Virtual Sensors for existing hardware replacement

Cut costs by reducing the number of hardware sensors in vehicles

Vehicles produce large amounts of data. These data can be utilized to virtualize legacy hardware sensors.

Less sensor hardware directly reduces product costs through:

  • Reduced bill of material
  • Reduced integration effort

How can Virtual Sensors replace hardware sensors?

Virtual Sensors are purely software-based and do not require any additional or specific hardware to run. Combining semi-physical approach, signal processing and machine learning algorithms, Virtual Sensors only use standardly available signals and data in series-production vehicles and accurately reproduce physical quantities out of them. They can be used to enhance the capabilities of existing vehicle software or to replace hardware sensors in any vehicle’s electrical and electronic (E/E) architecture, from current legacy architectures up to Software-Defined Vehicle (SDV) architecture.

Development and application examples of Virtual Sensors

When we replace a hardware sensor in a vehicle, we either take a generalized ready-to-use Virtual Sensor or utilize our proprietary Virtual Sensor pipeline for agnostic Virtual Sensor calibration, allowing us to deliver dedicated Virtual Sensors very fast for a broad spectrum of applications, thus reducing time-to-market.

A typical example of a generalized Virtual Sensor is our Virtual Mass Sensor, which computes the current vehicle mass with only a few kilograms of difference. Getting access to the current vehicle mass and payload is critical for overload detection in the commercial vehicle area or for a more accurate range estimation of electric vehicles.

For fleet managers, payload and mass are critical parameters regarding safety at work, or to assign the most appropriate vehicle to each driver based on their job profiles.

Hardware sensors for vehicle mass or vehicle payload exist either as a retrofit or already as a built-in by vehicle OEMs. Still, because of their high Total Costs of Ownership (TCO), such sensors are usually limited to high-end light commercial vehicles (LCV) or truck applications.

Our Virtual Mass Sensor makes payload measurement affordable for passenger cars, light commercial vehicles, trucks and passenger counting in busses. The Virtual Mass Sensor determines vehicle mass after a couple of minutes of driving. It is a reliable, convenient, and cost-efficient solution to prevent vehicle overloading.

Automotive OEMs and telematics solution providers can choose between different methods of implementation:

  • Embedded on existing electronics hardware – on the edge
  • Through our partner’s BlackBerry IVY platform
  • In a cloud-to-cloud architecture

Our Virtual Mass Sensor has already been applied to various vehicle platforms: passenger cars, light commercial vehicles, busses and trucks. COMPREDICT’s Virtual Mass Sensor plays for instance a key role in the electric fleet solution of WATÈA by Michelin.

Short study on the impact of payload on EV driving range

The smartest sensor is one which does not physically exist

When replacing hardware sensors with Virtual Sensors, you:

  • Reduce product costs
  • Reduce the amount of electronics in vehicles
  • Increase supply chain flexibility

Watch the demo video of our Virtual Sensor for suspension travel

Vehicle mass is just one example of a hardware device that can be replaced by a Virtual Sensor.


We work on other applications of Virtual Sensors replacing hardware sensors with major automotive OEMs and TIER 1 suppliers.