RAPIDFORT INC — Department of Defense STTR Phase I: AF21S-TCSO1

Description

OSAMs Need for Agile Software The impressive hardware advances of the satellite industry have far outpaced its software. However, OSAM's new mission provides a clear opportunity to innovate at the software layer. Historically, satellite software has been "simple and reliable" in that most of the complex processing was done by ground-based systems and relayed to the satellite. Programming languages used for these on-satellite systems are dated, yet stable, and were primarily invented 50 years ago: HAL (1970), ADA (1983), C (1970), and C++ (1985). These languages have several performance and interoperability restrictions as evidenced by the quote from Google founder Larry Page: "Python wherever we can, C++ where we must." In Customer Discovery Discussions with an Executive from SES, we learned that: "Historically, our software orientation has been towards simple, rudimentary, bare-bones deployments that are reliable. We prefer 'heritage software' proven, old-fashioned stuff that has flown before and has lots of mission time behind it. The satellite is expected to be in service for 10+ years and we can ill-afford an unrecoverable "bluescreen event" that loses control of the asset. But satellites now need increased intelligence and performance. So, we need to start using new technologies for performance and cost reasons. Alas, we are stuck between the trusted old and the much needed yet uncertain new." In particular, there are at least two key needs where modern software must be deployed to deliver the OSAM mission: 1. On-Orbit Object Approach: Satellites are moving at 16,000 MPH, so the docking (Rendezvous Proximity Operations Docking (RPOD)) component will require AI/ML frameworks to support autonomous docking. AI/ML frameworks tend to run on agile, open-source frameworks, so there may be a learning curve in adopting these technologies within the OSAM community. In the commercial AI/ML space, there is a significant movement towards agile open-source technologies where 50% to 70% of all academic papers on AI are based on open-source.  It is expected that 65% of all AI will use open-source by 2023. The aerospace industry is expected to follow the lead of its commercial counterparts and a key benefit of this proposal is that it enables software to be future-proof. 2. On-Orbit Object Acquisition: Capture and Control: Again, these are new capabilities for satellites, and there is a significant movement towards agile open-source in robotics with modern frameworks. For example, Astrobe Robots on the ISS are running on open-source operating system called ROS. Software optimization makes agile software more secure, faster and cheaper to build, easier to maintain, faster to load, and more affordable to run. RapidFort is the creator and leader of this new market category. RapidFort's solutions have a high probability of keeping pace with technological change and are closely tied to best practices in commercial technologies and solutions.