This chapter presents a description of a guidance and control system on robust pathfollowing flight for Unmanned Aerial Vehicle (UAV). Guidance and control systems on path-following flight for UAV, most based on a tracking-error-correction approach, have been studied for several years (Blajer, 1990; Baba et al., 1996, 2002; Kaminer et al., 1998; Boyle & Chamitoff, 1999; Ochi et al., 2002; Park et al., 2004; Rysdyk, 2006). However, these systems present the difficulty of following in large tracking-error situations, e.g. steep curved path-following or tracking under wind turbulence, which might cause control saturation or divergence because the control command tends to increase as the trackingerror becomes large. One coauthor proposed a variable gain method using fuzzy logic (Baba & Takano, 1998), which gathers and weighs on control laws according to tracking-error quantities. It performed well, but it was necessary to set some design points and to conduct several gain tunings before applying fuzzy logic. This chapter presents a novel, simple, yet robust guidance method for path-following UAV (Sato et al., 2006; Yamasaki et al., 2007). The methodology uses pure pursuit guidance instead of traditional tracking-error correction-based methods. Pure pursuit guidance (e.g. Machol et al., 1965) demands only one gain-tuning. It produces guidance commands that are not large, with no relation to tracking error quantities. It can avoid control divergence because the pure pursuit guidance system generates guidance commands in relation to the line-of-sight angle (the angle formed by the UAV velocity vector and the line-of-sight to the target point), which is, at most, π radian. That angle might be much less than the tracking error, e.g. 10 m. For that reason, a robust path-following UAV can be realized assisted by pure pursuit guidance (Park et al., 2007). A target point for the UAV to orient is necessary for applying pure pursuit guidance for path-following. We introduced a receding virtual waypoint, which is placed at a proper point along a desired trajectory. This is the point of difference from the error-based reference point described in the literature (Park et al., 2007). The desired path can be provided easily in a form of a function of the trajectory arc length using cubic-spline interpolation based on some waypoints through which the UAV is presumed to fly. Once the path is generated in the form of the function of the arc length, the receding virtual waypoint, which is the target point for pure pursuit guidance, is calculable using the cubic-spline function based on the UAV flight arc length added a few seconds ahead of the future flight path length, as inferred from the UAV dynamics. This added term