Collision typology and autonomous vehicle testing
Accident typology schemes describe the pre-crash manoeuvres of vehicles. No two typologies are the same and none is perfect. Typology is one of many variables defining future CAV collisions and there is a need to map out all the variables.
This work has created a translation between the different collision typologies (RAIDS, STATS19, FNoL, iRAP and FARS), and also created a ’tree‘ of variables that define an AV collision (considering vehicle, human and environmental factors).
By helping to inform an understanding of AV collisions, we can better define the effects, test standards and facilities required to do both testing and forensics. By working towards standards for AVs, we help improve compatibility, communication and harmonisation, all of which contribute to a Vision Zero goal for road traffic collisions.
Determinants of unsafe behaviour in autonomous vehicles
Following an expert workshop held with The University of Southampton, a number of key driving scenarios have been identified as likely to create safety risks, if different levels of automation are applied to the driving task. These scenarios will then underpin a detailed analysis which will inform an analysis of potential safety hazards, system and human errors and violations. This is a collaborative project with a range of experts in human factors, ergonomics, and system approach research techniques.
This work has implications for the development of new technologies and their mitigation measures as we transition from partially automated vehicles to fully automated vehicles.
Autonomous Vehicle Freight
The use of CAV for freight has many benefits including safety, increased business efficiency, reduced congestion and lower emissions. To ensure the UK is prepared for the deployment of AV freight, government investment is required to verify the robustness of the technology and identify what infrastructure must be in place, to calculate operational cost savings and how AV freight can optimise road space.
The Academy has so far published two papers:
- Future Transport Visions: Preparing highways for Connected & Autonomous Vehicles PPR864
And the 2018 Symposium topic was all about the Future of Freight in the UK.
Integrating connected and autonomous vehicles into the road network
In conjunction with the University of Southampton, this project is investigating how road infrastructure can sustainably support high numbers of automated vehicles on the network. The existing infrastructure is designed for non-automated vehicles. The impact ranges from the road space to the thousands of traffic sensors, the control infrastructure, refuelling infrastructure and integration between modes.
This project is investigating strategies to redesign the infrastructure, using a combination of newly developed control inputs and models, modelling of scenarios and various experiments with simulators.
