The UK government are committed to bringing forward legislation to allow the safe and secure deployment of self-driving vehicles, as set out in the recent policy paper, “Connected and automated mobility 2025: realising the benefits of self-driving vehicles”. As part of the Connected and Automated Vehicle Process for Assuring Safety and Security (CAVPASS) programme, TRL was commissioned to propose a concept for assuring the safety of Automated Vehicles (AVs) throughout their operational life.
TRL collaborated with ADA Innovation Lab, Reed Mobility Limited, and The Floow which brought together experts in safety assurance, in-vehicle data availability and analysis, real-world application of safety monitoring through trials and testing, and in-depth collision investigation.
In-use monitoring is the use of data to monitor whether desired safety performance is being achieved during the operational lifetime of an AV. This is necessary to continually validate safety throughout operational life, assess compliance with the traffic rules, support the investigation of incidents, identify emerging and previously unconsidered risks, and support continual learning.
The work involved developing technical, procedural and administrative approaches for safety incident identification, investigation and reporting based on an evidence review of current and proposed in-vehicle datasets, safety metrics and collision investigation methodologies and supported by expert judgment. Currently the scope of this framework would apply to Low Speed Automated Vehicles operated commercially but with the intent that this provides the groundwork for In-use monitoring to be applied to all AVs deployed in Great Britain.
This work is of interest to regulators looking to understand how in-use safety assurance may be achieved as well as AV manufacturers and operators looking to understand how data that they are collecting can help to prove the safety of their vehicles to regulators and other stakeholders.
7 reports were produced in this project, each addressing key aspects in building a function in-use safety assurance scheme:
- Road incident taxonomy – establishing a common framework of terminology for road incidents to allow common repeatable analysis of in-use monitoring data
- Minimum dataset specification – a proposal for the minimum set of data required to be collected by AV operators and manufacturers to provide a fundamental dataset to evaluate safety performance.
- In-use monitoring framework – sets out a proposal for the overall framework of an in-use monitoring scheme to support regulatory safety assurance of AVs in Great Britain. This reports also sets out proposals for how AVs can monitor their own compliance with traffic rules.
- Post-incident response framework – discusses how the in-use monitoring framework should coordinate with the proposed Road Safety Investigation Branch to ensure learnings from AV-involved collisions are captured and shared to promote continuous improvement of AV safety.
- Outcome reporting – details the requirements and considerations for aggregating and processing data to allow safety performance data to be evaluated against human driving baselines such as ‘an average human driver’.
- Data privacy – discusses the data privacy issues associated with collecting in-use monitoring data for tracking safety performance and gives recommendations on how they may be addressed.
- Change control – recommendations for managing changes and updates to the proposed in-use monitoring scheme to ensure it reflects best practice and is continually improving the ability to accurately monitor AV safety performance.
This work was presented at the National Highway Traffic Safety Administration’s 27th Enhanced Safety of Vehicles conference in Yokohama, Japan. This paper published as part of the conference proceedings provides a summary of the entire work package.
This work has been funded by the UK Government’s Department for Transport as part of their Connected and Automated Vehicles: Process for Assuring Safety and Security (CAVPASS) programme. The information contained within this work does not necessarily reflect the position of the Department or reflect any future policies or regulations.
