Animals that live and move primarily on land navigate over their natural terrains using a set of complex leg trajectories.
 

Individual animal leg trajectories are influenced by a variety of factors including the animal's posture, vision of leg and hip as well as ankle and shape of foot and actuation capabilities.
 

For robotic engineers mimicking these movements is often extremely difficult as there are so many possibilities. The task just got a little easier thanks to a team of researchers from Harvard University and the Wyss Institute for Biologically Inspired Engineering who have developed a computationally efficient framework for the estimation and control of leg trajectories on a quadrupedal microrobot. 

Read More at Interesting Engineering