Continuous and Parallel LiDAR Point-Cloud Clustering

Abstract

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 singlepass 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.

Adaptive Stream-based Shifting Bottleneck Detection in IoT-based Computing Architectures

“…AMBLE is such an example, that is presented here as a streaming operator. In line with approaches in [12]- [14], enabling the algorithmic implementation of AM-BLE through primitive streaming operators and enhancing the algorithmic implementations with parallelization of the data Problems related to predictable distributed processing using flexible real-time service-oriented approaches in factory automation, have also been studied in contexts of web-based infrastructures (e.g [5], [10]). Here we study the problems from a continuous processing perspective, working on the flows of data generated in IoT contexts with multiple sensors and show how the multi-tier processing architecture can be utilized for accurate, low-latency processing.…”

Painting Path Planning for a Painting Robot with a RealSense Depth Sensor

“…C2d(x, y) = C(i * ), Z2d(x, y) = X(i * ), [44]. This step uses the cluster method and k-dimensional tree search method to separate clusters [45][46][47][48].…”

Motivations and Challenges for Stream Processing in Edge Computing

“…Challenge 2: Hardware-, data-and system-awareness. Large distributed systems such as smart grids or vehicular networks are composed of heterogeneous devices and sensors [5,6,14,15,17,21], ranging from the small embedded ones found in smart meters to GPU-based platforms for AI-based self-driving cars [9]. Such heterogeneity must be taken into account for streaming applications to scale while sharing resources with the existing ecosystem of applications running in each device.…”