In August, Silicon Valley startup Metamoto will launch its new simulation service. The offering includes the three following components:
Director – A scalable, parameterized simulator enabling tests to run across a spectrum of environmental and hardware parameters and unique edge cases.
Designer – A first-of-its-kind tool to custom build parameterized scenarios via a visual editor. Scenarios define the training and testing behaviors of traffic, pedestrians, sensor configurations, etc, within a virtual scene.
Analyzer – A comprehensive analysis tool to assess how a vehicle performed in specific simulations. This tool is often referenced as an autonomous systems debugger.
“Simulation offers an immediate means to train and test the performance of automated systems in a risk-free environment, allowing auto makers to accelerate to their autonomous vehicle production timelines,” said Chad Partridge, CEO of Metamoto. “Metamoto’s offering is designed to give every player in the mobility space – not just the largest – a tool to test their automated vehicle in a virtual world, ensuring safe deployment in the real world.”
The solution provides the ability to run millions of tests in a single cycle to identify isolated outcomes, performance boundaries and system tolerances.
Simulations are executable across a spectrum of environmental (weather, traffic, road conditions, pedestrians, etc) and hardware (vehicle properties, sensor placement, latency, sensor settings, etc) parameters.
On-demand simulation is inserted into the AV product lifecycle with tools like Jenkins and Jira, enabling seamless regression testing, agile workflows, and version control, every time vehicle software, sensors and/or infrastructure change.