Out-of-Distribution Robustness with Deep Recursive Filters
This addresses safety-critical navigation for autonomous vehicles and robots, though it appears incremental as a hybrid of existing techniques.
The paper tackles robust state and uncertainty estimation for mobile robots in pedestrian-rich environments under out-of-distribution noise, combining deep neural networks with recursive filters. It demonstrates improved estimation and approximately 3x better computational efficiency on a pendulum problem and nuScenes pedestrian localization.
Accurate state and uncertainty estimation is imperative for mobile robots and self driving vehicles to achieve safe navigation in pedestrian rich environments. A critical component of state and uncertainty estimation for robot navigation is to perform robustly under out-of-distribution noise. Traditional methods of state estimation decouple perception and state estimation making it difficult to operate on noisy, high dimensional data. Here, we describe an approach that combines the expressiveness of deep neural networks with principled approaches to uncertainty estimation found in recursive filters. We particularly focus on techniques that provide better robustness to out-of-distribution noise and demonstrate applicability of our approach on two scenarios: a simple noisy pendulum state estimation problem and real world pedestrian localization using the nuScenes dataset. We show that our approach improves state and uncertainty estimation compared to baselines while achieving approximately 3x improvement in computational efficiency.