The objective of this Phase I STTR proposal is to develop a first prototype of a digital assistant to support NASA engineers in space mission engineering and design tasks in Pre-Phase A and Phase A mission studies. This technology will: 1) increase the quality and diversity of the designs explored during early stage mission concept studies; 2) decrease the time and resources it takes to perform such studies; and 3) reduce the cognitive workload of systems engineers while performing such studies. While digital assistants are ubiquitous today in our daily lives, they are still to be adopted by aerospace organizations. With the societal push from assistants such as Google Home, Alexa, Siri, etc., and given recent advances in machine learning and natural language processing, the time is ripe for such adoption. This technology has the potential to save NASA millions of dollars by reducing design effort and helping NASA identify better system designs faster. This proposal addresses topic T11.04 Digital Assistants for Science and Engineering of the 2019 SBIR/STTR solicitation.
While the proposed application for Phase I is early design of Earth observing missions, this technology has the potential to be applied throughout the project lifecycle and across NASA’s portfolio of projects, from small CubeSat missions to large flagship missions. Substantial code reuse should be possible across projects, as the required functionalities (searching across databases, using engineering models) stay the same and only the sources of information and interfaces with them change.
Mycroft’s technology is broadly applicable as a general intelligent assistant that can be adapted to use cases such as healthcare, automotive, room and building control, and more based upon the skills integrated into it. The NLP and embodiment for the assistant developed for this proposal are made to be general purpose implementations that are easily transferred to other use cases.