Selecting motors for robots using biomimetic trajectories: Optimum benchmarks, windings, and other considerations

Robotic actuators must frequently be designed to provide both substantial torque and acceleration in a lightweight package. Their ballistic trajectories require new optimization guidelines, in contrast to constant-speed or low-torque conventional motor design. A framework was derived in this study based on trajectories that incorporated the effect of velocity on torque and acceleration. This framework suggested that speed rate (or mechanical motor constant) was the best benchmark of robotic motor performance. Implications of this framework were evaluated using simulations of 37 optimized motor models. Simulation results confirmed the framework predictions, suggesting that speed rate was a better predictor of motor success than either motor constant or rated power. Optimum winding/gear ratio, supply voltage, and implications for design choices such as continuously variable transmissions are discussed in light of these findings.