Abstract
The automotive domain has witnessed a tremendous growth in the amount of software deployed in the cars. The car no longer contains mechanical components only but more and more functionality is controlled by embedded systems instead. Due to this, the field is a multi-disciplinary one involving engineers from mechanical, electrical, electronics and software disciplines. However, this is now changing as the problems being addressed in the domain are becoming more complex. Problems are now societal such as smart cities or green mobility. To solve such complex problems requires going beyond multi-disciplinarity and towards trans-disciplinarity. This involves including stakeholders that represents the users and also the societal interests. This paper discusses opportunities and challenges for trans-disciplinarity in the automotive domain.
1. Introduction
In [8], multi-disciplinarity is characterized by research involving complex problems that require researchers from several disciplines to coordinate to solve the problem. In multi-disciplinary research, theories and models from the different fields are brought together and each field only solves a piece of the bigger problem. The different disciplines merely complement each other usually in a sequential or juxtaposing manner, but no new knowledge is formed between them [8,10]. Trans-disciplinarity on the other hand, goes beyond just several disciplines collaborating to solve a problem. With trans-disciplinary research the focus is to synthesize and unify knowledge from the different disciplines in order to come up with solutions for complex problems [11]. The models and theories from the different disciplines can be altered and improved since the major aim of the research is to solve a practical problem.
Over the past 20 years in the automotive domain, there has been a rise in the amount of embedded systems that are deployed in the car [2]. An embedded system is a system whose critical function is not computational but which is controlled with by a computer embedded in it [12]. A current modern car contains approximately 10 million lines of code in total distributed over about 100 Electronic Control Units (ECUs) [3], which are the physical devices hosting the software part of embedded systems. Examples of embedded systems in a car are adaptive cruise control (which automatically controls braking and acceleration of the car), brake assist system, airbag control systems and navigation systems.
The development of such embedded systems has made the automotive domain multi-disciplinary in nature due to the different disciplines that are involved in making these systems. For example, a typical embedded system needs to be worked on by mechanical engineers, electronic engineers, electrical and software engineers. All these engineers bring their expertise to create the final system. This kind of development is known as Systems Engineering [13].
Recently, this paradigm of embedded systems in the automotive domain is shifting towards more complex problems for example autonomous driving. This means that using only engineering knowledge from the different disciplines is no longer enough and this field needs more appropriate research methods. Since trans-disciplinary research focuses more on problem solving rather than specific field enhancements, it is a better suite for the complex problems the automotive industry is now tackling [5]. In this paper, we will point on the opportunities for trans-disciplinary research and the challenges in the automotive industry.
The rest of the paper is structured as follows. Section 2 gives a brief overview of the evolution from multi-disciplinarity to trans-disciplinarity, Section 3 discusses the opportunities and Section 4 discusses the challenges for trans-disciplinarity. The paper ends with Section 5 which gives a brief reflection and conclusion.
2. From Multi-disciplinarity to Trans-disciplinarity: Bigger and More Complex Problems
In their review of the past 30 years of autonomous driving, the researchers Klaus et al. [1] came up with a time line of how the automotive industry has evolved in terms of the systems being developed and also how the goals for these systems have evolved. The authors show that the focus is moving from purely technical goals such as vehicle dynamic stabilization which is more on how to improve the car itself technically to broader goals that involve the society such as automated and cooperative driving. This change in focus also coincides with the experience report on systems engineering in the automotive domain in [13]. This shows that the problems that are now being tacked with both industry and academia in the field of autonomous driving are leaving the bound of multi-disciplinary and becoming more trans-disciplinary. For example to be able to have an autonomous vehicle on the street, several things such as safety, ethics, legal issues and even urban planning need to be taken into account. This can only be achieved if researchers and practitioners from these different disciplines work together.
3. Opportunities for Trans-disciplinary research
This section outlines opportunities for both researchers and industry practitioners brought about by trans-disciplinarity.
