BROWSE
Detail
Journal Article
Reinforced Intelligence Through Active Interaction in Real World: A Survey on Embodied AI
Cited 2 time in 
Share 
Share
- Authors
- Wookyong Kwon, Seungmin Baek, Jongchan Baek, WooSang Shin, Minseon Gwak, PooGyeon Park, Sangmoon Lee
- Issue Date
- 2025-06
- Citation
- International Journal of Control, Automation and Systems, v.23, no.6, pp.1597-1612
- ISSN
- 1598-6446
- Publisher
- Institute of Control, Robotics and Systems
- Language
- English
- Type
- Journal Article
- Abstract
- Embodied AI is a transformative field that enables intelligent systems to actively interact with and adapt to complex physical environments. This survey examines recent advancements, focusing on how embodied agents bridge the gap between cyber-physical systems and human-centered settings. Key developments include leveraging foundation models for high-level planning, affordance learning, and low-level control, alongside unifying data from internet-scale, simulation, and real-world sources. Reinforcement learning (RL), inverse RL, and imitation learning have been pivotal in advancing robotic control and skill acquisition. Additionally, the transition from transformers to deep state-space models (SSM) offers new possibilities for enhancing prediction and decision-making capabilities in embodied systems. We also discuss challenges and future directions, highlighting the importance of generalization, affordance learning, and the path toward artificial general intelligence (AGI). This survey provides a concise roadmap for researchers and practitioners shaping the future of embodied intelligence in real-world applications.
- KSP Keywords
- Artificial general intelligence, Imitation learning, Internet-scale, Level Control, Real-world applications, Reinforcement learning(RL), Robotic Control, Skill acquisition, cyber physical system(CPS), decision making, embodied intelligence