How Virtual Sensors Streamline Complex System Development

Systems tend to become smarter: a car can now interact with its environment, cameras on your smartphone can correct the image quality by taking environmental conditions into account, and augmented reality goggles can track your head, eyes, and gestures to display specific content. This trend of pervasive intelligence is only possible because sensors are everywhere. But when it comes to sensors, the physical variety have their limitations. Namely, they can be prohibitively expensive and laborious to implement in more and more complex systems.

Take the field of autonomous cars as a simple example. To operate effectively, a vehicle’s autonomous driving system requires millions of miles of driving to build an “understanding” of all the possible situations it may encounter, involving an extensive mapping of the environment captured by embedded sensors.

One way to do this is by manually deploying a large number of cars equipped with physical sensors to drive on the roads of a city such as San Francisco and record what they see, amounting to millions of images. To do this, you must factor in both significant time and financial outlay, running from fitting out the cars to hiring the drivers and covering vast stretches of road.

Alternatively, employing virtual sensors and leveraging high-performance computing technology minimizes the heavy lifting as well as the hazards associated with extensive real-world driving. Read on to discover more about the benefits of virtual sensors, potential applications, and more.

What is a Virtual Sensor?

To read the full article, click here

×
Semiconductor IP