EnergyShield: Provably-Safe Offloading of Neural Network Controllers for Energy Efficiency
Not on the map. Why?
Abstract (arXiv)
To mitigate the high energy demand of Neural Network (NN) based Autonomous Driving Systems (ADSs), we consider the problem of offloading NN controllers from the ADS to nearby edge-computing infrastructure, but in such a way that formal vehicle safety properties are guaranteed. In particular, we propose the EnergyShield framework, which repurposes a controller ''shield'' as a low-power runtime safety monitor for the ADS vehicle. Specifically, the shield in EnergyShield provides not only safety interventions but also a formal, state-based quantification of the tolerable edge response time before vehicle safety is compromised. Using EnergyShield, an ADS can then save energy by wirelessly offloading NN computations to edge computers, while still maintaining a formal guarantee of safety until it receives a response (on-vehicle hardware provides a just-in-time fail safe). To validate the benefits of EnergyShield, we implemented and tested it in the Carla simulation environment. Our results show that EnergyShield maintains safe vehicle operation while providing significant energy savings compared to on-vehicle NN evaluation: from 24% to 54% less energy across a range of wireless conditions and edge delays.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 24 of its 44 references that are in the map. First posted on arXiv 13 Feb 2023.
Builds on
24 of 44 references.
References · 24
- 2017CARLA: An Open Urban Driving Simulatorcited by 2,449
- 2017Safe Reinforcement Learning via Shieldingcited by 449
- 2017Safe Model-based Reinforcement Learning with Stability Guaranteescited by 366
- 2018A Lyapunov-based Approach to Safe Reinforcement Learningcited by 297
- 2019Efficient and Accurate Estimation of Lipschitz Constants for Deep Neural Networkscited by 293
- 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 289
- 2019Algorithms for Verifying Deep Neural Networkscited by 221
- 2017A General Safety Framework for Learning-Based Control in Uncertain Robotic Systemscited by 163
- 2022Safe Control with Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction methodscited by 151
- 2019Lyapunov-based Safe Policy Optimization for Continuous Controlcited by 145
- 2018Learning-based Model Predictive Control for Safe Explorationcited by 120
- 2020Learning Control Barrier Functions from Expert Demonstrationscited by 92
12 more
- 2018Neural Lander: Stable Drone Landing Control using Learned Dynamicscited by 85
- 2018Verisig: verifying safety properties of hybrid systems with neural network controllerscited by 84
- 2018Formal Verification of Neural Network Controlled Autonomous Systemscited by 41
- 2017Safe Learning of Quadrotor Dynamics Using Barrier Certificatescited by 35
- 2020Synthesis of Control Barrier Functions Using a Supervised Machine Learning Approachcited by 35
- 2016Robustness of Control Barrier Functions for Safety Critical Controlcited by 30
- 2021Rule-based Optimal Control for Autonomous Drivingcited by 21
- 2020ShieldNN: A Provably Safe NN Filter for Unsafe NN Controllerscited by 17
- 2021SAGE: A Split-Architecture Methodology for Efficient End-to-End Autonomous Vehicle Controlcited by 8
- 2018Reachable Set Estimation and Verification for Neural Network Models of Nonlinear Dynamic Systemscited by 7
- 2020A Control Barrier Perspective on Episodic Learning via Projection-to-State Safetycited by 7
- 2022Romanus: Robust Task Offloading in Modular Multi-Sensor Autonomous Driving Systemscited by 3
Built on by
Citing papers per year: 2023: 1.
1 paper.