Variational State-Space Models for Localisation and Dense 3D Mapping in 6 DoF
This addresses the problem of robust spatial understanding for robotics, particularly in UAV applications, though it appears incremental by combining existing techniques.
The paper tackles 6-DoF localization and dense 3D mapping by using approximate Bayesian inference in a deep state-space model, achieving performance close to state-of-the-art visual-inertial odometry systems on UAV flight data.
We solve the problem of 6-DoF localisation and 3D dense reconstruction in spatial environments as approximate Bayesian inference in a deep state-space model. Our approach leverages both learning and domain knowledge from multiple-view geometry and rigid-body dynamics. This results in an expressive predictive model of the world, often missing in current state-of-the-art visual SLAM solutions. The combination of variational inference, neural networks and a differentiable raycaster ensures that our model is amenable to end-to-end gradient-based optimisation. We evaluate our approach on realistic unmanned aerial vehicle flight data, nearing the performance of state-of-the-art visual-inertial odometry systems. We demonstrate the applicability of the model to generative prediction and planning.