•  
  •  
 

Abstract

In this research, the objective is to reduce energy (or power) consumption for a legged robot and demonstrate a correlation between energy (or power) consumption and center of mass (CoM) motion. The approach used is to experimentally measure and analyze the power consumption and center of mass motion of a legged robot carrying an elastically suspended load. A high-speed motion-tracking camera system was used to capture the robot body and load trajectories, and an analog-to-digital converter was used to record the robot’s average power. By analyzing the frequency content of the trajectories, we found that the center of mass dynamics exhibited multiple frequency components. Each frequency had a unique phase and amplitude that contributed to the average power. A multivariable fitting method was used to quantitatively define the contribution weight of each frequency on the overall center of mass dynamics. We found that the center of mass dynamics may be loosely correlated with the power consumption of locomotion and that multiple modes of motion complicate this relationship.

Share

COinS