I’m coordinating Prof. Pan Jia and Dr. Wang Weifu on a research topic with the Robot Learning section of Frontiers in Robotics and AI. All submissions are welcome and thank you for your attention.
About this Research Topic
Classical research on manipulation relies heavily on the rigid-body assumption throughout the design, modelling, and control of advanced robots. However, recent advancement in material science and rapid prototyping promotes an emerging field of soft robotics, where biological inspirations are drawn from the soft-bodied system for more natural behaviors among robots, humans, and the environment. On the other hand, leveraging recent development in computational resource and algorithm intelligence, the integration of machine learning techniques in robotic manipulation has demonstrated interesting and capable applications that were challenging to solve using classical modelling-centric methods. The combined effect makes one wonder: how can we better leverage classical theories in rigid-bodied design, modelling, and control in the interdisciplinary fields of soft robotics and robot learning to develop the next generation, advanced robots in exciting and challenging applications.
Rigid-soft interactions can be widely observed in manipulation problems involving biological systems and industrial applications, where learning-based methods show capable potentials to provide novel solutions. For example, endoskeletons such as human muscular systems and exoskeletons such as crustaceans combine both rigid and soft components in biological structure, showing different levels of dexterity while interacting with the physical environment. In applications such as fabric handling and food processing, learning-based methods also show promising solutions to deal with object-centric representations and interaction uncertainties.
This research topic calls for interdisciplinary contributions at the intersection of robot learning and soft robotics. Particular focus will be given to contributions proposing novel theories, models, and methods for learning the interaction problems between the rigid and soft components. We will welcome original technical contributions presenting learning-focused systems, algorithms, and computational methods tailored to rigid-soft interactions.
Below is a non-exhaustive list of topics:
– Robot design with both rigid and soft components for manipulation learning.
– Learning soft and fabric manipulation with rigid grippers.
– Machine learning for soft robot sensing and control.
– Fabric handling and manipulation for domestic and industrial use.
– Learning human-robot interaction for dexterous manipulation and assistance.
– Representation of rigid-soft interactions for manipulation learning.
– Learning soft end-effector manipulation of rigid objects.
– Tactile sensing for manipulation learning with deformable objects.
– Learning active or passive control with variable stiffness.
Keywords: Machine Learning, Rigid-Soft Interaction, Robotic Manipulation, Robot Learning, Soft Robotics
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.
