8 February 2024
Systems and Control
Proposes learning framework for robust underwater gyrocompassing
Learns to extract earth's rotation rate signal for heading estimation
Mitigates effects of ocean currents and disturbances on UUVs
Assesses performance via simulations of challenging conditions
Contributes a resilient heading solution for autonomous vehicles
Learning-based underwater heading determination
This paper introduces a machine learning framework to enable precise determination of heading angle for unmanned underwater vehicles. It focuses on mitigating environmental disturbances that degrade standard gyrocompassing methods. By analyzing inertial measurements, the framework learns to extract earth's rotation rate vector, even in dynamic conditions. Simulations assess its adaptability to challenges of underwater navigation.
No comments yet, be the first to start the conversation...
Sign up to comment on this paper