Title :
Application of stochastic optimization to collision avoidance
Author_Institution :
Appl. Phys. Laboratory, Johns Hopkins Univ., Baltimore, MD, USA
fDate :
June 30 2004-July 2 2004
Abstract :
This paper applies simulation-based optimization to the problem of vessel traffic management in a high vessel density environment. Specifically, a Monte Carlo simulation has been developed that models a relatively small craft operating in a high vessel density environment under poor visibility conditions. In this simulation, the small vessel maneuvers to keep all other vessels outside some acceptable range (one of the objectives of the optimization) subject to the requirements on the other vessels to obey pre-established traffic management rules. Several stochastic optimization algorithms- blind random search, simultaneous perturbation stochastic approximation, and simulated annealing are applied to this problem ´with conclusions drawn regarding their relative applicability and performance, as well as the practical implications of the results.
Keywords :
Monte Carlo methods; collision avoidance; marine vehicles; simulated annealing; stochastic processes; traffic; Monte Carlo simulation; blind random search; collision avoidance; high vessel density environment; simulated annealing; simultaneous perturbation stochastic approximation; stochastic optimization; vessel traffic management;
Conference_Titel :
American Control Conference, 2004. Proceedings of the 2004
Conference_Location :
Boston, MA, USA
Print_ISBN :
0-7803-8335-4