Basic Computers — Page 315, Lesson 371

I want to walk you through the world of basic computers as it applies to aircraft. We're starting fresh here, so let's build this up from the ground. First, let's talk about how we actually tell a computer what to do. We use programming languages, and there are two broad families you need to know. The first is a High Level Language. This is an advanced computer programming language that is not limited by the computer, or by one specific job, and it's more easily understood by the author than an assembly language. In plain terms, it's written in a way that's closer to human logic, so a programmer can read and write it more naturally. The examples you'll see in aviation and general computing are FORTRAN, ALGOL, BASIC, and C++. These are the tools you'd use to write a complex application. The second family is a Scripting Language. This is a programming language that allows control of one or more applications. Here's the key distinction: "scripts" are distinct from the core code of the application, and they apply to a specific program only. They're usually written in a different language, something similar to Basic, and they're often created, or at least modified, by the end-user. So think of the high level language as building the main engine, and the scripting language as the small set of instructions that tweaks how that engine behaves for a particular job. Now, here's where aviation gets serious. For a flight to be conducted safely, it's vital that any computer software must be safe to use and free of any errors that could endanger the flight. So a specification has been agreed to which all aircraft-related software must comply. That specification is known as EUROCAE ED12B. This defines 5 levels of software, and the level required depends entirely on the application. Let me give you the contrast that makes this click: the galley oven controller could accept a major failure without endangering the aircraft, whereas any failure within the Flight Control Computer could be dangerous to the flight. Same standard, different consequences, so different levels of rigor. Let me walk you through those five levels, because they're tied directly to failure conditions. Level A is for a Catastrophic failure condition, which in the aviation context means it would prevent continued safe flight or landing. Level B is Hazardous/Severe-Major, meaning there's potential for fatal injuries to a small number of occupants. Level C is Major, which impairs crew efficiency, causes discomfort, or possible injury to occupants. Level D is Minor, which reduces aircraft safety margins but stays well within crew capabilities. And Level E is No Effect — it doesn't affect the safety of the aircraft at all. So the more severe the potential failure, the higher the software assurance level required. Now let's look at where these computers actually live in the aircraft. The systems that are computer controlled include the Flight Management System, or FMS; the Digital Flight Guidance System, or DFGS; the Ground Proximity Warning System, or GPWS; and the Traffic Alert Collision Avoidance System, or TCAS. And of course, fly-by-wire aircraft take computer control much further — the whole flight envelope is controlled by computer process, with inputs from the crew when necessary. Here's an important design trend. Current design favours the use of dedicated computers for each separate system. Each one gets its own box. But in the future, we may see sharing of computer power in the form of an Integrated Hazard Warning System, or IHWS. In that concept, a powerful central processor, with appropriate back-up, handles inputs from the stall warning system, windshear detection, GPWS, TCAS, and even the Weather Radar. It processes all that information and prioritizes warnings to the crew. So instead of five separate boxes, one central brain sorts out which warning matters most at any given moment. Finally, let's talk about how these computers talk to the real world. Many aircraft sensors produce analogue information — varying voltages, pressures, temperatures, and so on. But digital computers use digital, binary information. So we need a device called an Analogue to Digital Converter, or A to D, in the interface between the sensor and the computer input device. It translates the continuous analogue signal into the discrete binary ones and zeros the computer understands. And the process works in reverse too. When a digital computer has to pass information to an analogue device, we use a Digital to Analogue Converter, or D to A. So the A to D gets sensor data in, and the D to A gets commands out to the actuators. That's the complete loop — sense, convert, compute, convert back, act.