Modeling and feedback linearization control of a nonholonomic wheeled mobile robot with longitudinal, lateral slips

Neural network-based adaptive tracking control for a nonholonomic wheeled mobile robot with unknown wheel slips, model uncertainties, and unknown bounded disturbances

“…We compared the tracking performance of the proposed control method with that of the feedback linearization control method [8] with the purpose of validating the advantages of the proposed control method. For illustration, the two following examples were implemented in MATLAB/Simulink software.…”

“…Additionally, another difference between our proposed control method and other methods, except for the one in [8], is that our proposed control method must check the invertible property of matrix h in Eq. (13) as in Remark 2 before designing the control law, and then the control law must always check and guarantee that matrix h is invertible in the implementation of the closed-loop control system.…”

“…Obviously, in Figures 4 and 5, we can easily see that when the accelerations and velocities of the unknown wheel slips were not measured and model uncertainties and unknown bounded disturbances existed, the control approach in [8] could not compensate the undesired effects while the proposed control method effectively dealt with the undesired effects. In addition, as can be seen in Figure 6, the control inputs and some of the weights have been bounded.…”

“…Regardless of the unknown wheel slips, model uncertainties, and unknown bounded disturbances, the proposed control method managed to compensate the harmful effects very effectively, while the control approach in [8] could not. It should be noted that the position tracking errors of the proposed control method almost converged to zero, as shown in Figures 5 and 8, whereas that of the feedback linearization control method [8] did not. As a consequence, the tracking performance of the former is better than that of the latter.…”

“…Methods based on gyros and accelerometers to deal with wheel slips in real time were also described in [6,7]. The authors of [8] proposed a feedback linearization controller for tracking a desired trajectory of a WMR in the presence of longitudinal and lateral slip at each driving wheel under ideal conditions where model uncertainties did not exist, such as unstructured unmodeled dynamic components and unknown bounded disturbances such as unknown bounded external forces, and the values of the accelerations and velocities of the wheel slip could be measured exactly. Nonetheless, it is impractical to achieve a good performance by using this feedback linearization controller in real applications as the ideal condition is unrealistic.…”