Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
Abstract Neural architectures have been proposed to navigate mobile robots within several environment definitions. In this paper a new neural modular constructive approach to navigate mobile robots in unknown environments is presented. The problem, in its basic form, consists of defining and executing a trajectory to a predefined goal while avoiding all obstacles, in an unknown environment. Some crucial issues arise when trying to solve this problem, such as an overflow of sensorial information and conflicting objectives. Most neural network (NN) approaches to this problem focus on a monolithic system, i.e., a system with only one neural network that receives. | Neural Comput Applic 2003 12 200-211 DOI 10.1007 s00521-003-0383-y ORIGINAL ARTICLE Catarina Silva Bernardete Ribeiro Navigating mobile robots with a modular neural architecture Received 28 March 2002 Accepted 3 July 2003 Published online 14 November 2003 Springer-Verlag London Limited 2003 Abstract Neural architectures have been proposed to navigate mobile robots within several environment definitions. In this paper a new neural modular constructive approach to navigate mobile robots in unknown environments is presented. The problem in its basic form consists of defining and executing a trajectory to a predefined goal while avoiding all obstacles in an unknown environment. Some crucial issues arise when trying to solve this problem such as an overflow of sensorial information and conflicting objectives. Most neural network NN approaches to this problem focus on a monolithic system i.e. a system with only one neural network that receives and analyses all available information resulting in conflicting training patterns long training times and poor generalisation. The work presented in this article circumvents these problems by the use of a constructive modular NN. Navigation capabilities were proven with the NOMAD 200 mobile robot. Keywords Constructive algorithms Mobile robot navigation Modular neural networks Sensor fusion 1 Introduction There has been an increasing interest in the development of intelligent mobile robots i.e. robots that are able to learn to navigate and act in complex possibly unknown environments. This interest grew with the realisation that a mobile robot s application environments are usually dynamic leading to the search for new solutions for mobile robot navigation. When attempting to satisfy the requirements of such a control system it is often difficult to find a unique C. Silva B. Ribeiro Centro de Informatica e Sistemas Universidade de Coimbra 3030 Coimbra Portugal E-mail catarina@dei.uc.pt control law that applies in all relevant .
