Best Development Practices for Rapid Deployment of Alt-PNT Systems
Alexander McNeil, Andrew Sosanya, Sriharsha Thoram, Mario Russo, Prasenjit Sengupta, Eric Shoup, Sandbox AQ
Location: Ballroom E
Date/Time: Wednesday, Jun. 3, 11:50 a.m.
Magnetic Anomaly Navigation (MagNav) represents a critical Alternative Position Navigation and Timing (Alt-PNT) capability, yet it presents significant challenges regarding rapid deployment and iteration. Drawing on Sandbox’s execution during the the Defense Innovation Unit’s (DIU) Transition of Quantum Sensing (TQS) program, this work demonstrates that a robust, cloud-centric infrastructure is a key determinant of operational efficiency and algorithmic maturation and can enable rapid deployment in situations where it was not previously possible.
Our development methodology leverages a cloud-native development environment for data analysis, algorithm development, and simulation. We detail a scalable pipeline for managing magnetic anomaly maps that integrates geophysical expertise with automated ingestion tools. This pipeline facilitates an interaction loop which produces high-quality magnetic anomaly maps and ensures that they are readily accessible by the team members. Concurrently, modern build systems were utilized to establish cross-platform and cross-language compilation and development. These build systems allowed developers to deploy code across a range of hardware and software stacks, including a recent focus on integrating containerized solutions.
These architectural decisions have directly enabled the navigation performance metrics achieved on the Anduril CPJ100 platform. We analyze specific flight test results, illustrating how rapid software delivery methods accelerated performance validation and algorithmic iteration. This analysis validates that data pipelines and modern build systems are foundational requirements for deploying operationally relevant MagNav systems.