A Hyundai Venue N Line runs down a well-lined access road, its sensors alive. Off to the left overhead, an Akasa Boeing 737 Max makes its way down gracefully, flaps extended, ready to land. Both machines are bristling with technology that allows them to operate with a degree of independence once unthinkable. The car can keep itself within a lane, regulate speed in relation to the car ahead, and brake in emergencies. The aircraft can fly itself across continents, execute complex routes, and even land in zero visibility. Welcome to the new world where machines are self-aware and can operate for long stretches on their own.

Today, we dive deep into tech and compare how Advanced Driver Assistance Systems (ADAS), found in modern cars like the Hyundai Venue work, and juxtapose them against how advanced autopilot systems on airliners like the Boeing 737 Max function. Want to know how this Hyundai Venue senses lane markings, what it picks up, how it processes the data, makes a decision and executes a steering manoeuvre to keep the car within the lane, or how aircraft like the Boeing 737 Max are programmed to fly themselves and how they can even land automatically? Read on; there’s plenty of fascinating tech ahead. 

Sensor fusion 

Walk around the attractive-looking Akasa 737 Max, now one of forty-three, and you notice the nose of the aircraft bristling with sensors. It’s from where a vast majority of the data for the Autopilot is collected. Showing me around and explaining things is Captain Svetlana from Akasa. 
First up, the speed sensors – elbow-shaped pitot tubes that stick out at an angle. This tube helps measure the airflow and, hence, the speed of the aircraft. For accuracy, this speed data is correlated with the static port, mounted flush on the side of the aircraft, which measures the real atmospheric pressure at the height the aircraft is flying. 

Then there’s the AOA or Angle Of Attack sensor that basically tells you how the air is hitting the wings or how much the wing is tilted into the oncoming airflow. This is essential to know just how much lift the wing is making and how close the aircraft is to stall. 

Then, inside the nose cone, apart from the weather radar, are antennas for the localiser and glideslope. Both pick up radio beams sent from transmitters positioned near the runway. The localiser tells the aircraft how far left or right of the runway centreline it is, and the glideslope tells if the aircraft is too high or too low. Together, they give the aircraft a complete three-dimensional picture of the ideal path to the runway. They feed data to the Instrument Landing System (ILS) and are what lets a 737 Max line up and descend precisely even when visibility is poor.

Further back are a pair of TCAS antennas, one on the belly and one on the top. They listen for nearby transponders (that also help flightradar24 funtion) to keep the aircraft clear of traffic. 

Positioning is done with the help of a GPS antenna on the roof. Another system, known as the Inertial Reference System, uses laser gyroscopes bolted to the airframe to sense every turn the aircraft makes. It uses accelerometers that sense every change in speed. A computer fuses the two, continuously tracking exactly how far and in which direction it’s travelled from a known starting point, which is set by the pilots at the gate.

Road and track 

Now swap the open skies for chaotic tarmac. Here, the Venue N Line’s ADAS Level 2 suite runs off far fewer sensors or data points. The difference, however, is that the car actually ‘sees’ what is ahead, correlating data from the radar and the high-resolution camera into a single picture. Sensor fusion is essential on both the 737 and the Venue. What’s different is that the reactions of the car have to be immediate, and that means Smart Sense on the Venue not only has to process gigabytes of data every second, it also has to be ready with, say, a braking solution in milliseconds, just in case that petulant biker cuts across the Venue’s nose. 

One of the primary sensors is the radar. Located behind the black square plastic panel under the number plate up front, it acts as a measuring tape, constantly tracking the exact distance and closing speed of the car ahead. Similar to a fruit bat, the radar sends out bursts of high-frequency radio waves (or chirps) that vary between 10,000-50,000 bursts per second. Just as important is how quickly the data is processed and passed on, which provides enough sharpness and resolution to differentiate between shapes ahead. This data is used by the Forward Collision-Avoidance Assist (FCA) to stop the Venue often before human reflexes can even respond to danger. In addition, the system uses a pair of rear-mounted radars, ultrasonic sensors and parking and proximity cameras for a more secure 360-degree ‘picture’. 

The other important data source is a high-resolution, high-speed camera at the top of the windshield. While its primary job is to look for lane markings and obstacles, it also works along with the radar up front, sharing the workload. The camera, for example, sees every detail but lacks depth perception (or 3D vision), and a radar tracks distance but cannot see the same amount of detail as the camera. The camera sends data via a special, automotive application-optimised CMOS sensor, similar to the one on your Sony mirrorless camera that also has a high dynamic range (for bright lights). Hyundai’s system can even detect faded lines with its edge-detection, texture-tracking, and colour-reading systems. 
Hyundai’s SmartSense also integrates Artificial Intelligence (AI) and Machine Learning. So, instead of using rigid “if-then” decision-making software, it uses embedded neural networking to interpret real-time sensor data. 

Together, these systems form a nervous system that feeds into the car’s brain, which then nudges the steering, modulates throttle, or applies brakes. Unlike aviation, redundancy is limited. A sensor failure usually hands control back to the driver, who remains the ultimate fallback.

