Search Menu
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information
Search Scholarly Publications

Title: Reconfigurable Framework for Resilient Semantic Segmentation for Space Applications

Journal Article · · ACM Transactions on Reconfigurable Technology and Systems
DOI: https://doi.org/10.1145/3472770 · OSTI ID:1983003
 [1];  [1];  [2]
  1. Univ. of Pittsburgh, PA (United States)
  2. NASA Goddard Space Flight Center (GSFC), Greenbelt, MD (United States)
+ Show Author Affiliations

Deep learning (DL) presents new opportunities for enabling spacecraft autonomy, onboard analysis, and intelligent applications for space missions. However, DL applications are computationally intensive and often infeasible to deploy on radiation-hardened (rad-hard) processors, which traditionally harness a fraction of the computational capability of their commercial-off-the-shelf counterparts. Commercial FPGAs and system-on-chips present numerous architectural advantages and provide the computation capabilities to enable onboard DL applications; however, these devices are highly susceptible to radiation-induced single-event effects (SEEs) that can degrade the dependability of DL applications. In this article, we propose Reconfigurable ConvNet (RECON), a reconfigurable acceleration framework for dependable, high-performance semantic segmentation for space applications. In RECON, we propose both selective and adaptive approaches to enable efficient SEE mitigation. In our selective approach, control-flow parts are selectively protected by triple-modular redundancy to minimize SEE-induced hangs, and in our adaptive approach, partial reconfiguration is used to adapt the mitigation of dataflow parts in response to a dynamic radiation environment. Combined, both approaches enable RECON to maximize system performability subject to mission availability constraints. We perform fault injection and neutron irradiation to observe the susceptibility of RECON and use dependability modeling to evaluate RECON in various orbital case studies to demonstrate a 1.5–3.0× performability improvement in both performance and energy efficiency compared to static approaches.

Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States). Los Alamos Neutron Science Center (LANSCE)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA); National Science Foundation (NSF)
Grant/Contract Number:
89233218CNA000001
OSTI ID:
1983003
Journal Information:
ACM Transactions on Reconfigurable Technology and Systems, Journal Name: ACM Transactions on Reconfigurable Technology and Systems Journal Issue: 4 Vol. 14; ISSN 1936-7406
Publisher:
Association for Computing Machinery (ACM)Copyright Statement
Country of Publication:
United States
Language:
English

References (34)

A survey on deep learning techniques for image and video semantic segmentation journal September 2018
Survey on semantic segmentation using deep learning techniques journal April 2019
Deep learning in neural networks: An overview journal January 2015
The Los Alamos Neutron Science Center Spallation Neutron Sources journal January 2017
Radiation effects in reconfigurable FPGAs journal March 2017
CREME96: a revision of the C_osmic R_ay E_ffects on M_icro-E_lectronics code journal January 1997
Fast Algorithms for Convolutional Neural Networks conference June 2016
Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference conference June 2018
Efficient Error-Tolerant Quantized Neural Network Accelerators
  • Gambardella, Giulio; Kappauf, Johannes; Blott, Michaela
  • 2019 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT) https://doi.org/10.1109/DFT.2019.8875314
conference October 2019
On the Reliability of Convolutional Neural Network Implementation on SRAM-based FPGA conference October 2019
Reconfiguration Control Networks for TMR Systems with Module-Based Recovery
  • Agiakatsikas, Dimitris; Nguyen, Nguyen T. H.; Zhao, Zhuoran
  • 2016 IEEE 24th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM) https://doi.org/10.1109/FCCM.2016.30
conference May 2016
Fine-grained module-based error recovery in FPGA-based TMR systems conference December 2016
Reliability Evaluation of Mixed-Precision Architectures conference February 2019
High-Reliability FPGA-Based Systems: Space, High-Energy Physics, and Beyond journal March 2015
Onboard Processing With Hybrid and Reconfigurable Computing on Small Satellites journal March 2018
Fault tolerance in neural networks: Neural design and hardware implementation conference December 2017
Comparative Analysis of Inference Errors in a Neural Network Implemented in SRAM-Based FPGA Induced by Neutron Irradiation and Fault Injection Methods conference August 2018
ReCoN: A Reconfigurable CNN Acceleration Framework for Hybrid Semantic Segmentation on Hybrid SoCs for Space Applications conference July 2019
The Near-Earth Space Radiation Environment journal August 2008
Effectiveness of Internal Versus External SEU Scrubbing Mitigation Strategies in a Xilinx FPGA: Design, Test, and Analysis journal August 2008
Near-Earth Space Radiation Models journal June 2013
Challenges in Testing Complex Systems journal April 2014
A Hybrid Approach to FPGA Configuration Scrubbing journal January 2017
Selective Hardening for Neural Networks in FPGAs journal January 2019
Understanding the Impact of Quantization, Accuracy, and Radiation on the Reliability of Convolutional Neural Networks on FPGAs journal July 2020
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation journal December 2017
Design and Evaluation of Hybrid Fault-Detection Systems journal May 2005
Voter insertion algorithms for FPGA designs using triple modular redundancy conference February 2010
Reconfigurable Fault Tolerance: A Comprehensive Framework for Reliable and Adaptive FPGA-Based Space Computing journal December 2012
Mitigation of Radiation Effects in SRAM-Based FPGAs for Space Applications journal January 2015
Automated Systolic Array Architecture Synthesis for High Throughput CNN Inference on FPGAs conference June 2017
Fine-Grained Module-Based Error Recovery in FPGA-Based TMR Systems
  • Zhao, Zhuoran; Nguyen, Nguyen T. H.; Agiakatsikas, Dimitris
  • ACM Transactions on Reconfigurable Technology and Systems, Vol. 11, Issue 1 https://doi.org/10.1145/3173549
journal January 2018
[DL] A Survey of FPGA-based Neural Network Inference Accelerators journal April 2019
Reconfigurable Framework for Environmentally Adaptive Resilience in Hybrid Space Systems
  • Sabogal, Sebastian; George, Alan; Wilson, Christopher
  • ACM Transactions on Reconfigurable Technology and Systems, Vol. 13, Issue 3 https://doi.org/10.1145/3398380
journal July 2020

Similar Records

Related Subjects

97 MATHEMATICS AND COMPUTING
Computer Science
dependability modeling
environmentally adaptive computing
single-event effects
space computing