Virtual environments offer significant advantages for autonomous vehicle development and will be the arena where a majority of testing and validation occur. Complex simulations deliver results faster than real-time testing, with easily repeatable test cases and safe replication of hazardous scenarios. To drive even 20 percent better than a human requires 11 billion miles of validation. That translates to more than 500 years of nonstop driving in the real world with a fleet of 100 cars — an impossible task.
Metamoto connects developers with those advantages through its cloud-based simulation platform, which can be tailored for a wide variety of research and development objectives.
Metamoto’s services include designer scenario creation with multi-sensor simulations across a broad spectrum of environmental parameters, as well as a direction of tests and results analysis. Synthetic annotated sensor data created from test scenarios also can be used to train perception and planning machine learning algorithms.
Take your simulations further by pairing with VectorZero RoadRunner
Pair with Atlatec mapping for real-world simulations
Check out the AutonomouStuff simulation suite for help deciding what is best for your simulation needs.
- Train, test, debug and validate your automated vehicle software
- Designer — scenario creation, script AV exercises, parameterized everything
- Director — cloud simulation, scalable, on demand
- Analyzer — simulation replay, result analysis, AV debugging
- Synthetic data
- Scalability — run millions of tests in a single cycle, drive billions of virtual test miles
- Multi-sensor simulation —precise, unified sensor simulation: lidar, radar, camera, ultrasound, GPS, IMU and others; responsive across diverse material properties
- Parameterization — run tests across a spectrum of environmental and hardware parameters and unique edge cases; identify isolated outcomes, performance boundaries and system tolerances