Regaining Sitational Awareness as a User in Charge: Responding to transition demands in automated vehicles

January2025 Department for TransportGreat Minster House33 Horseferry RoadLondonSW1P 4DRenquiries@ccav.gov.uk Control sheet This report has been prepared by Lacuna Agency, a strategic insights agency servingclients in public sector, automotive, luxury, and sport. We combine scientific techniqueswith evidence-based practices to design innovative research for hard-to-reach or difficult-to-study audiences. Our work also extends to the broader mobility and transport sectors,where we assist governments, manufacturers and network providers in planning for futureneeds. Executive Summary Background to the Project In 2022, the Law Commission of England and Wales and the Scottish Law Commission (TheLaw Commissions) published a report suggesting a substantial overhaul of the legalframework governing automated vehicles. This joint report conducted a comprehensivereview of the regulatory structure for automated vehicles on public roads and highways andintroduced a novel legal entity known as the User-in-Charge (UiC), an individual situatedwithin the vehicle and capable of operating the driving controls while a self-driving feature isengaged but not a driver (Law Commissions, 2022a). The UiC does not have to monitor theroad while the self-driving system is active. This project explores the transition from self-driving technology to human-operated driving,particularly focusing on the role of the UiC, who must be ready to take control when theautomatedsystem issues a takeover request.A takeover request occurs when theautomated driving system encounters emergencies or conditions outside its programming.The UiC receives the takeover request via visual, auditory, or haptic cues. The first approved system under these guidelines is the Automated Lane Keeping System(ALKS), which requires the UiC to take over within 10 seconds of a takeover request. Asself-drivingtechnologies become more prevalent,questions arise regarding thepermissibility of engaging in Non-Driving Related Activities (NDRA) while using vehicles withactivated self-driving features such as ALKS. There is potential for certain NDRAs to beallowed, provided they do not compromise the driver's ability to resume control of the vehiclesafely when a takeover request is issued by the system. Upon receiving a takeover request, the UiC must suddenly focus on building SituationalAwareness (SA) from the surrounding road environment to enable a safe takeover. SA iscritical for safe driving, including during the transition from automated to manual control. SAinvolves three levels: perception of the environment, comprehension of the situation, andprojection of future events. The time required to gain sufficient SA after a takeover requestis crucial, with studies showing response times ranging from 3 to 20 seconds. While simpletasks, like basic perceptual awareness, are processed quickly, more complex activities thatinvolvehigher levels of situational awareness—such as interpreting road signs oranticipating how current events will impact the road ahead—require more time. Project Aims This project focused on exploring the implications of NDRAs in vehicles with self-drivingcapabilities, when a UiC must respond to a transition demand. As the use of ALKS becomesmore prevalent, it is essential to understand which NDRAs can be performed (within the 10seconds mandated by ALKS regulations) without compromising the ability to safely resumecontrol of the vehicle. The project had the following objectives: •Investigate which non-driving related activities (NDRAs), if any, can be safelyperformed during non-driving periods in cars with self-driving features but requiring aUiC.•Establish mechanisms for measuring SA and determine appropriate thresholds toensure safe takeover and resumption of manual driving.•Understand potential variations in the impact of NDRAs across different scenarios toinform policy development. Toachieve these objectives,Lacuna Agency worked in partnership with UniversityCollege London (UCL) and Loughborough University (LU), combining our expertise inHuman Factors research and driving simulation. The project used a simulator-basedapproach to closely replicate real-life driving conditions. This collaboration, together with DfTandCCAV aimed to provide evidence-based insights that will inform future policydevelopment and contribute to safer implementation of automated driving technologies. Method Theresearch was conducted using high-fidelity driving simulators at two locations:University College London (UCL) and Loughborough University (LU), with 97 participantsrepresenting the general UK driving population. The study used a within-participants designwitheight trials per participant,involving two motorway scenarios—roadworks andcongestion—designed to simulate conditions that exceed the operational limits of the ALKSand trigger planned takeover requests due to speed changes: 1.Roadworks Scenario: The ego vehicle (which the participant controls in the drivingsi