BMW Builds Advanced Simulation Center in Munich

The virtual test-drives will be conducted not only by the development engineers but also by customers on a regular basis.

German auto-giant BMW announced construction of a facility for the simulation of real-world driving situations in the north of its headquarters-city Munich. One of the central features of the facility is the high-fidelity simulator, in which longitudinal, transverse and rotational movements of a vehicle can be represented simultaneously and, hopefully, more realistically. At the heart of the center are two driving simulators designed to meet requirements for testing complex automated-driving systems.

The high-dynamic simulator, able to generate longitudinal and transverse acceleration forces of up to 1.0 g, is used to test new systems and functions by replicating highly dynamic evading maneuvers, full braking and hard acceleration. The systems to be tested are fitted in a vehicle mock-up attached to a platform inside the dome of the driving simulator. Mounted on an electromechanical hexapod system, the dome can be moved both longitudinally and transversely by an electric drive while it’s also being turned. In order to give the drivers a realistic visual experience of the simulated driving situation, the dome housing the mock-up is equipped with a projection screen. The virtual test-drive scenario is completed by a sound simulation matched to the situation portrayed.

The automotive industry is reinventing itself, owing to the rise of interest in autonomous vehicles, and simulation is playing a critical role in their development. In order for autonomous vehicles to operate safely in the real world, they must be able to adapt to a multitude of changing conditions such as navigating curves, crossing bridges and climbing hills. Before advanced driver assistance systems (ADAS) or fully self-driving vehicles hit the road, they undergo a lengthy period of testing where the vehicle’s sensors and artificial intelligence are tested in a variety of simulated real-world environments.

Carmakers across the globe are betting big on simulation technology. Autonomous Intelligent Driving (AID), a subsidiary of German auto-giant Audi, announced in June that it was partnering with Cognata, an Israeli tech firm specializing in virtual tests for autonomous vehicle software. Cognata’s platform employs artificial intelligence, deep learning and computer vision within a realistic simulation environment in order to judge and validate AVs prior to physical roadway tests. The company uses patented computer vision and deep-learning algorithms to automatically generate a whole-city simulator that includes buildings, roads, lane marks, traffic signs and even trees and bushes.

That same month, Toyota’s autonomous-vehicle division in Silicon Valley invested $100,000 in CARLA, an open source simulation project, to take advantage of the collaborative-development potential of an open-source community. The project, administered by the Computer Vision Center (CVC) in Barcelona, counts Intel as its founding sponsor. CARLA includes open source code and protocols, as well as digital assets like virtual buildings, vehicles, traffic lights, trees and pedestrians that users can integrate into simulated environments.

In May, simulation specialist Ansys announced the acquisition of Optis, a French optical simulation and virtual-prototyping company. Using the photo-realistic virtual reality and closed-loop simulation platform Optis has developed in combination with other Ansys platforms, automakers will be able simulate the environment driverless vehicles are navigating, including road conditions, weather and one-way streets.

Meanwhile, Microsoft’s Project Road Runner team has been using photo-realistic simulation and deep learning to train autonomous-driving algorithms. The team gathers data and trains AI platforms through real-world simulation. Microsoft’s AirSim, an open-source simulation platform for drone research, proved particularly useful to the Road Runner team. Its photo-realistic environments, built using the Unreal Engine, are ideal for training autonomous-driving algorithms.

— Nathan Eddy is a filmmaker and freelance journalist based in Berlin. Follow him on Twitter.