Sensing Our Way to Connected Vehicle Innovation
Industry Insights

/

July 26, 2022

/

#

Min Read

Sensing Our Way to Connected Vehicle Innovation

This is an external post, click the button below to view.
View Post

What burns fossil fuels and is made of metal, rubber, and glass? A motor vehicle! This answer was true for a long time. Later on, other materials and methods entered the picture, like plastics, semiconductors, and lines of code. This change commenced slowly, to then accelerate. It has indeed been argued that the vehicles of 2022 differ more from 1972 vehicles than those of 1972 from 1922.

Note that these were 50-year periods. But today, things are radically shifting. What began with the first sensors and proceeded to printed circuit boards has now caused a revolution in car design, manufacture, and driver experience within less than a decade. Software costs are currently 10% of the vehicle bill of material but are predicted to reach 50% by 2030. 

This series of articles covers factors that will determine the future of the vehicle and the vehicle of the future. Last time we talked about Tight vs Loose Coupling—a rather abstract aspect. This time we will discuss something more concrete…sensors

Morning

Let us return to the year 1972 for a moment, to a cold and clear winter morning. Imagine being a motorist of those days: You leave the house and, without the assistance of a key fob, unlock the car. Being careful, you checked the oil level last night, using the dipstick, and tried to deduce from the color if an oil change was warranted. Somewhere there is a note about oil and coolant levels from the garage, but you don’t remember where it is. You walk around the car, kick a tire or two, and get in. You pull the choke, press the accelerator pedal once or twice to awaken the carburetor (don’t flood it!) and turn the key. After some extended turning—on subfreezing mornings the carburetor takes its time to produce an actual combustible mix—the engine awakens, slightly coughing. You let it idle to warm up for a while. The car interior remains cold, only after some miles will the engine be hot enough for the heating to become useful.

Try this in 2022? You jump in your car and press a button. The inside of the car is warm, for you have programmed the heating. The engine awakes immediately, thanks to electronic injection that considers factors like temperature, air pressure, and airflow. Last night you received an email that, according to your driving patterns and mileage, an oil change is due soon. You know the fluid levels in the car are fine because you’d get a message on the dashboard otherwise. Off you go. A beautiful, cold motorist morning in 2022.

Certainly, there are a lot more differences in cars and driving between 72 and 22. But the fact that, for example, cars do not just rust away beneath us anymore, thanks to better paint and galvanization, is more of an incremental improvement. Electronics, on the other hand, have been a true game changer, and this is, in particular, true for sensors and control unit combinations.

Sensing Things

What is a sensor? To keep it simple, we could begin with tire kicking as a rudimentary sensor action: the contact of foot and tire creates a, well, sensation about firmness, information which is then conveyed to and processed by the brain and may lead to further events. So a sensor is a device (foot) that detects qualities of or changes in its environment (tire) and sends the information to a processor (brain).

However, as venerable as the combo foot-tire-brain might be, it only creates a rather vague, one-off impression of conditions. We would not call this data. Compare this to the 4 camshaft position sensors in a modern DOHC V8 motor that produce at least four times four hundred data points about position per minute. While sensors are not new and oxygen sensors, for example, have been in cars from the ’80s on, a modern car may have up to 100 of them, to measure parameters like motion, rotation, angle, position, pressure, temperature, gas flow, oxygen, obstacle detection, and even occupant convenience and alertness.

Fun fact: Sensors may produce their own artifacts and effects. If you put an old-fashioned capillary thermometer in the coffee cup in front of you, the temperature shown will very briefly fall in the beginning (because the hull containing the liquid expands first) and in the process, the thermometer will cool down the coffee, if only infinitesimally (on account of its own heat capacity). So, to really exactly determine the temperature of your coffee you will have to (a) wait long enough for the instrument to fully adjust and (b) account for the initial temperature and thermal mass of the instrument. Enjoy your cup of joe!

Driving

But let us get back on the road: As a 72 driver you have a fuel gauge and a jerry can, for fuel gauges back then were notoriously unreliable. The jerrycan was a standard piece of car equipment of those days that has largely disappeared today. An amenity already in use in 72 was cruise control, which had been around since the sixties, but has developed from simple, vacuum-controlled mechanical systems to today’s integration into engine management systems and the wonders of adaptive cruise control.

In 2022, lane assist keeps us on the highway, blind spot warning watches out for us, and rain sensors adjust wiper speed. Thus we glide along and if the comfort of the drive makes us sleepy, eye movement or heartbeat detection can check if we are becoming drowsy.

And these are only the performances of sensors and ECUs we immediately perceive while driving. But much more is going on. Deep in the engine, knock sensors watch over the combustion, the manifold absolute pressure sensor monitors engine load, the NOx sensor measures the oxides of nitrogen in the exhaust. They all report to their respective ECUs and adjustments are made in milliseconds per precisely defined parameters.

Making Sense of It

We could now ask if this all makes, well, sense. Our driver from 1972, after all, had a decent chance to get the car started and arrive at their destination. 100 sensors in a vehicle means a considerable investment and the question is: does it pay off, for manufacturers and drivers alike?

