On-the-fly mobility event detection over aircraft trajectories

Abstract

We present an application framework that consumes streaming positions from a large fleet of flying aircrafts monitored in real time over a wide geographical area. Tailored for aviation surveillance, this online processing scheme only retains locations conveying salient mobility events along each flight, and annotates them as stop, change of speed, heading or altitude, etc. Such evolving trajectory synopses must keep in pace with the incoming raw streams so as to get incrementally annotated with minimal loss in accuracy. We also develop one-pass heuristics to eliminate inherent noise and provide reliable trajectory representations. Our prototype implementation on top of Apache Flink and Kafka has been tested against various real and synthetic datasets offering concrete evidence of its timeliness, scalability, and compression efficiency, with tolerable concessions to the quality of resulting trajectory approximations.

Review and classification of trajectory summarisation algorithms: From compression to segmentation

“…This parallelisation should be in the design of the online and offline compression algorithms of the future and can accelerate very expensive implementations by many orders of magnitude. Some examples of the parallelisation are the Patroumpas et al 73,74 approaches, graph speed-up by Deng et al, 59 Feldman et al 34 or Huang et al, 160 or the spin line detection of Feldman et al 34…”

“…Several algorithms follow this same principle. Others, such as online data compression algorithm for trajectories (OLDCAT) 26 or the proposals of Patroumpas et al 73,74 perform different decoupled conditions in parallel, by means of a concatenation of comparisons. These techniques perform online compression to simplify the trajectory and obtaining segments with semantic content.…”

“…OLDCAT, 26 Patroumpas et al, 73,74 Siddique and Ban 32,33 or Garcia and colleagues [35][36][37] are authors of approaches that detect the change of trajectory motion type. This implies that they are able to identify the type of motion between two points of change, categorising segments with a particular type of motion.…”

Optimizing Vessel Trajectory Compression

“…In this Section, we overview the mobility events detected along vessel trajectories and the parameters involved in their specification by the Synopses Generator. In depth analysis of the applied trajectory summarization is available in [11,12].…”

“…The resulting trajectory synopsis per vessel is derived from those judiciously annotated critical points and can approximately reconstruct its original course. This framework has been further enhanced to run against scalable streaming data in cluster infrastructure and enable Big Data Analytics [17] not only in the maritime domain, but also in aviation [12]. This methodology can provide reliable trajectory synopses with little deviations from the original course by discarding at least 70% of the raw data as redundant; and in case of frequent updates, even less than 1% of the data need be retained.…”