Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 33000000017463 | TJ211.35 A736 2019 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
The book offers an insight on artificial neural networks for giving a robot a high level of autonomous tasks, such as navigation, cost mapping, object recognition, intelligent control of ground and aerial robots, and clustering, with real-time implementations. The reader will learn various methodologies that can be used to solve each stage on autonomous navigation for robots, from object recognition, clustering of obstacles, cost mapping of environments, path planning, and vision to low level control. These methodologies include real-life scenarios to implement a wide range of artificial neural network architectures.
Includes real-time examples for various robotic platforms. Discusses real-time implementation for land and aerial robots. Presents solutions for problems encountered in autonomous navigation. Explores the mathematical preliminaries needed to understand the proposed methodologies. Integrates computing, communications, control, sensing, planning, and other techniques by means of artificial neural networks for robotics.Author Notes
Nancy Arana-Daniel received her B.Sc. Degree from the University of Guadalajara in 2000, and her M. Sc. And Ph.D. degrees in electric engineering with the special field in computer science from Research Center of the National Polytechnic Institute and Advanced Studies, CINVESTAV, in 2003 and 2007, respectively. She is currently a research fellow at the University of Guadalajara, in the Department of Computer Science México, where she is working at the Laboratory of Intelligent Systems and the Research Center for Control Systems and Artificial Intelligence. She is IEEE Senior member and a member of National System of Researchers ( SNI-1). She has published several papers in International Journals and Conferences, and she has been technical manager of several projects that have been granted by the National Council of Science and Technology (CONACYT). Also, she has collaborated in an international project granted by OPTREAT. She is Associated Editor of the Journal of Franklin Institute (Elsevier). Her research interests focus on applications of Geometric Algebra to Geometric computing, machine learning, bio-inspired optimization, pattern recognition, and robot navigation.
Carlos Lopez-Franco gained his Ph.D. in Computer Science in 2007 from the Center of Research and Advanced Studies, CINVESTAV Unidad Guadalajara, Jalisco, México. Currently, he is a full professor at the University of Guadalajara, México, Department of Computer Science. He is currently working with the Intelligent Systems group, and he is the head of the department of Computer Sciences at CUCEI, Universidad de Guadalajara. His current research interests include geometric algebra, computer vision, robotics, and pattern recognition.
Alma Y. Alanis received her B.Sc degree from Instituto Tecnologico de Durango (ITD, Durango Campus, Durango) in 2002, and her M.Sc. and Ph.D. degrees in electrical engineering from the Advanced Studies and Research Center of the National Polytechnic Institute (CINVESTAV-IPN, Guadalajara Campus, Mexico) in 2004 and 2007, respectively. Since 2008, she has been with the University of Guadalajara, where she is currently a Chair Professor in the Department of Computer Science. She is also a member of the Mexican National Research System (SNI-2). She has published papers in recognized International Journals and Conferences, along with two international books. She is a Senior Member of the IEEE and Subject and Associated Editor of the Journal of Franklin Institute (Elsevier) and Intelligent Automation & Soft Computing (Taylor & Francis); moreover, she is currently serving on a number of IEEE and IFAC Conference Organizing Committees. In 2013, she received the grant for women in science by L'Oreal-UNESCO-AMC-CONACYT-CONALMEX. In 2015, she received the Marcos Moshinsky Research Award. Since 2008, she has been a member of the Accredited Assessors record (RCEA-CONACYT), evaluating a wide range of national research projects. She has belonged to important project evaluation committees for national and international research projects. Her research interests are in neural control, backstepping control, block control, and their applications to electrical machines, power systems, and robotics.
