Environmental Adaptation of Robot Morphology and Control Through Real-World Evolution.

Robots operating in the real world will experience a range of different environments and tasks. It is essential for the robot to have the ability to adapt to its surroundings to work efficiently in changing conditions. Evolutionary robotics aims to solve this by optimizing both the control and body (morphology) of a robot, allowing adaptation to internal, as well as external factors. Most work in this field has been done in physics simulators, which are relatively simple and not able to replicate the richness of interactions found in the real world. Solutions that rely on the complex interplay among control, body, and environment are therefore rarely found. In this article, we rely solely on real-world evaluations and apply evolutionary search to yield combinations of morphology and control for our mechanically self-reconfiguring quadruped robot. We evolve solutions on two distinct physical surfaces and analyze the results in terms of both control and morphology. We then transition to two previously unseen surfaces to demonstrate the generality of our method. We find that the evolutionary search finds high-performing and diverse morphology-controller configurations by adapting both control and body to the different properties of the physical environments. We additionally find that morphology and control vary with statistical significance between the environments. Moreover, we observe that our method allows for morphology and control parameters to transfer to previously unseen terrains, demonstrating the generality of our approach.

The convergence of backpropagation trained neural networks for various weight update frequencies.

One of the problems concerning the backpropagation training of feed-forward neural networks is the effect of the weight update frequency. This aspect influences the efficiency of parallel implementations of the training algorithm where the training vectors are distributed among processors. In this paper the convergence of two applications for various weight update intervals is reported. Further, several models are proposed for describing convergence and learning rate aspects in the context of a set of weight update intervals. The results show that the convergence by updating the weights after each training vector leads to about 10 times less number of training iterations compared to updating the weights only ones for the whole training set.