Spelling suggestions: "subject:"articial aeural anetwork"" "subject:"articial aeural enetwork""
1 |
Sistema de classificação de plantas por meio de suas folhas usando uma arquitetura híbrida composta por algoritmos genéticos e rede neural artificial / Plants classication system through their leaves using a hybrid architecture composed of genetic algorithms and backpropagation neural networkBorges, Thiago Henrique 12 April 2013 (has links)
The number of plants at risk of extinction has increased gradually. With the purpose of
reducing the risk is necessary identify the species for planning protection methods. The
biodiversity of species existing in the plant kingdom make the use of traditional models of
recognition and taxonomy a process very complex and slow. The identification of a plant
can be performed observing his features, such as: fruits, seeds, flowers, roots, leaves and
stems. But the simplest feature used are the leaves.This paper presents a hybrid system
for identifying plant based on leaf image. This system is composed by Genetic Algorithm
(GA) and Artificial Neural Network (ANN). The role played by the GA is to perform
a preselection of plants forming a group that the answer of an unknown leaf is more
probable and the purpose of ANN, trained by backpropagation algorithm, is to classify the
unknown leaf performing the search only in the group calculated by the AG. Several tests
were conducted and the results obtained demonstrate that the hybrid system achieved a
recognition rate of 93,2%. / O número de plantas com risco de extinção tem aumentado gradativamente. Com a finalidade
de diminuir esse risco, faz-se necessário planejar métodos de proteção e identificação
das espécies. A grande biodiversidade de plantas existentes no reino vegetal torna os
modelos tradicionais de identificação e de taxonomia uma função muito complexa e lenta.
A identificação de uma planta pode ser realizada observando várias características, tais
com: frutos, sementes, ores, raízes, folhas e caule. A característica mais simples de ser
utilizada nessa identificação são as folhas. Este trabalho apresenta um sistema híbrido e
automático de identificação de plantas por meio de suas folhas. Esse sistema é composto
por Algoritmos Genéticos (AG) e pela Rede Neural Artificial (RNA). O objetivo do AG
é realizar uma pré-seleção de plantas formando um grupo de folhas desconhecidas que
seriam a resposta mais provável, enquanto que a finalidade da RNA, treinada pelo algoritmo
backpropagation, é classificar a folha considerando apenas o grupo calculado pelo
AG. Vários testes foram realizados e os resultados obtidos mostram que o sistema híbrido
atingiu uma taxa de reconhecimento de 93,2 %. / Mestre em Ciências
|
2 |
Adaptive Steering Behaviour for Heavy Duty VehiclesÅfeldt, Tom January 2017 (has links)
Today the majority of the driver assistance systems are rule-basedcontrol systems that help the driver control the truck. But driversare looking for something more personal and exible that can controlthe truck in a human way with their own preferences. Machine learningand articial intelligence can help achieve this aim. In this studyArticial Neural Networks are used to model the driver steering behaviourin the Scania Lane Keeping Assist. Based on this, trajectoryplanning and steering wheel torque response are modelled to t thedriver preference. A model predictive controller can be used to maintainstate limitations and to weigh the two modelled driver preferencestogether. Due to the diculties in obtaining an internal plant modelfor the model predictive controller a variant of a PI-controller is addedfor integral action instead. The articial neural network also containsan online learning feature to further customize the t to the driverpreference over time. / Idag används till största del regelbaserade reglersystem förförarassistanssystem i lastbilar. Men lastbilschaufförer vill ha någotmer personligt och flexibelt, som kan styra lastbilen på ett mänskligtsätt med förarens egna preferenser. Maskininlärning och artificiell intelligenskan hjälpa till för att uppnå detta mål. I denna studie användsartificiella neurala nätverk för att modellera förarens styrbeteende genomScania Lane Keeping Assist. Med användning av detta modellerasförarens preferenser med avseende på placering på vägbanan och momentpåslag på ratten. En modell prediktiv kontroller kan användas föratt begränsa tillstånd och för att väga de två modellerade preferensernamot varann. Eftersom det var mycket svårt att ta fram den internaprocessmodellen som krävdes för regulatorn används istället en variantav en PI-kontroller för att styra lastbilen. De artificiella neuralanätverken kan också tillåtas att lära sig under körning för att anpassasig till förarens preferenser över tid.
|
Page generated in 0.0708 seconds