User-centred design for type approval of AD(A)S

Roadmap towards a process audit
Goede, M. de; Jansen, R.J.; Grondelle (TUDelft), E.D. van

Increasing numbers of vehicles (i.e., cars and trucks) are equipped with ADAS (Advanced Driver Assistance Systems) and ADS (Automated Driving Systems). Several risks relating to the use of AD(A)S have been described in the literature, including confusion about driving task responsibility, overreliance on the system, drivers being caught off guard by sudden transitions between vehicle automation levels, and inappropriate levels of attention to the driving context. Well-designed interaction and communication between the driver and the system may reduce these human factors risks However, clear specifications for a system that fulfils such requirements do not exist. This provides a challenge for type approval processes. In the absence of unambiguous system requirements for the interaction between an AD(A)S and its users, evaluating how a system was developed, could provide an alternative way of auditing safe system interaction.

The aim of this assignment was to draw up an advice on whether and how an audit on the application of User-Centred Design processes (UCD) in AD(A)S development regarding the user-system interaction could contribute to the approval of safe AD(A)S. The advice is based on collecting answers to two main research questions:

  1. How is UCD currently implemented in the automotive industry?
  2. How do stakeholders react to the implementation of UCD as part of the type approval process?

First, a literature study was done to establish a preliminary indication as to what a UCD approach for AD(A)S, based on human factors criteria, should consist of. Subsequently seven interviews with seven OEMs and two discussions with the Netherlands Vehicle Authority (RDW) were conducted to investigate the research questions.


Application of UCD in practice

In general, it can  be concluded that whereas the industry considers UCD a valuable approach in AD(A)S development, no indications were found that UCD principles were applied all the time, and – when they were applied – they seemed to focus more on customer satisfaction than on human factors in general.

Support for audit of UCD processes

All OEMs reacted critically to a product-based UCD audit. Two parties specifically questioned the added value. Considering the vast impact, they would like to see proof that applying UCD indeed yields better products before introducing such an audit. There was moderate support for periodic UCD certification on a company/department level, as opposed to a product-related audit. Such an audit is believed to take less time. But, for such an audit to be workable, important conditions were mentioned: the workload should not be increased and the prescribed processes must be beneficial to the own business processes.

RDW considers it a promising approach to place more responsibility on the industry to ensure that AD(A)S are safe to use for all drivers. However, RDW indicates that this requires clear and substantiated specifications of human factors criteria that have to be met.


Substantiate added value

While the value of human factors in this context is undisputed, it is only an unsubstantiated assumption that UCD is an effective implementation method. Since support from the industry will most probably contribute to the quality of the prescribed approaches, it is important to substantiate the added value of a specific approach such as UCD.

Optimally combine flexibility and specificity

Seeing that specifications for safe driver-AD(A)S interactions are lacking and the industry does not want to be restricted, a prescribed approach should be flexible wherever possible. This requires proper consideration of parts that have to be strictly followed and parts that can be approached more flexible. Flexibility is also required since each system, operating under different conditions and in different situations, should fit within the approach.

Focus on human factors guidelines

Concerning the driver-AD(A)S interaction, it seems that more attention should be paid to human factors guidelines. Based on the interviews, it seems that the industry mainly focuses on customer demands, expectations and experiences, i.e., only a part of the user experience that represents compliance to human factors requirements.


Gepubliceerd door
SWOV, Den Haag


Dit is een publicatie van SWOV, of waar SWOV een bijdrage aan heeft geleverd.