Search

Automated vehicles are entering the roads and automation is applied to cars, trucks, buses, and even motorcycles today. High automation foresees transitions during driving in both directions. The driver and rider state become a critical parameter since reliable automation allows safe intervention and transit control to the automation when manual driving is not performed safely anymore. When the control transits from automation to manual an appropriate driver state needs to be identified before releasing the automated control. The detection of driver states during manual and automated driving and an appropriate design of the human-machine interaction (HMI) are crucial steps to support these transitions. State‐of‐the‐art systems do not take the driver state, personal preferences, and predictions of road conditions into account. The ADAS&ME project, funded by the H2020 Programme of the European Commission, proposes an innovative and fully adaptive HMI framework, able to support driver/rider state monitoring‐based transitions in automated driving. The HMI framework is applied in the target vehicles: passenger car, truck, bus, and motorcycle, and in seven different use cases.

Context Emerging paradigms like Internet of Things and Smart Cities utilize advanced sensing and communication infrastructures, where heterogeneity is an inherited feature. Applications targeting such environments require adaptability and context-sensitivity to uncertain availability and failures in resources and their ad-hoc networks. Such heterogeneity is often hard to predict, making the deployment process a challenging task. Objective This paper proposes GoalD as a goal-driven framework to support autonomous deployment of heterogeneous computational resources to fulfill requirements, seen as goals, and their correlated components on one hand, and the variability space of the hosting computing and sensing environment on the other hand. Method GoalD comprises an offline and an online stage to fulfill autonomous deployment by leveraging the use of goals. Deployment configuration strategies arise from the variability structure of the Contextual Goal Model as an underlying structure to guide autonomous planning by selecting available as well as suitable resources at runtime. Results We evaluate GoalD on an existing exemplar from the self-adaptive systems community – the Tele Assistance Service provided by Weyns and Calinescu [1]. Furthermore, we evaluate the scalability of GoalD on a repository consisting of 430,500 artifacts. The evaluation results demonstrate the usefulness and scalability of GoalD in planning the deployment of a system with thousands of components in a few milliseconds. Conclusion GoalD is a framework to systematically tackle autonomous deployment in highly heterogeneous computing environments, partially unknown at design-time following a goal-oriented approach to achieve the user goals in a target environment. GoalD has demonstrated itself able to scale for deployment planning dealing with thousands of components in a few milliseconds.

Self-adaptive software-intensive cyber-physical systems (sasiCPS) encounter a high level of run-time uncertainty. State-of-the-art architecture-based self-adaptation approaches assume designing against a fixed set of situations that warrant self-adaptation. As a result, failures may appear when sasiCPS operate in environment conditions they are not specifically designed for. In response, we propose to increase the homeostasis of sasiCPS, i.e., the capacity to maintain an operational state despite run-time uncertainty, by introducing run-time changes to the architecture-based self-adaptation strategies according to environment stimuli. In addition to articulating the main idea of architectural homeostasis, we introduce four mechanisms that reify the idea: (i) collaborative sensing, (ii) faulty component isolation from adaptation, (iii) enhancing mode switching, and (iv) adjusting guards in mode switching. Moreover, our experimental evaluation of the four mechanisms in two different case studies confirms that allowing a complex system to change its self-adaptation strategies helps the system recover from run-time errors and abnormalities and keep it in an operational state.

Driver models help improve and evaluate systems for road crash mitigation and avoidance. As systems develop and address increasingly complex scenarios. Driver models also need to be developed to be able to account for the interactions among these road users. Even as we improve driver modeling with control-theory models and actual data-driven implementations, existing driver models fail to sufficiently take interaction among road users into consideration. This paper addresses this insufficiency by proposing a new operational framework to computationally model interactions among road users. For this purpose, we introduce a definition for interaction among road users. The modeling framework is demonstrated by a specific driving scenario: the overtaking of a cyclist when an oncoming vehicle may be present. In this scenario, modeling driver interaction using Unified modeling language within our framework can lead to improved crash mitigation and avoidance through tailored system activation of automated emergency braking

Smart Cyber--Physical Systems (sCPS) are modern CPS systems that are engineered to seamlessly integrate a large number of computation and physical components; they need to control entities in their environment in a smart and collective way to achieve a high degree of effectiveness and efficiency. At the same time, these systems are supposed to be safe and secure, deal with environment dynamicity and uncertainty, cope with external threats, and optimize their behavior to achieve the best possible outcome. This "smartness" typically stems from highly cooperative behavior, self--awareness, self--adaptation, and selfoptimization. Most of the "smartness" is implemented in software, which makes the software one of the most complex and most critical constituents of sCPS. As the specifics of sCPS render traditional software engineering approaches not directly applicable, new and innovative approaches to software engineering of sCPS need to be sought. This paper reports on the results of the Second International Workshop on Software Engineering for Smart Cyber--Physical Systems (SEsCPS 2016), which specifically focuses on challenges and promising solutions in the area of software engineering for sCPS.

