The Surprising Effectiveness of Visual Odometry Techniques for Embodied PointGoal Navigation
This addresses a critical limitation for personal robots in noisy real-world settings, though it is incremental as it builds on existing navigation policies.
The paper tackled the problem of PointGoal navigation in realistic noisy environments without perfect localization, showing that integrating visual odometry improves success rates from 64.5% to 71.7% and speeds up execution by 6.4 times.
It is fundamental for personal robots to reliably navigate to a specified goal. To study this task, PointGoal navigation has been introduced in simulated Embodied AI environments. Recent advances solve this PointGoal navigation task with near-perfect accuracy (99.6% success) in photo-realistically simulated environments, assuming noiseless egocentric vision, noiseless actuation, and most importantly, perfect localization. However, under realistic noise models for visual sensors and actuation, and without access to a "GPS and Compass sensor," the 99.6%-success agents for PointGoal navigation only succeed with 0.3%. In this work, we demonstrate the surprising effectiveness of visual odometry for the task of PointGoal navigation in this realistic setting, i.e., with realistic noise models for perception and actuation and without access to GPS and Compass sensors. We show that integrating visual odometry techniques into navigation policies improves the state-of-the-art on the popular Habitat PointNav benchmark by a large margin, improving success from 64.5% to 71.7% while executing 6.4 times faster.