Line-of-sight path-following control utilizing an extended Kalman filter for estimation of speed and course over ground from GNSS positions

Abstract Path-following control systems for ships can be designed using both heading and course angle autopilots in conjunction with a proportional line-of-sight (LOS) guidance law. Ships are usually equipped gyrocompass from safety perspective since magnetic compasses susceptible to disturbances. Unfortunately, the is an expensive device, smaller vessels boats cannot afford use this as primary device steering. An alternative solution compute over ground (COG) speed (SOG) global navigation satellite (GNSS) these signals feedback control. This article presents autopilot design path following five-state extended Kalman filter (EKF) estimate COG SOG efficiently. Even though many algorithms exist computation of SOG, it advantageous EKF state estimator include other sensory such Doppler Velocity Log (DVL), radar, attitude rate sensors, computer vision systems, etc. contrast proprietary that do not allow user modify software. The convergence accuracy significantly improved by target-tracking models combination kinematic equations. A high-fidelity model MARINER class cargo ship used path-following case study. From simulation study, concluded successfully estimates GNSS measurements during following. remarkably robust accurate, when combined autopilot, need compass eliminated