Overtaking cyclists is challenging for drivers because it requires a well-timed, safe interaction between the driver, the cyclist, and the oncoming traffic. Previous research has investigated this manoeuvre in different experimental environments, including naturalistic driving, naturalistic cycling, and simulator studies. These studies highlight the significance of oncoming traffic—but did not extensively examine the influence of the cyclist’s position within the lane. In this study, we performed a test-track experiment to investigate how oncoming traffic and position of the cyclist within the lane influence overtaking. Participants overtook a robot cyclist, which was controlled to ride in two different lateral positions within the lane. At the same time, an oncoming robot vehicle was controlled to meet the participant’s vehicle with either 6 or 9 s time-to-collision. The order of scenarios was randomised over participants. We analysed safety metrics for the four different overtaking phases, reflecting drivers’ safety margins to rear-end, head-on, and side-swipe collisions, in order to investigate the two binary factors: 1) time gap between ego vehicle and oncoming vehicle, and 2) cyclist lateral position. Finally, the effects of these two factors on the safety metrics and the overtaking strategy (either flying or accelerative depending on whether the overtaking happened before or after the oncoming vehicle had passed) were analysed. The results showed that, both when the cyclist rode closer to the centre of the lane and when the time gap to the oncoming vehicle was shorter, safety margins for all potential collisions decreased. Under these conditions, drivers—particularly female drivers—preferred accelerative over flying manoeuvres. Bayesian statistics modelled these results to inform the development of active safety systems that can support drivers in safely overtaking cyclists.

Context: Adaptive systems are expected to cater for various operational contexts by having multiple strategies in achieving their objectives and the logic for\ud matching strategies to an actual context. The prediction of relevant contexts at\ud design time is paramount for dependability. With the current trend on using data\ud mining to support the requirements engineering process, this task of understanding context for adaptive system at design time can benefit from such techniques\ud as well.\ud Objective: The objective is to provide a method to refine the specification of\ud contextual variables and their relation to strategies for dependability. This refinement shall detect dependencies between such variables, priorities in monitoring\ud them, and decide on their relevance in choosing the right strategy in a decision\ud tree.\ud Method: Our requirements-driven approach adopts the contextual goal modelling structure in addition to the operationalization values of sensed information\ud to map contexts to the system’s behaviour. We propose a design time analysis process using a subset of data mining algorithms to extract a list of relevant contexts\ud and their related variables, tasks, and/or goals.\ud Results: We experimentally evaluated our proposal on a Body Sensor Network\ud system (BSN), simulating 12 resources that could lead to a variability space of\ud 4096 possible context conditions. Our approach was able to elicit subtle contexts that would significantly affect the service provided to assisted patients and\ud relations between contexts, assisting the decision on their need, and priority in\ud monitoring.\ud Conclusion: The use of some data mining techniques can mitigate the lack\ud of precise definition of contexts and their relation to system strategies for dependability. Our method is practical and supportive to traditional requirements\ud specification methods, which typically require intense human intervention.

Autonomous vehicles are considered to be the next big thing. Several companies are racing to put self-driving vehicles on the road by 2020. Regulations and standards are not ready for such a change. New technologies, such as the intensive use of machine learning, are bringing new solutions but also opening new challenges. This chapter reports the state of the art, future trends, and challenges of autonomous vehicles, with a special focus on software. One of the major challenges we further elaborate on is using machine learning techniques in order to deal with uncertainties that characterize the environments in which autonomous vehicles will need to operate while guaranteeing safety properties.

Software engineering practice has shifted from the development of products in closed environments towards more open and collaborative efforts. Software development has become significantly interdependent with other systems (e.g. services, apps) and typically takes place within large ecosystems of networked communities of stakeholder organizations. Such software ecosystems promise increased innovation power and support for consumer oriented software services at scale and are characterized by a certain openness of their information flows. While such openness supports project and reputation management, it also brings requirements engineering-related challenges within the ecosystem, such as managing dynamic, emergent contributions from the ecosystem stakeholders, as well as collecting their input while protecting their IP. In this paper we report from a study of requirements communication and management practices within IBM(R)'s Collaborative Lifecycle Management(R) product development ecosystem. Our research used multiple methods for data collection, including interviews within several ecosystem actors, on-site participatory observation, and analysis of online project repositories. We chart and describe the flow of product requirements information through the ecosystem, how the open communication paradigm in software ecosystems provides opportunities for 'just-in-time' RE - and which relies on emergent contributions from the ecosystem stakeholders-, as well as some of the challenges faced when traditional requirements engineering approaches are applied within such an ecosystem. More importantly, we discuss two tradeoffs brought about by the openness in software ecosystems: i) allowing open, transparent communication while keeping intellectual property confidential within the ecosystem, and ii) having the ability to act globally on a long-term strategy while empowering product teams to act locally to answer end-users' context specific needs in a timely manner. A sufficient level of openness facilitates contributions of emergent stakeholders. The ability to include important emergent contributors early in requirements elicitation appears to be a crucial asset in software ecosystems.

Context: Existing knowledge in agile software development suggests that individual competency (e.g. skills) is a critical success factor for agile projects. While assuming that technical skills are important for every kind of software development project, many researchers suggest that non-technical individual skills are especially important in agile software development. Objective: In this paper, we investigate whether non-technical individual skills can predict the use of agile practices. Method: Through creating a set of multiple linear regression models using a total of 113 participants from agile teams in six software development organizations from The Netherlands and Brazil, we analyzed the predictive power of non-technical individual skills in relation to agile practices. Results: The results show that there is surprisingly low power in using non-technical individual skills to predict (i.e. explain variance in) the mature use of agile practices in software development. Conclusions: Therefore, we conclude that looking at non-technical individual skills is not the optimal level of analysis when trying to understand, and explain, the mature use of agile practices in the software development context. We argue that it is more important to focus on the non-technical skills as a team-level capacity instead of assuring that all individuals possess such skills when understanding the use of the agile practices., 18 pages, 1 figure

In the past three editions of RESACS, submitted papers and workshop discussions have proven useful to presenting authors and attending audience alike, resulting in new impulses in shaping this still novel and wide-spread field. For the fourth edition of RESACS, the tradition of RESACS was continued. In a highly interactive workshop the arising challenges in the emergent field of self-adaptive, collaborative, and cyber physical systems were discussed, positioned, and addressed. Thought-provoking and highly constructive discussions among a broader audience and presenters jointly identified properties of these types of system, what their properties entail, which challenges emerge due to these properties, and how traditional, contemporary, and novel approaches might address these challenges for requirements engineering in particular, but also software engineering in general. We hope that the results of the RESACS workshop will spawn research ideas, and standardization of terminology in SAS and CPS research, validate existing research avenues, foster empirical studies, and facilitate collaboration between industry and academia for software engineering as a whole.

The size of a software artifact influences the software quality and impacts the development process. In industry, when software size exceeds certain thresholds, memory errors accumulate and development tools might not be able to cope anymore, resulting in a lengthy program start up times, failing builds, or memory problems at unpredictable times. Thus, foreseeing critical growth in software modules meets a high demand in industrial practice. Predicting the time when the size grows to the level where maintenance is needed prevents unexpected efforts and helps to spot problematic artifacts before they become critical. Although the amount of prediction approaches in literature is vast, it is unclear how well they fit with prerequisites and expectations from practice. In this paper, we perform an industrial case study at an automotive manufacturer to explore applicability and usability of prediction approaches in practice. In a first step, we collect the most relevant prediction approaches from literature, including both, approaches using statistics and machine learning. Furthermore, we elicit expectations towards predictions from practitioners using a survey and stakeholder workshops. At the same time, we measure software size of 48 software artifacts by mining four years of revision history, resulting in 4,547 data points. In the last step, we assess the applicability of state-of-the-art prediction approaches using the collected data by systematically analyzing how well they fulfill the practitioners' expectations. Our main contribution is a comparison of commonly used prediction approaches in a real world industrial setting while considering stakeholder expectations. We show that the approaches provide significantly different results regarding prediction accuracy and that the statistical approaches fit our data best.

The technology in the area of automated vehicles is gaining speed and promises many advantages. However, with the recent introduction of conditionally automated driving, we have also seen accidents. Test protocols for both, conditionally automated (e.g., on highways) and automated vehicles do not exist yet and leave researchers and practitioners with different challenges. For instance, current test procedures do not suffice for fully automated vehicles, which are supposed to be completely in charge for the driving task and have no driver as a back up. This paper presents current challenges of testing the functionality and safety of automated vehicles derived from conducting focus groups and interviews with 26 participants from five countries having a background related to testing automotive safety-related topics.We provide an overview of the state-of-practice of testing active safety features as well as challenges that needs to be addressed in the future to ensure safety for automated vehicles. The major challenges identified through the interviews and focus groups, enriched by literature on this topic are related to 1) virtual testing and simulation, 2) safety, reliability, and quality, 3) sensors and sensor models, 4) required scenario complexity and amount of test cases, and 5) handover of responsibility between the driver and the vehicle., Comment: 8 pages

The amount of software in modern vehicles is constantly growing. However, the risk for functional and quality deficiencies increases simultaneously with size. This results in industry for example in inevitable and unexpected refactorings of software models, which is slowing down development processes in turn. In this industrial case study, we evaluate model growth predictors applied to foresee critical model size developments. We present five approaches and systematically compare them regarding prediction accuracy.