Meysar Zeinali

Application of Adaptive Sliding Mode Control for Regenerative Braking Torque Control

In air hybrid vehicles, there are two independent braking systems: frictional and regenerative. Since the regenerative braking torque is proportional to the parameters such as tank pressure and engine speed, a controller is needed for the control of the regenerative braking torque generated by internal combustion engine, based on the driver preference. In this work, a nonlinear control approach based on adaptive sliding-mode control (ASMC) is employed to tackle the problem of engine torque control during regenerative mode. To this end, a novel mean value model for a recently proposed cam-based air hybrid engine is derived for the regenerative mode and employed for designing the controller. The adaptive sliding-mode controller incorporates the approximately known inverse dynamic model output of the engine as a model-base component of the controller, and an estimated uncertainty term to compensate for the unmodeled dynamics, external disturbances (e.g., gear shifting), and time-varying system parameters such as tank pressure. The robustness and performance of the controller for this particular application is investigated and compared with that of a high-gain PID controller and a smooth sliding-mode controller numerically and experimentally. The results show that the controller performs remarkably well in terms of the robustness, tracking error convergence, and disturbance attenuation. Chattering effect is also removed by utilizing the ASMC scheme.

Development of an adaptive fuzzy logic-based inverse dynamic model for laser cladding process

The precision, performance, and robustness of model-based controllers depend, to a large extent, on the accuracy of the inverse dynamic model which is incorporated in the design of the controller. Due to complex nature of the laser cladding process and presence of time-varying uncertainties, derivation of an accurate mathematical inverse dynamic model of the process is very difficult, and involves many unknown parameters. The inverse dynamic model of the complex nonlinear laser cladding process, which is difficult to be described mathematically, can be described by a fuzzy logic-based inverse dynamic model constructed form input-output data. In this paper, the development of an adaptive fuzzy inverse dynamic model of the laser cladding process, using a systematic fuzzy modelling approach is presented. In a closed-loop laser cladding process, the scanning speed of the substrate is required to produce a clad with desired geometry and quality. In this paper, a fuzzy inverse dynamic model that describes the scanning speed as a function of the cladding parameters in particular the clad height is developed. The developed fuzzy model is validated by comparing the model output with experimental data. The results are very promising and show that fuzzy models can accurately describe the process dynamics.

Height Control in Laser Cladding Using Adaptive Sliding Mode Technique: Theory and Experiment

A closed-loop control of the laser cladding process is desired due to difficulties encountered in depositing a layer with acceptable quality from both geometrical and metallurgical point of views. One of the main parameters to achieve the desired geometry in laser cladding process is the height of the deposited layers. In this paper, a real-time measurement and control of the clad height is presented. Due to complex nature of the process and presence of uncertainties, a robust and adaptive sliding mode control is proposed and implemented to control the clad height. The velocity of the substrate is used as a control input while the molten pool height, which is obtained using a charge-coupled device (CCD) camera and an image processing algorithm is used as a feedback signal. Stability of the controller is proven in the presence of time-varying uncertainties and the performance of the closed-loop system is validated by simulation and experiments. The experimental results are promising and show that the geometrical accuracy of the deposited layers can be improved significantly.

A systematic method of adaptive fuzzy logic modeling, using an improved fuzzy c-means clustering algorithm for rule generation

Complex dynamical systems, which are difficult to be mathematically modeled, can be described by a fuzzy model. This paper attempts to improve and to address the problems concerning the systematic fuzzy-logic modeling, by introducing the following concepts: 1) an effective theoretical base method to identify the optimum fuzziness parameter (weighting exponent) m instead of the heuristic selection method mainly reported in the literature; 2) an additional criterion to choose the optimum number of clusters (rules) using fuzzy model output variation with number of clusters; 3) a generalized and parameterized reasoning mechanism constructed based on the weighted sum of the normalized defuzzified output value of each individual rule. Fuzzy model with this reasoning mechanism is suitable for online learning and real-time control applications; and 4) a gradient-descent based parameter adjustment to tune the parameters of reasoning mechanism instead of the existing heuristic parameter identification in the literature. The proposed systematic method of fuzzy modeling has the advantages of simplicity, flexibility, and high accuracy. The two example data, which have been widely used in the textbooks and literature as benchmark, are used to evaluate the performance of the proposed method

A Real-Time Object Distance Measurement using a Monocular Camera

Distance measurement for moving object using image processing is one of the important research areas in robotics and computer vision. In the current paper, an improved method is proposed to compute the distance of the moving object using a monocular camera and based on the feature extraction method. First, the specific object is tracked by a camera and then the distance from the object to the camera is measured using a computationally efficient algorithm that is implemented in MATLAB/ SIMULINK environment. Experimental results show that this method is fast, efficient, and accurate and can be used for real-time robotic applications.

