Table 2.

Comparison between OS-ELM and other methods.

Approaches	Number of Calibration Points (Offline + Online)	Training Time (s)	Testing Time (s)	Accuracy (m)
KNN	40 + 0	-	0.108	3.098
Fuzzy KNN	40 + 0	-	0.117	2.728
Batch ELM	40 + 0	11.574	0.113	2.581
OS-ELM	30 + 0	8.101	0.113	2.615
	30 + 2	1.204	0.111	2.487
	32 + 2	1.213	0.114	2.346
	34 + 2	1.182	0.109	2.219
	36 + 2	1.226	0.115	2.091
	38 + 2	1.229	0.117	1.973