Mola

MOLA is a modular optimization framework for localization and mapping (SLAM), purpose-built for LiDAR-based robotics and surveying. It provides a comprehensive suite of tools including LiDAR Odometry (LO), LiDAR-Inertial Odometry (LIO), full 3D SLAM, localization-only modes, geo-referencing, and map management. The system is designed as a collection of interoperable ROS 2 packages, enabling researchers and engineers to mix and match components for custom pipelines. MOLA targets both academic research and commercial applications. It is available under an open-core model: the Community edition is free and fully featured for non-commercial use, while Pro/enterprise licenses are required for commercial deployment. The framework emphasizes production-grade reliability and has been validated on real-world robotics and surveying tasks. Architecturally, MOLA separates state estimation (e.g., MP2P-ICP for point cloud registration) from mapping back-ends (e.g., the central SLAM map module 'Mapper'). This allows swapping estimators or fusing multiple sensor inputs without rewriting core logic. The system also supports dataset import from common formats like KITTI, Mulran, and EuRoC, and can replay ROS 1 bags. Key differentiators include: first-class ROS 2 integration (Humble, Jazzy, Rolling), a focus on modularity over monolithic SLAM, and built-in tools for geo-referencing and survey-grade mapping. The project is actively maintained with regular releases and a public changelog. Unlike turnkey SLAM solutions, MOLA gives you full control over every pipeline stage, making it ideal for advanced users who need to customize instead of being locked into a black box.