1:30 PM - 2:10 PM
[2I4-OS-9a-01] (OS invited talk) Distributed Ship Collision Avoidance Algorithms
Keywords:Ship Collision Avoidance, Distributed Constraint Optimization, Deep Reinforcement Learning
Distributed ship collision avoidance (DisSCA) algorithms are one of the applied studies of distributed constraint satisfaction/optimization that realizes automatic collision avoidance between autonomous ships through ship-to-ship communication under many-to-many situations. In this work, we introduce the Distributed Stochastic Search Algorithm (DSSA), which is one of the DisSCA algorithms, and its advanced version, DSSA+. We also introduce the latest project on DisSCA, called DSSQ, which aims at realizing more efficient collision avoidance by incorporating deep reinforcement learning into DSSA+.
Authentication for paper PDF access
A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.