3.1 Research Funding Opportunities
The rise of autonomous driving brings along some opportunities for researchers as well. For example there are now opportunities to apply for grants that support trans-disciplinary research [9]. Such grants are focused on solving societal problems more than scientific advancement of particular disciplines. For instance one societal problem is environmental pollution. For such a problem, autonomous vehicles can be seen as a solution to minimize waste caused by having so many cars produced and used. A more real example can be seen in the European Union Framework Program for Research and Innovation from 2014 to 2020 in which they have set aside funding to investigate three societal problems which are: 1. Smart, Green \& Integrated Transport,
2. Secure, Clean \& Efficient Energy and 3. Climate Action, Environment, Resource Efficiency and Raw Materials [6].
3.2 Opportunity to Show Applicability of Research Results
Trans-disciplinary research can also be seen as an arena for researchers to show case the application of their research in different context. This is because for research focused on advancement of a particular discipline, the research usually ends at a proof of concept phase where theoretically or in very restricted environments, the advancements such as models and concepts are proved to work. Working in a trans-disciplinary research project is however a different story as researchers need to apply their knowledge to a complex problem that is also practical and in a practical environment. This facilitates more learning and synthesis of knowledge from various contexts.
4. Challenges of Trans-disciplinary research
This section describes the challenges of trans-disciplinary research in general and explains how they apply to projects in the automotive domain.
4.1 Coordinating large projects
Trans-disciplinary research typically involves very large projects [4]. This is because the project involves people from separate disciplines. Coordinating such projects is more complex as it requires the skills to be able to manage people from the different disciplines. It is also important to make sure that everyone in the project feels that they are important and needed and most of all feel that they are contributing to the final goal of the project. Depending on the project there might also be several goals and it is difficult for all the stakeholders to keep in mind the final goal. As in every large project, it is likely that the objectives of the different stakeholders might be conflicting. For instance for autonomous driving from the context of software engineers efficiency of the program might be an objective that is of importance but for safety engineers this may not be the case. Also research in mono-disciplinary projects typically has traditions that differ among disciplines. Bringing them together may lead to the conflicts between different traditions of conducting a project, presenting the results etc.
4.2 Communication problems
When different disciplines need to coordinate, communication problems usually arise. This is because the different disciplines have their own vocabulary that may mean something else in other disciplines. For trans-disciplinary research project this is also challenge. Researchers suggest that before the beginning of the project the different stakeholders involved need to meet and discuss the objectives of the project. Also during the course of the project regular contact in form of meetings or workshops is advised in order to always ensure that you are all on the same page. It is important to communicate the expected outcomes as early as possible. Expectations of all the different disciplines should also be put to the table early on.
4.3 Applied research vs. Fundamental research
Fundamental research or basic research refers to the type of research where scientists act out of curiosity and gather knowledge for the sake of knowledge without a specific problem in mind that this knowledge will solve [7]. On the other hand applied research takes a more practical approach where scientists seek knowledge that will solve a particular problem.
As mentioned before, what differentiates trans-disciplinary research from other types of research is that it is mainly aimed at solving an identified practical problem. For research this usually translates to mostly applied research rather than fundamental or basic research.
This means that in the end the contribution is not a new advancement in any particular field but rather unified knowledge gained and a practical problem solved. While publishing such results is not a problem as many publication venues accept such papers, for researchers, it can be hard to have an identity of belonging and contributing to a particular discipline (something which is expected in academia). This can also be reflected in how the universities hire researchers based on needs of a certain discipline.
5. Conclusion
Based on the discussion of the challenges and opportunities for trans-disciplinary research in the automotive domain, there are two ways in which trans-disciplinarity can advance in the automotive domain. One is that using trans-disciplinary research, the automotive industry as a whole can be advanced by solving more complex problems. It is therefore an opportunity for advancement that would not have been possible using mono-disciplined research approaches. On the other hand, applying the different research methods suggested to be suitable for trans-disciplinary research gives an opportunity for such methods to be used and evaluated in the automotive industry. This does not only benefit researchers, but also practitioners and the society in the end.
References
1. K. Bengler, K. Dietmayer, B. Farber, M. Maurer, C. Stiller, and H. Winner.Three decades of driver assistance systems: Review and future perspectives. IEEE Intelligent Transportation Systems Magazine, 6(4):6–22,2014
