Prathap H.,College of Engineering, Trivandrum |
Proceedings of the IEEE International Conference on Control Applications | Year: 2013
Due to the inherent nonlinearities of Reusable Launch Vehicle (RLV) dynamics, its changing properties during flight and the engineering difficulties to predict its aerodynamics with high levels of fidelity, flight control requires strategies that allow to cope up with the non-linearity of the model and assure robustness in the presence of inaccuracies and changes in configuration. This paper presents a flight control strategy based on dynamic inversion controller which is designed for the re-entry phase of Reusable Launch Vehicle. In order to solve the robustness problem of regular explicit Nonlinear Dynamic Inversion (NDI) control law, the Incremental Nonlinear Dynamic Inversion (INDI) control law is proposed. Sensitivity to model mismatch is eliminated by feeding back state acceleration in INDI approach. The improved control law design is validated for re-entry phase of RLV for nominal and aerodynamic perturbation cases. Analysis of simulation results reveal that the robustness of the control law is increased. © 2013 IEEE.
Anjaly P.,College of Engineering, Trivandrum |
2016 Indian Control Conference, ICC 2016 - Proceedings | Year: 2016
An adaptive integrated guidance and control scheme is developed for the entry phase of winged RLV. The reference profiles are generated off-line using constrained optimization techniques to take care of all possible variations in initial conditions and path constraints. Lateral and longitudinal guidance commands are generated in-flight with the help of stored reference profiles to meet the boundary conditions. Non-linear dynamic inversion is used to design inner loop control. Unpredicted failures like control surface saturations, aerodynamic uncertainties, chattering of control surfaces are addressed by the bandwidth adaptation logic. The integration of the guidance and control loop is achieved by a feedback from control to guidance section, thereby making the guidance loop adaptive to changes in the control bandwidth. In this way, the guidance subsystem generates feasible trajectories that ensure safe flight of RLV despite unpredicted scenarios. The integrated scheme is simulated extensively in 6 Degrees of freedom (DOF) environment along with non-linear actuator dynamics to validate the algorithm. Robustness of the proposed method is demonstrated using Monte Carlo simulations with various failure modes and plant parameter uncertainties. © 2016 IEEE.
International Journal of ChemTech Research | Year: 2015
Multi-gate devices like FinFETs can be used in the nanometer regime to reduce short channel effects encountered by planar MOSFETs. Along with the scaling advantages, FinFETs face the problem of width quantization. To overcome this, an advanced architecture, Asymmetric Drain Spacer Extension (ADSE) FinFET was proposed. This is a comparatively new device and the modeling and applicability analysis is still in its infancy. Till now no compact SPICE model is proposed for this device which helps circuit designers to analyze its performance at circuit level. This paper is trying to fill that literature gap. In this paper, a new compact SPICE model is proposed for ADSE FinFET. The model was verified against the published results in the literature. Obtained simulation results agreed with the published results from analytical models. This model was later used for the performance analysis of the device at circuit level. This is followed by a comparison of its performance against planar devices at the circuit level. This was done by evaluating the performance of ring oscillator circuit. This was done for a fixed underlap length. The major observations from this analysis are: FinFET is better than MOSFET in terms of drain current and frequency response and the power dissipation of underlap device is less than that of corresponding device with no underlap. The propagation delay of underlap device is less than that of normal device. © (2015), International Journal of ChemTech Research. All rights reserved.
Skariya S.E.,College of Engineering, Trivandrum |
Sebastian B.,VSSC |
IFAC Proceedings Volumes (IFAC-PapersOnline) | Year: 2014
Reusable Launch Vehicle (RLV) is used to carry payload from earth's surface to the outer space for more than once.The digital autopilot (DAP) used to control the RLV has an actuation system as its inner loop. The actuation system deflects the control surfaces (eg. fins) to bring about the necessary control action. The actuator is a complex system with many elements which work in co-ordination to bring about the necessary action. But in the design of autopilot, actuator is usually approximated as a second order system having the same bandwidth as that of the actual one. Due to this approximation, an understanding of the entire plant cannot be achieved. During implementation, the approximation of the order of the actuator may lead to deviation of the performance of the system from its specifications. It may also lead to unnecessary control of subsystems which would have been corrected without any control in an integrated model. In this paper, an integrated design of the RLV and its actuation system is developed. The actuation system consists of an electro hydraulic actuator and the associated elements. It is designed to actuate the booster fin of an RLV for the purpose of pitch control. The Linear Quadratic Regulator (LQR) control technique is used to design DAP for the integrated system. LQR is a modern control technique which ensures optimality in the design of the integrated system. A comparison of the performance of the integrated design of RLV and the design with an approximated second order actuator is made. The advantages of LQR technique over the conventional gain design technique are also mentioned. © 2014 IFAC.
Jaison H.,Karunya University |
Anas S.R.,National Institute of Technology Calicut |
Gopinath A.,VSSC |
Subathra M.S.P.,Karunya University
2011 International Conference on Emerging Trends in Electrical and Computer Technology, ICETECT 2011 | Year: 2011
This paper presents, an Electro-Mechanical Actuator (EMA) based position control system with redundancy technique is being presented. The servo system is composed of two chains, a prime chain and a redundant chain, which includes the position servo system of the EMA. Only one chain works at a time, probably the main chain operates during normal condition of the system. A new, simpler fault detection and isolation (FDI) technique is being developed, which will be monitoring the prime chain continuously. If any type of fault occurs in the prime chain, the FDI isolates the prime chain from the system and switches to the redundant chain. Thereafter redundant chain will be driving the system. The system model is then simulated using MATLAB/Simulink, with trapezoidal waveforms of backemf. Using the DSP controller chip, TMS320F2812 as a control core, hardware implementation has been carried out. © 2011 IEEE.