Designing a modern aircraft requires as much knowledge about sensor technology as aerodynamics. Without the right sensors and software, many of the aircraft we now take for granted would never get off the ground. Commercial and military aircraft alike rely on a mind-numbing array of sensors to facilitate takeoff, landing, and everything in between.
As you might imagine, how sensors are configured on aircraft makes a big difference. Configurations are so important that researchers at Texas A&M University have developed a mathematical framework they believe can create better sensor configurations that are more reliable and require less hardware.
Solving the Human Problem
The impetus behind the Texas A&M research is human error. Unfortunately, human beings are not perfect. We create problems through the mistakes we make. And when you are talking aerospace, even the simplest of mistakes can lead to devastating consequences. Anyone who has been affected by a commercial aircraft crash knows this all too well.
As things currently stand, engineers designing new aircraft have to create sensor configurations manually. They use their engineering skills to make educated guesses about which sensors to use and how to arrange them within the design.
This way of doing things works fairly well. Plane crashes are more rare than they’ve ever been, and planes are safer even when crashes do occur. The point of the research is to make things even better by reducing the need to rely on human assumptions, knowledge, and skill.
Modeling Sensor Configurations
So what have the researchers done? They have created a complex mathematical framework that they can use to model how certain sensor configurations would work in the real world. Like most other modeling technologies, they create a scenario involving a specific aircraft and its available sensors. They run the scenario and collect the data.
Analyzing the data tells researchers how well the sensors performed. Their analysis shows that where improvements can be made in both sensor selection and placement. In theory, the researchers say their mathematical framework can reduce the total number of sensors a modern aircraft would need through proper selection of sensor type and placement.
Modeling and Signal Processing
What the Texas A&M researchers are doing is rather fascinating from a signal processing perspective. For purposes of illustration, consider California-based Rock West Solutions. In addition to other things, Rock West develops sensor and signal processing technologies for the aerospace sector.
When working on a signal processing project, they are attempting to develop systems and tools that can collect data from a given signal, then filter out unwanted data (otherwise known as noise) so that only wanted data remains. This is exactly what the Texas A&M researchers are doing with their mathematical framework.
Their modeling formula looks at a ton of data from a full range of sensors. It then analyzes that data to determine what is necessary and what is superfluous. Furthermore, it analyzes weaknesses in the sensor array and points the way to better configurations.
The whole thing is rather complex and difficult to wrap one’s brain around if you are not familiar with how sensors and signal processing work. Suffice it to say that some of the most complicated sensors in the world are found on aircraft. They are measuring hundreds of data variables in real time and feeding that data to the on-board computers that keep the aircraft flying.
Perhaps one day better sensor configurations will all but eliminate plane crashes. That day may never come, but it is at least worth trying to get there. The Texas A&M research is a big step in that direction.