Table of Contents
Preface | p. xi |
Abbreviations | p. xvii |
1 Recurrent High Order Neural Networks for Rough Terrain Cost Mapping | p. 1 |
1.1 Introduction | p. 1 |
1.1.1 Mapping background | p. 3 |
1.2 Recurrent High Order Neural Networks, RHONN | p. 5 |
1.2.1 RHONN order | p. 6 |
1.2.2 Neural network training | p. 8 |
1.2.2.1 Kalman filter | p. 8 |
1.2.2.2 Kalman filter training | p. 8 |
1.2.2.3 Extended Kalman filter-based training algorithm, EKF | p. 8 |
1.3 Experimental Results: Identification of Costs Maps Using RHONNs | p. 10 |
1.3.1 Synthetic dynamic environments | p. 11 |
1.3.1.1 Synthetic dynamic random environment number 1 | p. 12 |
1.3.1.2 Synthetic dynamic random environment number 2 | p. 18 |
1.3.1.3 Synthetic dynamic random environment number 3 | p. 18 |
1.3.2 Experiments using real terrain maps | p. 27 |
1.3.2.1 Real terrain map: grove environment | p. 27 |
1.3.2.2 Real terrain map: golf course | p. 29 |
1.3.2.3 Real terrain map: forest | p. 29 |
1.3.2.4 Real terrain map: rural area | p. 33 |
1.4 Conclusions | p. 33 |
2 Geometric Neural Networks for Object Recognition | p. 37 |
2.1 Object Recognition and Geometric Representations of Objects | p. 37 |
2.1.1 Geometric representations and descriptors of real objects | p. 40 |
2.2 Geometric Algebra: An Overview | p. 41 |
2.2.1 The geometric algebra of n-D space | p. 42 |
2.2.2 The geometric algebra of 3-D space | p. 44 |
2.2.3 Conformal geometric algebra | p. 45 |
2.2.4 Hyperconformal geometric algebra | p. 47 |
2.2.5 Generalization of G 6,3 into G 2n,n | p. 48 |
2.3 Clifford SVM | p. 49 |
2.3.1 Quaternion valued support vector classifier | p. 53 |
2.3.2 Experimental results | p. 53 |
2.4 Conformal Neuron and Hyper-Conformal Neuron | p. 55 |
2.4.1 Hyperellipsoidal neuron | p. 55 |
2.4.2 Experimental results | p. 57 |
2.5 Conclusions | p. 57 |
3 Non-Holonomic Robot Control Using RHONN | p. 61 |
3.1 Introduction | p. 61 |
3.2 RHONN to Identify Uncertain Discrete-Time Nonlinear Systems | p. 63 |
3.3 Neural Identification | p. 63 |
3.4 Inverse Optimal Neural Control | p. 64 |
3.5 IONC for Non-Holonomic Mobile Robots | p. 66 |
3.5.1 Robot model | p. 66 |
3.5.2 Wheeled robot | p. 68 |
3.5.2.1 Controller design | p. 68 |
3.5.2.2 Neural identification of a wheeled robot | p. 69 |
3.5.2.3 Inverse optimal control of a wheeled robot | p. 70 |
3.5.2.4 Experimental results | p. 71 |
3.5.3 Tracked robot | p. 71 |
3.5.3.1 Controller design | p. 75 |
3.5.3.2 Results | p. 75 |
3.6 Conclusions | p. 122 |
4 NN for Autonomous Navigation on Non-Holonomic Robots | p. 123 |
4.1 Introduction | p. 123 |
4.2 Simultaneous Localization and Mapping | p. 124 |
4.2.1 Prediction | p. 125 |
4.2.2 Observations | p. 126 |
4.2.3 Status update | p. 126 |
4.3 Reinforcement Learning | p. 127 |
4.4 Inverse Optimal Neural Controller | p. 129 |
4.4.1 Planning-Identifier-Controller | p. 129 |
4.5 Experimental Results | p. 131 |
4.6 Conclusions | p. 145 |
5 Holonomic Robot Control Using Neural Networks | p. 151 |
5.1 Introduction | p. 151 |
5.2 Optimal Control | p. 154 |
5.3 Inverse Optimal Control | p. 156 |
5.4 Holonomic Robot | p. 159 |
5.4.1 Motor dynamics | p. 159 |
5.4.2 Neural identification design | p. 160 |
5.4.3 Control design | p. 161 |
5.4.4 Omnidirectional mobile robot kinematics | p. 162 |
5.5 Visual Feedback | p. 162 |
5.6 Simulation | p. 164 |
5.7 Conclusions | p. 167 |
6 Neural Network-Based Controller for Unmanned Aerial Vehicles | p. 169 |
6.1 Introduction | p. 169 |
6.2 Quadrotor Dynamic Modeling | p. 170 |
6.3 Hexarotor Dynamic Modeling | p. 172 |
6.4 Neural Network-Based PID | p. 175 |
6.5 Visual Servo Control | p. 176 |
6.5.1 Control of hexarotor | p. 177 |
6.6 Simulation Results | p. 178 |
6.6.1 Quadrotor simulation results | p. 178 |
6.6.2 Hexarotor simulation results | p. 178 |
6.7 Experimental Results | p. 180 |
6.7.1 Quadrotor experimental results | p. 180 |
6.7.2 Hexarotor experimental results | p. 183 |
6.8 Conclusions | p. 184 |
Bibliography | p. 195 |
Index | p. 209 |