Graph Optimality-Aware Stochastic LiDAR Bundle Adjustment with Progressive Spatial Smoothing

Jianping Li1 Thien-Minh Nguyen1 Muqing Cao2 Shenghai Yuan1 Tzu-Yi Hung3 Lihua Xie1 1 Nanyang Technological University 2 Carnegie Mellon University 3 Delta Research Center

What can PSS-GOSO do?


Graph optimality-aware stochastic optimization with progressive spatial smoothing for a robust, efficient, and scalable LBA. The Progressive Spatial Smoothing (PSS) module extracts robust LiDAR feature association exploiting the prior structure information obtained by the polynomial smooth kernel. The Graph Optimality-aware Stochastic Optimization (GOSO) module first sparisifies the graph according to optimality for an efficient optimization. GOSO then utilizes stochastic clustering and graph marginalization to solve the large-scale state estimation problem for a scalable LBA.

Navigation in indoor-to-outdoor campus scenes using the optimized point clouds for automatic last-mile delivery



Optimized point cloud in large-scale scenes

UAV mapping



Sea port mapping



Multi-session campus indoor-to-outdoor mapping