About this Research Topic
Neural computing is a powerful paradigm that has revolutionised machine learning. However, robotics brings stringent constraints on size, weight and power constraints (SWaP), which challenges the developers of these technologies in new ways. Equally, the fact that learning and adaptation might happen on a real physical platform implies a significantly greater emphasis on ensuring safety and rendering automated decisions verifiable and explainable. RAI-NR is a 3 day meeting featuring invited and contributed talks exploring issues of reliability, safety and resource efficiency in neurorobotic systems.
To this end, this Research Topic welcome submissions of Original Research, Reviews and previous published articles (with at least 30% extensions) in line with the following themes:
- Robotics and Robot Learning: How are neural network-based methods and neuromorphic design principles driving advances in robotics?
- Hardware for AI: How are new advances in neurally inspired computing and electronic sensing hardware enabling these advances?
- Safety, verification and explainability: How are we developing the new techniques required to ensure the safety of learning, perception and control in real world settings?
- Neuroscience for perception and action: How can perspectives on how animals/humans solve problems be used to drive advances in robotics?
Keywords: Robot learning, neuromorphic systems, neuroscience of perception and action, Safe AI, Explainability
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.