How To Establish a Flight Dynamics and Control Institute at PAC Kamra

Author Profile: Syed Aseem Ul Islam is PhD candidate at the University of Michigan, Ann Arbor, USA, specializing in adaptive and model-predictive flight control systems. He received his bachelor’s degree in aerospace engineering from the Institute of Space Technology, Islamabad, and his master’s degree in flight dynamics and control from the University of Michigan.

In a previous article we examined the importance of a lynch-pin technology that is flight dynamics and control (FDC). We analyzed all the ways in which expertise in this area manifests itself in aircraft and munitions programs. We ended the previous article with the assertion that the establishment of a Flight Dynamics and Control Institute (FDCI) is the need of the hour.

In this article, we will attempt to outline broad goals of this FDCI, and the ingredients needed to establish such an institute keeping in mind various practical aspects.

We will examine the objectives of the FDCI and discuss how to achieve those objectives.

Note: this is not an exhaustive list, but a list of key tasks for which FDCI is needed.

Developing, Instrumenting, and Operating Wind Tunnels

Wind tunnels are often large, building-sized devices used to model aerodynamic flows that a flight vehicle may experience without having to build and fly an actual prototype.

Scale models are tested in wind tunnels to verify predictions made using aerodynamic models and computational fluid dynamics (CFD) simulations. This is needed as airflow around flight vehicles is a complex phenomenon that can only ever be approximated, and never exactly modeled.

Therefore, physical experiments are always needed to investigate and verify predictions made by theory. Furthermore, data from wind tunnels is used to construct accurate mathematical flight models for flight vehicles. This is done by measuring constants called “stability derivatives” that quantify how the aircraft behaves in flight.

Not surprisingly, wind tunnels are heavily instrumented and are equipped with sensitive force and moment balances, high-speed cameras, accurate mounting systems, and power systems that supply the large amounts of power needed for the wind tunnel’s operation.

Wind tunnels are usually classified by size of their test sections and the speed at which they can move air through them. The size of the test section determines the size of the model to be tested. As a rule of thumb, larger models are better at replicating full scale effects, and thus, wind tunnels with larger test sections are preferred. However, the larger a wind tunnel gets, the more expensive it is to build and operate.

Airspeed in wind tunnels is classified by Mach number, which is the ratio of the speed to the local speed of sound. Wind tunnels are specialized for specific ranges of Mach number and FDCI will need wind tunnels that can simulate everything from 0 to up to Mach 4. It can be envisioned that FDCI will need the following wind tunnels:

A low-subsonic tunnel (Mach 0-0.3): A simple and cheap to build and operate facility. These are usually open circuit (open at both ends) with powerful fans moving air at the required speeds. This type of wind tunnel is useful for testing flight characteristics at take-off/landing speeds and constructing models of slow flying flight vehicles such as most UAVs.

A high-subsonic and transonic wind tunnel (Mach 0.3 – 1.1): A complicated and usually closed-circuit (a continuous-tube) tunnel. For this type of wind tunnel the test gas is not air but a refrigerant like R-134a. Refrigerants are used as they allow simulation of higher Mach numbers at relatively lower airspeeds.

Unfortunately, these wind tunnels require specialized equipment to handle the test gas, maintain seals, and have high power requirements (in the range of thousands of horsepower).This kind of wind tunnel is used to test flutter (aerodynamic forces causing structural flex that negatively effects flight) and other compressible flow and transonic effects. Like all wind tunnels, this type of wind tunnel is also used to create mathematical models of flight vehicles at high-subsonic and transonic speeds.

A supersonic wind tunnel (Mach 1-4): A wind tunnel that also uses refrigerants as the test gas but with a much smaller test section than a transonic wind tunnel.This wind tunnel might have two test sections where one is used for Mach numbers as high as 4 while the other one is used for Mach numbers between 1 and 2.

This type of wind tunnel is used for testing supersonic flight conditions.As the FGFA will almost certainly fly at supersonic speeds, the design will need to be tested in supersonic wind tunnels. Additionally, the development of flight vehicles like high-speed target drones, air-to-air missiles, supersonic cruise missiles will require wind tunnels that can achieve up to Mach 4 in their test sections.

FDCI would need to construct and operate these wind tunnels utilizing lots of time and resources. It is almost certainly true that such facilities are found in organizations like AWC and DESTO already, but these facilities are not easily accessible, and as such of limited use to PAC.

PAC needs its own dedicated wind tunnels that can be used extensively by its engineers without the need for travelling to other organizations, and working through security clearances and bureaucratic red tape. As an economic offset, time on these wind tunnels can be sold to researchers from universities in Pakistan (and allied countries) who currently have extremely limited access to wind tunnels. This time can also be sold to other strategic organizations in the country that may be saturating their own wind tunnel testing capacities.

These wind tunnels at PAC will primarily be designed and operated by aerodynamicists and mechanical engineers. Building and maintaining them will require large teams of civil engineers, electrical engineers, and technicians as wind tunnels are usually housed in large purpose-built buildings with dedicated grid-stations to provide the large quantities of power.

Operating wind tunnels also requires spending resources on accurate scale model construction that are appropriately instrumented. Computer numerical control (CNC) machining and 3D printing will play an important role in the construction of accurate models for wind tunnels and thus, will need special attention.

Building Scale Models

Any aerospace setup needs a scale model construction facility. Scale models are of two broad kinds: scale models used in wind tunnel testing, and scale flight models used for flight tests.

Wind tunnel testing requires accurate models of flight vehicles and even small parts of the flight vehicles separately. Most wind tunnel tests are done primarily to measure the aerodynamic flows, and are thus, not about quantifying structural and dynamic effects.

