Uncertainty-driven Planner for Exploration and Navigation

Abstract

We consider the problems of exploration and point-goal navigation in previously unseen environments, where the spatial complexity of indoor scenes and partial observability constitute these tasks challenging. We argue that learning occupancy priors over indoor maps provides significant advantages towards addressing these problems. To this end, we present a novel planning framework that first learns to generate occupancy maps beyond the field-of-view of the agent, and second leverages the model uncertainty over the generated areas to formulate path selection policies for each task of interest. For point-goal navigation the policy chooses paths with an upper confidence bound policy for efficient and traversable paths, while for exploration the policy maximizes model uncertainty over candidate paths. We perform experiments in the visually realistic environments of Matterport3D using the Habitat simulator and demonstrate: 1) Improved results on exploration and map quality metrics over competitive methods, and 2) The effectiveness of our planning module when paired with the stateof-the-art DD-PPO method for the point-goal navigation task.

Hallucinated Neural Radiance Fields in the Wild

“…Uncertainty estimation for neural networks is relevant to assessing confidence, information gain and capturing outliers [58]- [60]. Bayesian neural networks [58], [61], [62] learn the posterior distribution of network weights and estimate it using variational inference, which can be approximated by Dropout or Deep Ensembles [63]- [66].…”

“…For evaluating the potential contribution of each synthetic sample, we consider not only the rendering quality (reflected by uncertainties from U-NeRF) but also how informative the sample is for learning SCR. To this end, the epistemic uncertainty [59], [60], [74], [75] can be adopted to reflect the information gain the sample brings to the network. Compared to aleatoric uncertainty which represents the inherent randomness in data that cannot be explained away, epistemic uncertainty captures the uncertainty over network parameters and describes the confidence of the prediction [58].…”