Info on Philip Venturelli

Resume:

BS Physics from California Institute of Technology

PhD ABD from UCLA in Computer Science, focus on developing computer models for binary black holes

Pursued Aerospace Modeling and Simulation in the context of Spacecraft Systems Engineering at Caltech

Worked at Raytheon as a Systems Engineer (El Segundo) for 7 years

Aerospace Program Experience:

NASA: Visible Infrared Imaging Radiometer Suite
DoD: Space Based Infrared Surveillance
DoD: Space Tracking and Surveillance System
iRAD: Space debris detection, observation, and removal
DoD: Multiple Hypothesis Tracking Algorithm Development

Some thoughts on applications of AI to Radiometry:

In space debris observation (i.e. in amateur astronomy), the field known as characterization is often more challenging than simple target tracking. Target tracking involves determining what orbit or trajectory a space object is likely to follow. Characterization is the identification of physical characteristics of an object based on its radiant properties. AI might be able to help us solve this problem.

One idea to explore is to use the information present in a spectral signature to solve backwards for what characteristics of the target might be present.

If you look at the problem as having many possible solutions, what you are really saying is there are a number of combinations of material type, texture, lighting, shape, and intervening atmosphere or media that could produce the observed radiation spectrum. Given known constraints, some of these solutions may be more likely than others. You could look at this set of possible solutions as a superposition state with associate probabilities of measurement. Is there some way that the Everett interpretation of quantum mechanics could be applied to perform a kind of annealing algorithm? This might take the form of a cost function minimization, where the cost is essentially the associated probability of the superposition state.

I would love to thank my lucky stars for my inspiration this day!