A5030365199- Neural Networks and Applications
- Advanced Adaptive Filtering Techniques
- Distributed Control Multi-Agent Systems
- Robotic Path Planning Algorithms
- Analog and Mixed-Signal Circuit Design
- Smart Parking Systems Research
- Control Systems and Identification
Yokohama National University
2022-2024
Although various studies have been conducted on automatic berthing, including offline optimization and online control, real-time berthing control remains a difficult problem. Online methods without reference trajectories are promising for control. We used reinforcement learning (RL), which is type of machine learning, to obtain an law trajectories. As difficult, obtaining appropriate with naive difficult. Furthermore, almost all existing do not consider port geometries. This study proposes...
The covariance matrix adaptation evolution strategy (CMA-ES) is a powerful optimization method for continuous black-box problems. Several noise-handling methods have been proposed to bring out the performance of CMA-ES on noisy objective functions. adaptations population size and learning rate are two major approaches that perform well under additive Gaussian noise. reevaluation technique another evaluates each solution multiple times. In this paper, we discuss difference between those from...