Depth control for biomimetic and hybrid unmanned underwater vehicles

Abstract

Unmanned underwater vehicles which use biomimetic mechanisms are becoming increasingly useful in the realisation of tasks requiring silent and efficient propulsion. Complex fish kinematics are simplified to some extent and implemented in such vehicles. One of the essential fish behaviours is their ability to adjust their buoyancy using a swim bladder. This paper covers the issues concerning the implementation of artificial swim bladders as well as depth regulators in two underwater vehicles: biomimetic and hybrid. The control of vehicle depth through buoyancy change was examined in the computer simulation and in the experiment. Two types of artificial swim bladder were tested – a rigid cylinder with a piston and an elastic container with a water pump.

Analysis of Neural Network Training Algorithms for Implementation of the Prescriptive Maintenance Strategy

“…The issues discussed in this article may be of interest to a wide audience using data-driven data analysis methods, including in materials sciences [16][17][18][19], machining [20][21][22], construction of internal combustion engines [23], energy control [24,25], mechanics of solids [27,28], civil engineering [29,30], and even in military equipment development [31,32]. The provided analysis may be used in analogous situations requiring the creation of appropriate predictive approximators [33][34][35] and classifiers [36,37].…”

Analysis of Neural Network Structure for Implementation of the Prescriptive Maintenance Strategy

“…The issues discussed in this article may be of interest to a wide audience using data-driven data analysis methods, including in materials sciences [9][10][11], machining [12][13][14], thermodynamics of internal combustion engines [15], energy [16,17], mechanics of solids [18,19], BIM [20,21], road traffic engineering [22,23], and even in military equipment [24,25]. The developed identification (learning) techniques of neural networks can be used in analogous situations requiring the creation of appropriate predictive approximators [26][27][28] and classifiers [29].…”

The use of Computer Simulation Techniques in Production Management

“…The simplest way to describe the role of simulation in the decision-making process is as follows: simulation is an experiment and a simplified imitation (with the help of a computer) of a specific action [10]. Process simulation requires prior construction of appropriate predictive models [11,12], especially in the case of heavy industries [13][14][15], a high risk of contamination [16][17][18][19][20] or a risk of injuries [21][22][23][24]. The construction of predictive models requires the application of many data analysis methods to previously collected data, both parametric [25][26][27] and non-parametric [28][29][30][31].…”