Tom Keane is the corporate vice president and general manager of the Azure AI and Research group. With a background in engineering, computer science, and management, he has helped build the cloud computing platform of the future with teams across Microsoft.
For those not in the know, Microsoft offers a cloud platform called Azure. Companies and organizations primarily use them to create private clouds and offer a more reliable computing infrastructure. However, with new technology and a lot of hard work from Microsoft’s engineers, the company decided to develop a service that allows users to harness the power of Azure to launch satellite missions into space. Tom Keane states that even more surprising is that the Azure-based satellite will lift off from the edge of space and reach its destination without making a single trip to Earth.
A few years back, Microsoft also announced that Azure would aid NASA in its efforts to launch its Orion spacecraft on deep space missions.
Testing AI for Astronaut Safety at the ultimate edge
According to Tom Keane, it is great to see companies working with the government and non-profit organizations to push the limits of technology further, so along with NASA and HPE, Microsoft has deployed an AI application in space.
The main objective is to create software that will enable astronauts to check the condition of their equipment as Tom Keane recalls. Astronauts’ space suits and gloves are a major weakness when surviving in space. If they are damaged or punctured, the astronaut could get killed in a matter of minutes because the entire suit is pressurized and filled with oxygen.
The AI app leverages computer vision to identify the condition of the space suits and gloves and thus automatically detect damage in real time. Tom Keane finally adds. Once they had trained the app with relevant data taken from their on-ground systems, NASA and HPE uploaded it to a computer on board the ISS. After a few months of operation, the AI system has already trained itself with all the relevant data to detect damage.