Abstract
We have modeled load carrying robots by applying the physical engine (PhysX) to a virtual environment. This study aims at acquiring an autonomous and cooperative behavior for load carrying robots which moves from a depot to a destination place through the virtual force propagation. The load carrying robots are controlled by ANN (Artificial Neural Network), which is evolved by learning proper actions by use of GA (Genetic Algorithm). Simulation results show that the load carrying robots acquire autonomous and cooperative behaviors under the virtual physical environment.
