Collaborative Research: Advancing Human-Robot Synergy in Dynamic Environmental Conditions Through Bi-directional Adaptive Feedback Systems

About This Grant

This award supports research to improve collaboration between humans and robots in dynamic industrial environments such as construction and manufacturing settings. Environmental factors such as temperature, lighting, and noise can affect each worker's performance and reaction time differently, depending on their unique physical and mental characteristics. These variations can impact on the overall effectiveness and safety of human-robot collaboration, particularly in scenarios where the robot assists without taking over control. Current robotic systems do not adequately account for such differences, limiting their ability to adapt to changing environments or individual workers’ needs. This research will create a new generation of human-centered robotic systems that monitor both environmental conditions and human well-being in real time and adjust their behavior to support smoother and more efficient teamwork. The robots will interpret indicators of workers’ mental states and use that information to adjust how they interact. They will also communicate their status to human partners intuitively. These innovations are expected to enhance workplace safety, task accuracy, and worker satisfaction, especially in labor-intensive jobs. Broader impacts of the project include the integration of research outcomes into university curricula and outreach programs aimed at inspiring and educating students from high school through graduate levels. This research aims to develop an adaptive human-robot collaboration framework. The approach integrates real-time human state with adaptive robot control, allowing the robot to respond intelligently to changes in human and environmental conditions. Human states are inferred through physiological sensing methods and used within a closed-loop system to guide collaborative behaviors. The robot communicates its internal state and intentions through intuitive multimodal feedback, promoting seamless coordination and reducing cognitive effort for the human operator. The control strategy leverages deep reinforcement learning techniques to optimize decision-making in complex, high-dimensional environments. The system will be tested in both laboratory and real-world settings to evaluate its robustness and effectiveness. Results from this work are expected to contribute to the advancement of control, perception, and learning in human-centered robotics, with broad implications for future collaborative systems across various industries. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.