LogoAutomotive Agile PEP Europe|November 16 – 17, 2020, Online
10:00 AM - 10:30 AM

Automated vehicles require a reliable and safe view of the current driving situation to make safe decisions about future maneuvers. From an algorithmic point of view, an environmental model combines data of different sensors and sensor modalities to provide a unified representation of the environment. At the same time, the ISO 26262 provides guidelines for the functional safety of road vehicles, including guidelines for their development. The presentation will give an overview on the typical approaches for functional safety of environmental models on an architectural level.

Key Take aways:

  • Decomposition into diverse path offers one way to address the required integrity of an environmental model
  • The approach taken on the decomposition of an autonomous system depends on its operational design domain
  • Combining machine learning with probabilistic approaches increase diversity of the environmental model of an automated vehicle
  • Dynamic Grid algorithms can be used as a unified probabilistic representation for environmental models
Testing + Validation

Eric Richter

Director Customer Relations

Dr. Eric Richter is the Director of Customer Relations and co-founder at BASELABS. With his strong market perspective, he brings in new requirements for the development of BASELABS´ products such as data fusion for embedded and series devices. He holds a PHD in the field of data fusion for automated driving.

BASELABS provides software products for sensor fusion for automated driving. Its embedded library for production vehicles helps the customer to efficiently develop and adapt data fusion systems for advanced driver assistance systems (ADAS). It allows the user to focus on the development of the driving function and minimizes risks and cost in the development projects. Our customers are leading OEM, tier1, and tier2 suppliers. BASELABS offers interesting job opportunities in software and data fusion application development to continuously grow our team of currently about 40 team members.