Software Driven Design

Software-driven design principles go beyond merely selecting the tools used in the aerospace system design process such as Autodesk Inventor or ANSYS. They are a key component of our low-cost strategy for software solutions throughout development and operations.

Our strategy begins at the design table. We employ industry-standard tools during system definition and detailed design to produce verifiable end products. This reduces the overall development cycle time and manufacturing cost. Additionally, it enables us to create designs that are ready for manufacture using standard commercial industrial-grade processes and tooling.
Another important aspect that we recognize is that in order for a space vehicle to pass rigorous certifications or obtain a launch license, many of the tools used to create the vehicle must also be verified and/or certified. A space vehicle can only be as reliable as the tools and processes used to create it. Our design work may be passed to regulatory agencies, insurers, clients and other stakeholders for analysis and verification much more readily than if all the work was performed by hand or custom software.

However, we carry the concept software-driven design throughout the lifecycle. In part, we leverage licensing with NASA and other third parties to acquire software, at low or no cost that has already been verified. Examples of this include Rocket Engine Transient Simulation (ROCETS), ML_POGO for POGO analysis, Ground Operations Aerospace Language (GOAL) for ground processing and launch automation and other software packages.

Focusing on software processes and leveraging what is already available reduces our development cycles by months or years. It’s also critical to maintaining a relatively small team and high level of automation.

We extend our focus on software design to operations as can be seen in the method of thrust vector control of our Phoenix launch vehicle. Primary objectives are to have minimal cost, mass and moving parts. Rather than implementing gimbaled engines or other “active” thrust vectoring, we made the decision to push as much of it into the software as possible. The avionics “knows” the current state of the vehicle and any steering that needs to be executed. It’s also a given that each of the combustion chambers may be independently throttled. In our solution, rather than the system driving mechanical components like gimbals, each combustion chamber is throttled to create a thrust offset that causes a change in attitude of the vehicle and steers it on the proper trajectory.

By moving complexity into avionics software, we are able to reduce the mass and size of the launch vehicle while increasing overall reliability. This is just one area where we have applied software design methodologies to drive the hardware and operational design for our launch vehicle. This has an added benefit of lowering the cost of manufacture as well as streamlining launch operations.