A smarter habitat for deep space exploration

In order to explore the Moon or Mars, astronauts need smart habitats that will support life and remain operational when they are vacant. To advance the design of autonomous systems for space habitats, NASA is funding a multi-university space research institute called Habitats optimized for exploration missions, or more specifically HOME.

Mario Bergés, Associate Professor of Civil and Environmental Engineering (CEE), leads the research team at Carnegie Mellon University under the auspices of HOME. Their task is to provide a complete knowledge of the housing situation by providing the ability to process and interpret their own data and to make decision recommendations that can be transmitted to robotic or suggested systems. to human occupants.

"Space is tough and mistakes can be catastrophic, so we need high performance standalone systems," said Bergés. One of the ideas explored by the team is the use of artificial intelligence to analyze equipment data to understand electricity consumption in the home. By knowing how energy is consumed, the condition of all electrical systems in the habitat can be monitored.

Scientists have experience analyzing data for equipment on Earth, but Mr. Bergés said that we needed to learn to translate this knowledge into the context of space and specifically in the systems of these habitats.

For example, on Earth, there are a variety of air conditioning systems and we can determine where potential defects are and how they occur. But in new habitats, all systems will be punctual.

"How to perform automated detection and diagnosis of errors without a lot of system data is where artificial intelligence comes in," said Bergés. "We have machines that learn by themselves if you give them enough data, but we do not have many machines that can reason using existing technical knowledge, which can reduce the amount of data they need. "

To limit the amount of data needed to detect equipment failures in the habitat, the team will collect electrical measurements. These data will be shared with robotic systems that will process them, act on the physical environment and, in theory, allow the habitat to maintain itself.

The CMU team is composed of Bergés, an expert in data detection and analysis for infrastructures; Burcu Akinci, professor of CEE and expert in information modeling; and Stephen Smith and Artur Dubrawski of the CMU Robotics Institute, who will conduct research on machine learning and robotic systems. CMU research will feed into other ongoing projects in the institute.

HOME receives funding of about $ 15 million over five years, and NASA can implement the recommendations from the institute before the end of the funding period. According to Bergés, CMU research could potentially illuminate the design of a gateway station that will orbit the Moon as part of NASA's far-distant space exploration plans.

Bergés thinks that civil engineers will play a role in space exploration.

"From the beginning, civil engineers have been the guardians of the infrastructure that supports modern life," he said. "If humanity moves in space, it makes sense that civil engineers are part of it."