Shubhi Purwar

On-line system identification of complex systems using Chebyshev neural networks

This paper proposes a computationally efficient artificial neural network (ANN) model for system identification of unknown dynamic nonlinear discrete time systems. A single layer functional link ANN is used for the model where the need of hidden layer is eliminated by expanding the input pattern by Chebyshev polynomials. Thus, creation of nonlinear decision boundaries in the multidimensional input space and approximation of complex nonlinear systems becomes easier. These models are linear in their parameters and nonlinear in the inputs. The recursive least squares method with forgetting factor is used as on-line learning algorithm for parameter updation. The good behaviour of the identification method is tested on Box and Jenkins Gas furnace benchmark identification problem, single input single output (SISO) and multi input multi output (MIMO) discrete time plants. Stability of the identification scheme is also addressed.

Adaptive output feedback tracking control of robot manipulators using position measurements only

In this paper, a new adaptive neuro controller for trajectory tracking is developed for robot manipulators without velocity measurements, taking into account the actuator constraints. The controller is based on structural knowledge of the dynamics of the robot and measurements of joint positions only. The system uncertainty, which may include payload variation, unknown nonlinearities and torque disturbances is estimated by a Chebyshev neural network (CNN). The adaptive controller represents an amalgamation of a filtering technique to generate pseudo filtered tracking error signals (for the elimination of velocity measurements) and the theory of function approximation using CNN. The proposed controller ensures the local asymptotic stability and the convergence of the position error to zero. The proposed controller is robust not only to structured uncertainty such as payload variation but also to unstructured one such as disturbances. Moreover the computational complexity of the proposed controller is reduced as compared to the multilayered neural network controller. The validity of the control scheme is shown by simulation results of a two-link robot manipulator. Simulation results are also provided to compare the proposed controller with a controller where velocity is estimated by finite difference methods using position measurements only.

Real-time implementation of Chebyshev neural network observer for twin rotor control system

This paper addresses the problem of observer design for the twin rotor multi-input-multi-output (MIMO) system which is a nonlinear system. Exact knowledge of the dynamics of twin rotor MIMO system (TRMS) is difficult to obtain but it is highly desired that the observer can dominate the effects of unknown nonlinearities and unmodeled dynamics independently to prevent the state estimations from diverging and to get precise estimations. The unknown nonlinearities are estimated by Chebyshev neural network (CNN) whose weights are adaptively adjusted. Lyapunov theory is used to guarantee stability for state estimation and neural network weight errors. A comparative experimental study is presented to demonstrate the enhanced performance of the proposed observer.

A Nonlinear State Observer Design for 2-DOF Twin Rotor System Using Neural Networks

In this paper, a stable neural network based observer for twin rotor multi-input-multi-output (MIMO) system is proposed. The twin rotor MIMO system (TRMS) is a highly nonlinear system. First, a simple local state observer for TRMS is presented. The efficiency of this observer will depend on the accuracy of the model. Then, a neural network – based observer is proposed. This observer can be applied to TRMS system without any a priori knowledge about the system dynamics. A two – layer neural network is used to approximate the nonlinearities of the system. A learning rule for neural network is given which guarantee robustness of the observer. Simulation results are carried out to exemplify the performance of the proposed observers.

Optimal control of twin rotor MIMO system using output feedback

In this paper an optimal state regulator is designed for the twin rotor multi-input-multi-output (MIMO) system. The twin rotor MIMO system (TRMS) exemplifies a high order nonlinear system with significant cross couplings. From the nonlinear model of TRMS a linearised model is obtained and a controller is designed to regulate the states. The controller gain is updated iteratively, until optimal value is reached. Finally simulation results are presented to show the effectiveness of the proposed controller for the TRMS.

Sliding mode state observer for 2-DOF twin rotor MIMO system

This paper presents a sliding mode state observer for the 2-DOF twin rotor MIMO (multi-input-multi-output) system which belongs to a class of inherently nonlinear systems. Design parameters are selected such that on the defined switching surface, asymptotically stable sliding mode is always generated. Robust sliding and global asymptotic stability conditions are derived by using Lyapunov method. The unknown nonlinearities are estimated and the state estimation errors tend to zero asymptotically.

Backstepping Control of Discrete-Time Nonlinear System Under Unknown Dead-zone Constraint

This paper proposes the adaptive back stepping controller for a class of nonlinear discrete-time systems in strict-feedback form with unknown dead-zone using neural networks. The control design is attained by introducing the dead-zone nonlinearity and using it in the controller design with back stepping technique. A dead-zone inverse is developed to compensate the dead-zone effect in nonlinear systems. In this scheme, Chebyshev Neural Network (CNN) is used to approximate the unknown nonlinear functions and also used to compensate the dead-zone nonlinearity. New weight updates laws are derived to guarantee uniform ultimate boundedness (UUB) for all signals in closed loop system.

Higher Order Sliding Mode Controller for Robotic Manipulator

This paper proposes higher order sliding mode controller for robotic manipulator. The scheme is used to compensate for the influence of unmodeled dynamics and to reduce chattering. Simulation results show that the proposed controller gives better performance compared to fuzzy sliding mode control in the face of uncertain system parameters and external disturbances.

Adaptive Control Of Robot Manipulators Using CNN Under Actuator Constraints

In this paper, a stable neuro adaptive controller for trajectory tracking is developed for robot manipulators without velocity measurements, taking into account the actuator constraints. The controller is based on structural knowledge of the dynamic equations of the robot and measurements of joint positions only. The gravity torque which may include payload variation and disturbances etc represent system uncertainty, which is estimated by a single layer Chebyshev neural network (CNN). The adaptive controller represents an amalgamation of a filtering technique to eliminate velocity measurements and the theory of function approximation using CNN to estimate the gravity torque. The proposed controller ensures the local asymptotic stability and the convergence of the position error to zero. The proposed controller is robust not only to structured uncertainty such as payload parameter variation but also to unstructured one such as disturbances. The validity of the control scheme is shown by simulation studies on a two link robot manipulator.

Nonlinear Controllers for a Light-Weighted All-Electric Vehicle Using Chebyshev Neural Network

Two nonlinear controllers are proposed for a light-weighted all-electric vehicle: Chebyshev neural network based backstepping controller and Chebyshev neural network based optimal adaptive controller. The electric vehicle (EV) is driven by DC motor. Both the controllers use Chebyshev neural network (CNN) to estimate the unknown nonlinearities. The unknown nonlinearities arise as it is not possible to precisely model the dynamics of an EV. Mass of passengers, resistance in the armature winding of the DC motor, aerodynamic drag coefficient and rolling resistance coefficient are assumed to be varying with time. The learning algorithms are derived from Lyapunov stability analysis, so that system-tracking stability and error convergence can be assured in the closed-loop system. The control algorithms for the EV system are developed and a driving cycle test is performed to test the control performance. The effectiveness of the proposed controllers is shown through simulation results.