کنفرانس بین المللی فناوری اطلاعات و دانش

صفحه اصلی / شانزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

A Data-Driven Hybrid Algorithm for 2D Path Planning via Modeling and Metaheuristic-Based Identification

نویسندگان :

Vahid Safari Dehnavi¹ Masoud Shafiee²

1- دانشگاه صنعتی امیرکبیر 2- دانشگاه صنعتی امیرکبیر

کلمات کلیدی :

Discrete two-dimensional systems،Identification،Metaheuristic algorithms،Hybrid learning algorithm

چکیده :

This paper presents the modeling and identification of discrete two-dimensional causal systems with applications in rehabilitation path planning. Initially, two methods were proposed to estimate the order of model. Then, two models for identification have been used. In the second model, a new hybrid algorithm used to optimize the parameters. In the hybrid algorithm, the linear parameters initialized by least square (LS) and optimized by recursive least square (RLS) with adaptive local forgotten factor in white noise condition. In colored noise conditions, by estimating the filter of noise, colored noise converted to white noise. The hybrid algorithm optimizes nonlinear parameters by global nonlinear optimization algorithms such as metaheuristic algorithms. In metaheuristic algorithms, the grey wolf optimizer (GWO) used to optimize the nonlinear parameters. Finally, nonlinear parameters optimized by using backpropagation algorithm with a new algorithm proposed for determining learning rate. Simulations demonstrate the effectiveness and accuracy of the proposed methods, contributing to advancements in rehabilitation technologies.