Figure 6. Progress of the neural learning
Observing the gait behavior of the robots trained using neural networks, they are much more stable and robust than the gait behavior when controlled directly by the half ellipse, because the neural network uses sensor information (bumpers) and have the capacity to generalize to non expected situations. The Figure 7 shows the gait control generated using locus based gait, and The Figure 8 shows the gait control generated using the neural network3 .
Figure 7. Half ellipse robot gait