Lisco: A Continuous Approach in LiDAR Point-cloud Clustering
This work addresses the need for efficient, real-time data processing in cyberphysical systems like fog architectures, though it is incremental as it builds on existing Euclidean-distance-based clustering approaches.
The paper tackles the problem of clustering LiDAR point clouds in a continuous, streaming fashion to support automated systems, achieving up to 3x improvements in processing efficiency compared to baseline methods.
The light detection and ranging (LiDAR) technology allows to sense surrounding objects with fine-grained resolution in a large areas. Their data (aka point clouds), generated continuously at very high rates, can provide information to support automated functionality in cyberphysical systems. Clustering of point clouds is a key problem to extract this type of information. Methods for solving the problem in a continuous fashion can facilitate improved processing in e.g. fog architectures, allowing continuous, streaming processing of data close to the sources. We propose Lisco, a single-pass continuous Euclidean-distance-based clustering of LiDAR point clouds, that maximizes the granularity of the data processing pipeline. Besides its algorithmic analysis, we provide a thorough experimental evaluation and highlight its up to 3x improvements and its scalability benefits compared to the baseline, using both real-world datasets as well as synthetic ones to fully explore the worst-cases.