Reproduce “Observability of Strapdown INS Alignment: A Global Perspective” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuanxin Wu, Hongliang Zhang, Meiping Wu, Xiaoping Hu, Dewen Hu
This paper presents a global, constructive observability analysis for strapdown INS alignment, demonstrating that static alignment is fundamentally unobservable and single-axis rotations typically result in two indistinguishable states for sensor biases. It finds that full observability for initial attitude and sensor biases is achieved only through successive rotations about two or more linearly independent axes, challenging prior linearization-based findings.
Key insights
- Observability of nonlinear strapdown INS alignment can be fundamentally understood without linearization, revealing inherent properties and ambiguities that linearization might obscure.
- The specific orientation and type of attitude maneuvers critically dictate the estimability of sensor biases and initial attitude, often leading to multiple indistinguishable solutions under single-axis rotations.
- Full observability, guaranteeing unique determination of all unknown initial states and sensor biases, inherently requires the system to undergo successive rotations about at least two linearly independent axes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.