ABSTRACT

The research objective of this project is to develop an upper extremity exoskeleton and dedicated controller that can enhance agility of arm motions while retaining stability and safety of the coupled human-robot system. The exoskeleton design utilizes a novel mechanism and passive slip device to align the exoskeleton with the human user's shoulder and elbow kinematics. In addition, the project will characterize the 3D impedances of the human shoulder and elbow joints, thereby contributing to a fundamental understanding of the biomechanics of the human upper extremity. A new, user-adaptive, variable impedance controller with safety supervisor will manage the tradeoff between agility and coupled stability in the physical human-robot system while avoiding awkward postures that could lead to musculoskeletal injury. If successful, the technology has potential to positively impact society and the national well-being by reducing work-related musculoskeletal disorders and their undesirable impacts on the productivity and healthcare costs of workers and employers. Broader impacts of the work include mentorship, educational, and outreach activities that focus on inclusion for underrepresented minorities.

This project will design and control a high-performing and stable upper-extremity exoskeleton robot. A novel mechanical design integrates parallel and serial actuation mechanisms with a passive slip interface to improve upper extremity mobility while alleviating mechanical interference (misalignment) between the human user's joints and the robot's joints. A key innovation will be a characterization of the 3D impedances of the human shoulder and elbow joints, which will contribute both to a fundamental understanding of the biomechanics of the human upper extremity and to the development of a novel robotic controller. The robotic controller will incorporate the estimates of human mechanical impedance and a measure of user intent to improve the agility of the human-robot system beyond state-of-the-art passivity-based controllers. In addition, a high-level supervisory controller based on the synthesis of robust controlled invariant safety sets will prevent the coupled human-robot system from reaching unsafe or awkward configurations that could cause musculoskeletal injury. Human subject experiments to evaluate the performance of the system and its controller in comparison with existing robotic controllers are planned. If successful, the "user-adaptive variable impedance controller with safety guarantees" could provide a generalizable example of how to manage the tradeoff between agility and coupled stability in physical human-robot systems.

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.

Please report errors in award information by writing to: awardsearch@nsf.gov.