Pursuing a Career in Flight Control Engineering

Flight Control Engineer is certainly one of those jobs you would struggle to explain to your grandparents and by the end there is a high chance that they think “this kid should rather be a doctor” (unless they are aerospace engineers themselves). But which are exactly the skills expected from such kind of engineers? Which are the crucial differences between a confirmed/mid-career and a senior Flight Control Engineer (and more specifically Flight Control Laws Engineer)? In order to try to answer these questions, I went through the typical development process of the Flight Control Systems (FCS) and searched for the main required skills in job positions advertised by important aerospace companies. These are my two cents.

What do Flight Control Laws stand for?

We can start this investigation by trying to answer briefly why do we insert flight control laws in our modern flying machines. This is a kind of philosophical question which I invite even the most experienced engineers to do. Luckily we know a possible clear answer since at least 1971 (by McRuer) who says that a flight control system provides basically the following features:

• Stability (either general or at specified times).
• Desired responses to specified inputs;
• Suppression of the effects of undesired inputs.

This can be translated basically as safety and economic operation, as summarized by Magni, Bennani and Terlouw, in their book Robust Flight Control — A Design Challenge. For example, we can build an aircraft with a relaxed static stability in order to obtain a reduction in fuel consumption using active control techniques; or we can decrease the pilot workload and make the aircraft behavior uniform throughout the flight envelope, as taught in the very first classes of Aircraft Control and Stability at university; ultimately we can make the aircraft completely autonomous.

Theoretical Background

The successful achievements of Flight Control Systems (FCS) in the aeronautical history so far, allowed the increasing of safety requirements and economical and performance demands. As expected this was followed by an increasing complexity of these systems such that, still according to Magni et al, this made the work proportionally challenging from both technical and management perspectives.

There is a very informative and still effective compilation of best practices in Flight Control Design done by NATO/RTO (Technical Report 29, Flight Control Design — Best Practices, 2000) which I consider a must-read for younger engineers (at least the chapter 4). You may be surprised how a document from 19 years ago is still very applicable nowadays. This document (referring to Shmul et al) says that the required skills for flight control design cover the following engineering areas:

• control theory,
• control system architecture,
• aerodynamics,
• aircraft dynamics,
• aero- (and aero-servo-) elasticity,
• aircraft loads,
• weight balance,
• simulation and modelling methods,
• digital signal processing,
• software engineering.

We can add to this list some rising requirements that may become usual in a near future like, for example: the knowledge of the engine and propulsion in the eVTOLs (Electric Vertical and Take-off Landing vehicles), which may be entangled in the inner loop of the control; or possible demands from outer-loops like in a context of intelligent Flight Management Systems.

Hands-on Experience

Besides the huge amount of theory to be aware of, as in any engineering field, the Flight control design career demands a long time to accumulate experience, whose acquired knowledge is not always organically passed to the new generation of designers.

Following the flight control design subprocesses listed in the Best Practices (by the aforementioned NATO report), we will try to address the industry needs from a flight control designer engineer. Here is important to state a little note: the comments and opinions for each item below are totally based on this author thoughts, solely.

1. Establishing the aerodynamic design and system performance requirements. In the very first phases of the aircraft development, the challenge in the control design can be dramatically simplified if an experienced FCS engineer is able to point out, for example, non-linearities in the aerodynamic data that can be avoided, reducing the number of design cases to be tackled in the future.

• Industry may require: anticipation in preliminary designs are massively based on the history of lessons learned by other aircraft programs, in other words: this basically requires experience.

2. Modelling and analysis of the unaugmented vehicle. For any control design task is of uttermost importance to be fully aware about the plant we are trying to control. This is not different for FCS. Discovering the dynamics of the aircraft in each part of flight envelope, including its non-linearities and understanding how each part of the plant (including sensors, for example) will be modeled, provides the designers with a clear picture of the size of the problem they are supposed to solve.

• Industry may require: good understanding of the different types of sensors (air data, inertial, etc), their architecture, and how they are usually modeled (filters, delays, conversion calculations); familiarity with tools for visual modeling of dynamic systems (Simulink, Scade, Xcos); good understanding of the aircraft dynamic and how the main stability derivatives affect its overall behavior; understanding of all the available actuation system (surfaces or possibly engines, for example).

3. Design criteria and flying qualities specifications. No matter if the flight control is for a military or civil aircraft, manned or unmanned vehicle, setting up the design criteria and flying qualities specifications since the early stages of development can avoid a lot of rework later on, mainly when we will have to show our products to the certification authorities. Obviously that these requirements will be iteratively refined throughout the product development. Knowing the regulations rules will help to choose which requirements can be relaxed or not, for example.

• Industry may require: knowledge of the market segment and customers needs; knowledge of certification regulations (EASA Certification Specifications, FAA Federal Aviation Regulations), the MIL-STD specifications (if applicable); understanding of stability margins, PIO rating, flying qualities standard criteria and pilot rating (like Cooper-Harper rating scale), etc.

4. Control laws design and development. In a traditional approach, after having a non-linear model we produce linearised models throughout the flight envelope; we decide the structure of our controller; we calculate the gains for each chosen points in flight envelope; we choose the way we will schedule the gains; we translate our linear controller structure to the non-linear model; we will eventually include non-linear structures like rate and position limiters in the structure; we may anticipate the clearance documentation that will be required by the authorities for proving the fulfilling of design criteria.

