Helping you assure the safety of your autonomous system
Whatever domain you work in, your autonomous system must be safe. Those who use it, design new features for it, regulate it or might be impacted by it need confidence that the system will behave safely and as expected.
On this site, you will find expert, impartial guidance to help you to create a credible and compelling assurance case for your autonomous system.
The guidance is written by the Assuring Autonomy International Programme at the University of York and is peer-reviewed by industry experts. The guidance will cover the core technical issues that must be considered for the safe development and introduction of an autonomous system.
Our methodology for the Assurance of Machine Learning for use in Autonomous Systems (AMLAS) is the first of its kind and can help you to justify the safety of your machine learnt components.
Despite the interest from government and industry, there is currently no established process for certifying and assuring an AI component. AMLAS fills this gap by providing us with a framework to integrate safety assurance into our development process and build a compelling argument for our safety case. We are very much looking forward to trialling the AMLAS process in our upcoming flight controller development project.
Co-Founder and Chief Executive Officer, Luffy AI
AMLAS has made an important contribution to safety case patterns and provided valuable guidance on how to proceed to justify the safety of machine learning components.Research Engineer, Robert Bosch GmbH
The AMLAS methodology really helped to shape our thinking and approach to the assurance of ML/AI. The lifecycle structure mirrors our development phases and it is well presented and easy to read. AMLAS helped me to structure my thinking about strategies for verification and validation and the examples given are useful.Systems/Software Engineer, Critical Systems Labs Inc
Introducing our safety assurance methodologies
We have created a series of processes which enables the creation of a safety assurance case for both autonomous systems, and machine learning autonomous systems. These are AMLAS (Assurance of Machine Learning for us in Autonomous Systems) and SACE (Safety Assurance of autonomous systems in Complex Environments).
If you are new to these processes, download our AMLAS and SACE factsheets to learn how and why your organisation can benefit from our guidance.
AMLAS Factsheet
SACE Factsheet
Safe machine learning
We have developed the first methodology for the Assurance of Machine Learning for use in Autonomous Systems (AMLAS).
AMLAS has six stages, which complement the machine learning (ML) development process. It incorporates a set of safety case patterns and a process for systematically integrating safety assurance into your development of ML components.
Browse AMLAS
Future guidance
Our guidance provides you with proven, accessible methodologies and processes for assuring the safety of your autonomous system.
We are writing five essential pieces of guidance that will be published on this website and free to use. These will help you to create a credible and compelling assurance case for your autonomous system.
Future guidance
