Charles River Analytics will develop a probabilistic modeling system for NASA researchers to evaluate flight systems with a focus on aviation safety.
NASA analysts will work to estimate the remaining useful life of components with Charles River’s prediction technique under the Programming Useful Life Prediction program, Charles River said Wednesday.
“We will be developing a new approach to RUL prediction by using probabilistic programming to combine both expert knowledge and historical data to reasoning about RUL,” said Avi Pfeffer, principal investigator for the PULP effort.
“We’ll apply it to unmanned aircraft, in a way that we expect will be more accurate than using either approach on its own,” added Pfeffer.
NASA partner Unmanned Experts will also build a physics model for the probabilistic program with the Figaro probabilistic programming language.
“We will also apply Figaro’s machine learning techniques to learn RUL predictors from NASA’s data,” Pfeffer said.