Sensor Fusion of Structure-from-Motion, Bathymetric 3D, and Beacon-Based Navigation Modalities

This paper describes an approach for the fusion of 3D data underwater obtained from multiple sensing modalities. In particular, we examine the combination of imagebased Structure-From-Motion (SFM) data with bathymetric data obtained using pencil-beam underwater sonar, in order to recover the shape of the seabed terrain. We also combine image-based egomotion estimation with acousticbased and inertial navigation data on board the underwater vehicle. We examine multiple types of fusion. When fusion is performed at the data level, each modality is used to extract 3D information independently. The 3D representations are then aligned and compared. In this case, we use the bathymetric data as ground truth to measure the accuracy and drift of the SFM approach. Similarly we use the navigation data as ground truth against which we measure the accuracy of the image-based ego-motion estimation. To our knowledge, this is the first quantitative evaluation of image-based SFM and egomotion accuracy in a large-scale outdoor environment. Fusion at the signal level uses the raw signals from multiple sensors to produce a single coherent 3D representation which takes optimal advantage of the sensors’ complementary strengths. In this paper, we examine how lowresolution bathymetric data can be used to seed the higherresolution SFM algorithm, improving convergence rates, and reducing drift error. Similarly, acoustic-based and inertial navigation data improves the convergence and drift properties of egomotion estimation.