TECHNOLOGY AREA(S): Ground SeaOBJECTIVE: To provide opportunity for scientific exploration of next-generation robotic and physical human augmentation performance systems and associated controls through development of actuation technologies, and the associated framework of predictive modeling.DESCRIPTION: Currently human-scale robots and devices employed in human-scale physical augmentation devices and prosthetics employ mostly rotary-motion electric motors or hydraulics. We have made significant advances and demonstrations through design, fabrication methods, and controls in each of these technologies over the past several years, but they are still lacking in terms of performance, cost, and fundamental physics-based criteria for systems that are used in human-scale dynamic limb-based locomotion and whole-body manipulation. These forms of actuation have also been seen as limiting factors in development of machine morphologies which can replicate the degrees of freedom of human motion and human performance needed for human prosthetics and exoskeletons. There are examples of efficient and dynamic limb-based mechanisms which have been achieved through means of iterative design in which the systems mechanics and morphology are expertly matched with highly customized and optimized forms of actuation and unique electronic controllers. They represent a state of the art which has yet to be accepted as suitable for machines that are expected to perform as physical teammates to Soldiers in high-OPTEMPO missions.This proposal seeks to continue further and spawn new research and commercialization for forms of robotic actuation and promote a mechanism design paradigm compatible with that of open, modular software development recently being adopted within the Department of Defense and academia. The goal is to provide examples of scalable forms of actuation (size and number) which may be seen as viable options for improving the performance and efficiency of next generation robotics mechanisms. New forms actuation which can deliver human scale forces and moments in lightweight and energy efficient configurations such as hydraulically amplified self-healing electrostatic (HASEL) soft actuators or other less common electrostatic-based actuators which have potential for making systems with less mass, less cost, and compatible with morphologies requiring distributed actuation are examples. Some of these forms of actuation may be seen as complementary to established actuators such as electric rotary actuators and hydraulics. For example, this may include actuated structures which have adaptive compliance characteristics. New electric rotary actuator and hydraulics based concepts may be considered as well. For example, limb-based human-scale dynamic locomotion and whole-body manipulation requires high-torque with high-frequency control. New scalable actuator designs addressing these requirements may be considered. For new rotary electric motor designs capable of offering improved suitable performance this could mean novel coil and magnet configurations combined with new motor controller sensing techniques which optimize force generated from magnetic field interaction.PHASE I: In Phase I, the following shall be accomplished: a) Survey current design and approach for developing scalable actuator technology that may be employed for efficient dynamic human-scale whole-body manipulation and dynamic locomotion. Review typical applications and regimes of interest, and identify relevant physical, electronic, software specifications and parameters to demonstrate the feasibility of an analytic and engineering infrastructure for their design, fabrication, and control. b) Analyze and identify useful families of robotic morphology and/or structures in which the actuators may be employed. c) Develop concept(s) through which the actuators or combinations of actuators may be employed and controlled feasibly to improve performance of human-scale robotic systems. d) Implement the concept(s) numerically and conduct the appropriate proof-of-concept computations. e) After the concept has been numerically demonstrated, use to fabricate a prototype or demonstration which validates numerical simulation.PHASE II: In Phase II, the following shall be accomplished: a) The actuator technology (actuator, actuator controller, actuator feedback) from Phase I will be tested, validated, and implemented. Aspects of efficient scalable performance and fabrication for efficient custom design will be demonstrated and characterized. b) The actuator performance characterization models and control algorithm software and from will be tested, validated, and implemented as a documented software package that can be shared or distributed. The models should have compatibility with modern physics-based simulation software such that their performance may be predicted in a mechanical device. c) Numerically demonstrate that models characterizing the actuator and performance are compatible with a modern physics-based robot simulation and that the information feedback from the actuators and/or actuation controller is suitable for whole-body manipulation control. d) Numerically and in device tests, demonstrate that the actuator and controls software performs as predicted. This should be demonstrated at multiple scales (2x, 3x) or in the case of distributed actuators possibly different numbers (2x, 3x) of actuators. e) Generalize the methodology in a-d to provide a range or families of actuators which may be readily simulated and fabricated for near-term and future human-scale robot use. f) Develop and demonstrate fabrication method for the scalable range of actuators described above. Transition the developed methods and software, including documentation, to interested users in academia, industry and government (e.g. ARL) under appropriate licensing agreement.PHASE III: The actuators, numerical techniques, performance and control models, and fabri-cation techniques developed under this topic will aid in further advancement of robotic technologies for dynamic human-scale whole-body manipulation and lo-comotion. The results will be corroborated by prototype fabrication. In addition this will demonstrate a model and paradigm for robotic actuator development which is synergistic to modern dynamic robot design, modular open robotic soft-ware development and DoD interoperability protocols. This will lead to a much needed methodology for actuator design for dynamic limbed systems which is technically sound, facilitates advancements to state of the art robot design, and is commercially and fiscally viable.KEYWORDS: actuator, robot, exoskeleton, prosthetics, control, dynamicsReferences:E. Acome, S. K. Mitchell, T. G. Morrissey, M. B. Emmett, C. Benjamin, M. King, M. Radakovitz, C. Keplinger, Hydraulically amplified self-healing electrostatic actuators with muscle-like performance. Science, Vol. 359, Jan 2018, pp. 61-65.; M. Mazzara, RSJPO Interoperability Profiles, Robotic Systems Joint Project Office, US Army TARDEC, Warren MI, September 2012.; P. Wensing, A. Wang, S. Seok, D. otten, J. Lang, S. Kim, Proprioceptive Ac-tuator Design in the MIT Cheetah: Impact Mitigation and High-Bandwidth Physical Interaction for Dynamic Legged Robots, IEEE Transactions on Ro-botics, Vol. 33, Issue: 3, June 2017.
