Type-2 FLSs were able to model and control the cooling water flow to achieve the target coiler entry temperature in an HSM. It showed a substantial improvement in performance and stability of type-2 FLS over type-1 FLS. The experiments and results presented in [11] show that type-2 FLSs were able to model and control the cooling water flow to achieve the target coiler entry temperature in an HSM. It showed a substantial improvement in performance and stability of type-2 FLS over type-1 FLS. Throttle Rudder Ailerons Elevators In [12] a type-2 FLS was applied to airplane altitude control for the propulsion based RC airplane shown in Fig. 3, where the throttle is used to regulate the speed of the airplane by varying the rotational speed of the propeller. The elevator is used to control the airplane's ascent and descent, the ailerons are used for airplane's lateral stabilization and mid air turning, and the rudder is used for the on-ground taxiing of the airplane. The work compared type-1 and type-2 FLS in airplane control. It was found that under high uncertainty levels the type-2 FLS outperformed the type-1 FLS. Specifically, the type-1 FLS showed oscillatory behavior around the reference altitude set-points as shown in Fig. 4. C. Type-2 FLSs Application in the Modeling of Micromilling Cutting Forces In [13], the authors present the use of a type-2 FLS for the estimation of the dynamic cutting forces. Ren et al. note: "The type-2 fuzzy estimation not only filters the noise and estimates the instantaneous cutting force in micromilling using observations acquired by sensors during cutting experiments, but also assesses the uncertainties associated with the prediction caused by the manufacturing errors and signal 700 700 600 600 Altitude (m) Altitude (m) FIGURE 3 Basic airplane controls [12]. 500 Reference Actual Altitude 500 400 400 300 300 0 100 200 300 400 500 600 700 800 900 Time (s) (a) B. Type-2 FLSs Application to Airplane Altitude Control Reference Altitude Error 0 100 200 300 400 500 600 700 800 900 Time (s) (b) FIGURE 4 (a) Results of the type-1 FLS in the simulation setup with uncertainties [bottom blocks are the magnified steady state responses, (RMSE = 3.58 m)]. (b) Results of the type-2 FLS in the simulation setup with uncertainties [bottom blocks are the magnified steady state responses (RMSE = 0.43 m)] [12]. 18 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | AUGUST 2012