Computers have already been proven to solve certain problems faster and more accurately than the human brain. However, so far even the simplest walking animals are far more agile and stable than any robot built. It would be ideal to understand how biological systems operate in order to implement what is useful for robots. Animals agility is partly due to their complex morphology, and this can be imitated to some extend. For instance, the many degrees of freedom or the geometrical arrangement can be replicated. Nonetheless, sensory integration, decision-making and motor control are very complex as well. These are responsible for most of the animals' successful locomotion. Unfortunately, at present more problems focus on understanding the information processing within the nervous system. Nowadays there is still debate about how biological neural networks operate. The nervous system operation is not easily understood because it does not only depend on its topology, dynamic parameters inside each neuron are strongly temporally dependent.
On the other hand, just as one can use biological systems as inspiration or target systems, one could also use robots as platforms for testing biological hypotheses. Provided both systems operate with similar rules, analogous information, and a proper validation is performed.