Deploying safe autonomous systems is extraordinarily challenging.
We have gathered the world's best safety experts with the right tools and techniques to help you deploy with confidence.
Safer Perception for Autonomy
ECR is piloting Hologram to make machine learning safe for important applications from advanced driver assistance systems to air taxis, from manufacturing to face recognition. Hologram is a platform under development that will help find and help you fix edge cases in perception algorithms that trip up machine learning. Hologram makes testing far more efficient by turning every mile driven by a real vehicle into millions of potential scenarios in an effort to root out the “unknown unknowns” as quickly and safely as possible.
reliable software for autonomy
No matter how many tests you run on a prototype or small-scale deployment, the real world has a way of presenting scenarios that you can't imagine or can't test for. Switchboard helps you find problems before your software is fielded. Switchboard is a scalable, cloud-based automated robustness testing tool that finds ways in which complex software is unsafe or unreliable.
Autonomous system safety accelerated
Autonomy poses unique challenges to conventional safety approaches, and we know because ECR has helped clients plan and execute a number of system safety programs in defense, automotive, and other applications. These programs depend on the ability to iterate and converge on a successful safety concept. Informed by our decades of expertise in both safety and autonomy, MISSA was developed to more efficiently and completely identify & model safety risks posed by complex systems.