Year: 2026
Type: preprint
Abstract: Biomimicry has inspired the evolution of robotics toward greater autonomy, adaptability, and symbiosis with humans and dynamic environments. However, current robotic systems still face major challenges in recapitulating the high-efficiency decision-making capabilities of the human brain under complex and dynamic conditions. Here, we present Brainobot, a biohybrid robotic system that establishes a brain organoid controller as a high-level robotic decision-making layer for closed-loop embodiment. By leveraging brain organoid reservoir computing, Brainobot interacts with dynamic environments by receiving and processing sensory inputs and generating motor actions. As a proof-of-concept demonstration, Brainobot is implemented in a humanoid robotic system to perform real-world tasks, including object grasping and laser chasing. Interestingly, Brainobot exhibits unique features, including cross-task adaptivity, high computing efficiency, and low energy consumption. Thus, our approach may provide insights for advancing robotic embodiment and understanding biological decision-making.
Institutions: Indiana University Bloomington, Indiana University, Brigham and Women's Hospital, Harvard University, Indiana University Indianapolis, Indiana University School of Medicine, Indiana University – Purdue University Indianapolis, Washington University in St. Louis, Cincinnati Children's Hospital Medical Center, University of Cincinnati, University of California, Los Angeles, Hastings Center, National University of Singapore
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: National Science Foundation
OpenAlex ID: W7213317022
Cites:
Cited by:
Related to: