Search results for: M. Khadka

Authors: A. Paz, M. Khadka, N. Veeramisti, B. Morris

Abstract:

There is a continuous large number of crashes involving pedestrians in Nevada despite the numerous safety mechanisms currently used at roadway crossings. Hence, additional as well as more effective mechanisms are required to reduce crashes in Las Vegas, in particular, and Nevada in general. A potential mechanism to reduce conflicts between pedestrians and vehicles is a High-intensity Activated crossWalK (HAWK) signal. This study evaluates the effects of such signals at a particular site in Las Vegas. Video data were collected using two cameras, facing the eastbound and westbound traffic. One week of video data before and after the deployment of the signal were collected to capture the behavior of both pedestrians and drivers. T-test analyses of pedestrian waiting time at the curb, curb-to-curb crossing time, total crossing time, jaywalking events, and near-crash events show that the HAWK system provides significant benefits.

Keywords: Pedestrian crashes, HAWK signal, traffic safety, pedestrian danger index.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2248

Authors: Anita Khadka, Gregory Epiphaniou, Carsten Maple

Abstract:

There is a dramatic surge in the adoption of Machine Learning (ML) techniques in many areas, including the nuclear industry (such as fault diagnosis and fuel management in nuclear power plants), autonomous systems (including self-driving vehicles), space systems (space debris recovery, for example), medical surgery, network intrusion detection, malware detection, to name a few. Artificial Intelligence (AI) has become a part of everyday modern human life. To date, the predominant focus has been developing underpinning ML algorithms that can improve accuracy, while factors such as resiliency and robustness of algorithms have been largely overlooked. If an adversarial attack is able to compromise the learning method or data, the consequences can be fatal, especially but not exclusively in safety-critical applications. In this paper, we present an in-depth analysis of five adversarial attacks and two defence methods on a crack detection ML model. Our analysis shows that it can be dangerous to adopt ML techniques without rigorous testing, since they may be vulnerable to adversarial attacks, especially in security-critical areas such as the nuclear industry. We observed that while the adopted defence methods can effectively defend against different attacks, none of them could protect against all five adversarial attacks entirely.

Keywords: Resilient Machine Learning, attacks, defences, nuclear industry, crack detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 404