Adaptive Integrated Control for Omnidirectional Mobile Manipulators Based on Neural-Network

An omnidirectional mobile manipulator, due to its large-scale mobility and dexterous manipulability, has attracted lots of attention in the last decades. However, modeling and control of such systems are very challenging because of their complicated mechanism. In this article, an unified dynamic model is developed by Lagrange Formalism. In terms of the proposed model, an adaptive integrated tracking controller, based on the computed torque control (CTC) method and the radial basis function neural-network (RBFNN), is presented subsequently. Although CTC is an effective motion control strategy for mobile manipulators, it requires precise models. To handle the unmodeled dynamics and the external disturbance, a RBFNN, serving as a compensator, is adopted. This proposed controller combines the advantages of CTC and RBFNN. Simulation results show the correctness of the proposed model and the effectiveness of the control approach. DOI: 10.4018/jcini.2009062303 IGI PUBLISHING This paper appears in the publication, International Journal of Cognitive Informatics and Natural Intelligence, Volume 3, Issue 4 edited by Yingxu Wang © 2009, IGI Global 701 E. Chocolate Avenu , He shey PA 17033-1240, USA Tel: 717/533-8845; Fax 717/533-8661; URL-http://www.igi-global.com ITJ 5460 International Journal of Cognitive Informatics and Natural Intelligence, 3(4), 34-53, October-December 2009 35 Copyright © 2009, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. greatly improves the functionality of the mobile manipulator. Because of these distinct advantages, mobile manipulators have been applied more and more extensively. A number of related works have been developed in this field in the last decades. In literatures (Campion and Bastin, 1996; Betourne and Campion, 1996; SicilianO, Wit, &Bastin, 1996), modeling and control of omnidirectional mobile robots were analyzed in details. The dynamic model and kinematic model of an omnidirectional mobile robot with three castor wheels were presented in (Chung, et al., 2003; Yi and Kim, 2001). Tan and Xi (2001) proposed a unified dynamic model for a mobile manipulator consisting of a Nomadic XR4000 and a Puma 560 robot arm. Holmberg and Khatib (2000), Khatib, et al. (1996), Khatib (1987) developed a holonomic mobile robot and presented a dynamic control method for a parallel redundant system. On the whole, existing works can be approximately divided into two groups: the holonomic mechanical system and the nonholonomic mechanical system. The omnidirectional mobile manipulator is a typical example of holonomic mechanical system. Compared to nonholonomic mechanical systems such as differential-driven mobile robots, the holonomic system has several advantages. Firstly, avoidance of nonholonomic constraints made the holonomic system more dexterous and easy to be controlled. Secondly, it can completely use the null space motions to improve the workspace and overall dynamic endpoint properties, and the redundant degrees of freedom can be used to accomplish accessory tasks. However, while the omnidirectional mobile manipulator provides a lot of advantages, we suffer a lot from its complicated mechanical structure. Firstly, it is very difficult to derive the model of an omnidirectional mobile manipulator, especially the dynamic model. Secondly, more redundant degrees of freedom also implies more motors need to be controlled, therefore, path planning and motion control of such a system are very challenging. Moreover, many intelligent control methods are unable to be applied online due to the computational complexity of the model. Existing control strategies of the omnidirectional mobile manipulator can be approximately divided into two groups: One is separate control of the mobile platform and the robot arm, i.e., the mobile manipulator is regarded as two subsystems. In this way, Liu and Lewis presented a decentralized robust controller in (Liu and Lewis, 1990). Chung and Velinsky (1998) also derived the dynamic model of the holonomic mobile platform and the manipulators separately. The other is control the mobile manipulator as a whole, that is, we consider the mobile platform as a multiple-DOFs joint. For example, Yamamoto and Yun (1994) studied a two-linked planar mobile manipulator subject to nonholonomic constraints. Considering the characteristics of our experimental setup, we adopt the latter control strategy in this article. As for these methods, CTC is an effective motion control strategy. However, requirements of precise model are very rigor in practical applications. Furthermore, this kind of controllers becomes unstable when the unmodeled dynamics or the external disturbance is significant. To handle these difficulties, Song and Yi (2005) adopted a fuzzy approach to compensate the uncertainties. In literatures (Kwan, et al., 1998, Wai, 2003), generalized fuzzy neural network, wavelet neural network were introduced in their controller respectively. In this article, a robust controller, based on CTC and RBFNN, is presented. As pointed in (Berni, et al.,2003; Er and Gao, 2003; Li, 2006;Hou and Tan, 2004; Fang, Zhao, &Li, 2005), artificial neural networks have been adopted extensively for their ability to perform nonlinear mappings. It has been used widely in modeling and control, artificial intelligence, cognitive informatics (Wang, Y., 2007; Wang, Y., 2006). Because neural networks provide a fast method of autonomously learning the relation between a set of output states and a set of input states, we introduce a RBFNN into the controller to approximate the unstructured or structured uncertainties of the proposed model in this article. 18 more pages are available in the full version of this document, which may be purchased using the "Add to Cart" button on the product's webpage: www.igi-global.com/article/adaptive-integrated-controlomnidirectional-mobile/37573?camid=4v1 This title is available in InfoSci-Journals, InfoSci-Journal Disciplines Computer Science, Security, and Information Technology. Recommend this product to your librarian: www.igi-global.com/e-resources/libraryrecommendation/?id=2