An approach to clearing Space Debris using Artificial Intelligence to Direct a Modular Satellite

The MARS Aerospace Company vehicle design proposal is meant to provide a platform to solve any-and-all On-orbit Servicing, Assembly, & Manufacturing (OSAM) applications. Through Modular Assembly & Rapid Servicing (MARS) techniques. A heavy focus is put on vehicle longevity and modifiable components to encompass a variety of mission profiles. Removal of space debris by this design will performed with "De-Orbiter" payload in the main vehicle Hub. These MARS Vehicle de-orbiters provide debris removal capabilities of up to 1600kg, and and operational altitudes of up to 30,000km. Using A.I. to simulate a variety of orbital encounter scenarios, the MARS vehicle software will be capable of rendezvous with a wide range of capturable objects. The software solution, that integrates with the hardware solution, is an agent that determines the optimal trajectory to capture objects. This agent will be trained as an AI by evaluating an array of simulated captures across a database of objects with varying size and shape. The agent is then utilized for in situ operations by having the on-board sensors provide information about an object of interest, followed by the AI providing the optimal estimate of how best to capture that object. The team will use the UF HiPerGator-AI supercomputer, which is tailored for AI applications, to ensure the agent is trained for a huge database within the 4-month timeframe for the Phase-I effort.  In application, once the AI indicates the optimal location to intercept an object, the craft will initiate the appropriate maneuver to achieve capture - a completely autonomous docking or capture operation with foreign objects in the space domain.