analiza aiqbench
This research advances the safety of Deep Neural Network-controlled robots by developing an explainability-guided method to generate robustness tests that effectively transfer from simulation to physical hardware. By leveraging Integrated Gradients to focus on influential image regions, the study demonstrates a significant improvement in identifying adversarial failures across different operational observations. The findings highlight that while simulation can efficiently prioritize transferable threats, physical validation remains essential for ensuring real-world safety in cyber-physical systems.
Generalizable Robustness Testing of DNN-Based Robotic Navigation Systems via XAI-Guided Search
arXiv:2610.06862v1 Announce Type: new Abstract: **Context:** Deep Neural Networks (DNNs) increasingly control Cyber-Physical Systems (CPSs), yet small input perturbations can cause unsafe system-level behavior. Existing approaches often optimize perturbations for individual images and evaluate them only in simulation, limiting their generalizability and practical validity. **Objectives:** This work aims to generate robustness tests that remain effective across operational observations and to evaluate whether the resulting failures transfer from simulation to a physical robot. **Methods:** We propose an explainability-guided multi-objective evolutionary approach that generates sparse perturbations over representative images selected through visual and behavioral clustering. Aggregated Integrated Gradients guide mutations toward influential image regions. We evaluate the approach on a DNN-controlled LeoRover in Gazebo, conduct an ablation study, and validate a stratified subset of perturbations on the physical robot. **Results:** The approach achieved a median success rate of 70.0%, compared with 53.85% for unguided search, and increased median hypervolume from 0.65 to 0.73. Multi-image optimization improved the success rate from 50.0% to 57.5%, while XAI guidance further increased it to 70.0%. In the sim-to-real evaluation, simulation achieved 0.95 precision and 0.67 recall, and simulated and physical failure times showed a significant positive correlation of 0.617. **Conclusion:** Combining multi-image optimization with explainability-guided search improves robustness testing for DNN-controlled robotic systems. Simulation effectively identifies and prioritizes transferable failures, but physical validation remains necessary because some real-world failures are not reproduced in simulation.
Abstract:**Context:** Deep Neural Networks (DNNs) increasingly control Cyber-Physical Systems (CPSs), yet small input perturbations can cause unsafe system-level behavior. Existing approaches often optimize perturbations for individual images and evaluate them only in simulation, limiting their generalizability and practical validity. **Objectives:** This work aims to generate robustness tests that remain effective across operational observations and to evaluate whether the resulting failures transfer from simulation to a physical robot. **Methods:** We propose an explainability-guided multi-objective evolutionary approach that generates sparse perturbations over representative images selected through visual and behavioral clustering. Aggregated Integrated Gradients guide mutations toward influential image regions. We evaluate the approach on a DNN-controlled LeoRover in Gazebo, conduct an ablation study, and validate a stratified subset of perturbations on the physical robot. **Results:** The approach achieved a median success rate of 70.0%, compared with 53.85% for unguided search, and increased median hypervolume from 0.65 to 0.73. Multi-image optimization improved the success rate from 50.0% to 57.5%, while XAI guidance further increased it to 70.0%. In the sim-to-real evaluation, simulation achieved 0.95 precision and 0.67 recall, and simulated and physical failure times showed a significant positive correlation of 0.617. **Conclusion:** Combining multi-image optimization with explainability-guided search improves robustness testing for DNN-controlled robotic systems. Simulation effectively identifies and prioritizes transferable failures, but physical validation remains necessary because some real-world failures are not reproduced in simulation.
Robotics (cs.RO)
arXiv:2610.06862 [cs.RO]
(or arXiv:2610.06862v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.06862 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 314 cuvinte
