About this Research Topic
Intelligence and autonomy are among the most extraordinary capacities blossomed by human evolution. Yet, endowing humanoid robots with these two crucial capabilities is still one of the biggest problems for the robotics community, despite decades of research. On the software side, algorithms for artificial intelligence are still at an embryonic stage. On the hardware side, robotic actuators are a far cry from the muscular human system in terms of flexibility and adaptability, which in turn reduces autonomy and robustness. Underneath the nature of algorithms for intelligence and technology for autonomy, the importance of efficient, scalable implementations of robust software goes without saying.
Among the large variety of humanoid robots, the iCub has emerged as one of the most diffused research platforms. It has been developed as part of the RobotCub EU project and subsequently adopted by more than 35 laboratories worldwide. Collaborations across laboratories are encouraged by writing code and libraries openly available. As a consequence, iCub is considered to be the ideal platform for experimenting and advancing open-source software for research in several domains, ranging from motor control to cognitive systems.
Thereby, this Research Topic calls for papers detailing open source software devoted to the development of the intelligence and autonomy of the humanoid robot iCub, either physical or simulated. As part of the publication process, the code presented in the submissions must be accessible to the iCub community, by means of references to GitHub or similar repositories.
Additionally, the code paper requirements follow:
– The submitted software must be well documented.
We suggest to organize the documentation at various levels, comprising: i) the description of methods (APIs); ii) tutorials facilitating the use of the software. One way of providing documentation is using GitHub Pages, and a general template for this can be found at https://github.com/robotology/how-to-document-modules.
– Possibility of reproducing the presented experiments in any iCub compatible environment, i.e. physical and/or simulated. The accepted simulators are: iCubSim and iCubGazebo.
– Video tutorials detailing how to launch the associated software.
– Datasets and/or CAD files of the materials required to reproduce the experiments.
Contributions to this special issue can address a wide variety of topics, including, but not limited to the following:
• Robot behavior frameworks integrating various modular components.
• Algorithms related to the visual, tactile, auditory and proprioceptive perception.
• Methods for grasping.
• New control methods and their applications.
• Frameworks for Human-Robot Interaction.
• Helper applications that ease the use of iCub.
• Machine learning and reasoning methods.
• Multi-robot frameworks for transferring skills among diverse platforms.
• Virtual or augmented reality applied to iCub.
Submissions of code papers to this Research Topic must include original research or negative studies (in which scientists conduct careful experiments disproving previous theories). Authors should prepare their manuscript according to the “Author Guidelines” available at the journal homepage. In addition to the ”Manuscript Guidelines”, authors should also comply with the ”Code” submission section as additional requirements.
Keywords: iCub, open-source, intelligence, autonomy, code
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.