Analyzing and Improving Fault Tolerance of Learning-Based Navigation Systems

Learning-based navigation systems are widely used in autonomous applications, such as robotics, unmanned vehicles and drones. Specialized hardware accelerators have been proposed for high-performance and energy-efficiency for such navigational tasks. However, transient and permanent faults are incre...

Full description

Saved in:
Bibliographic Details
Published in2021 58th ACM/IEEE Design Automation Conference (DAC) pp. 841 - 846
Main Authors Wan, Zishen, Anwar, Aqeel, Hsiao, Yu-Shun, Jia, Tianyu, Reddi, Vijay Janapa, Raychowdhury, Arijit
Format Conference Proceeding
LanguageEnglish
Published IEEE 05.12.2021
Subjects
Data models
Fault tolerance
Fault tolerant systems
Inference algorithms
Navigation
Training
Unmanned vehicles
Online AccessGet full text
DOI10.1109/DAC18074.2021.9586116

Cover

More Information
Summary:Learning-based navigation systems are widely used in autonomous applications, such as robotics, unmanned vehicles and drones. Specialized hardware accelerators have been proposed for high-performance and energy-efficiency for such navigational tasks. However, transient and permanent faults are increasing in hardware systems and can catastrophically violate tasks safety. Meanwhile, traditional redundancy-based protection methods are challenging to deploy on resource-constrained edge applications. In this paper, we experimentally evaluate the resilience of navigation systems with respect to algorithms, fault models and data types from both RL training and inference. We further propose two efficient fault mitigation techniques that achieve 2 \times success rate and 39% quality-of-flight improvement in learning-based navigation systems.
DOI:10.1109/DAC18074.2021.9586116