Towards the Verification of Safety-critical Autonomous Systems in Dynamic Environments

Autonomous systems are becoming pervasive in everyday life, and many of these systems are complex and safety-critical. Formal verification is important for providing performance and safety guarantees for these systems. In particular, Hamilton–Jacobi (HJ) reachability is a formal verification tool for nonlinear and hybrid systems; however, it is computationally intractable for analyzing complex systems, and computational burden is in general a difficult challenge in formal verification. In this review, we begin by briefly presenting background on reachability analysis with an emphasis on the HJ formulation. We then present recent work showing how high-dimensional reachability verification can be made more tractable by focusing on two areas of development: system decomposition for general nonlinear systems, and traffic protocols for unmanned airspace management. By tackling the curse of dimensionality, tractable verification of practical systems is becoming a reality, paving the way for more pervasive and safer automation.

Download Full-text

Quadrotor navigation in dynamic environments with deep reinforcement learning

Assembly Automation ◽

10.1108/aa-11-2020-0183 ◽

2021 ◽

Vol ahead-of-print (ahead-of-print) ◽

Author(s):

Jinbao Fang ◽

Qiyu Sun ◽

Yukun Chen ◽

Yang Tang

Keyword(s):

Reinforcement Learning ◽

Short Term Memory ◽

Autonomous Systems ◽

Dynamic Environments ◽

Training Data ◽

Content Type ◽

Distributed Training ◽

Spatiotemporal Information ◽

Cloud Robotics ◽

Experience Replay

Purpose This work aims to combine the cloud robotics technologies with deep reinforcement learning to build a distributed training architecture and accelerate the learning procedure of autonomous systems. Especially, a distributed training architecture for navigating unmanned aerial vehicles (UAVs) in complicated dynamic environments is proposed. Design/methodology/approach This study proposes a distributed training architecture named experience-sharing learner-worker (ESLW) for deep reinforcement learning to navigate UAVs in dynamic environments, which is inspired by cloud-based techniques. With the ESLW architecture, multiple worker nodes operating in different environments can generate training data in parallel, and then the learner node trains a policy through the training data collected by the worker nodes. Besides, this study proposes an extended experience replay (EER) strategy to ensure the method can be applied to experience sequences to improve training efficiency. To learn more about dynamic environments, convolutional long short-term memory (ConvLSTM) modules are adopted to extract spatiotemporal information from training sequences. Findings Experimental results demonstrate that the ESLW architecture and the EER strategy accelerate the convergence speed and the ConvLSTM modules specialize in extract sequential information when navigating UAVs in dynamic environments. Originality/value Inspired by the cloud robotics technologies, this study proposes a distributed ESLW architecture for navigating UAVs in dynamic environments. Besides, the EER strategy is proposed to speed up training processes of experience sequences, and the ConvLSTM modules are added to networks to make full use of the sequential experiences.

Download Full-text

Challenges in Assessing Safety Critical Autonomous Systems

The Proceedings of Mechanical Engineering Congress Japan ◽

10.1299/jsmemecj.2017.j1810204 ◽

2017 ◽

Vol 2017 (0) ◽

pp. J1810204

Author(s):

Joachim IDEN

Keyword(s):

Autonomous Systems ◽

Safety Critical

Download Full-text

AI and Reliability Trends in Safety-Critical Autonomous Systems on Ground and Air

2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W) ◽

10.1109/dsn-w50199.2020.00024 ◽

2020 ◽

Cited By ~ 1

Author(s):

Jyotika Athavale ◽

Andrea Baldovin ◽

Ralf Graefe ◽

Michael Paulitsch ◽

Rafael Rosales

Keyword(s):

Autonomous Systems ◽

Safety Critical

Download Full-text

Autoencoder-Based Semantic Novelty Detection: Towards Dependable AI-Based Systems

Applied Sciences ◽

10.3390/app11219881 ◽

2021 ◽

Vol 11 (21) ◽

pp. 9881

Author(s):

Andreas Rausch ◽

Azarmidokht Motamedi Sedeh ◽

Meng Zhang

Keyword(s):

Artificial Intelligence ◽

System Development ◽

Novelty Detection ◽

Autonomous Systems ◽

Training Data ◽

Safety Measure ◽

False Negatives ◽

Safety Critical ◽

Error Calculation

Many autonomous systems, such as driverless taxis, perform safety-critical functions. Autonomous systems employ artificial intelligence (AI) techniques, specifically for environmental perception. Engineers cannot completely test or formally verify AI-based autonomous systems. The accuracy of AI-based systems depends on the quality of training data. Thus, novelty detection, that is, identifying data that differ in some respect from the data used for training, becomes a safety measure for system development and operation. In this study, we propose a new architecture for autoencoder-based semantic novelty detection with two innovations: architectural guidelines for a semantic autoencoder topology and a semantic error calculation as novelty criteria. We demonstrate that such a semantic novelty detection outperforms autoencoder-based novelty detection approaches known from the literature by minimizing false negatives.

Download Full-text