Object Goal Navigation using Goal-Oriented Semantic Exploration
This addresses the problem of inefficient exploration and long-term planning in object goal navigation for robotics and AI systems, representing a strong incremental improvement over existing modular and learning-based methods.
The paper tackles object goal navigation in unseen environments by proposing a modular system called Goal-Oriented Semantic Exploration, which builds episodic semantic maps for efficient exploration based on goal object categories. Empirical results show it outperforms various baselines in simulation, winning the CVPR-2020 Habitat ObjectNav Challenge, and transfers effectively to real-world mobile robots with similar performance.
This work studies the problem of object goal navigation which involves navigating to an instance of the given object category in unseen environments. End-to-end learning-based navigation methods struggle at this task as they are ineffective at exploration and long-term planning. We propose a modular system called, `Goal-Oriented Semantic Exploration' which builds an episodic semantic map and uses it to explore the environment efficiently based on the goal object category. Empirical results in visually realistic simulation environments show that the proposed model outperforms a wide range of baselines including end-to-end learning-based methods as well as modular map-based methods and led to the winning entry of the CVPR-2020 Habitat ObjectNav Challenge. Ablation analysis indicates that the proposed model learns semantic priors of the relative arrangement of objects in a scene, and uses them to explore efficiently. Domain-agnostic module design allow us to transfer our model to a mobile robot platform and achieve similar performance for object goal navigation in the real-world.