In previous experiment, another interesting phenomenon was observed (Figure 4).
Same system setup as that in previous experiment is used.
Figure 4 shows average classification accuracy measured at different SIR levels for all GMM and MLP models tested in previous experiment.
The choice of the kernel function of the algorithm fell on Radial Basis Function (RBF), as shown in previous experiment on the same chemical set [20].
Due to the Space Colonization algorithm, both robots recalculate the path in order to deviate and achieve the goal, as in previous experiment.
The loading of the catalyst were 0.3 mg cm-2 Au and 60 μg cm-2 Pt which are three times as much as those in previous experiment.
And in previous experiment in simulator, we found the change of robot yaw angel (robot moving direction) is very small in designed tasks (as shown in following parts: "rough terrain passing" and "getting over obstacle").
The second table presents the quality of language modeling using gender-dependent acoustic models (optimized to male and female speech) trained on the APD1, APD2 and PAR databases, giving the best results in previous experiment.
