The project objective is to create a class of modular and distributed electromechanical actuators and their power network that will enable robots to be agile, efficient, and capable of reproducing biological motions that today are impossible. Animals have the innate capability to move and maneuver effectively in complex and unstructured environments. Recent advances in bio-inspired robots have conceptualized collaborative robots or "cobots", which will interact with humans in multiple settings. Although state-of-the-art bio-inspired robots have achieved exquisite maneuvers, such systems have yet to closely replicate the grace, fluidity, and agility of their biological counterparts. There is a critical need to re-imagine these robots not only as an embodiment of mechanical linkages and artificial intelligence but also as a complex network of electromechanical actuators. The project aims to emulate a biological spine. A distributed actuator mimicking the spine mechanism will improve mobility, efficiency, and stability of robots in search, rescue, and recovery making them the first line of defense for disaster relief as well as surveillance reconnaissance, inspection, and exploration applications. The proposed research trajectory will catapult hardware advances in these robots to converge with the exploding capability of artificial intelligence and autonomous control, saving human lives and enhancing national security. Further, an active synthetic spine opens up opportunities to design a life-like exoskeleton, arrest spine deformities in children, augment upper-body rehabilitation therapy for stroke patients, perform whole-body robotic teleoperation, and add non-verbal capability in social robots. The integrated education and outreach plan aims to ignite curiosity in K-12 students about electromechanics and power electronics--foundations of our modern civilization--by using robotics as the catalyst.
A spine is fundamental to providing flexibility and balance in animals while allowing efficient locomotion. Construction of a synthetic spine is remarkably different from other standard robotic mechanisms, such as arms and legs, due to the presence of multiple single-joint segments, each with a limited range of motion. The design strategy will take advantage of the limited displacement requirement to increase the actuator's torque-to-weight ratio. Instead of employing conventional electric motors that utilizes shear stress to generate motion, the proposed gearless design will use normal stress. Integrated design of mechanical springs and electromagnetics will enable a customized torque-displacement characteristic to achieve compliance and high efficiency--similar to muscles. Deployment of appropriate control and estimation techniques is proposed to vary output torque and compliance. The design methodology will be validated by constructing a hardware prototype of a synthetic spine. The project plans to construct demonstration kits using research results that connect math and theory to the craft of real-world systems such as robots and automated systems. These demo kits will attract and inspire K-12 students, underrepresented groups, and a broader audience about electrical power and energy processing. The demo kits blueprint will also be shared with K-12 educators to help them teach their STEM clubs. In summary, the proposed framework is the basis to build a multi-disciplinary understanding of distributed actuators and their power network in robots and automated systems and to advance the robotics workforce through educational pathways.
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.
