Research Article

Dynamics Model Abstraction Scheme Using Radial Basis Functions

Figure 5

RMSE values for predicted data of sensors 𝑆 1 and 𝑆 2 for sparse velocity-driven control scheme. The darker colour in the top of the columns indicates the increment of RMSE in the test stage. (a) Using a different class (model) for each single object. (b) Using a different class (model) for each set of two objects. Cases (objects): 𝐴 (0 and 1), 𝐵 (0 and 2), 𝐶 (0 and 3), 𝐷 (0 and 4), 𝐸 (1 and 2), 𝐹 (1 and 3), 𝐺 (1 and 4), 𝐻 (2 and 3), 𝐼 (2 and 4), and 𝐽 (3 and 4). (c) Using a different class (model) for each set of three objects. Cases (objects): 𝐴 (0, 1, and 2), 𝐵 (0, 1, and 3), 𝐶 (0, 1, and 4), 𝐷 (0, 2, and 3), 𝐸 (0, 2, and 4), 𝐹 (0, 3, and 4), 𝐺 (1, 2, and 3), 𝐻 (objects 1, 2, and 4), 𝐼 (1, 3, and 4), and 𝐽 (2, 3, and 4). (d) Using a different class for each set of four objects. Cases (objects): 𝐴 (0, 1, 2, and 4), 𝐵 (0, 1, 2, and 3), 𝐶 (0, 1, 3, and 4), 𝐷 (0, 2, 3, and 4), 𝐸 (1, 2, 3, and 4). (e) Using a single class for the only set of five objects. In this case the increment of RMSE in the test stage is neglectable. Cases (objects): (0, 1, 2, 3, and 4).
761019.fig.005a
(a)
761019.fig.005b
(b)
761019.fig.005c
(c)
761019.fig.005d
(d)
761019.fig.005e
(e)