SASRA: Semantically-aware Spatio-temporal Reasoning Agent for Vision-and-Language Navigation in Continuous Environments
arXiv:2108.11945
Abstract
This paper presents a novel approach for the Vision-and-Language Navigation (VLN) task in continuous 3D environments, which requires an autonomous agent to follow natural language instructions in unseen environments. Existing end-to-end learning-based VLN methods struggle at this task as they focus mostly on utilizing raw visual observations and lack the semantic spatio-temporal reasoning capabilities which is crucial in generalizing to new environments. In this regard, we present a hybrid transformer-recurrence model which focuses on combining classical semantic mapping techniques with a learning-based method. Our method creates a temporal semantic memory by building a top-down local ego-centric semantic map and performs cross-modal grounding to align map and language modalities to enable effective learning of VLN policy. Empirical results in a photo-realistic long-horizon simulation environment show that the proposed approach outperforms a variety of state-of-the-art methods and baselines with over 22% relative improvement in SPL in prior unseen environments.
10 pages, 4 figures
References in corpus (9)
- Sequence to Sequence Learning with Neural Networks
- Professor Forcing: A New Algorithm for Training Recurrent Networks
- Learning to Explore using Active Neural SLAM
- Neural Map: Structured Memory for Deep Reinforcement Learning
- Unifying Map and Landmark Based Representations for Visual Navigation
- A Recurrent Vision-and-Language BERT for Navigation
- The Regretful Agent: Heuristic-Aided Navigation through Progress Estimation
- Evolving Graphical Planner: Contextual Global Planning for Vision-and-Language Navigation
- Topological Planning with Transformers for Vision-and-Language Navigation