Walking turns out to be a challenging problem because the system is under-actuated. There is limited or no actuation between each foot and the ground, unlike robotic arm manipulators where the base of the system is rigidly attached to some immobile or heavy base, and an actuator at this base (and any other joints) allows for arbitrary trajectories of the end-effector to be achieved.
The LittleDog Learning Locomotion project seeks to develop a robust control strategy to enable a quadruped robot to traverse rough terrain.
In walking robots (or animals), if there is limited or no actuation between the base and the feet — for example during heel-strike, toe-off, or running — then arbitrary trajectories of the center of body can not be attained.
We started out thinking that his comment was humorous, but the bot is obviously going into a tidy configuration for carrying. As we watched, we saw that [Evan] seemed very correct. Littlebot completely collapses at the end of each run, toward the end of the video, it doesn’t even bother to fold up nicely, sprawling out in a very lifelike pose of utter exhaustion. We don’t know if this is intentional, but we think it should be.