Muutke küpsiste eelistusi

E-raamat: Robot Learning by Visual Observation

,
  • Formaat: PDF+DRM
  • Ilmumisaeg: 13-Jan-2017
  • Kirjastus: John Wiley & Sons Inc
  • Keel: eng
  • ISBN-13: 9781119091783
Teised raamatud teemal:
  • Formaat - PDF+DRM
  • Hind: 128,38 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
  • Raamatukogudele
Robot Learning by Visual ObservationSuurem pilt
  • Formaat: PDF+DRM
  • Ilmumisaeg: 13-Jan-2017
  • Kirjastus: John Wiley & Sons Inc
  • Keel: eng
  • ISBN-13: 9781119091783
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

The book presents the methodology for robot learning from observations. The focus of the book is on trajectory level of task abstraction. The content is divided into several chapters, with each chapter addressing the methods for tackling individual steps in the observational learning problem. The book describes several methods for mathematical modelling of observed trajectories, such as hidden Markov models, conditional random fields, Gaussian mixture models, and dynamic motion primitives. Methods for generalization of the demonstrated trajectories and for generation of a trajectory for task reproduction are presented. Furthermore, the authors describe methodology for robust execution of generated plans for task reproduction, which employs vision based robot control.

Preface x
List of Abbreviations
xiii
1 Introduction
1(42)
1.1 Robot Programming Methods
2(1)
1.2 Programming by Demonstration
3(1)
1.3 Historical Overview of Robot PbD
4(2)
1.4 PbD System Architecture
6(15)
1.4.1 Learning Interfaces
8(2)
1.4.1.1 Sensor-Based Techniques
10(3)
1.4.2 Task Representation and Modeling
13(1)
1.4.2.1 Symbolic Level
14(2)
1.4.2.2 Trajectory Level
16(2)
1.4.3 Task Analysis and Planning
18(1)
1.4.3.1 Symbolic Level
18(1)
1.4.3.2 Trajectory Level
19(1)
1.4.4 Program Generation and Task Execution
20(1)
1.5 Applications
21(4)
1.6 Research Challenges
25(7)
1.6.1 Extracting the Teacher's Intention from Observations
26(1)
1.6.2 Robust Learning from Observations
27(1)
1.6.2.1 Robust Encoding of Demonstrated Motions
27(2)
1.6.2.2 Robust Reproduction of PbD Plans
29(1)
1.6.3 Metrics for Evaluation of Learned Skills
29(1)
1.6.4 Correspondence Problem
30(1)
1.6.5 Role of the Teacher in PbD
31(1)
1.7 Summary
32(11)
References
33(10)
2 Task Perception
43(6)
2.1 Optical Tracking Systems
43(1)
2.2 Vision Cameras
44(2)
2.3 Summary
46(3)
References
46(3)
3 Task Representation
49(8)
3.1 Level of Abstraction
50(1)
3.2 Probabilistic Learning
51(1)
3.3 Data Scaling and Aligning
51(4)
3.3.1 Linear Scaling
52(1)
3.3.2 Dynamic Time Warping (DTW)
52(3)
3.4 Summary
55(2)
References
55(2)
4 Task Modeling
57(16)
4.1 Gaussian Mixture Model (GMM)
57(2)
4.2 Hidden Markov Model (HMM)
59(5)
4.2.1 Evaluation Problem
61(1)
4.2.2 Decoding Problem
62(1)
4.2.3 Training Problem
62(1)
4.2.4 Continuous Observation Data
63(1)
4.3 Conditional Random Fields (CRFs)
64(4)
4.3.1 Linear Chain CRF
65(1)
4.3.2 Training and Inference
66(2)
4.4 Dynamic Motion Primitives (DMPs)
68(2)
4.5 Summary
70(3)
References
70(3)
5 Task Planning
73(56)
5.1 Gaussian Mixture Regression
73(1)
5.2 Spline Regression
74(43)
5.2.1 Extraction of Key Points as Trajectories Features
75(5)
5.2.2 HMM-Based Modeling and Generalization
80(1)
5.2.2.1 Related Work
80(1)
5.2.2.2 Modeling
81(2)
5.2.2.3 Generalization
83(4)
5.2.2.4 Experiments
87(13)
5.2.2.5 Comparison with Related Work
100(7)
5.2.3 CRF Modeling and Generalization
107(1)
5.2.3.1 Related Work
107(1)
5.2.3.2 Feature Functions Formation
107(2)
5.2.3.3 Trajectories Encoding and Generalization
109(2)
5.2.3.4 Experiments
111(4)
5.2.3.5 Comparisons with Related Work
115(2)
5.3 Locally Weighted Regression
117(4)
5.4 Gaussian Process Regression
121(1)
5.5 Summary
122(7)
References
123(6)
6 Task Execution
129(60)
6.1 Background and Related Work
129(3)
6.2 Kinematic Robot Control
132(2)
6.3 Vision-Based Trajectory Tracking Control
134(7)
6.3.1 Image-Based Visual Servoing (IBVS)
134(1)
6.3.2 Position-Based Visual Servoing (PBVS)
135(6)
6.3.3 Advanced Visual Servoing Methods
141(1)
6.4 Image-Based Task Planning
141(15)
6.4.1 Image-Based Learning Environment
141(1)
6.4.2 Task Planning
142(1)
6.4.3 Second-Order Conic Optimization
143(1)
6.4.4 Objective Function
144(2)
6.4.5 Constraints
146(1)
6.4.5.1 Image-Space Constraints
146(3)
6.4.5.2 Cartesian Space Constraints
149(1)
6.4.5.3 Robot Manipulator Constraints
150(2)
6.4.6 Optimization Model
152(4)
6.5 Robust Image-Based Tracking Control
156(27)
6.5.1 Simulations
157(1)
6.5.1.1 Simulation 1
158(3)
6.5.1.2 Simulation 2
161(1)
6.5.2 Experiments
162(4)
6.5.2.1 Experiment 1
166(7)
6.5.2.2 Experiment 2
173(4)
6.5.2.3 Experiment 3
177(1)
6.5.3 Robustness Analysis and Comparisons with Other Methods
177(6)
6.6 Discussion
183(2)
6.7 Summary
185(4)
References
185(4)
Index 189
ALEKSANDAR VAKANSKI is a Clinical Assistant Professor in Industrial Technology at the University of Idaho, Idaho Falls, USA. He received a Ph.D. degree from the Department of Mechanical and Industrial Engineering at Ryerson University, Toronto, Canada, in 2013. The scope of his research interests encompasses the fields of robotics and mechatronics, artificial intelligence, computer vision, and control systems.

FARROKH JANABI-SHARIFI is a Professor of Mechanical and Industrial Engineering and the Director of Robotics, Mechatronics and Automation Laboratory (RMAL) at Ryerson University, Toronto, Canada. He is currently a Technical Editor of IEEE/ASME Transactions on Mechatronics, an Associate Editor of The International Journal of Optomechatronics, and an Editorial Member of The Journal of Robotics and The Open Cybernetics and Systematics Journal. His research interests include optomechatronic systems with the focus on image-guided control and planning.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97811190917832e.html
Märksõnad: