We would like to invite you to submit your research paper to Robotic Applications Based on Deep Learning, a Special Issue to be published by the International Journal of Advanced Robotic Systems.
Deep learning has been widely used in computer vision for recognition. Although the proposed approaches have been validated on typical datasets, it is still questionable how well these methods would perform under practical conditions. Besides, the recognition task could only be considered as an intermediate result considering robotic applications, and further reasoning and decision making is required. Meanwhile, one-stroke training strategy, widely used by computer vision researchers, may not be suitable for robotic applications. It is more suitable to use online learning algorithms to allow the system to improve performance and to increase confidence after each operation. Besides, Q-learning models have been adapted to deep neural networks.
Recent works have reported a successful utilization of CNN with Q-learning for human-level control. Results show that the proposed system performs well when dealing with problems with simple states. When it comes to problems that require more reasoning, the performance of the system gets poorer.
The objective of this Special Issue is to seek high quality research works that significantly contribute to putting deep learning into practical robotic applications, including the introduction of novel software libraries, novel applications and novel state representations. Submitted papers should contain design, implementation, validation, evaluation and discussion of the proposed algorithm or library. The invited topics include but are not limited to:
- Library of deep learning for robotics
- Robotic applications using deep learning
- Reinforcement learning for robotics
- Hierarchical learning theory and application
- Datasets/database for deep learning
- Novel network training techniques
- Novel metric for validation
- Visual learning
- Learning-based servoing and control
- Performance evaluation for learning in robotics
- Artificial brain
Guest Editors
Ming Liu, City University of Hong Kong, Hong Kong
Hesheng Wang, Shanghai Jiaotong University, China
Simon X. Yang, University of Guelph, Canada
Haoyao Chen, Harbin Institute of Technology, China
Submission deadline: June 30, 2016
Submit your paper
All details regarding this Special Issue can be found on its webpage.
Manuscript preparation guidelines are available on the following link.IJARS applies an Article Processing Charge to all accepted papers.
For all questions regarding the journal, this Special Issue or submissions, contact us atijars@intechopen.com.