Get Started
Home
Topics
Search
Library
Research questionHow can vessel tracking select one informative sensor at each step in noisy, nonlinear maritime environments?A tracker must estimate a vessel’s state from noisy measurements while deciding which sensor to activate. Activating every sensor increases sensing cost, while repeatedly calculating expected information gain can be computationally expensive.
AI
Machine Learning
Reinforcement Learning
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Reinforcement Learning for Heterogeneous Sensor Selection in Maritime SurveillanceEvidence comes from a georeferenced simulation of the CMMI Smart Marina testbed in Ayia Napa Marina, using its actual five-sensor configuration and a Bayesian sequential Monte Carlo tracker. A learned policy selected one sensor per decision epoch; comparisons included random selection, always-on sensing, and an expected-information-gain baseline. Zero-shot tests covered ten moderately perturbed layouts, so the evidence is limited to this maritime tracking setting and these evaluated configurations.research paper · Sep 2, 2026
Related questions
How can maritime LiDAR detect both small sparse vessels and large-vessel structure amid sea clutter?How can remote robots maintain reliable control over bandwidth-limited, changing wireless links without transmitting every high-dimensional observation?How can multiple aerial robots identify and track individual wildlife using onboard RGB cameras under decentralized, low-bandwidth coordination?How can complex neural-network graphs be mapped across heterogeneous SoCs to balance inference latency and throughput?