Search

Abstract This article is concerned with the state estimation problem for linear systems with linear state equality constraints. We re-examine constrained Kalman filter variations and propose an alternative derivation of the optimal constrained Kalman filter for time variant systems. This results in an oblique state projection that gives the smallest error covariance. A simple example illustrates the performance of the different Kalman filters. [ABSTRACT FROM AUTHOR]

The open wording of the traffic rules of the sea, COLREGS, and the existence of unwritten rules, make it essential for an autonomous ship to understand the intentions of other ships. This article uses a dynamic Bayesian network (DBN) to model and infer the intentions of other ships in open waters based on their observed real-time behavior. Multiple intention nodes are included to describe the different ways a ship can interpret and conflict with the behavioral rules outlined in COLREGS. The prior probability distributions of the intention nodes are adapted to the current situation based on observable characteristics such as location and relative ship size. The resulting model is able to identify situations that are prone to cause misunderstandings and infer the state of multiple intention variables that describe how the ship is likely to behave. Different collision avoidance algorithms can use the resulting intention information to better know if, when, and how to act. • Intention modeling and inference of ships for use in collision avoidance systems. • Intentions of ships are modeled and inferred using a dynamic Bayesian network. • Multiple different intention states are considered. • Underlying cases are inferred by combining information over time. • Prior information is used to adapt the model to the current situation. [ABSTRACT FROM AUTHOR]

Sensor fusion plays a key part in autonomous surface vehicles, however, the high cost of sensors makes the barrier of entry in this research field quite high. In this work, we present a complete system for sensor fusion on the milliAmpere autonomous ferry research platform as well as an open sensor fusion dataset for maritime tracking across two environments. Individual sensors and their detection pipelines are evaluated across various detection metrics. We also evaluate the tracking performance of the sensors both individually and in fusion using a multi-sensor extension of the JIPDA multi-target tracker. We find that the different environments have distinct challenges precluding the use of only a single sensor. Utilizing multiple sensors, either individually or in fusion, can mitigate these issues increasing the safety margins of the situational awareness system. • Experimental validation of a multi-sensor extension of the JIPDA multi-target tracker. • In-depth description of a complete sensor fusion pipeline from sensor data to detection and tracking. • Open sensor fusion dataset for maritime multi-target tracking with multiple environments and scenarios. • Multiple sensors improve safety margins across differing environments. [ABSTRACT FROM AUTHOR]