Dr. Shyam Chetty was the Former Director of CSIR - National Aerospace Laboratories, Bangalore, and held additional responsibilities as Director CSIR-4PI, Bangalore, Project Director of the National Control Law team for LCA, and Chairman of the Systems Engineering Cluster of NAL. He has over 40 years of experience in the field of Aircraft Flight Mechanics & Control. His research interests include Flight Control System Design & Development, Aircraft Simulation & Modelling, Handling Qualities & Aircraft Pilot Coupling, Computer Aided Flight Control Design & Rapid Prototyping Techniques.
He also has the distinction of having served as the Chairman / Technical Expert of Review Committees on most of the Major National Aerospace Programmes of ISRO, DRDO and ADA. He has won several awards, prominent among them being the Jawahar Lal Nehru Memorial Award in 1976, Sir C.V. Raman Distinguished Young Scientist Award in 1998, DRDO Award for Path Breaking Research in 2002 and many more.
CSIR-NAL has also provided significant value-added inputs to all the Indian national aerospace programmes. Its contributions over the last five decades have enabled it to create a niche for itself in advanced aerospace research and technology development.
Question 3: The control laws were developed with the aid of real time simulators at ADE, Bangalore and BAE, UK. A second series of in-flight simulation tests of flight control software took place in July 1996 at Calspan, USA on an F-16D VISTA (variable in-flight stability aircraft), why was this done and how did it benefit the Tejas?
If one were to look at the block schematic diagram for the development of flight control laws one would notice there are several types of 6 degree of freedom nonlinear simulation that the designer uses to develop, test and clear the control laws. Due to the strong coupling between the pilot and the airframe, to assess the performance and stability of the outer loop which gets closed by the visual & motion cues the pilot receives when flying, extensive pilot in loop tests are carried out in realtime fixed base simulators.
The pilot in loop simulators have a representative cockpit, high fidelity external visuals and dynamic functional models of the airframe, engine, landing gear, other associated onboard systems and the atmosphere. When we started the development of the CLAW in late 1992, the realtime pilot in the loop simulator being developed by ADE was not yet fully functional. Hence it was decided to hire time & convert the BAeS Experimental Aircraft Programme fixed base ground simulator at Warton, UK for the initial pilot assessment of the LCA control laws.
Ground based flight simulators for high performance aircraft provide high fidelity visual cues but cannot provide realistic motion cues. A pilot when flying these high performance aircraft reacts to both visual and motion cues and hence assessing the performance and fine tuning the parameters of the control laws especially under maneuvering conditions becomes difficult in the absence of realistic motion cue feedback to the pilot.
Since both the design and flight test teams had no previous project experience it was considered prudent to carry out campaigns using the USAF owned NT-33 inflight simulator in 1995 and F-16 VISTA in 1996 to test and evaluate the control laws in the presence of realistic motion cues for the critical flight phases and tasks.
Both the ground based and inflight simulators also allow the pilots to assess the performance of the control laws in the presence of system failures thereby giving them the required confidence for handling flight emergencies. Further due to the presence of experienced safety pilots in the rear seat of the IFS, high workload flight tests like offset landings and closed formation flying can be planned and carried out under highly stressed conditions to ensure that there no hidden defects or sudden deterioration in the control law performance.
Question 4: How were the initial designs of the Tejas developed and how did it evolve from a Technology Demonstrator in 2001 into a 4.5 Generation state of the art Fighter Aircraft with Serial Production aircrafts now with the Indian Air Force to the next versions in MK1A and MK2?
The flight control law designers face two major challenges during development The first is to get an in depth understanding the highly nonlinear coupled dynamics of the airframe which changes rapidly during flight. The only way to capture the nonlinear characteristics over the entire flight envelope and start design is by trimming the aircraft & generating linear models for controller design at over nearly 30,000 flight conditions to cover the multidimensional flight envelopes.
It is essential to do a detailed flight mechanics parameter analysis prior to design to determine the deficiencies in the airframe characteristics, synthesise the control laws at each of these equilibrium points and then stitch the multiple designs back together to realise a unified CLAW.
The second challenge is the tight coupling between the pilot & machine and the inability of the designer to model the human pilot thereby restricting the designers inability to guarantee the stability and performance of the outer loop other than by extensive testing using several pilots in a realtime simulation.
The design of the inner loops of the stability and command flight control laws are done in the linear domain and the performance and handling are initially verified by closing the loops and numerically solving the mathematical models of the airframe and associated onboard systems using a proven 6 DoF offline nonlinear simulation software on desk top computers used for control law synthesis.
Due to the constantly changing dynamic characteristics of the airframe the designer relies heavily on the accuracy of the parameters derived from the airdata system in addition to other onboard sensor measurements to determine the current model of the airframe at each flight instant.
Unlike inertial sensor measurements which give accurate body rates and accelerations, the externally mounted airdata sensors measure values which are distorted by the local flow around the fuselage and hence are erroneous and therefore have to be corrected in flight in real time before use. The corrections in airdata parameters are initially determined using data derived from wind tunnel tests and CFD techniques. However it has been found that unless these values are validated in actual flight one can have surprises.
The initial block of LCA flights were therefore flown in the fixed gain mode by operating the standby gain switch in the cockpit. This mode which consists of two fixed sets of control law gains & parameters are selected based on the position of the undercarriage lever. This allows initial flights to be carried out over a reasonably sized flight envelope that permits not only safe takeoff and landing but also allows the performance of the airdata system to be verified using actual flight measurements.
The flight envelope is then expanded in small increments, simultaneously carrying out inflight calibration of the airdata system and the validation of the aerodynamics within the boundaries of the fixed gain envelopes using specialised parameter identification techniques.
Once the performance in this restricted flight envelope are found to be satisfactory, the scheduled gain control laws are invoked. In the scheduled gain mode, the control law parameters – over 45 in number are then selected automatically from a stored database based. This then allows systematic and safe expansion over the complete flight envelope.
When the inner loop stability and command system performance is found to be satisfactory the other control law features for carefree maneuvering including the boundary limiters and pilot relief outer loop autopilot modes are designed, integrated and then cleared.
In addition, the control law also has reversionary modes to cater for Angle of Attack, Angle of sideslip and other airdata system failures. This provision for reversion is necessary as the redundancy and integrity of these sensors are lower and the chances of the probes and vanes getting damaged are higher because they are mounted external to the airframe. The incorporation of all these features ultimately leads to the final set of control laws which are required for flying a 4.5 generation fighter.