SoMaSLAM: 2D Graph SLAM for Sparse Range Sensing with Soft Manhattan World Constraints
Jeahn Han, Zichao Hu, Seonmo Yang, Minji Kim, Pyojin Kim
AI summary
Problem
Tiny robots must use lightweight range sensors that yield sparse data, causing existing SLAM methods to accumulate drift and lose accuracy. Rigid structural assumptions like the strict Manhattan world further fail when real environments deviate from ideal geometric rules.
Approach
The method extends 2D graph SLAM by introducing soft landmark-landmark constraints that encourage nearby walls to align parallel or orthogonal without enforcing hard geometric rules, preserving flexibility while correcting drift.
Key results
- Lowers absolute translation and rotation error across multiple datasets compared to EG-SLAM and FL-SLAM
- Successfully maps mixed Manhattan and non-Manhattan environments without structural failure
- Proves soft constraints preserve continuous optimization, reducing computational complexity over strict formulations
- Publicly releases source code and evaluation datasets
Why it matters
Enables reliable, drift-free navigation for payload-constrained micro-drones and nano-robots in diverse real-world indoor spaces.
Abstract
We propose a novel graph SLAM algorithm for sparse range sensing that incorporates a soft Manhattan world utilizing landmark-landmark constraints. Sparse range sensing is necessary for tiny robots that do not have the luxury of using heavy and expensive sensors. Existing SLAM methods dealing with sparse range sensing lack accuracy and accumulate drift error over time due to limited access to data points. Algorithms that cover this flaw using structural regularities, such as the Man- hattan world (MW), have shortcomings when mapping real-world environments that do not coincide with the rules. We propose SoMaSLAM, a 2D graph SLAM designed for tiny drones with sparse range sensing. Our approach effectively maps sparse range data without enforcing strict structural regularities and maintains an adaptive graph. We implement the MW assumption as soft constraints, which we refer to as a soft Manhattan world. We propose novel soft landmark-landmark constraints to incorporate the soft MW into graph SLAM. Through extensive evaluation, we demonstrate that our proposed SoMaSLAM method improves localization accuracy across diverse datasets and is flexible enough to be used in the real world. We release our source code and dataset on our project page https://SoMaSLAM.github.io/.