Design Optimization of a Novel Networked Electromagnetic Soft Actuators System Based on Branch and Bound Algorithm

Abstract

This paper presents a design optimization framework based on Branch and Bound Algorithm for novel networked electromagnetic soft actuators. The soft actuators work based on the operating principle of solenoids but are made of intrinsically soft materials. We confirmed that by scaling down the size of the soft actuators, their force to volume ratio increases. This suggests that by miniaturizing the actuators size and attaching them as a network based on the arrangement of actin and myosin filaments in skeletal muscles, the output force can be enhanced. In order to achieve the maximum output force, design parameters of a single soft actuator as well as those of a network are considered as design variables. The maximum available volume (thickness, width and length) and deflection range of the network are considered design constraints. The cost function, i.e. the output force is a non-linear mixed-integer function. A Branch and Bound optimization algorithm based on interval analysis is then proposed to solve the optimization problem. Numerical simulations are presented for a representative example of an active soft brace for the human elbow joint. The results suggest that an electromagnetic soft actuator network can provide sufficient torque to be used as a drive train for an active elbow brace for both flexion and extension over a range of around 93 degrees.