Lifelong Localization in Dynamic Indoor Environments Combining Odometry with Sparse Distance Sampling
Michael M. Bilevich, Tomer Buber, Dan Halperin
AI summary
Problem
Pre-determined maps quickly become inaccurate in dynamic indoor environments due to unforeseen obstacles, making lifelong localization and the kidnapped robot problem challenging without expensive or bandwidth-heavy sensors.
Approach
The method deterministically computes all feasible robot poses from sparse range samples, then fuses these candidates with odometry via Bayesian filtering to continuously update and converge on the true location.
Key results
- Fuses odometry dead reckoning with sparse distance sampling for robust lifelong localization
- Proves convergence to ground truth pose in both static and learned dynamic environments
- Achieves SLAM-comparable accuracy using only 16 distance samples instead of full LiDAR
- Provides an open-source C++ library with Python bindings and ROS2 deployment packages
Why it matters
Enables cost-effective, privacy-preserving, and bandwidth-efficient robot navigation in changing indoor spaces without relying on heavy LiDAR or camera systems.
Abstract
Localization is a key task in robot navigation, and many techniques exist for it. In many plausible scenarios, a robot might face unforeseen, dynamic obstacles, rendering any pre-determined map inaccurate for localization. In this work, we propose a robust lifelong localization framework in dynamic planar indoor environments, using the robot’s odometry and sparse distance sampling. We demonstrate how distance samples can be used to provide a robust prior on the robot’s location. This technique can solve the kidnapped robot problem in real time, up to symmetries. Based on insights from real-world recorded data, we also account for dynamic obstacles. We then fuse this prior, over time, with the odometry to converge to the robot’s location. A central property of our method is that it provably converges to the robot’s ground truth pose even in large indoor environments when the environment is static. We further show that this guarantee also holds in dynamic environments, as long as the nature of those changes has been correctly learned. We demonstrate the effectiveness of our approach in different real-world indoor environments. In particular, we achieve a localization comparable to SLAM with merely a few (sixteen) distance samples, as opposed to the full LiDAR range. Sufficing with only sparse distance sampling is advantageous in terms of sensor cost, privacy, storage space, and transmission bandwidth.