Alternative Metrics to Select Motors for Quasi-Direct Drive Actuators
This work addresses motor selection for robotic systems like legged robots, exoskeletons, and prosthetics, offering incremental improvements over traditional methods.
The paper tackled the problem of selecting motors for quasi-direct drive actuators in legged locomotion by proposing alternative metrics that are invariant to transmission ratio, resulting in a better balance of properties such as minimizing inertia and maximizing torque without focusing on gap radius, as demonstrated through a case study comparing the T-Motor RI50 and U8.
Robotic systems for legged locomotion -- including legged robots, exoskeletons, and prosthetics -- require actuators with low inertia and high output torque. Traditionally, motors have been selected for these applications by maximizing the motor gap radius. We present alternative metrics for motor selection that are invariant to transmission ratio. The proposed metrics reward minimizing the motor inertia while maximizing the torque and motor constants without special consideration for gap radius, providing a better balance of properties for legged locomotion applications. We rigorously characterize the T-Motor RI50 and demonstrate the use of the metrics by comparing the RI50 to the widely-used T-Motor U8 as a case study.