S. N. Balakrishnan

Neuroadaptive Model Following Controller Design for Non-affine Non-square Aircraft System

Aircrafts with high performance requirements or structural damage, often operate in conditions characterized by rapidly changing nonlinear aerodynamic phenomena. A neural network based model-following adaptive control design is proposed in this paper for stable control of aircrafts in such scenarios. These systems have control dependent nonlinearities making their mathematical characterization non-affine in control and they are usually nonsquare too. The uncertainties that occur during the flight are estimated online with weight update adaptation laws derived by using the Lyapunov theory to ensure boundedness of the estimation errors and the weights. The nonsquare system model is re-structured by using slack variables to facilitate the design of control using the dynamic inversion technique. Typically, when large changes appear in the aerodynamic derivatives, fast adaptation is required to ensure the system stability. But the drawback with using high adaptive gains is that they could result in high frequency oscillations in the control signal that excite the unmodeled dynamics of the plant. A novel observer structure is employed in this paper that enables fast adaptation without inducing any high frequency oscillations. The proposed method is applied to the short period dynamics of a fighter aircraft and is compared with the recently popular L1 adaptive control technique. Simulation results demonstrate the potential of the proposed method.

Robust Adaptive Control of a General Aviation Aircraft

In this paper, an application of newly developed modified state observer (MSO) based adaptive controller for the control of longitudinal dynamics of a general aviation (GA) aircraft is considered. The proposed controller structure uses nonlinear dynamic inversion to decouple the flight controls and to modify the handling qualities of the aircraft. The inversion error caused by modeling inaccuracies is compensated for by a real time robust adaptive control algorithm. The modeling inaccuracies can stem from the differences between the model used for the inversion controller and the actual aircraft which may have some impairment such as inadequate thrust delivery, improper functioning of the control surfaces or in-flight hardware failure. In all the cases the adaptive control algorithm should be able to rapidly adapt to the changing aircraft behavior and restore acceptable aircraft performance. In order to tackle this problem, the recently developed MSO methodology is employed which allows for fast adaptation without inducing any high frequency oscillations in the control signals. Simulations were carried out by introducing unanticipated failures in both trimmed and controlled flight conditions. Aeroelastic modes have also been included. This paper presents the results obtained and the analysis carried out.

Study to assess the changing pattern of clinical profile and determine the prognosis in hepatic encephalopathy

Background: Hepatic encephalopathy (HE) is a common complication of liver disease that requires intensive care management. The prevalence of HE is increasing during recent period. The most important factors of HE are alcohol consumption, chronic hepatitis, hepatotoxic drugs and unhealthy changes in life style. There were only relatively few studies from our region on the changing profile of hepatic encephalopathy under the background of life style changes. This study was conducted with the aim to detect the changing pattern of clinical profile, precipitants and to assess the prognosis of patients with hepatic encephalopathy. Methods: This was a prospective study for a period of 18 months since January 2012 at Academy of Medical Sciences, Pariyaram, Kannur, a tertiary care centre situated in the northern part of Kerala. Patients admitted in the medical and gastroenterology wards and intensive care units that fulfilled the inclusion criteria were enrolled in this study. Results: Among the 76 patients with HE, 60 were suffering from CLD and 16 due to acute liver failure. The common etiologies for HE in CLD patients were Alcoholic cirrhosis (63%), Cryptogenic cirrhosis (17%) and cirrhosis due to chronic HBV (10%) and HCV hepatitis (7%) respectively. Among the CLD patients at the start of observation majority were in Child Pugh class B and C. Based on West Haven grading most of them had Grade 2 and 3 HE. Majority with Grade 1, 2 and 3 improved where as those with Grade 4 and Grade 3 in Child Pugh class C worsened. The common precipitants of HE were GI bleed, dyselectrolemia, constipation and infections. Among these precipitants a statistically significant association for a worse outcome was present only with infection. Leptospirosis and deliberate self-harm due to ingestion of hepatotoxic rodenticide and paracetamol were the leading cause of hepatic encephalopathy in acute liver failure Conclusions: In present study HE was most commonly seen in patients with alcoholic liver disease. Cryptogenic cirrhosis associated with other life style diseases was the second common condition. Among all precipitating factors infection appeared as a statistically significant factor predicting a worse outcome. Health education among alcoholic patients and life style modifications to prevent cryptogenic cirrhosis are of paramount importance in curtailing the increase in incidence of HE in this region.

A Novel Integrator-less Robust Adaptive Control Scheme for Aerospace Tracking Applications

This paper presents an optimal control scheme to help track a reference command and if necessary a dynamic model with desirable transient characteristics. A nominal modelfollowing controller for the linearized system is designed using LQR with an error feedback. Typically, integrators are augmented to the state space for nonzero reference tracking by erro r feedback. In this case, the control at steady state based on the known system dynamics is precalculated and is used to modify the performance index. This formulation allows error feedback without necessitating the addition of an integral state and the solution of the algebraic Riccati equation can be used to obtain the tracking controller gains. However in the presence of unmodeled dynamics and nonlinearities, the steady state control needs to be redefined. This is done by using neural network based observers to identify the unknown quantities and update the control. The system stability is analyzed and Lyapunov based weight update rules are applied to the neural networks to guarantee the boundedness of the error in estimation and weights. In order to demonstrate the effectiveness of the proposed method, the simulation results for a 2�DOF helicopter tracking problem and the short period dynamics of a fighter aircraft are presented .