Advanced Robotics

Next Generation Cognitive Robotics: Nurturing Embodied Intelligence for a Symbiotic Future with Humans and AI

Today, cognitive robotics is being reshaped by two complementary forces: (i) longstanding embodied, developmental approaches that emphasize active, motivated learning through interaction, and (ii) rapid progress in generative AI and foundation models that scale learning and reasoning from large datasets. Despite these advances, a key open challenge remains: integrating foundation-model capabilities with embodied autonomy, social interaction, and human-centered requirements—trust, safety, value alignment, and ethical, legal, and societal implications (ELSI)—to realize a symbiotic future with humans and AI.

This special issue solicits the latest research results and comprehensive surveys on next-generation cognitive robotics, including interdisciplinary perspectives spanning robotics, machine learning, neuroscience, psychology, philosophy, affective computing, and ELSI-related research. We particularly welcome works that combine embodied robot experiments with modern AI approaches (LLMs, vision-language-action models, robot foundation models, and world models), and works that address multi-party social interaction as well as methodologies using virtual spaces/metaverse to complement physical embodiment.

Prospective contributed papers may cover, but are not limited to, the following topics:

• Embodied AI and embodied intelligence for service robotics and socially assistive robotics
• World models and predictive world modeling for long-horizon autonomy (model-based RL, imagination, prospection)
• Predictive coding, active inference, and the free energy principle (FEP) in world modeling and perception–action loops
• Large Language Models (LLMs) and multimodal foundation models for cognitive robotics (reasoning, planning, dialogue, tool use)
• Vision–Language–Action (VLA) models and robot foundation models for manipulation and mobile tasks in open-world settings
• Autonomy in intelligent systems: intrinsic motivation, curiosity-driven exploration, and active learning
• Language and cognitive development: grounded semantics, symbol emergence, and emergent communication
• Beyond dyadic HRI: multi-party social interaction, group dynamics, and AI systems that support social interaction
• Virtual embodiment and metaverse methodologies for scalable studies of social interaction with physical constraints
• Trust in HRI: transparency, explanation, legibility, trust calibration and trust repair
• Affective computing and computational psychiatry for modeling human cognitive/affective states (emotion, interoception, neurodiversity)
• Safety, robustness, and failure-aware behaviors; verification/validation for human-centered cognitive robots
• AI alignment with human values and societal norms; ethical, legal, and societal implications (ELSI) of human–robot–AI systems
• Benchmarks, evaluation protocols, datasets, and reproducible methodologies for embodied cognitive robotics

This Special Issue also welcomes submissions that are related to the RSJ Special Section on Cognitive Development and Symbol Emergence. It is intended to serve as an additional and complementary publication venue for papers that fall within the scope of that Special Section, and such contributions are explicitly encouraged.