Spatiotemporal Dynamics Modeling of Brain Activity for Human-Robot Cognitive Interaction: A Distributed-Lumped...

Area of Science:

• Neuroscience
• Robotics
• Artificial Intelligence

Background:

• Human-robot cognitive interaction (HRCI) presents complex system modeling challenges.
• Understanding the dynamical process of human brain activity during HRCI is crucial for developing advanced human-robot systems.

Purpose of the Study:

• To develop a computational model for human brain activity in HRCI.
• To accurately identify and monitor the spatiotemporal dynamics of human brain activity in real-time HRCI.

Main Methods:

• Proposed a human-distributed robot-lumped parameter system (HDRLPS) model.
• Developed a deterministic learning (DL)-based spatiotemporal dynamics identification scheme.
• Designed a spatiotemporal dynamics estimator using a radial basis functional neural network (RBF NN) model.

Main Results:

• The proposed method accurately identifies and evaluates human brain activity dynamics in real-time HRCI.
• Validated through mathematical proofs, simulations, and brain-computer interface (BCI) experiments.
• Outperformed state-of-the-art methods (LGGNet, EEGNet, etc.) in recognition accuracy and macro-F1 scores.

Conclusions:

• The developed HDRLPS model and DL-based identification scheme are effective for understanding human brain dynamics in HRCI.
• The proposed spatiotemporal dynamics estimator provides accurate real-time monitoring capabilities.
• This research advances the field of brain-computer interfaces and intelligent human-robot systems.