University of Warwick: China welcomes world’s largest scenario database for autonomous vehicle safety

CATARC-ADC is China’s principal research and technical organisation for the automotive industry, and is at the cutting edge of the country’s innovation and regulation for connected and autonomous vehicles (CAV). Meanwhile, CATARC-ADC actively participates in ISO, ASAM and UNECE and other international organizations, with international influence.

This significant new major collaboration, launched today (9 September 2021), means that Safety Pool™ Scenario Databases will enable road simulations including use cases from China, broadening the scope of this global safety platform – thus supporting the growing CAV industry, informing emerging regulatory policies, and enhancing the safety of millions more drivers across China and beyond.

“Safety of Autonomous Vehicles needs to be a collaborative mission. No one organisation or country can achieve this mission on their own. With this in mind we created Safety PoolTM Scenario Database to enable global collaboration on scenario sharing.

We are delighted that CATARC-ADC have joined Safety PoolTM Scenario Database which reinforces our mission of international collaboration on CAV safety. With a diverse set of scenarios, the database caters to a large number of autonomous vehicle applications, many of which will be relevant for our stakeholders in China.”

Ego vehicle (the red vehicle) is following an accelerating agent vehicle (white car) with a safe distance on a straight road in a residential area, and the sun rises in front of ego vehicle.Nicola Croce, Technical Program Manager, Deepen AI, said:

“Safety Pool™ has all the ingredients to be the reference platform and initiative for AV safety assurance worldwide. What’s very unique about it is its global scope, the incentive-based mechanisms engineered to attract and provide value to every different industry stakeholder, and the deep engagement with regulators, everything based on a common foundation of data sharing.

We are excited to welcome CATARC-ADC to the Safety Pool initiative. CATARC-ADC is the major player in scenario-based testing and scenario databases in China. CATARC-ADC’s entry into Safety Pool™ provides a key stepping stone in international collaborations in the scenario-based testing landscape of AVs, and a major leap forward to help companies improve their adaptability in China-oriented testing for ADS.”

Since the launch of this pioneering project in March 2021, WMG at the University of Warwick and Deepen AI have collaborated with stakeholders around the world: to date, over 200 organisations have enrolled in the Safety PoolTM Scenario Database

Bolin Zhou, Global Business General Manager, CATARC – ADC said:

“As a founding member of Safety PoolTM Scenario Database in China and the leading third-party company for ADS validation in China, Automotive Data of China will use the great opportunity of Safety PoolTM Scenario Database to tackle the global autonomous vehicle safety issues with its own strength. Safety PoolTM Scenario Database is a crucial, open platform that aligns well with ADC positioning in China and around the world. Through Safety PoolTM Scenario Database, a global safety tool, China will continue to provide data and tool services for automated driving system validation”

Tim Dawkins, Global Impact Strategy Lead, World Economic Forum said:

“Initiatives like Safety Pool are key to making safe autonomous vehicles a reality – we should not be making safety a competitive advantage. This shared scenario library will allow developers to learn from one another’s datasets to increase the robustness of their systems through exposure to a diverse scenario set. CATARC’s support for Safety Pool represents a vital commitment to a level playing field for the development of autonomous vehicles in the name of safety”

Richard Morris, Innovation Lead – Autonomous & Connected Vehicles, Innovate UK said:

“Innovate UK is glad to have supported the creation and the development of the Safety PoolTM Scenario Database. We would also like to encourage more organisations and countries to contribute scenario content. Scenarios kept in private siloes will not help the mass acceptance of vehicle automation. We all need to share safety knowledge and make best practice widely available. The more comprehensive the Safety PoolTM Scenario Database becomes, the more useful it is for any developer wanting to deploy CAVs anywhere around the world.”

The database provides a diverse set of scenarios in different operational design domains (ODDs i.e. operating conditions) that can be leveraged by governments, industry and academia alike to test and benchmark Automated Driving Systems (ADSs) and use insights to inform policy and regulatory guidelines.

The scenarios have been generated using a novel hybrid methodology developed by WMG, at the University of Warwick, using both knowledge-based and data-based approaches. The Safety Pool™ Scenario Database allows organisations to create scenarios in their own libraries, collaborate with other organisations via both shared and public libraries and enable the public to submit challenging real-world scenarios.

Enabling scenarios to be matched to specific environments and operating conditions means that trials and tests can be undertaken in the simulated environment, controlled test facilities and on public roads, with evidence from each environment being used to inform our understanding of safe behaviours, bringing Autonomous Vehicles closer to market at pace.

It has been suggested that in order for CAV to be safe for the average driver, they will need to be tested on 11 billion miles of road – an insurmountable goal in the physical world. The Safety Pool™ scenario based virtual simulations not only offer the necessary quantity of testing, but also the complexity and quality of countless ‘real-world’ road scenarios.

The development of the Safety Pool™ Scenario Database was funded by UK’s Centre for Connected & Autonomous Vehicles (CCAV), Innovate UK and Zenzic funded Midlands Future Mobility project led by WMG, University of Warwick.