Siemens has launched a startup venture for simulating machine learning systems in autonomous vehicles and industrial automation
Simulytic is headed by CEO Andrea Kollmorgen, who was previously vice president and head of connected eMobility at Siemens and Andy Gill, the former director of connected mobility.
The startup is focused on accelerating autonomous mobility deployment at scale with reliable synthetic data. Simulytic aims to use simulation tools and digital twins to create insight into the impact and safety of autonomous driving.
Simulytic, based in Munich, is already applying Siemens’ experience in the simulation of complex, automated systems and in the use of artificial intelligence in safety-critical applications. Siemens already builds digital twins of vehicles with simulation through the Pave 360 tools and Siemens EDA technology.
This enables the venture to make competitive, comprehensive and independent assessments of incident probabilities, changing traffic flows and congestion patterns, the effects of weather and road conditions, and many other localized factors.
It is building an ecosystem of libraries of simulated driving data in digital replicas of deployment areas that can then be augmented by driving activity in the real world. Using digital tools to understand the consequences is the only way to get around the challenge and equip companies such as insurers with the data necessary to understand a complex, dynamic risk environment, says the company.
The starting point is not likely to be comparing risk profiles of individual AV companies, but rather in identifying important risk factors for different operational domains. This starts with a digital twin of the geographical deployment area that is created from a 3D model built from high definition maps.
The model is populated with a representative traffic mix of cars, pedestrians, cyclists together with the actual infrastructure elements that are present such as traffic signs, traffic lights and road markings. The local behaviours of the traffic and vulnerable road users are used to create their respective behavioural models. All static and dynamic elements in this digital twin can be varied according to the actual variety seen in the location, including velocities, the extent to which traffic rules are obeyed, the quality of the signs and markings, the weather conditions. Meaningful edge cases and complex scenarios for the type of deployment location are also generated.
Next: Fleets of driverless cars in a digital twin
In these highly detailed environmental models, fleet of driverless cars can be implemented, with these vehicles finding their way from A to B. Simulation makes it possible to run these A to B trajectories under a massive variation of conditions. But even more valuable is the capability to actively search for the corner cases, the conditions in which the AV will fail or cause an issue for the non-AV actors,
This data can equip the insurer in discussions with city governments and AV providers to demand specific infrastructure or vehicle design improvements for the local environment.
The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
The technical storage or access that is used exclusively for statistical purposes.The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.