Grant: Discovery Grants Program - Individual
The University of British Columbia
$36,000.00
Fault-Tolerant Real-Time CPS for Emerging Low-Cost Consumer Devices
FY 2024-25
Summary
High-stakes domains, such as commercial aircraft manufacturers, embrace costly reliability engineering practices to ensure the safety of their users. Their goal is to engineer ultra-reliable CPS that are safe against design flaws and can tolerate stochastic faults due to defects or hazardous environment with an extremely high likelihood. Hence, critical avionic systems use custom fault-tolerant hardware with complex redundancy management protocols, undergo extensive testing and rigorous mathematical analysis of residual failure scenarios, and are subject to expensive software audits to comply with safety standards.
However, such practices are not affordable for the emerging class of low-cost consumer devices, such as consumer drones or service robots, which also rely heavily on CPS. Millions of CPS are flooding the markets today without adequate built-in safety mechanisms, compromising human safety. My long-term goal is to bring avionics-style ultra-reliability to the masses. Instead of custom fault-tolerant hardware and expensive, excruciatingly slow development processes, I propose to use highly reliable and formally analyzable safety mechanisms in software. My research will contribute novel techniques, tools, and analyses that will help developers implement and reason about such mechanisms with ease.
In the next 5 years, I will focus exclusively on the problem of fault-tolerant, real-time computing for production-grade consumer CPS, such as ArduPilot for unmanned aerial vehicles and Robot Operating System (ROS) for robots, which currently exhibit soft real-time performance but without offering any guarantees on timing, and without any fault tolerance. My research will address 4 challenges.
O1: In the absence of well synchronized clocks, most distributed systems resolve distributed agreement by assuming partial synchrony and prioritizing soundness over liveness. I will use clock synchronization, predictable Ethernet standards, and formal timing analyses to propose fault-tolerant, real-time computing systems with precise, quantifiable, and tight bounds on liveness.
O2: Engineers with domain-specific skills cannot be tasked with redesigning their systems for fault tolerance. My research will design synthesis and compilation frameworks so that widely used CPS middleware like ROS can provide safety mechanisms out of the box.
O3: Computations in CPS are increasingly being enhanced by ML, but ML runtimes like TensorFlow are neither designed for fault tolerance nor timeliness. My research will provide both reliable and timely ML inference serving frameworks for next-generation CPS, such as infrastructure-assisted driving.
O4: Formal methods historically were expensive to use; they were used when the cost of a failure justified the expense. With advances in verification tools, certification companies are increasingly accepting formal methods artifacts as certificates. I will use this opportunity to expand the scope of certification to more CPS.
Program
Discovery Grants Program - Individual
Recipient
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2024-04737
Name nom
Gujarati, Arpan
Department département
Computer Science
Organization id
2
Institution établissement
The University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
36000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
2720
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Computer systems software
Research subject fr
Logiciel de systèmes informatiques
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Fault-Tolerant Real-Time CPS for Emerging Low-Cost Consumer Devices
Keywords
real-time systems; embedded systems; cyber-physical systems; system reliability and fault tolerance; distributed systems; middleware
Application summary
High-stakes domains, such as commercial aircraft manufacturers, embrace costly reliability engineering practices to ensure the safety of their users. Their goal is to engineer ultra-reliable CPS that are safe against design flaws and can tolerate stochastic faults due to defects or hazardous environment with an extremely high likelihood. Hence, critical avionic systems use custom fault-tolerant hardware with complex redundancy management protocols, undergo extensive testing and rigorous mathematical analysis of residual failure scenarios, and are subject to expensive software audits to comply with safety standards.
However, such practices are not affordable for the emerging class of low-cost consumer devices, such as consumer drones or service robots, which also rely heavily on CPS. Millions of CPS are flooding the markets today without adequate built-in safety mechanisms, compromising human safety. My long-term goal is to bring avionics-style ultra-reliability to the masses. Instead of custom fault-tolerant hardware and expensive, excruciatingly slow development processes, I propose to use highly reliable and formally analyzable safety mechanisms in software. My research will contribute novel techniques, tools, and analyses that will help developers implement and reason about such mechanisms with ease.
In the next 5 years, I will focus exclusively on the problem of fault-tolerant, real-time computing for production-grade consumer CPS, such as ArduPilot for unmanned aerial vehicles and Robot Operating System (ROS) for robots, which currently exhibit soft real-time performance but without offering any guarantees on timing, and without any fault tolerance. My research will address 4 challenges.
O1: In the absence of well synchronized clocks, most distributed systems resolve distributed agreement by assuming partial synchrony and prioritizing soundness over liveness. I will use clock synchronization, predictable Ethernet standards, and formal timing analyses to propose fault-tolerant, real-time computing systems with precise, quantifiable, and tight bounds on liveness.
O2: Engineers with domain-specific skills cannot be tasked with redesigning their systems for fault tolerance. My research will design synthesis and compilation frameworks so that widely used CPS middleware like ROS can provide safety mechanisms out of the box.
O3: Computations in CPS are increasingly being enhanced by ML, but ML runtimes like TensorFlow are neither designed for fault tolerance nor timeliness. My research will provide both reliable and timely ML inference serving frameworks for next-generation CPS, such as infrastructure-assisted driving.
O4: Formal methods historically were expensive to use; they were used when the cost of a failure justified the expense. With advances in verification tools, certification companies are increasingly accepting formal methods artifacts as certificates. I will use this opportunity to expand the scope of certification to more CPS.