Learning to Explore using Active Neural SLAM
arXiv:2004.05155
Abstract
This work presents a modular and hierarchical approach to learn policies for exploring 3D environments, called `Active Neural SLAM'. Our approach leverages the strengths of both classical and learning-based methods, by using analytical path planners with learned SLAM module, and global and local policies. The use of learning provides flexibility with respect to input modalities (in the SLAM module), leverages structural regularities of the world (in global policies), and provides robustness to errors in state estimation (in local policies). Such use of learning within each module retains its benefits, while at the same time, hierarchical decomposition and modular training allow us to sidestep the high sample complexities associated with training end-to-end policies. Our experiments in visually and physically realistic simulated 3D environments demonstrate the effectiveness of our approach over past learning and geometry-based approaches. The proposed model can also be easily transferred to the PointGoal task and was the winning entry of the CVPR 2019 Habitat PointGoal Navigation Challenge.
Published in ICLR-2020. See the project webpage at https://devendrachaplot.github.io/projects/Neural-SLAM for supplementary videos. The code is available at https://github.com/devendrachaplot/Neural-SLAM
Cited by in corpus (45)
- Object Goal Navigation using Goal-Oriented Semantic Exploration
- ViNG: Learning Open-World Navigation with Visual Goals
- AllenAct: A Framework for Embodied AI Research
- Embodied Visual Navigation with Automatic Curriculum Learning in Real Environments
- Where to Look Next: Learning Viewpoint Recommendations for Informative Trajectory Planning
- Learn to Navigate Maplessly with Varied LiDAR Configurations: A Support Point-Based Approach
- Learning to Set Waypoints for Audio-Visual Navigation
- Learning Robotic Navigation from Experience: Principles, Methods, and Recent Results
- Causal Navigation by Continuous-time Neural Networks
- A Survey of Embodied AI: From Simulators to Research Tasks
- Focus on Impact: Indoor Exploration with Intrinsic Motivation
- Collaborative Visual Navigation
- FILM: Following Instructions in Language with Modular Methods
- Move to See Better: Self-Improving Embodied Object Detection
- Unsupervised Visual Odometry and Action Integration for PointGoal Navigation in Indoor Environment
- Deep Learning for Embodied Vision Navigation: A Survey
- Learning Affordance Landscapes for Interaction Exploration in 3D Environments
- Topological Planning with Transformers for Vision-and-Language Navigation
- The ThreeDWorld Transport Challenge: A Visually Guided Task-and-Motion Planning Benchmark for Physically Realistic Embodied AI
- Out of the Box: Embodied Navigation in the Real World
- Implicit Obstacle Map-driven Indoor Navigation Model for Robust Obstacle Avoidance
- MVP: Unified Motion and Visual Self-Supervised Learning for Large-Scale Robotic Navigation
- Leveraging Large Language Model-based Room-Object Relationships Knowledge for Enhancing Multimodal-Input Object Goal Navigation
- No RL, No Simulation: Learning to Navigate without Navigating
- Deep Reinforcement Learning for Autonomous Ground Vehicle Exploration Without A-Priori Maps
- Unsupervised Domain Adaptation for Visual Navigation
- SOON: Scenario Oriented Object Navigation with Graph-based Exploration
- SASRA: Semantically-aware Spatio-temporal Reasoning Agent for Vision-and-Language Navigation in Continuous Environments
- Learning to Map for Active Semantic Goal Navigation
- Shaping embodied agent behavior with activity-context priors from egocentric video
- Spot the Difference: A Novel Task for Embodied Agents in Changing Environments
- Learning Efficient Multi-Agent Cooperative Visual Exploration
- Learning to Navigate Sidewalks in Outdoor Environments
- MaAST: Map Attention with Semantic Transformersfor Efficient Visual Navigation
- Beyond Tabula-Rasa: a Modular Reinforcement Learning Approach for Physically Embedded 3D Sokoban
- SeanNet: Semantic Understanding Network for Localization Under Object Dynamics
- Semantic Audio-Visual Navigation
- A Few Shot Adaptation of Visual Navigation Skills to New Observations using Meta-Learning
- Learning Composable Behavior Embeddings for Long-horizon Visual Navigation
- Realistic PointGoal Navigation via Auxiliary Losses and Information Bottleneck
- Learning Synthetic to Real Transfer for Localization and Navigational Tasks
- Building Intelligent Autonomous Navigation Agents
- Pushing it out of the Way: Interactive Visual Navigation
- Cooperative and Asynchronous Transformer-based Mission Planning for Heterogeneous Teams of Mobile Robots
- Graph Convolutional Memory using Topological Priors