About COMPREDICT
COMPREDICT provides virtual sensor solutions for the automotive industry, specifically targeting Original Equipment Manufacturers (OEMs) and Tier 1 suppliers. The platform replaces expensive or complex physical hardware sensors with purely software-based alternatives. By leveraging existing vehicle data—such as wheel speed, torque, and battery voltage—the tool uses machine learning and signal processing to estimate values that would normally require dedicated physical instrumentation. This transition to software-defined mobility helps manufacturers reduce vehicle weight, simplify the bill of materials, and lower overall production costs.
The technology works through a three-step process: data acquisition from standard vehicle buses, calibration of the specific virtual sensor model, and finally, deployment either in the cloud or directly within the vehicle's embedded system stack. Key use cases include monitoring wheel forces, damper forces, driveshaft torque, and vehicle mass. By using sensor fusion and embedded AI, COMPREDICT allows for real-time monitoring of component health. This enables right-sizing of components during development and provides the necessary data for predictive maintenance strategies once the vehicle is on the road.
This tool is designed for vehicle architecture teams, software-defined vehicle (SDV) engineers, and digital business units within the automotive sector. It is particularly valuable for fleet management companies looking to optimize maintenance schedules and reduce consumption costs. Because the solutions are deployable across both passenger cars and heavy commercial vehicles, it serves a wide range of industrial applications. Tier 1 equipment suppliers also benefit by adding measurement capabilities to their components without the added cost of physical sensor integration.
What sets COMPREDICT apart is its deep integration of automotive engineering with data science. Founded as a spin-off from TU Darmstadt, the company combines academic rigor with industrial-scale application. Unlike general-purpose AI platforms, it focuses exclusively on chassis and powertrain domains, providing highly specialized models for tire and brake wear. Their partnership with major industry players like the Renault Group and backing from Toyota's Woven Capital underscores the reliability and scalability of their virtual sensor technology in a production environment.
COMPREDICT FAQs
How do virtual sensors work without hardware?
They use machine learning and signal processing to transform existing vehicle signals, like speed and battery voltage, into virtual measurements. This allows the system to estimate physical forces or wear levels without needing a dedicated physical hardware component.
Where can these virtual sensors be deployed?
The technology is highly flexible and can be integrated directly into a vehicle's embedded SDV stack for real-time edge processing. Alternatively, it can be hosted in a cloud environment to provide insights for fleet-wide analysis and aftersales business.
What types of vehicles are supported?
COMPREDICT's solutions are compatible with a wide range of vehicles, including standard passenger cars and heavy-duty commercial trucks. The software is designed to work across different powertrain domains and chassis configurations.
Can these sensors replace physical testing equipment?
Yes, they are specifically designed to replace traditional hardware sensors like wheel force transducers during the development and testing phases. This reduces the complexity of automotive testing and allows series vehicles to become as smart as development vehicles.
