there and the energy requirements to reach these trajectories are low if gravitational assists can be used. Unfortunately such orbits are impractical because of high cost, power source constraints (inability to use solar cells), communication delays, and a long travel time before data return. Additionally, the energy requirements to reach our low and high energy optimal trajectories are respectively half as much and equivalent to a direct trip to 5AU with no planetary gravitational assists.