Skip-SCAR: Hardware-Friendly High-Quality Embodied Visual Navigation

TL;DR

Skip-SCAR is a novel framework that enhances embodied visual navigation by reducing computational costs through skipping redundant steps and using a hybrid network, achieving high performance with lower resource usage.

Contribution

It introduces a hardware-friendly optimization framework that improves navigation efficiency and accuracy by skipping redundant computations and employing a hybrid network architecture.

Findings

Minimizes hardware resources while maintaining high navigation quality.

Sets new benchmarks on HM3D ObjectNav datasets.

Effective on real-world robotic hardware.

Abstract

In ObjectNav, agents must locate specific objects within unseen environments, requiring effective perception, prediction, localization and planning capabilities. This study finds that state-of-the-art embodied AI agents compete for higher navigation quality, but often compromise the computational efficiency. To address this issue, we introduce "Skip-SCAR," an optimization framework that builds computationally and memory-efficient embodied AI agents to accomplish high-quality visual navigation tasks. Skip-SCAR opportunistically skips the redundant step computations during semantic segmentation and local re-planning without hurting the navigation quality. Skip-SCAR also adopts a novel hybrid sparse and dense network for object prediction, optimizing both the computation and memory footprint. Tested on the HM3D ObjectNav datasets and real-world physical hardware systems, Skip-SCAR not only minimizes hardware resources but also sets new performance benchmarks, demonstrating the benefits of optimizing both navigation quality and computational efficiency for robotics.