SIMPLE Mission Control
Bear Systems is building on its next generation cybersecurity platform to provide a range of SIMPLE (Secure Integrated ML Platform Leveraging Expertise) solutions for the world’s most pressing needs. This blog entry is one of a series of introductions showing how Bear can provide a SIMPLE solution for specific vertical challenges.
We all have heard and/or seen, Mission Control with all of those computer terminals, really smart-looking people cheering, and something magical occurring somewhere in space. Any fan of science fiction films probably remembers some astronaut calling mission control to ask what to do in response to a given impending disaster…
The reality is that the modern mission control is comprised of more automated systems that ever before. The truth is that distance, along with our current inadequate speed of transmission being so slow, results in more and more instructions being programmed into space vehicles. While mission control is still essential, if those operators have to quickly respond to some event then it is already too late.
There are a couple of major issues associated with this approach that are making even this model untenable. The first is the requirement to clone expertise into every new astronaut as each person has to be an expert in so many systems. The high barrier of entry, much less success, is not sustainable if space travel is ever going to move from limited NASA launches into the commercialization world presented by companies such as SpaceX.
The second major issue is that there are just too many unpredictable variables to ever completely program into some remote system. Forcing major software payload updates is too slow and still cannot handle the unknown in any realistic manner.
SIMPLE MC is a new approach to controlling space assets that embeds machine learning and AI directly into each asset. This intelligent system runs a series of workflow engines that control everything from position to power consumption to processing secure communications. The behaviors of each of these workflow engines is controlled in policy. Policies are small, millions can be stored in a very small space and most important, those workflows contain Domain Expert responses to any situation.
The second part of SIMPLE MC consists of a new type of virtualized simulator that can Live Mirror any system and include any type of dynamic data into that virtual system. AI systems drive these simulations with one side representing adversarial or disadvantageous options and the other side representing possible responses. This simulation environment can iterate through millions of scenarios in a continuous manner, be guided by domain experts, and have its results analyzed for optimal results by secondary machine learning algorithms. This latter function provides important training models for the machine learning systems in the space assets.
Instead of transmitting large patches and/or updates, SIMPLE MC transmits a series of Expert Responses and small model updates to the remote SIMPLE MC Agent. Now a given satellite can self-position as part of a virtual private constellation of approved satellites – all with no ground communications. As some meteorite appears out of nowhere, the SIMPLE MC satellite can instantly move out of the way sufficient to clear the obstacles but not so far as to disrupt its primary functionality.
There is obviously more, much more, to our SIMPLE MC solution and, if you are interested, please contact us to learn more!