Interpretable Detection of Distribution Shifts in Learning Enabled Cyber-Physical Systems

Yang, Yahan; Kaur, Ramneet; Dutta, Souradeep; Lee, Insup

doi:10.1109/iccps54341.2022.00027

Cited by 9 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The role of the myopic action selector is to decide whether to switch from the performant controller to the safety controller, i.e., the forward switch. A number of varied works have proposed different potential methods to trigger the decision logic, e.g., safety verification methods such as reachability analysis, which computes a set of states reachable within some number of time steps and then checks if this reachable set contains states outside the system's safe region [29]; and OOD detection methods, which detect the data that are not similar to the data used for training [37].…”

Section: Myopic Action Selectormentioning

confidence: 99%

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Luo¹,

Ramakrishna²,

Pettet³

et al. 2023

Preprint

View full text Add to dashboard Cite

Learning Enabled Components (LEC) have greatly assisted cyberphysical systems in achieving higher levels of autonomy. However, LEC's susceptibility to dynamic and uncertain operating conditions is a critical challenge for the safety of these systems. Redundant controller architectures have been widely adopted for safety assurance in such contexts. These architectures augment LEC "performant" controllers that are difficult to verify with "safety" controllers and the decision logic to switch between them. While these architectures ensure safety, we point out two limitations. First, they are trained offline to learn a conservative policy of always selecting a controller that maintains the system's safety, which limits the system's adaptability to dynamic and non-stationary environments. Second, they do not support reverse switching from the safety controller to the performant controller, even when the threat to safety is no longer present. To address these limitations, we propose a dynamic simplex strategy with an online controller switching logic that allows two-way switching. We consider switching as a sequential decision-making problem and model it as a semi-Markov decision process. We leverage a combination of a myopic selector using surrogate models (for the forward switch) and a non-myopic planner (for the reverse switch) to balance safety and performance. We evaluate this approach using an autonomous vehicle case study in the CARLA simulator using different driving conditions, locations, and component failures. We show that the proposed approach results in fewer collisions and higher performance than state-ofthe-art alternatives.

show abstract

Section: Myopic Action Selectormentioning

confidence: 99%

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Luo¹,

Ramakrishna²,

Pettet³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The study in [95] trains a generative model and evaluates the likelihood of OOD inputs under that model at run time. The work in [96] detects OOD using kernel density estimation. Distance-based methods use distance metrics to detect OOD.…”

Section: Related Workmentioning

confidence: 99%

Improving security of autonomous cyber-physical systems against adversarial examples

Song¹

View full text Add to dashboard Cite

show abstract

Automating Weak Label Generation for Data Programming with Clinicians in the Loop

Park,

Pugh,

Sridhar

et al. 2024

2024 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE)

View full text Add to dashboard Cite

Interpretable Detection of Distribution Shifts in Learning Enabled Cyber-Physical Systems

Cited by 9 publications

References 14 publications

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Improving security of autonomous cyber-physical systems against adversarial examples

Automating Weak Label Generation for Data Programming with Clinicians in the Loop

Contact Info

Product

Resources

About