Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - L3Pilot (Piloting Automated Driving on European Roads)

Teaser

Automated driving (AD) technology has matured to a level motivating one of the final road tests to answer the key questions before market introduction. These newly-attained levels of maturity will ensure an appropriate assessment of the impact of AD, both on the vehicle and...

Summary

Automated driving (AD) technology has matured to a level motivating one of the final road tests to answer the key questions before market introduction. These newly-attained levels of maturity will ensure an appropriate assessment of the impact of AD, both on the vehicle and for the traffic in general, how vehicle security can be ensured, and eventually, making the first evaluation on societal impacts and emerging business models.
Consequently, the overall objective of the L3Pilot project is to test and study the viability of automated driving as a safe and efficient means of transportation, explore and promote new service concepts to provide mobility for all. This objective will be met through more specific targets as follows:

1. Coordinate activities and create a functional methodology to acquire the required data with high quality and efficiency.
2. Pilot, test and evaluate automated driving functions and connected automation.
3. Innovate and promote AD for wider awareness and market introduction.
4. Provide a comprehensive guideline with best practices for the development of automated driving functions (ADF) as a „Code of Practice for Automated Driving” (CoP).

The project objectives are well in line with current policies and trends at European level. In particular, the procedures set up need to be be coordinated and carried out in the similar way on each test area. Furthermore, the piloting needs to generate data that enables us to make the first data-led guesses on the impacts of AD on different levels from a single driver reactions to possible societal level implications. The project provides the first pan-European public data and results on the way automated driving would influence on driving, mobility and safety of transport across the continent.

Work performed

The first 18 months of the project have been focused on setting the piloting methodology in place and preparing a fleet of test vehicles.
There are no commonly agreed methods yet to study automated driving functions (ADF) and to make the test results comparable across test sites and manufacturers. There is also no common understanding of the requirements which new ADFs need to fulfil to be considered to be safe, reliable and certified. L3Pilot generates the expected common testing methodology, based on considerations regarding the questions to be answered, the data to be acquired, and their processing.
The development of a common methodogy was not a straightforward process just to draw up a testing plan but first, to get different cultures in intelligent vehicles development to “speak the same language” and then commit to the common procedures and implement it across the test areas and sites. Another methodology challenge was the data management, so how to collect the data and process it all the way to the analysis phase, with a view to industrial interests and to the needs of the research community at the same time.
The AD fleet has been built and is in place. After the pre-tests the piloting started in March 2019 and is under way. One of the project’s challenges was to take the test vehicles on public roads. This presupposes both internal approval inside companies and the permission from the road authorities. The safety of testing has the main priority in piloting, and several constraints to prepare the vehicles to public traffic have been taken into account.
Today, the data from piloting is accumulating, and the preparations for the data processing and analysis have been initiated with the data chain tested and validated.

Final results

L3Pilot is focusing on close-to-market evaluations, rather than enabling technologies. However, extensive piloting will give insight into the improvement of on-board equipment such as sensor systems, positioning, cyber-security, 5G communication in supporting environment perception and HMI. An important area where valuable information is accumulated regards Human Factors, as for instance acceptance and user preferences for different automated functions. This will help in developing credible business cases and understanding of the viability of AD functions. Driver behaviour data will provide valuable new information on key topics such as the ability to control AD systems, transitions between levels of automation, and shared control. The pilot tests will also investigate how the driver can be supported in the event of reducing errors and dealing with emergency situations.
In this context, the knowledge to be acquired in L3Pilot is critical for the development of a comprehensive Code of Practice (CoP) to establish guidelines for development and validation of applications at various levels of automation. The CoP is a key activity for the partners to support a reliable process for product development. This can only be obtained through a large-scale testing of real user experience in field conditions. The project will be the first of its kind, globally, to deliver such a critical step forward in AD implementation.
At this stage of the project until 18 months of duration, the L3Pilot developed:
• A coherent methodology for large scale AD testing.
• A fleet of vehicles for data acquisition.
• An overall method on how to manage data all the way from logging to processing up to a comprehensive analysis by function.
• A framework including principles and plans for the Code of Practice.

Website & more info

More info: https://www.l3pilot.eu/.