Requirements engineering for autonomous vehicles

Ribeiro, Quelita A. D. S.; Ribeiro, Moniky; Castro, Jaelson

doi:10.1145/3477314.3507004

Cited by 10 publications

(5 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research has also looked specifically at RE for AD, e.g., providing an overview of AD RE techniques [19], Riberio et al identified AD RE challenges addressed by the literature, and identified the languages and description styles used to describe AD requirements, with special attention given to NFRs [17]. Heyn et al investigated challenges with context and ODD definition in ML-enabled perception systems [11], including a lack of standardisation for context definitions, ambiguities in deriving ODDs, missing documentation, and lack of involvement of function developers while defining the context.…”

Section: Related Workmentioning

confidence: 99%

Requirements Engineering for Automotive Perception Systems: an Interview Study

Habibullah¹,

Heyn²,

Gay³

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: Driving automation systems (DAS), including autonomous driving and advanced driver assistance, are an important safety-critical domain. DAS often incorporate perceptions systems that use machine learning (ML) to analyze the vehicle environment. Aims: We explore new or differing requirements engineering (RE) topics and challenges that practitioners experience in this domain. Method: We have conducted an interview study with 19 participants across five companies and performed thematic analysis. Results: Practitioners have difficulty specifying upfront requirements, and often rely on scenarios and operational design domains (ODDs) as RE artifacts. Challenges relate to ODD detection and ODD exit detection, realistic scenarios, edge case specification, breaking down requirements, traceability, creating specifications for data and annotations, and quantifying quality requirements. Conclusions: Our findings contribute to understanding how RE is practiced for DAS perception systems and the collected challenges can drive future research for DAS and other ML-enabled systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Requirements Engineering for Automotive Perception Systems: an Interview Study

Habibullah¹,

Heyn²,

Gay³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In the development of Autonomous Vehicles (AV), vehicles can execute a broad range of tasks without human intervention or partial intervention, such as controlling the car's speed and switching lanes [51]. There has yet to be a consensus about the complete autonomy of these vehicles, as some researchers have proposed strategies for controlling their level of automation according to their location (e.g., highway, commercial street, residential street), application concerns (e.g., safety, security, improvement in fuel economy); or even to the driving style (e.g., aggressive, normal, calm).…”

Section: A Scenario A: Automated Vehiclesmentioning

confidence: 99%

“…Ribeiro et al [51] present a literature review about the requirements involved in the development of AVs, identifying different types of autonomous vehicles with varying levels of autonomy. Based on this literature review, we identified and classified the factors that can make AVs assume different degrees of autonomy.…”

Section: A Scenario A: Automated Vehiclesmentioning

confidence: 99%

Identifying Factors That Impact Levels of Automation in Autonomous Systems

et al. 2023

View full text Add to dashboard Cite

The need to support complex human and machine collaboration has increased because of recent advances in the use of software and artificial intelligence approaches across various application domains. Building applications with more autonomy has grown dramatically as modern system development capability has significantly improved. However, understanding how to assign duties between humans and machines still needs improvement, and there is a need for better approaches to apportion these tasks. Current methods do not make adaptive automation easy, as task assignments during system operation need to take knowledge about the optimal level of automation (LOA) into account during the collaboration. There is currently a lack of explicit knowledge regarding the factors that influence the variability of human-system interaction and the correct LOA. Additionally, models have not been provided to represent the adaptive LOA variation based on these parameters and their interactions and interdependencies. The study, presented in this paper, based on an extensive literature review, identifies and classifies the factors that affect the degree of automation in autonomous systems. It also proposes a model based on feature diagrams representing the factors and their relationships with LOAs. With the support of two illustrative examples, we demonstrate how to apply these factors and how they relate to one another. This work advances research in the design of autonomous systems by offering an adaptive automation approach that can suggest levels of automation to facilitate human-computer interactions.INDEX TERMS Software engineering, levels of automation, adaptive system, autonomous systems, software design.

show abstract

“…Among various AV safety requirements used in the literature [35,44,45], considering the capability of CARLA and the major functionality of our target MLC (i.e., the object detection module), we focus on the following safety requirement: "AV should have a distance no less than 𝑑 𝑚𝑖𝑛 from the vehicle in front." To detect safety violations, if any, during the simulation of a complete solution (i.e., the combination of a scenario and an MLC behavior), we measure the distance between the ego vehicle and the vehicle in front for each simulation time step.…”

Section: Evaluation Subjectsmentioning

confidence: 99%