LEARNING CONTROLLER DESIGN USING FUZZY MODELING, SLIDING MODE CONTROL, AND PID CONTROLLER FOR ROBOTS

This paper presents a novel approach to combine three powerful methods of control theory, namely, chattering free adaptive sliding mode control, fuzzy logic-based modeling method, PID control and fuzzy c-means clustering algorithm to build a robust learning control for nonlinear uncertain systems such as advance robot manipulators. The combination not only encompasses the features and capabilities of its components but also the limitations attributed to these techniques may be remedied by each other. To date different combinations of the above mentioned methods presented in the literature each having its own merits and limitations. But, for advance robotics applications, there is a pressing need for the control systems that are able to learn systematically and efficiently during the course of operation, from its own experience, from the demonstration and through supervised fashion as well. The controller proposed in this paper aims at building such a control system. The global stability and robustness of the proposed controller are established using Lyapunov’s approach and fundamentals of sliding mode control theory. Based on the simulations and experimental results, the proposed controller performs remarkably well in terms of the tracking error convergence and robustness against uncertainties.

System identification and height control of laser cladding using adaptive neuro-fuzzy inference systems

Adaptive neuro-fuzzy inference systems (ANFIS) are utilized to identify and control the clad height in the laser cladding process. The scanning speed of the substrate is used as the control action in the controller. A feedback signal is obtained using a CCD camera. First, the process is identified by means of an ANFIS network through a hybrid learning algorithm. The inverse dynamics of the ANFIS plant is later obtained in an ANFIS inverse learning scheme. The inverse dynamics is used in a neuro-fuzzy structure to obtain an ANFIS controller for the process. A complete control system is designed by tuning the ANFIS controller as a combined unit. Satisfactory results are obtained both in process modeling and process control.Adaptive neuro-fuzzy inference systems (ANFIS) are utilized to identify and control the clad height in the laser cladding process. The scanning speed of the substrate is used as the control action in the controller. A feedback signal is obtained using a CCD camera. First, the process is identified by means of an ANFIS network through a hybrid learning algorithm. The inverse dynamics of the ANFIS plant is later obtained in an ANFIS inverse learning scheme. The inverse dynamics is used in a neuro-fuzzy structure to obtain an ANFIS controller for the process. A complete control system is designed by tuning the ANFIS controller as a combined unit. Satisfactory results are obtained both in process modeling and process control.

A Novel Approach to Build a Learning Controller by Combining Fuzzy Modeling, Sliding Mode, and PID Control for Robotic Applications

This paper presents a novel approach to combine three powerful methods of control theory and fuzzy c-means clustering algorithm to build a robust learning control for nonlinear uncertain systems such as advance robot manipulators. It is worth noting that the combination not only encompasses the features and capabilities of its components but also the limitations attributed to these techniques (e.g, stability issues of fuzzy controller and poor performance of PID controller in the presence of time-varying uncertainties) may be remedied by each other. To date different combinations of the above mentioned methods presented in the literature each having its own merits and limitations. But, for advance robotics applications such as medical robot, and space applications and with increasing complexity of the robot’s tasks, there is a pressing need for the control systems that are able to learn systematically and efficiently during the course of operation, from its own experience, from the demonstration and also in an unsupervised fashion. The controller proposed in this paper aims at building such a control system using a novel approach to combine adaptive fuzzy modeling algorithm, fuzzy c-means clustering algorithm, adaptive sliding mode control and PID controller. Based on the simulations and experimental results, the proposed controller performs remarkably well in terms of the tracking error convergence and robustness against uncertainties.Copyright

Air Hybrid Engine Torque Control Using Adaptive Sliding Mode Control

In this work, a new air hybrid engine configuration is introduced in which two throttles are used to manage the engine load in three modes of operation i.e. braking, air motor, and conventional mode. A Mean Value Model (MVM) of the engine is developed at braking mode and a new Adaptive Sliding Mode Controller (ASMC), recently proposed in the literature, is applied to control the engine torque at this mode. The results show that the controller performs remarkably well in terms of the robustness, tracking error convergence and disturbance attenuation. Chattering effect is also removed by utilizing the ASMC scheme.Copyright

Design and Application of Chattering-Free Sliding Mode Controller to Cable-Driven Parallel Robot Manipulator: Theory and Experiment

High-performance robust controller design for nonlinear uncertain dynamical systems such as cable-driven parallel robot manipulators is a challenging work. In this paper, a new and systematic approach to combine sliding mode control, adaptive control design techniques and PID control for tracking control of cable-driven parallel robot manipulators, in the presence of model uncertainties is presented. In the proposed method, structured (parametric) and unstructured (un-modeled) uncertainties are lumped into one term and one uncertain parameter (term) is considered corresponding to each degrees of freedom of robot manipulator. Therefore, the problem of computation burden and large number of parameters, which are not addressed in the literature, is solved to a large extent. The global uniform ultimate boundedness stability is obtained in the presence of fast time-varying uncertainties. The simulation and experimental results revealed that the proposed method is robust against uncertainties and its simplicity makes the approach attractive for industrial applications.Copyright