Therefore, these models are machined from stiff and heavy materials like steel or aluminum using CNC machines, or 3D printed. These models are also machined for very small holes called “taps” through which air pressures are measured all across the model’s surface. 3D printing allows for many more and much finer taps to be included in these models compared to machined parts, and is thus, of great value.

Other wind tunnel tests are done to test dynamic and aeroelastic effects like flutter. For these tests, the models must be carefully constructed to match the structural properties of the designed flight vehicle. This can either be done by skilled technicians using traditional hobby modeling techniques or by using 3D printed parts, the internal structure of which can be tailored to achieve desired structural properties.

On the other hand, flight test models are primarily used to verify control systems designs using actual flights of the model. These models are scale replicas of the designed aircraft in every possible way, and thus, require fully accurate construction.

As these models are flown physically, they need onboard electronics to implement the flight software. Furthermore, experienced hobby pilots are needed to fly these scale models. Experts in hobby aircraft construction and flying would be vital for this setup.

Equally important will be technicians and experts in CNC machining and 3D printing. The large existing setup of CNC machining at PAC can be utilized for making scale models for wind tunnels, while investment in metal 3D printing machines and technologies will go a long way in facilitating Project Azm.

The envisioned aero structures design institute (ASDI) at PAC will likely provide expertise in composite manufacturing for these models as well.

Developing Computational Fluid Dynamics and Reduced Order Models

It is often expensive and time-consuming to test every configuration and condition for a flight vehicle in a wind tunnel. Therefore, computational fluid dynamics (CFD) based models are developed in conjunction with wind tunnel testing programs.

CFD simulations replicate the wind tunnel tests in simulation but are prone to errors due to the complexity of all fluid flow. This leads to the strategy in which CFD models are simulated and their results are verified in wind tunnels for a few carefully selected conditions. The data from these wind tunnel tests is then used to verify, and if necessary, correct the CFD models.

These corrected CFD models can now be run for various configurations and conditions at a fraction of the time and cost needed for wind tunnel testing. This allows for a much more extensive testing regime for flight vehicles.

Unfortunately, these CFD simulations are computationally expensive and powerful banks of processors are needed to run them in any reasonable amount of time.

The models obtained from CFD can be simplified using reduced order modelling techniques which produce models that are simple enough to be implemented in near real-time. These reduced order models together with wind tunnel testing data are used to construct a library of mathematical flight models of varying complexity and configurations that the rest of FDCI can use.

CFD models will be developed by experts in CFD and in finite element methods (FEM) in general. Furthermore, researchers that specialize in reduced order modelling will be needed to reduce the complexity of the models obtained from CFD to make them amenable for use in mathematical flight models that need to run in near real-time.

Finally, experts in flight dynamics will be needed to construct mathematical flight models which incorporate the reduced order models and measurements made in wind tunnel tests.

Developing control systems for aircraft subsystems and munitions

Arguably the most important role FDCI will have is the development of control systems for all the sub-projects envisioned under Project Azm.

Consider for example the high-speed target drone project currently being undertaken at PAC. This is essentially an autonomous aircraft that needs to fly preprogrammed mission profiles. Assuming that flight control models are already developed for this drone using wind tunnels, CFD, and reduced order modeling, then control systems engineers will need to develop control systems that will enable the drone to be safely launched from a catapult while maintaining stable flight and then fly the drone for the entirety of its mission.

Control systems design is not limited to the design of autopilots for aircraft. Control systems will need to be developed for the regulation of UAV engines, flight control of various types of missiles, stabilization of electro-optical payloads, control of submunitions, and control of missile seeker heads, to name a few use cases.

The design of control systems will be undertaken by experts in the field of control systems as detailed in the previous article. Additionally, to avoid the pitfalls of tunnel vision, experts in various subfields of control like robust control, model-predictive control, and adaptive control would need to work together to find the best solution for each application.

Developing Flight Hardware

The process of control systems design also involves developing the requirements for sensors, processors, and actuators. Software must be written to implement the envisioned control system on the hardware.

The sensors and actuators must be modeled, tested, and the models verified with testing. Models of these sensors and actuators form an integral part of the complete mathematical flight models needed for control design. The digital processor requirements for implementing the control system must also be specified. The control system may be implemented on a microcontrollers or FPGA hardware.

The choice of systems is determined by cost, size, and power requirements. The complete flight control system is a complex cyber-physical system for which detailed hardware and software architectures must be conceived, and in which different pieces of hardware and software must talk to each other seamlessly.

A large team of technicians and experts in electronics, embedded systems, aerospace sensors, avionics, and computer science would be needed to develop the flight hardware on which the control systems are to be implemented. Much of the expertise and hardware needed for these tasks can be found already at the avionics production factory (APF) at PAC.

Conclusions
As detailed in the previous article, the establishment of FDCI is of utmost importance for the success of Project Azm and all that it hopes to achieve.

This article has detailed what this would mean in terms of infrastructure and human resource development. The proposed investment in time and people might appear to be very large, but nevertheless it is justified.

It may be tempting to believe that PAC can get away without establishing FDCI, e.g., save time and money by outsourcing FDCI’s tasks to a foreign partner. However, this is living in a fool’s paradise as no fourth-generation aircraft (let alone fifth-generation aircraft) can be developed without a mastery over flight dynamics and controls (FDC).

This assertion is not due to some ill-conceived desire for indigenization, but a consequence of how interwoven flight control systems are for modern (fourth generation onwards) aircraft. On top of this, expertise in FDC is vital for munitions development and integration. The freedom to develop and integrate munitions is something PAF has always strived towards, and it will be shame if it does not have this freedom for its FGFA under Project Azm.

Finally, and most importantly, it would be quite unfortunate if an oversight such as this leads to significant delays and cost overruns in the development of FGFA or leads to a design that does not meet PAF’s requirements.