Animal Locomotion

Introduction
Any legged device that moves through it a natural environment must solve some hard mechanical and control problems, whether it is made of living tissue or metal. Therefore, one can both learn to understand animal movement and to design legged robots by working on both issues at the same time.

The locomotor system of slowly walking insects is well suited for coping with highly irregular terrain and therefore might represent a paragon for an artificial six-legged walking machine.

Website: Biologically inspired robotics lab, EPFL, Switserland

Animal Locomotion

Creatures
Legged walkers and swimmers constructed in the standard arsenal of Framsticks creatures with bend and rotation muscles. Creatures are most often controlled by neural networks that mimic biological neural mechanisms.

Modelling and Projects
Experiments in animal locomotion are likely to involve the construction of biologically plausible neural mechanisms. This includes different types of networks (e.g. spiking neural network, cyclic topological networks, pattern generating networks, etc.). Imporant also is the exploitation of physical features of the creatures morphology (in interplay with its neural control system).
Project in this area do not focus on evolutionary processes. However, the training for the neural networks is often done through exogenous evolution (i.e. standard GAs).