This paper explores the use of machine learning and deep learning artificial intelligence (AI) techniques as a means to integrate multiple sensor modalities into a cohesive approach to navigation for autonomous ships. Considered is the case of a fully autonomous ship capable of making decisions and determining actions by itself without active supervision on the part of onboard crew or remote human operators. These techniques, when combined with advanced sensor capabilities, have been touted as a means to overcome existing technical and human limitations as unmanned and autonomous ships become operational presently and in upcoming years. Promises of the extraordinary capabilities of these technologies that may even exceed those of crewmembers for decision making under comparable conditions must be tempered with realistic expectations as to their ultimate technical potential, their use in the maritime domain, vulnerabilities that may preclude their safe operation; and methods for development, integration and test. The results of research performed by the author in specific applications of machine learning and AI to shipping are presented citing key factors that must be achieved for certification of these technologies as being suitable for their intended purpose. Recommendations are made for strategies to surmount present limitations in the development, evaluation and deployment of intelligent maritime systems that may accommodate future technological advances. Lessons learned that may be applied to improve safety of navigation for conventional shipping are also provided. http://www.transnav.eu the International Journal on Marine Navigation and Safety of Sea Transportation Volume 13 Number 3

A need exists for fast, accurate, and inexpensive identification of failed integrated circuits (ICs) on electronic circuit boards. Presently, testing of these boards is accomplished through the use of automatic test equipment (ATE) controlled by test program sets (TPS). The circuit board interfaces with the ATE through the use of a hardware device called an interface test adapter (ITA), which is developed in conjunction with the test program. Application of nanotechnology techniques into automatic test equipment area creates an opportunity to significantly reduce or entirely eliminate ATE and TPS costs associated with the maintenance of shop replaceable unit (SRU) circuit boards by designing a means for ICs to continuously test themselves during normal operation and to provide a visual indication that they have failed. This article describes original research and development in the use of nanotubebased chemical and electronic sensors as embedded test equipment within the architecture of integrated circuits (ICs) to detect failures that may affect IC reliability and performance. This R&D focuses on the monitoring of electromigration processes occurring within integrated circuits as well as electrical performance of the ICs to provide a reliable identification of failures.

No abstract

No abstract

No abstract