Page 107 - Read Online

Li et al. Intell Robot 2022;2(1):89–104 I http://dx.doi.org/10.20517/ir.2022.02 Page 101

Figure 10. The varying RMSE with the increasing iterations: (A) Joints 1 and 3; and (B) joints 2 and 4.

Table 3. The tracking errors under non-disturbance and disturbance
Max Error (deg) Mean Error (deg)
Joint 1 2 3 4 1 2 3 4
Non-disturbance 0.71 0.0021 0.61 0.0016 0.27 0.0004 0.24 0.0003
Disturbance 0.76 0.087 0.68 0.089 0.32 0.039 0.30 0.057

Figure 11. Different pick-and-place trajectories within the workspace.

Table 4. The tracking errors along with different paths within the workspace
Max Error (deg) Mean Error (deg)
Joint Path 1 2 3 4 1 2 3 4
Path 1 0.71 0.0021 0.61 0.0016 0.27 0.0004 0.24 0.0003
Path 2 0.62 0.0001 0.67 0.0001 0.22 0.00003 0.24 0.00004
Path 3 0.27 0.0023 0.76 0.0015 0.069 0.0006 0.19 0.0003
Path 4 0.53 0.0009 0.33 0.0017 0.16 0.0002 0.10 0.0004

5.3. Overall performance analysis
To evaluate the overall performance of ILC in the workspace, multiple pick-and-place trajectories are selected,
as displayed in Figure 11. Table 4 shows the maximum and mean tracking errors of the joints along with
different paths, from which it can be seen that all the joint errors along with the selected trajectories can
converge to a value after iterative learning, and the converged magnitudes are quite close.