Steering wheels and yokes

Once behind the wheel of the Venue, engaging the ADAS is seamless. You need to be above 30kph and on a well-lined road. To engage the system, you have to zoom into the left spoke of the steering wheel. Here you find the four buttons that help activate and operate the system. The first activates the Smart Cruise Control, the second allows you to increase or decrease the speed you want to run at, the third helps you set the desired safe distance to the car ahead and the fourth activates Lane Keep Assist (LKA). 

Activating LKA for me comes first, and you do this when you are running in a lane on a well-marked road. You know the system is engaged when the green steering icon lights up on the instrument panel and the car icon shows it running between the lines. Then you engage the smart cruise control and set the speed you want to run at, stepping off the accelerator simultaneously. The display on the instrument cluster lights up with the speed set, and you also get a horizontal bar showing the distance to the car ahead; which can be adjusted by the button with the distance icon. 

Unseen by the driver and passengers in the car, the camera immediately begins scanning lane markings, and radar modules measure the distance to the vehicle ahead. The steering wheel then nudges itself gently to follow curves, while the throttle and brakes modulate to maintain speed and distance. The system makes split‑second decisions to help keep you out of harm’s way. 

Climbing into the Akasa Boeing 737 Max cockpit is an immediate sensory overload. Hundreds of buttons, switches, levers and dials just overwhelm you. They are ahead of you, to your left, your right, and even above. There are, in fact, so many switches, buttons and dials above, I have to duck-walk in. And I’m not exactly Joel Garner or Kareem Abdul-Jabbar. Once seated, I slide the seat to the left and forward by depressing a lever, no powered seats here. And forget about comparing screen size, this is a proper ‘glass cockpit’ with wall-to-wall screens. My feet rest on the pedals that work the rudder in the tail, and ahead of me is Boeing’s famous yoke; unlike a regular joystick, you bank by turning the semi-circular bit of the ‘wheel’. 

In contrast to our chaotic driving environment, the autopilot in the 737 Max operates on different principles. The environment is structured and regulated. There are no sudden jaywalkers or rogue bikers at thirty‑five thousand feet. So, unlike the Venue, the 737 doesn’t use optical cameras to see where it’s going. It doesn’t need to look for lanes or lines at 800kph. Instead, it relies on a combination of GPS, Inertial Reference Systems, antennas and sensors in the aircraft’s nose and cheek to get a fix on the exact heading, speed and altitude.

After taking a good look at the real cockpit and understanding some of the controls that help initiate the autopilot sequence, we shift to a full-scale moving platform simulator used to train airline pilots around the world. The cockpit is identical, down to the last detail, and here we have Captain Sanaya Chothia to help us conduct a simulated flight. Can’t help noticing, the 737 also has a tyre pressure monitor; of course, it does. 

First up, a full-bore take off in the sim, Capt. Sanaya working the rudder pedals to keep the aircraft running down the centre line. Wow, this feels real, the sound, the visuals, the motion all combined. What a thing! After climbing out, we take a quick look at the sequence used to initiate the autopilot, most of the steps done on the long horizontal panel that sits between the pilots, directly in their line of vision. The first step involves checking if the auto throttle is engaged, and then all the pilot needs to do is press a single button on the autopilot panel. Unlike ADAS, the route the aircraft has to take is already pre-programmed into the navigation computer, a fairly long and detailed process. The speed, the height, the course, the heading, everything. And the autopilot just follows this.  

Once the system is initiated, the throttles move on their own, adjusting thrust; the yoke makes deliberate corrections and the screens display the aircraft banking smoothly along its GPS track. As the flight progresses, pilots can keep the aircraft on autopilot and make plenty of alterations to the route, either to avoid weather, get a more direct route or as directed by Air Traffic Control. There are knobs on the panel to alter course, speed, heading and altitude, with even the rate of descent and ascent adjustable. 

We even execute an autoland, where the aircraft lands itself. The Instrument Landing System frequency is first inputted into the Multi-Mode navigation panel, the final approach course is set for the runway on the Mode Control or autopilot panel (272 for runway 27 in this case), and then the APP or approach mode is selected once the aircraft is positioned correctly. Doing this arms the localiser and glideslope systems that, as we’ve mentioned earlier, have their sensors in the nose cone. Seeing the aircraft adjust, manoeuvre, manage its path, height and distance, and land itself is an eye-opener. What an incredible piece of technology! 

What’s fascinating is that both systems have boundaries. In aviation, autopilot is trusted to fly for hours, even land. Pilots monitor, manage and intervene when necessary, but the system is designed to shoulder the bulk of the workload. Take offs, however, are always manual. In cars, ADAS is an assistant with a much narrower scope. It reduces fatigue, takes a load off the driver and adds a safety net, but ultimately the driver is always on call at short notice and is the one in control. 

Sharing the workload

Similar in execution but very different in how they function, ADAS and Autopilot both overlap but also exist in isolation. One is designed to keep you from bumping into a hatchback on the Western Express Highway, while the other is designed to guide an 80-tonne metal tube across oceans. Both are glimpses into a future where machines will shoulder more of the burden of control and blow us away with the technology on offer. But for now, if you step into an aircraft, know that while autopilot may be doing a bulk of the flying, a pilot is always watching, and when you engage ADAS to ease your commute, keep your eyes on the road, your hands on the wheel.