Quantitative Modeling of Failure Propagation in Intelligent Transportation Systems
-
2014-08-01
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:Unmanned vehicles are projected to reach consumer use within this decade - related legislation has already passed in California. The most significant technical challenge associated with these vehicles is their integration in transportation environments with manned vehicles. Abnormal or incorrect manipulation of the manned vehicles by their human drivers creates a highly non-deterministic environment that is difficult to consider in the control algorithms for unmanned vehicles. The objective of this project was to develop a model that can capture stochastic elements of this environment, in particular failure propagation from manned to unmanned vehicles and vice versa. A general analytical model reflecting the effect of cyber or physical failures on reliability of a large-scale cyber-physical system was developed in the course of project activities. This model was validated through simulation of related applications an intelligent power grid and water distribution network, respectively. Both examples are topologically and conceptually analogous to an intelligent transportation system. A qualitative model was developed for intelligent transportation systems, and work was commenced on development of a quantitative Petri-net model and cyber-physical simulation environment for such systems. Five refereed conference publications [1{5] and several presentations resulted from this project. Two related journal publications are under final submission and will be submitted in the near future. One MS thesis [6] was completed in conjunction with work related to the project. One undergraduate student, two doctoral students, and two MS students contributed to the research.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: