ReSonAte: A Runtime Risk Assessment Framework for Autonomous Systems

Charles Hartsell, Shreyas Ramakrishna, Abhishek Dubey, Daniel Stojcsics, Nag Mahadevan, Gabor Karsai

16th International Symposium on Software Engineering for Adaptive and Self-Managing Systems, SEAMS 2021 2021

PDF

Why This Matters

Autonomous cyber-physical systems must operate in unpredictable real-world environments where design-time assumptions may not hold. Traditional static risk assessment approaches are insufficient for handling the dynamic hazards and state-dependent failure modes that emerge at runtime. This work is innovative because it bridges design-time safety analysis with runtime operations, enabling systems to dynamically adjust their risk assessments based on current conditions and maintain safe operation in the face of uncertainties.

What We Did

The ReSONAte framework presents a runtime risk assessment methodology for autonomous cyber-physical systems that handles dynamic uncertainties in operating environments. The framework uses design-time hazard analysis information combined with system state observations at runtime to dynamically estimate risk using Bow-Tie Diagram models. The approach extends traditional safety assurance techniques by incorporating runtime monitoring data and state-dependent risk calculations to maintain system safety during continuous operation.

Key Results

The ReSONAte framework successfully demonstrates dynamic risk estimation on autonomous vehicle examples using Carla simulations with 600 executions. The approach shows that risk calculations can be performed with minimal computational overhead (0.3 milliseconds) while accurately tracking state-dependent hazard rates. The framework proves effective for handling uncertainty in system operations and provides practical mechanisms for autonomous systems to support self-adaptation based on dynamically computed risk values.

Full Abstract

Cite This Paper

@inproceedings{resonate2021,
  author = {Hartsell, Charles and Ramakrishna, Shreyas and Dubey, Abhishek and Stojcsics, Daniel and Mahadevan, Nag and Karsai, Gabor},
  booktitle = {16th {International} Symposium on Software Engineering for Adaptive and Self-Managing Systems, {SEAMS} 2021},
  title = {ReSonAte: A Runtime Risk Assessment Framework for Autonomous Systems},
  year = {2021},
  abstract = {Autonomous Cyber-Physical Systems (CPSs) are often required to handle uncertainties and self-manage the system operation in response to problems and increasing risk in the operating paradigm. This risk may arise due to distribution shifts, environmental context, or failure of software or hardware components. Traditional techniques for risk assessment focus on design-time techniques such as hazard analysis, risk reduction, and assurance cases among others. However, these static, design time techniques do not consider the dynamic contexts and failures the systems face at runtime. We hypothesize that this requires a dynamic assurance approach that computes the likelihood of unsafe conditions or system failures considering the safety requirements, assumptions made at design time, past failures in a given operating context, and the likelihood of system component failures. We introduce the ReSonAte dynamic risk estimation framework for autonomous systems. ReSonAte reasons over Bow-Tie Diagrams (BTDs), which capture information about hazard propagation paths and control strategies. Our innovation is the extension of the BTD formalism with attributes for modeling the conditional relationships with the state of the system and environment. We also describe a technique for estimating these conditional relationships and equations for estimating risk-based on the state of the system and environment. To help with this process, we provide a scenario modeling procedure that can use the prior distributions of the scenes and threat conditions to generate the data required for estimating the conditional relationships. To improve scalability and reduce the amount of data required, this process considers each control strategy in isolation and composes several single-variate distributions into one complete multi-variate distribution for the control strategy in question. Lastly, we describe the effectiveness of our approach using two separate autonomous system simulations: CARLA and an unmanned underwater vehicle. },
  category = {selectiveconference},
  contribution = {colab},
  acceptance = {30},
  project = {cps-middleware,cps-reliability},
  tag = {ai4cps},
  keywords = {autonomous systems, runtime risk assessment, cyber-physical systems, safety assurance, hazard analysis, Bow-Tie Diagrams, dynamic uncertainty}
}
Quick Info
Year 2021
Keywords
autonomous systems runtime risk assessment cyber-physical systems safety assurance hazard analysis Bow-Tie Diagrams dynamic uncertainty
Research Areas
CPS Explainable AI scalable AI
Search Tags

ReSonAte, Runtime, Risk, Assessment, Framework, Autonomous, Systems, autonomous systems, runtime risk assessment, cyber-physical systems, safety assurance, hazard analysis, Bow-Tie Diagrams, dynamic uncertainty, CPS, Explainable AI, scalable AI, 2021, Hartsell, Ramakrishna, Dubey, Stojcsics, Mahadevan, Karsai