However, without even getting technical, some things are apparent. The 1972 no-sensor car would not sell today, lacking basic comforts and security features. Nor would it get the approval of traffic authorities around the world, not being safe and clean enough for modern standards. 

And while even in 1972 your daily drive usually worked out just fine, many car troubles from back then have vanished, thanks to sensors and electronics. Carburetors being flooded, distributors acting erratically, spark plugs full of soot—that kind of stuff may make some of us feel a little nostalgic, but few of us want these little mundanities back. While an old issue of Popular Mechanics may quite fascinatingly enlighten you about various afflictions of spark plugs, these problems are gone, fuel injection and electronic engine control having eradicated them.

As we have seen, the reasons to put sensors in cars are as manifold as their tasks. Sensors improve performance and reduce pollution (an engine that burns cleaner runs better, in many aspects), enhance safety, create comfort, and improve reliability. On a second level, a vehicle that offers these benefits will be attractive to buyers, reduce long-run costs for driver and OEM alike, and improve the perception of the brand.

Today, the sensor game is reaching a new level, with new use cases rapidly becoming feasible. This not only means that cars make further progress on the above named paths but will also help to generate ROI from value added services and to redefine the value creation for OEMs—from once at the point of sale to a lifetime via added services and subscriptions. Advanced driver assistance systems, for example, are rapidly gaining ground and rely on CMOS image sensors for driver support and safety. These sensors in turn need precise calibration, an aspect our blog article about Maintaining ADAS offers further reading about.

All this entails that more and more data is produced and will need to be leveraged. This is not without its challenges. The appropriate data has to be collected and a bidirectional flow of data to and from the car has to be managed. The data stream for an autonomous car may reach 25 GB per hour. To deal with such amounts of data at rest and in transit is not trivial and aspects like proper compression come into play to reduce rising transmission costs, a field we have covered in our blog article about OPEX Costs.

The Sibros Contribution

With Sibros’ Deep Connected Platform, consisting of Deep Updater, Deep Logger, and Command Manager, OEMs and fleet managers can leverage sensor data for a whole bevy of use cases like vehicle health monitoring and predictive maintenance. The DCP allows OEMs to update, select, store, compress, and transmit data with high precision and safety, little implementation effort, close control, and minimized driver interruption. Reliability, flexibility, and scalability have the potential to reach new levels.

The Sibros solution is hardware-agnostic and certified to the highest international safety, cybersecurity, and data protection standards. The following is a selection of important features of the Sibros DCP:

  • Better use of sensor data by precise and flexible data logging
  • Reduced OPEX costs through high compression
  • Over-the-air updates to all ECUs
  • Updates, logging, and commands on one platform
  • Hardware agnostic and cloud-ready solution
  • Powerful cloud portal
  • Rapid integration by API
  • 360-degree view via millisecond-level data collection

OEMs receive a single and reliable source of truth for connected vehicle data. This not only means better use of sensor data but also new insights from real-world driving to answer questions like:

  • How do engines age? How do different drivers of the same car and engine model influence load, performance, wear, and tear? Two cars with 50k miles on them may have completely different engine and powertrain conditions.
  • How are vehicles really used? This could affect the calibration of the suspension, for example, and even influence interior design, with data on how many passengers a certain vehicle type regularly transports and the general use of functions and amenities (see our article about Reductive Design). The same type of van with 8 seats may see, depending on region, completely different uses.
  • In connection with precise weather (and traffic) data: How does certain weather actually influence driving behavior? Drivers may rely on rain sensors or overrule them, depending on very small differences in conditions.

Answers to such questions may not only change the vehicles we build and use, but might be a game changer for the cooperation between OEMs and tier 1 suppliers.

Evening

Arriving at the destination, the 2022 driver relies on park assist; our 72 motorist parks the old school way, with a little trial and error, including occasional scratching of shiny chrome bumpers. After a long highway drive in 1972, it was still a good idea to let the engine idle for a while to cool down. Then a turn of the key switched off the engine, which meant the car was dead, unresponsive until the key got turned again in the ignition the next morning.

In 2022, turning and removing the key only means a change of state of the vehicle and in no way switching it off. The fan and other components may run for a while, to then fall dormant, but many systems remain latent.

The entire automotive sector senses the rapid transition from hardware to software, a process that has to be proactively addressed. Consumers today expect their vehicle brand to be innovative. OEMs that are in the driver’s seat with attractive features and value-added services will be the winners in this race.

Sensors and sensor data will see a new level of responsibility and face challenges of a new depth. This in turn will accelerate the cycle of innovation. Precise quantification and near-constant data streams open up great opportunities—if the collection, filtering, compression, transmission, and analysis of data are mastered. Sibros delivers on the requirements that OEMs have in this rapidly developing sector, and our total system design approach seamlessly connects every vehicle to the cloud. The days of kicking tires and relying on oil color are gone. Talk to us today.

Max Reinhold
Max Reinhold
Max Reinhold has been in the writing and technical communication space for more than 15 years. He spent the better part of his twenties disassembling motorcycle engines—sometimes he was even able to assemble them back again. Model airplanes make him feel nostalgic. He graduated from Humboldt University of Berlin.