Sign in Subscribe
Artificial intelligence

Robotics Research Highlights Shift towards General Embodied Artificial Intelligence

Strategic Insights

— 1 min read

Robots are increasingly being equipped with general embodied artificial intelligence, driven by the proliferation of large language models. This shift has sparked interest in combining traditional learning methods with foundation models, leading to potential real-life applications. A recent study published in Neurocomputing has comprehensively reviewed the state-of-the-art foundation models in robot learning, identifying future potential areas and highlighting critical issues that are neglected in current literature.

Key Takeaways:

  • The study, conducted by researchers from Tongji University, highlights the shift in robot learning from automation towards general embodied artificial intelligence, fueled by the proliferation of large language models.
  • The research community has increasingly adopted foundation models together with traditional learning methods for robot learning, showing potential for real-life application.
  • The review identified four mainstream areas of robot learning: manipulation, navigation, task planning, and reasoning, demonstrating how foundation models can be adopted in these scenarios.
  • Critical issues neglected in the current literature, including robot hardware and software decoupling, dynamic data, generalization performance in the presence of humans, etc., were discussed.
  • Future research should focus on multimodal interaction, especially dynamics data, robotics-specific foundation models, AI alignment, etc.

Statistics:

  • 638: The designation of the research paper "Robot Learning In the Era of Foundation Models: a Survey" in Neurocomputing.
  • 2025: The publication year of the research paper "Robot Learning In the Era of Foundation Models: a Survey" in Neurocomputing.
  • 29: The number of Radarweg, the address of Elsevier, the publisher of Neurocomputing.
  • 1043 Nx: The postal code of Amsterdam, Netherlands, where Elsevier is located.
  • 3,337: The page number of the research paper "Researchers from Tongji University Report on Findings in Robotics (Robot Learning In the Era of Foundation Models: a Survey)" in Journal of Engineering.

Sources:

  • VerticalNews - 2025 JUL 14
  • Journal of Engineering - July 14, 2025
  • Neurocomputing - 2025;638
  • Elsevier - www.elsevier.com
  • Tongji University - Natl Key Lab Autonomous Intelligent Unmanned Syst, Shanghai 201210, People's Republic of China

Read more