| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 8.76 MB | Adobe PDF |
Authors
Abstract(s)
Os veículos autónomos receberam nas últimas duas décadas a atenção de
muitos investigadores e cientistas, o que fez com que a sua evolução fosse notória.
Assim, tem ocorrido um largo desenvolvimento na área dos veículos autónomos
submarinos o que leva a um aumento da complexidade e a um aproximar dos modelos
matemáticos à realidade.
A criação de modelos matemáticos realistas ganhou muita preponderância,
pois é possível a realização de testes em ambientes controlados sem ser necessária a
utilização do veículo em si, diminuindo custos e riscos desnecessários. Os modelos
matemáticos são muito diversos e estão dependentes dos graus de liberdade que os
veículos possuem. Após uma verificação e aprovação do modelo matemático este é
aplicado ao veículo.
Paralelamente ao desenvolvimento dos veículos autónomos no geral, também os veículos autónomos biomiméticos têm ganho destaque. Um estudo mais
aprofundado da natureza e uma aproximação da mesma revela vários desafios na
modelação destes veículos.
Um dos objetivos da dissertação é criar um simulador de um veículo autónomo biomimético e estudar o comportamento deste tendo em conta diferentes
estímulos. A criação deste simulador reúne vários desafios e como tal a identificação
de parâmetros dinâmicos surgiu como um desses desafios. Nesta dissertação é abordado um modelo de veículo autónomo biomimético e a respectiva identificação de
parâmetros dinâmicos. A identificação de parâmetros desse modelo permite, a partir
de dados reais, estimar os valores correspondentes de cada um desses parâmetros,
tais como os hidrodinâmicos ou até mesmo os momentos de inércia.
Existem vários parâmetros importantes a retirar do movimento do veículo,
sendo que para os obter os valores de posição, velocidade e aceleração são muito
importantes pois são estes que “projetam” o veículo no espaço.De forma a identificar estes parâmetros utilizou-se o método dos mínimos
quadrados. De forma a aperfeiçoar o método de identificação de parâmetros, os
valores dos parâmetros são conhecidos, permitindo assim calcular o erro entre o
valor real e o valor obtido pela experiência de movimento. Em ambiente simulado
são realizadas diversas experiências para validar o algoritmo de identificação.
The last freelancers received in the last two decades and many researchers and researchers, which was with their notorious evolution. Thus, there has been a broad development in the area of autonomous underwater vehicles which leads to an increase in complexity and the approximation of mathematical models to reality. The creation of realistic mathematical models has gained a lot of preponderance, because it would be possible to carry out tests in controlled environments without having to the use of the vehicle itself, reducing unnecessary costs and risks. The models mathematicians are very diverse and are dependent on the degrees of freedom that the vehicles have. After checking and approving the mathematical model, it is applied to the vehicle. In parallel with the development of autonomous vehicles, vehicles also biomimetic autonomous glasses have gained prominence. A more in-depth study of nature and an approximation of it reveals several challenges in the modeling of these vehicles. One of the objectives of the dissertation is to create a simulator of an automatic biomimetic tone and study its behavior taking into account different stimuli. The creation of this simulator brings together several challenges and as such the identification of dynamic parameters emerged as one of these challenges. In this dissertation it is addressed a biomimetic autonomous vehicle model and its identification of dynamic parameters. The identification of parameters of this model allows, from real data, estimate the corresponding values of each of these parameters, such as hydrodynamics or even moments of inertia. There are several important parameters to take away from the vehicle movement, being that to obtain them the values of position, velocity and acceleration are very important and are these that "project"the vehicle in space.In order to identify these parameters, the minimum method was used. squares. Initially, in order to improve the method of identifying parameters, the parameter values are known, thus allowing the calculation of the error between the real value and the value obtained by the movement experience. In simulated entity, several experiments are carried out to validate the algorithm of identification.
The last freelancers received in the last two decades and many researchers and researchers, which was with their notorious evolution. Thus, there has been a broad development in the area of autonomous underwater vehicles which leads to an increase in complexity and the approximation of mathematical models to reality. The creation of realistic mathematical models has gained a lot of preponderance, because it would be possible to carry out tests in controlled environments without having to the use of the vehicle itself, reducing unnecessary costs and risks. The models mathematicians are very diverse and are dependent on the degrees of freedom that the vehicles have. After checking and approving the mathematical model, it is applied to the vehicle. In parallel with the development of autonomous vehicles, vehicles also biomimetic autonomous glasses have gained prominence. A more in-depth study of nature and an approximation of it reveals several challenges in the modeling of these vehicles. One of the objectives of the dissertation is to create a simulator of an automatic biomimetic tone and study its behavior taking into account different stimuli. The creation of this simulator brings together several challenges and as such the identification of dynamic parameters emerged as one of these challenges. In this dissertation it is addressed a biomimetic autonomous vehicle model and its identification of dynamic parameters. The identification of parameters of this model allows, from real data, estimate the corresponding values of each of these parameters, such as hydrodynamics or even moments of inertia. There are several important parameters to take away from the vehicle movement, being that to obtain them the values of position, velocity and acceleration are very important and are these that "project"the vehicle in space.In order to identify these parameters, the minimum method was used. squares. Initially, in order to improve the method of identifying parameters, the parameter values are known, thus allowing the calculation of the error between the real value and the value obtained by the movement experience. In simulated entity, several experiments are carried out to validate the algorithm of identification.
Description
Keywords
UUV BUV modelação simulação identificação