Dr. Jah Argues for Decision-Intelligent Processes in Collision Avoidance
May 17, 2021 | Space Security
In a recent op-ed written for SpaceWatch.Global, Dr. Moriba Jah, Associate Professor of Aerospace Engineering and Engineering Mechanics at UT Austin and Strauss Center Distinguished Scholar, addressed the space community’s use of “probability of collision” to asses risk of collision on orbit. First, Dr. Jah discusses how operators asses the possibility of their satellite colliding with another object on orbit, delineating the rules under which we presume the null hypothesis—that a satellite is safe—holds. Noting the inherent degree of uncertainty involved in knowing the precise location and movement of objects in orbit, Dr. Jah highlights the necessity of educated inference, noting the four things that influence what we infer about space object in motion. Dr. Jah also defines decision intelligence as the ability to “use and manage information in such a way that leads to desired outcomes” in an environment of uncertainty and incomplete information. He then outlines his methodology for a collision risk assessment process which enshrines decision intelligence, highlighting the role of abductive reasoning and maximum entropy which is an application of Occam’s razor. Once maximum entropy is achieved, Dr. Jah’s process calls for multiple hypothesis testing, which occurs concurrently with an application of Karl Popper’s Falsifiability Principle. After these steps have been taken, we are then able to reject or accept the null hypothesis and align our actions accordingly, i.e. maneuver the satellite to avoid collision or continue operating the satellite as normal. He concludes by reminding that the metric of “probability of collision,” used often by people in the space community, describes a scenario which does exist. The occurrence of a collision is a binary event which must be avoided by way of accepting and modeling our ignorance. Read the full article here.