Making the Case for Safety of Machine Learning in Highly Automated Driving

Burton, Simon; Gauerhof, Lydia; Heinzemann, Christian

doi:10.1007/978-3-319-66284-8_1

Cited by 117 publications

(94 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These elements may vary between the training and target execution environments leading to the trained function becoming dependent on hidden features of the training environment not relevant in the target system. Previous work by the authors as well as others have introduced concepts of applying assurance case structures to arguing the performance of an MLM within a safety-critical context [8], [17], [22]. Figure 1 describes a generic assurance case pattern for arguing the safety properties of a machine learning function (derived from the description in [23] using GSN [2]).…”

Section: Safety Case Patterns For Machine Learningmentioning

confidence: 99%

“…A contract-based approach to specifying safety properties of the MLM was proposed in [8], by which the MLM is specified as a component within its sys-tem context and defined by a set of assumptions on its operating environments under which certain safety guarantees (for example formulated as benchmark performance requirements) must hold. These performance requirements could include definitions of accuracy and failure rates to be achieved by the function.…”

Section: Safety Case Patterns For Machine Learningmentioning

confidence: 99%

See 1 more Smart Citation

Confidence Arguments for Evidence of Performance in Machine Learning for Highly Automated Driving Functions

Burton

Gauerhof

Sethy

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Due to their ability to efficiently process unstructured and highly dimensional input data, machine learning algorithms are being applied to perception tasks for highly automated driving functions. The consequences of failures and insufficiencies in such algorithms are severe and a convincing assurance case that the algorithms meet certain safety requirements is therefore required. However, the task of demonstrating the performance of such algorithms is non-trivial, and as yet, no consensus has formed regarding an appropriate set of verification measures. This paper provides a framework for reasoning about the contribution of performance evidence to the assurance case for machine learning in an automated driving context and applies the evaluation criteria to a pedestrian recognition case study.

show abstract

Section: Safety Case Patterns For Machine Learningmentioning

confidence: 99%

Section: Safety Case Patterns For Machine Learningmentioning

confidence: 99%

Confidence Arguments for Evidence of Performance in Machine Learning for Highly Automated Driving Functions

Burton

Gauerhof

Sethy

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…In Burton, Gauerhof, and Heinzemann (), an example is mapped on the above requirements with respect to a deep learning component. The problem space for the component is pedestrian detection with CNNs.…”

Section: Safety Of Deep Learning In Autonomous Drivingmentioning

confidence: 99%

“…In Burton, Gauerhof, and Heinzemann (2017), an example is mapped on the above requirements with respect to a deep learning where ASIL D represents the highest and ASIL A the lowest risk. If an element is assigned to quality management (QM), it does not require safety management.…”

Section: Safety Of Deep Learning In Autonomous Drivingmentioning

confidence: 99%

A survey of deep learning techniques for autonomous driving

Grigorescu

Trasnea

Cocias

et al. 2019

Journal of Field Robotics

1,102

495

View full text Add to dashboard Cite

The last decade witnessed increasingly rapid progress in self‐driving vehicle technology, mainly backed up by advances in the area of deep learning and artificial intelligence (AI). The objective of this paper is to survey the current state‐of‐the‐art on deep learning technologies used in autonomous driving. We start by presenting AI‐based self‐driving architectures, convolutional and recurrent neural networks, as well as the deep reinforcement learning paradigm. These methodologies form a base for the surveyed driving scene perception, path planning, behavior arbitration, and motion control algorithms. We investigate both the modular perception‐planning‐action pipeline, where each module is built using deep learning methods, as well as End2End systems, which directly map sensory information to steering commands. Additionally, we tackle current challenges encountered in designing AI architectures for autonomous driving, such as their safety, training data sources, and computational hardware. The comparison presented in this survey helps gain insight into the strengths and limitations of deep learning and AI approaches for autonomous driving and assist with design choices.

show abstract

“…As the success stories of AI/ML mount and AI/ML is increasingly used as an integral part of more and more real-word systems (e.g., for medical diagnosis, decision making in the legal system), the in-depth evaluation of newly developed learning models has become a top priority. For example, while in areas such as computer vision, researchers have began evaluating their learning algorithms with an eye towards bias or fairness [13,19], in the area of autonomous vehicles/driving, the learning models are increasingly being evaluated with respect to their safety and robustness [14,25]. In the networking area, the researchers' efforts to evaluate their learning models are already jeopardized by the amount of time they have to spend on running experiments and collecting data.…”

Section: Key Impedimentsmentioning

confidence: 99%