Particle Filter Networks with Application to Visual Localization

Abstract

Particle filtering is a powerful approach to sequential state estimation and finds application in many domains, including robot localization, object tracking, etc. To apply particle filtering in practice, a critical challenge is to construct probabilistic system models, especially for systems with complex dynamics or rich sensory inputs such as camera images. This paper introduces the Particle Filter Network (PFnet), which encodes both a system model and a particle filter algorithm in a single neural network. The PF-net is fully differentiable and trained end-to-end from data. Instead of learning a generic system model, it learns a model optimized for the particle filter algorithm. We apply the PF-net to a visual localization task, in which a robot must localize in a rich 3-D world, using only a schematic 2-D floor map. In simulation experiments, PF-net consistently outperforms alternative learning architectures, as well as a traditional model-based method, under a variety of sensor inputs. Further, PF-net generalizes well to new, unseen environments.

arXiv commentCoRL 2018 camera ready

References in corpus (9)

• Tutorial on Variational Autoencoders
• Building Generalizable Agents with a Realistic and Rich 3D Environment
• DESPOT: Online POMDP Planning with Regularization
• Task-based End-to-end Model Learning in Stochastic Optimization
• Auto-Encoding Sequential Monte Carlo
• Variational Sequential Monte Carlo
• Path Integral Networks: End-to-End Differentiable Optimal Control
• Active Neural Localization
• Learning to Search with MCTSnets

Cited by in corpus (8)

• Efficient Bird Eye View Proposals for 3D Siamese Tracking
• DynaNet: Neural Kalman Dynamical Model for Motion Estimation and Prediction
• Integrating Algorithmic Planning and Deep Learning for Partially Observable Navigation
• DOB-Net: Actively Rejecting Unknown Excessive Time-Varying Disturbances
• Instance based Generalization in Reinforcement Learning
• Multiplicative Gaussian Particle Filter
• A2: Extracting Cyclic Switchings from DOB-nets for Rejecting Excessive Disturbances
• Deep Samplable Observation Model for Global Localization and Kidnapping