RGB-D cameras have attracted much attention in the fields of robotics and computer vision, especially for object modeling and environment mapping. A key problem in all these applications is the registration of sequences of RGB-D images. In this paper, we present an efficient yet reliable approach to align pairs and sequences of RGB-D images that makes use of local surface information. We extend previous works on 3D mapping with micro aerial vehicles to sequences of RGB-D images. The resulting alignment is based on a robust surface-to-surface error metric and uses multiple surface-to-surface patch matches between pairs of RGB-D images. Quantitative evaluations show that our approach is competitive with state-of-the-art approaches.