Multiple Thinking Achieving Meta-Ability Decoupling for Object Navigation
This work addresses object navigation for robotics and AI systems, presenting a novel paradigm with interpretability, though it appears incremental as it builds on existing methods.
The paper tackles object navigation by proposing a meta-ability decoupling paradigm and a multiple thinking model to abstract and enhance various meta-abilities, achieving state-of-the-art performance on AI2-Thor and RoboTHOR benchmarks.
We propose a meta-ability decoupling (MAD) paradigm, which brings together various object navigation methods in an architecture system, allowing them to mutually enhance each other and evolve together. Based on the MAD paradigm, we design a multiple thinking (MT) model that leverages distinct thinking to abstract various meta-abilities. Our method decouples meta-abilities from three aspects: input, encoding, and reward while employing the multiple thinking collaboration (MTC) module to promote mutual cooperation between thinking. MAD introduces a novel qualitative and quantitative interpretability system for object navigation. Through extensive experiments on AI2-Thor and RoboTHOR, we demonstrate that our method outperforms state-of-the-art (SOTA) methods on both typical and zero-shot object navigation tasks.