acm-header
Sign In

Communications of the ACM

ACM TechNews

Robot Dog Learns to Walk in One Hour


View as: Print Mobile App Share:

Researchers at the Max Planck Institute for Intelligent Systems built the four-legged, dog-sized robot Morti as part of a study aimed at learning how animals learn to walk.

Credit: Felix Ruppert/Dynamic Locomotion Group

Researchers at Germany's Max Planck Institute for Intelligent Systems (MPI-IS) built a four-legged, dog-sized robot to study how newborn animals learn to walk.

The robot, named Morti, learned to walk in one hour with the help of a Bayesian optimization algorithm that adapts the control parameters of a Central Pattern Generator (GPG), a lightweight computer that acts as a virtual spinal cord.

The algorithm can change the length and speed of each leg’s swing and how long the leg is on the ground in response to any stumbling.

Said MPI-IS's Felix Ruppert, "Our robot is practically 'born' knowing nothing about its leg anatomy or how they work."

Ruppert added, "Changing the CPG output while keeping reflexes active and monitoring the robot stumbling is a core part of the learning process."

From Max Planck Institute for Intelligent Systems (Germany)
View Full Article

 

Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account