Adaptive Environmental Sampling for Underwater Vehicles Based on Ant Colony Optimization Algorithm

The high-precision oceanographic environment parameters are important for the study of underwater acoustic channels. However, the oceanographic environment is complex and changeable. The underwater vehicles are useful for collecting data in the ocean, but their energy is limited. Thus, it is meaningful to study adaptive sampling methods of underwater vehicles to collect data in the ocean more efficiently, so as to make the estimation results of the forecast ocean environment more accurate. In the paper, an adaptive sampling method based on the ant colony algorithm (ACA) is proposed and simulated on the temperature field calculated by Regional Ocean Model System (ROMS) model. Firstly, the assimilation results of high resolution multistage spectral interpolation (HRMSI) technique are used to compare the effects of the ACA and the genetic algorithm (GA) in two-dimensional adaptive sampling path planning. Then, the ACA is extended to three-dimensional space considering the changes in the oceanographic environment in space-time domain. The simulation results show that this adaptive sampling method is beneficial for underwater vehicles to collect more efficient information in the ocean.