The committee has developed a preliminary set of observations about cyber-physical systems (CPS) education based on the presentations and briefing summarized in Chapter 1 as well as from its own expertise and deliberations thus far. The committee will continue its information gathering and deliberations and will issue its final report providing its findings and recommendations later in 2015.
- • CPS skills and experience are in demand. The anecdotal reports from industry representatives point to two areas of demand. First, as CPS becomes an increasingly important aspect of products and services, some companies are looking specifically for people with an educational concentration in CPS—who some dubbed the “CPS engineer.” Second, CPS skills are becoming increasingly important across a wide range of engineering specialties. Although it is difficult to quantify the demand, more than one presenter observed that his firm was hampered in developing new products because it was not able to secure sufficient CPS talent. Others noted difficulties in hiring people with the CPS skills they were looking for; one response has been for some organizations to focus on using on-the-job training to develop additional CPS skills internally.
- • A lack of familiarity with the term “cyber-physical systems” itself may impede student interest in the field. Although cyber-physical systems have been widely used and designed for some time now, perspective students may not perceive CPS as an attractive area of study and work in part because they do not recognize the label. By contrast, students are often
eager to enroll in courses with “robotics” or the Internet of things in their title. Both of these overlap with CPS, although they may not cover all of the foundations needed for CPS. Also, large information technology firms like Facebook and Google have captured the imagination of students in a way that many CPS-intensive firms have not, even though the latter provide significant opportunities to innovate and have considerable real-world impact.
- • “CPS engineering”1may be emerging as a distinct field of engineering. CPS draws knowledge and approaches from multiple areas of engineering and may indeed have significant overlap with other areas of engineering; however, CPS has begun to take on a distinctive character.
Given that CPS is multidisciplinary and draws from many areas, it indeed has significant overlap with those areas. However, CPS is distinctive due to several qualities. For example, embedded systems often concentrate on low-cost, simple devices or large and complex systems embedded in closed or very controlled environments. Today, most new embedded systems are in environments that are open via connections (giving us the “cyber” in CPS) to the wireless world and the Internet. Systems engineering also contributes heavily to CPS. However, systems engineering typically concentrates on the organization, management, and integration required for large systems but does not deeply address the detailed technological needs that arise in combining the physical with the cyber. Several examples of CPS, such as autonomous vehicles, could be considered “robotics.” However, classical robots do not necessarily draw on the CPS principles needed for autonomous vehicles: real-time, safety-critical, large-scale, wireless communication environments and operation in unconstrained environments.
- • There is growing agreement on the core elements of CPS, but a diverse set of approaches to fashioning CPS programs is likely and appropriate. It seems possible to outline a set of core concepts, principles, and themes, and the committee will be endeavoring to use this to lay out a model undergraduate curriculum in CPS for its final report. That said, CPS is new and its applications are continuing to emerge and evolve, suggesting that an even higher degree of variation among approaches will be found than already exists across engineering and computer science programs at different institutions. Multiple pathways toward CPS-focused programs are also likely. Programs have already grown organically from computer science and electrical engineering programs; these programs will continue to
1 The committee uses the terms “CPS engineering” and “CPS engineer” to mean a set of skills and knowledge needed to design and build a CPS, and a person with those skills; the terms are not limited to a set of credentials or to someone who has a degree or certification in CPS.
grow organically and follow somewhat distinct trajectories. At the same time, engineering programs in areas such as such civil, mechanical, and aerospace have been placing increasing emphasis on CPS—a trend that seems likely to continue and spread.
- • The need for CPS skills is becoming pervasive across engineering and computer science. CPS are deployed in a variety of domains, including civil, mechanical, and aerospace. CPS engineers will not replace the need for engineering that focuses on deep understanding of topics specific to these fields, but people with such skills will increasingly be central to engineering teams. The other engineers and computer scientists on those teams will increasingly need at least some CPS knowledge and skills.
- • It will likely not be sufficient to simply bundle existing courses to create a CPS program. Although many of the topics in CPS can also be found in current engineering and computer science courses, the emphasis on the interaction of the cyber and the physical is unique to CPS. For example, control theory classes in electrical engineering typically focus on electromagnetic principles such as those for circuits. A CPS control theory course would need to better incorporate topics related to networks, human-in-the-loop models, security, software, and real-time and hybrid control. Similarly, a computer science department’s software design course tends to employ non-physical world applications; a CPS course would need to emphasize physical constraints.
- • CPS education programs need to include a hands-on component. These opportunities can be provided to students via an interdisciplinary capstone course, extensive course-specific laboratories, an engineering elective course focused on a single project, or internships and industry-academic partnerships.
- • Other paths to CPS knowledge will be important. Although this study will focus on 4-year undergraduate curricula, other paths for students and the workforce to gain CPS knowledge are also important. These paths include minors or certificates and graduate programs within other disciplines, opportunities for community and vocational schools and postgraduate professional studies, the use of online education or massive open online courses, and the role of K-12 education in preparing students for a range of careers involving CPS.
The committee also identified several challenges in creating and supporting cyber-physical systems programs at universities. The committee will explore possible solutions in its final report.
- • Working across disciplinary and department boundaries can create significant challenges. CPS exists at the intersection of multiple disciplines. This has significant impact on how a curriculum will be developed and
taught. Organically developed CPS activities have generally grown out of electrical engineering or computer science departments; however, to develop the best students, a degree program would need to incorporate both of these departments, plus inclusion of domain-specific departments such as aerospace or civil engineering.
- • A significant challenge will be identifying, recruiting, and educating appropriate faculty members to teach new and different courses. Many universities currently have a limited handful of faculty who can teach CPS courses. If schools are to develop and offer a full CPS curriculum, additional faculty members will be needed. Additionally, the interdisciplinary nature of CPS can create challenges for faculty in securing research funding and obtaining tenure. Alternative paths, such as hiring non-tenure-track professors from industry, may need to be explored.
- • Textbooks, curricular materials, and laboratory facilities will need to be developed. Developing these critical resources requires both time and financial support.
This page intentionally left blank.