Semiparametric Deep Learning Manipulator Inverse Dynamics Modeling Method for Smart City and Industrial Applications

Table 2

Prediction results of RBD, NDL, and SDL models.

		Axis 1	Axis 2	Axis 3	Axis 4	Axis 5	Axis 6	All axes
Cross 1	RBD	2.2899	2.3417	2.3683	0.6523	0.5889	0.5290	1.7024
	NDL	2.6107	2.7715	1.8227	0.3418	0.4487	0.4577	1.7486
	SDL	1.8310	1.9460	1.3504	0.3450	0.4220	0.4524	1.2560
Cross 2	RBD	2.3316	2.3162	2.2893	0.6642	0.6498	0.5525	1.6937
	NDL	2.3342	2.5439	2.5428	0.3864	0.5162	0.4722	1.7807
	SDL	1.6721	1.8492	1.4528	0.3435	0.4633	0.4654	1.2162
Cross 3	RBD	2.4861	2.6723	2.2126	0.6417	0.5924	0.3582	1.7846
	NDL	2.2477	2.4820	1.1689	0.4081	0.4487	0.3977	1.4779
	SDL	1.4031	1.5533	0.9656	0.3294	0.4022	0.3428	0.9748
Cross 4	RBD	2.7325	2.7323	2.5119	0.6625	0.6386	0.3708	1.9246
	NDL	2.3239	2.1714	1.6374	0.3745	0.4245	0.4280	1.4889
	SDL	1.4009	1.6816	1.1613	0.3435	0.4399	0.3696	1.0478
Cross 5	RBD	2.8011	2.7946	2.2741	0.6326	0.5795	0.3620	1.9015
	NDL	1.6768	2.2989	1.6690	0.4346	0.3893	0.3393	1.3746
	SDL	1.5326	1.6847	1.0004	0.3385	0.3775	0.3058	1.0439
Cross average	RBD	2.5282	2.5714	2.3313	0.6506	0.6098	0.4345	1.8014
	NDL	2.2387	2.4535	1.7682	0.3891	0.4455	0.4190	1.5741
	SDL	1.5679	1.7430	1.1861	0.3400	0.4210	0.3872	1.1077