Artificial Intelligence for Future Generation Robotics offers a vision for potential future robotics applications for AI technologies. Each chapter includes theory and mathematics to stimulate novel research directions based on the state-of-the-art in AI and smart robotics. Organized by application into ten chapters, this book offers a practical tool for researchers and engineers looking for new avenues and use-cases that combine AI with smart robotics. As we witness exponential growth in automation and the rapid advancement of underpinning technologies, such as ubiquitous computing, sensing, intelligent data processing, mobile computing and context aware applications, this book is an ideal resource for future innovation.
Author(s): Rabindra Nath Shaw, Ankush Ghosh, Valentina Emilia Balas, Monica Bianchini
Publisher: Elsevier
Year: 2021
Language: English
Pages: 178
City: Amsterdam
Artificial Intelligence for Future Generation Robotics
Copyright
Contents
List of contributors
About the editors
Preface
one Robotic process automation with increasing productivity and improving product quality using artificial intelligence and...
1.1 Introduction
1.2 Related work
1.3 Proposed work
1.4 Proposed model
1.4.1 System component
1.4.2 Effective collaboration
1.5 Manufacturing systems
1.6 Results analysis
1.7 Conclusions and future work
References
two Inverse kinematics analysis of 7-degree of freedom welding and drilling robot using artificial intelligence techniques
2.1 Introduction
2.2 Literature review
2.3 Modeling and design
2.3.1 Fitness function
2.3.2 Particle swarm optimization
2.3.3 Firefly algorithm
2.3.4 Proposed algorithm
2.4 Results and discussions
2.5 Conclusions and future work
References
three Vibration-based diagnosis of defect embedded in inner raceway of ball bearing using 1D convolutional neural network
3.1 Introduction
3.2 2D CNN—a brief introduction
3.3 1D convolutional neural network
3.4 Statistical parameters for feature extraction
3.5 Dataset used
3.6 Results
3.7 Conclusion
References
four Single shot detection for detecting real-time flying objects for unmanned aerial vehicle
4.1 Introduction
4.2 Related work
4.2.1 Appearance-based methods
4.2.2 Motion-based methods
4.2.3 Hybrid methods
4.2.4 Single-step detectors
4.2.5 Two-step detectors/region-based detectors
4.3 Methodology
4.3.1 Model training
4.3.2 Evaluation metric
4.4 Results and discussions
4.4.1 For real-time flying objects from video
4.5 Conclusion
References
Five Depression detection for elderly people using AI robotic systems leveraging the Nelder–Mead Method
5.1 Introduction
5.2 Background
5.3 Related work
5.4 Elderly people detect depression signs and symptoms
5.4.1 Causes of depression in older adults
5.4.2 Medical conditions that can cause elderly depression
5.4.3 Elderly depression as side effect of medication
5.4.4 Self-help for elderly depression
5.5 Proposed methodology
5.5.1 Proposed algorithm
5.5.2 Persistent monitoring for depression detection
5.5.3 Emergency monitoring
5.5.4 Personalized monitoring
5.5.5 Feature extraction
5.6 Result analysis
References
six Data heterogeneity mitigation in healthcare robotic systems leveraging the Nelder–Mead method
6.1 Introduction
6.1.1 Related work
6.1.2 Contributions
6.2 Data heterogeneity mitigation
6.2.1 Data preprocessing
6.2.2 Nelder–Mead method for mitigating data heterogeneity
6.3 LSTM-based classification of data
6.4 Experiments and results
6.4.1 Data heterogeneity mitigation using Nelder–Mead method
6.4.2 LSTM-based classification of data
6.5 Conclusion and future work
Acknowledgment
References
SEVEN Advance machine learning and artificial intelligence applications in service robot
7.1 Introduction
7.2 Literature reviews
7.2.1 Home service robot
7.3 Uses of artificial intelligence and machine learning in robotics
7.3.1 Artificial intelligence applications in robotics [6]
Assembly [7]
Packaging [7]
Customer service [7]
Open source robotics [7]
7.3.2 Machine learning applications in robotics [10]
7.4 Conclusion
7.5 Future scope
References
Eight Integrated deep learning for self-driving robotic cars
8.1 Introduction
8.2 Self-driving program model
8.2.1 Human driving cycle
Perception
Scene generation
Planning
Action
8.2.2 Integration of supervised learning and reinforcement learning
Supervised learning
Reinforcement learning
8.3 Self-driving algorithm
8.3.1 Fundamental driving functions
White lane detection
Signals
8.3.2 Signals
Traffic signs
Laneless driving
8.3.3 Hazards
YOLO and detection of objects
Collision avoidance
Estimation of risk level for self-driving
8.3.4 Warning systems
Driver monitoring
Pedestrian hazard detection
Sidewalk cyclists’ detection
8.4 Deep reinforcement learning
8.4.1 Deep Q learning
Learning rate
Discount factor
8.4.2 Deep Q Network
8.4.3 Deep Q Network experimental results
8.4.4 Verification using robocar
8.5 Conclusion
References
Further reading
NINE Lyft 3D object detection for autonomous vehicles
9.1 Introduction
9.2 Related work
9.2.1 Perception datasets
9.3 Dataset distribution
9.4 Methodology
9.4.1 Models
9.5 Result
9.6 Conclusions
References
TEN Recent trends in pedestrian detection for robotic vision using deep learning techniques
10.1 Introduction
10.2 Datasets and artificial intelligence enabled platforms
10.3 AI-based robotic vision
10.4 Applications of robotic vision toward pedestrian detection
10.4.1 Smart homes and cities
10.4.2 Autonomous driving
10.4.3 Tracking
10.4.4 Reidentification
10.4.5 Anomaly detection
10.5 Major challenges in pedestrian detection
10.5.1 Illumination conditions
10.5.2 Instance size
10.5.3 Occlusion
10.5.4 Scene specific data
10.6 Advanced AI algorithms for robotic vision
10.7 Discussion
10.8 Conclusions
References
Further reading
Index
