Embodied Intelligence in Physical Systems
Embodied Intelligence in Physical Systems
Embodied Intelligence in Physical Systems (EIPS) is an international, peer-reviewed, open-access journal dedicated to publishing high-quality original research articles, authoritative reviews, and per...
Embodied Intelligence in Physical Systems (EIPS) is an international, peer-reviewed, open-access journal dedicated to publishing high-quality original research articles, authoritative reviews, and perspectives. It serves as a premier forum for researchers and practitioners to advance the field of embodied intelligence, which focuses on the deep integration of perception, cognition, and action within physical systems.
The journal's core mission is to explore how intelligent behaviors emerge from and are shaped by the dynamic interaction between an agent's physical body and its environment. We welcome submissions addressing the fundamental challenges of enabling autonomous systems—from robots and autonomous vehicles to ubiquitous computing devices—to perceive, learn, reason, and act adaptively in complex, unstructured real-world settings.
Key areas of interest include, but are not limited to:
Theoretical Foundations of Embodied Intelligence: Research on embodied cognition, morphological computation, perception-action loops, and principles of intelligence arising from agent-environment interaction.
Embodied AI and Cognitive Architectures: Studies on large language models (LLMs), vision-language models (VLMs), and Vision-Language-Action (VLA) models as cognitive engines for physical systems. This includes world models, task planning, symbolic reasoning, and embodied foundation models that bridge digital and physical intelligence.
Human-Robot-Physical World Integration: Technologies and frameworks enabling seamless collaboration between humans, robots, and the environment to support efficient and natural teamwork.
Learning, Adaptation, and Evolution in Physical Systems: Applications of deep reinforcement learning, imitation learning, evolutionary algorithms, and self-supervised methods that allow systems to learn and improve through interaction.
Simulation to Reality (Sim2Real) and Virtual Training Platforms: Development of high-fidelity simulators, digital twins, and transfer learning techniques to overcome data efficiency and safety barriers in real-world training.
Sensing, Actuation, and Embodied Morphology: Advances in sensor fusion (visual, tactile, auditory), actuator design, and the role of physical morphology in facilitating intelligent behavior and control.
Edge Computing and Autonomous System Architectures: Computing paradigms supporting low-latency, high-reliability demands of real-time perception-decision-action cycles at the edge.
Security and Privacy of Embodied AI Systems: Addressing vulnerabilities unique to intelligent physical systems, including robustness against adversarial attacks on sensors or models, secure multi-agent collaboration, privacy-preserving sensing in IoT environments, and behavioral safety to prevent physical harm.
Applications and Deployment: Reports on real-world applications in smart manufacturing, intelligent logistics, service robots, smart homes, healthcare, autonomous driving, and other domains.
Ethics, Safety, and Human-Agent Alignment: Frameworks to ensure the safety, reliability, trustworthiness, and ethical behavior of autonomous embodied systems operating in human-centric environments.
EIPS is committed to advancing intelligence grounded in physical reality, providing a critical platform for interdisciplinary research that converges artificial intelligence, robotics, cognitive science, and cyber-physical systems on the path toward Artificial General Intelligence (AGI).