• Industry may require: design control techniques and performance analysis (stability margins, for example); understanding of model trimming and linearization processes and how to use the tools to perform it as desired; linear aircraft analysis (dynamic modes and state-space representation); dynamic system modeling (Matlab/Simulink, C/C++, Fortran or whichever programming language used); understanding of flight control computer architecture and functioning; understanding of configuration control systems (SVN, GIT); understanding of requirements management systems (Rational Doors, Caliber, Jama, Pearls, etc); understanding of issue tracking and project management tools (Jira, Clarizen, Zenhub, GitScrum, etc).

5. Control laws functional specification, implementation and verification. Having the flight control software implemented in the real hardware requires an exhaustive testing plan to assure that the code is doing what the designer intended it to do.

• Industry may require: understanding of the translation process of the control laws algorithms to the actual code to be embedded in the real hardware; awareness of the main constraints of the hardware; awareness of the integration details of the flight control system with the other systems; knowledge of the main recommendations of the international standards for critical software/system development (ARP4754, DO-178B, DO-178C).

6. Piloted simulation and handling qualities. Prior to expose to the pilots the non-linear model with the implemented control laws, we have to assure that the simulator facility is representative enough for the handling qualities assessment. Once this job is done, a series of pilot evaluations shall be conducted which will eventually lead to design changes.

• Industry may require: understanding of the representativeness of the simulator environment and its limitations; knowledge of standard pilot techniques at least in the main missions of the aircraft; understanding of PIO rating and flying qualities rating (Cooper-Harper, for instance).

7. Aeroservoelasticity and structural mode filter design. The flexible modes of the aircraft structure may cause instability when coupled with the flight control laws response. The main trade-off of the flight control designers is to make the structural mode filters effective enough to avoid the control law to respond in higher frequencies (nearby the flexible modes frequencies) and at the same time not compromising the lower frequencies of the “rigid body” response, keeping the targeted stability margins of the control law. This has been a great challenge for some newest aircraft, in which the flexible modes are closer and closer to the “rigid” modes.

• Industry may require: understanding of the flexible and rigid modes of the aircraft structures; knowledge of control system performance and stability analysis; understanding of filter design (mainly band-stop and notch filters); understanding of possible issues related to the discretization of these filters.

8. Design robustness and flight clearance. The increasing capacity of simulation and the reliability of the aircraft models have improved significantly the robustness of the prototypes prior to the actual flight. The costs of software simulations are much lower than real flights and the gains in maturity level of the product implies safer flights, such that the test capacity of each test environment (software-in-the-loop, hardware-in-the-loop, iron bird) shall be extensively used. After that, the flight clearance process to certify the aircraft will be as smooth as the certification plan allows. The effectiveness of this plan will rely on the engineer knowledge of the regulations and experience to deal with clearing (or redesigning) non-compliant cases.

• Industry may require: knowledge of robust control theory; understanding of the uncertainty associated with aerodynamic model and the various sensors and ability to assess the robustness of linear and non-linear designs; understanding of the certification procedures and requirements (see item 3).

9. Developments during Flight Testing. Issues will be found during the flight test campaign. This requires a very good level of readiness from the flight control engineers to fully understand and solve (if this is the case) the raised problems. Of course, before flying the solution, a very carefully planned cycle of validation shall be performed.

• Industry may require: readiness and resourcefulness to solve the problems safely, quickly, effectively and again safely.

10. Management aspects. In this myriad of different areas interfacing to each other, with many inputs and outputs to be provided, with a cycle to be iterated many, many times, we do need someone to at least prioritize the activities, take the main decisions, and to shield the technical engineers from unnecessary burdens.

• Industry may require: global view of the development process and team coordination skills to deliver results in a very fast pace, indisputably fulfilling the safety requirements but balancing quality and deadline constraints.

Walking through a career path

As a fresh graduate engineer it is very unlikely that you will start your career as a Flight Control Laws Engineer following the development process since the very first beginning until the end, i.e., following approximately the sequence of subprocesses enumerated in the previous section. As you might have noticed, in the preliminary design phase, for example, the main contributions will be done by the more experienced engineers. Also, you may start working in an aircraft program that is already flying and sometimes flying throughout many, many years.

Normally, in the beginning of the career, the engineer will be involved closely in one phase of the development cycle (again, you may find an approximate description among the ones described in the previous section). He/she may iterate few times over the same tasks and/or get involved in other subprocesses later. Either way, by being exposed to all the interface areas, the engineer will become more and more familiar with the macro process of the development. By this time it might be clear if a technical or a management career is desired. Trying to summarize very roughly, in a technical side, you will go deeper in each subprocess of the development, and in a management side, whilst developing the leadership skills (which I consider extremely important), you might improve your view of the process as whole and how it is integrated in the full product.

Of course, nothing said here is absolutely frozen and deterministic. The boundaries limiting the perimeter of a Flight Control Laws team activities in an aircraft development context are not absolutely clear themselves. Such that you may start your career as a System Engineer working in the logics for the consolidation of the sensors signals, for example; or as a Modelling and Simulation Engineer, building the unaugmented vehicle model, for instance. There are countless possibilities. I saw System Engineer becoming Senior Flight Control Laws Engineer, I saw Embedded Software Engineer becoming Flight Control Laws Engineer, I even saw System Engineer becoming Neuroscientist (I truly did). But that is another story.