2018/12/11 by Edgar Bolívar, Siavash Rezazadeh, Bolívar, Edgar +5
Engineering · #Prosthetics and Rehabilitation Robotics #Biomedical and Engineering Education
paper · pdf · doi:10.48550/arxiv.1812.04771
Design of robotic systems that safely and efficiently operate in uncertain\noperational conditions, such as rehabilitation and physical assistance robots,\nremains an important challenge in the field. Current methods for the design of\nenergy efficient series elastic actuators use an optimization formulation that\ntypically assumes known operational conditions. This approach could lead to\nactuators that cannot perform in uncertain environments because elongation,\nspeed, or torque requirements may be beyond actuator specifications when the\noperation deviates from its nominal conditions. Addressing this gap, we propose\na convex optimization formulation to design the stiffness of series elastic\nactuators to minimize energy consumption and satisfy actuator constraints\ndespite uncertainty due to manufacturing of the spring, unmodeled dynamics,\nefficiency of the transmission, and the kinematics and kinetics of the load. In\nour formulation, we express energy consumption as a scalar convex-quadratic\nfunction of compliance. In the unconstrained case, this quadratic equation\nprovides an analytical solution to the optimal value of stiffness that\nminimizes energy consumption for arbitrary periodic reference trajectories. As\nactuator constraints, we consider peak motor torque, peak motor velocity,\nlimitations due to the speed-torque relationship of DC motors, and peak\nelongation of the spring. As a simulation case study, we apply our formulation\nto the robust design of a series elastic actuator for a powered prosthetic\nankle. Our simulation results indicate that a small trade-off between energy\nefficiency and robustness is justified to design actuators that can operate\nwith uncertainty.\n