Fig 1.
Electromechanical model of PMDC motor.
Fig 2.
Smooth Dead Zone region of motors’ characteristic curve.
Fig 3.
Hard Dead zone of PMDC motor characteristics.
Fig 4.
Characteristic curve for Coulomb and viscous friction.
Fig 5.
Structure for deployment of PID controller.
Fig 6.
Procedure for simulation and experimentation.
Fig 7.
Reference input and Actuator Output.
Fig 8.
Actual and Noisy angular position outputs.
Fig 9.
Output estimation using Standard EKF.
Fig 10.
Output estimation using Adaptive EKF.
Fig 11.
Comparison of estimation results for standard and adaptive EKFs.
Fig 12.
Comparison of state error covariance for standard and adaptive EKFs.
Fig 13.
Experimental Setup for noise reduction in position and velocity scenarios.
Note in third part of Fig 14 that an overshoot is noted due to inertia when motor reach its reference angle when it is reversed. Also the controller output/effort change its polarity as motor reverses its direction. Furthermore, when noisy output and reference input is applied to the nonlinear variants of Kalman estimators, estimation results from standard and adaptive EKF are obtained that are shown in Fig 15. Fig 15 shows that the proposed adaptive EKF more comprehensively estimate the true reference value as well as provide less overshoot at reversal points. Thus the relatively better performance of proposed EKF variant is validated. The state.
Fig 14.
Reference input, controller effort and noise corrupted measured position signals.
Fig 15.
Noise reduction performance of Standard and Adaptive EKFs for position output.
Fig 16.
Comparison of State error covariance of Standard and Adaptive EKFs for position.
Fig 17.
Reference input, controller effort and noise corrupted measured Velocity signals.
Fig 18.
Performance comparison of Standard and Adaptive EKFs for velocity output.
Fig 19.
Comparison of State error covariance of Standard and Adaptive EKFs for velocity.
Table 1.
Statistical performance analysis KF variants for noise estimation and rejection.