BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211208T164507Z
LOCATION:3016
DTSTART;TZID=America/Los_Angeles:20211209T133000
DTEND;TZID=America/Los_Angeles:20211209T150000
UID:dac_DAC 2021_sess102@linklings.com
SUMMARY:Automating Autonomy: From GPU to Traffic Control
DESCRIPTION:Research Manuscript\n\nThis session features a range of exciti
 ng results for the design and analysis of autonomous systems. At the appli
 cation level, several papers address key challenges of real-time performan
 ce, fault tolerance, and distributed control of autonomous vehicles. On th
 e software side, two papers present novel state-sensitive fuzzing techniqu
 es and a robot runtime that can exploit environment heterogeneity to signi
 ficantly improve performance and resource usage. Finally, at the hardware 
 level, one paper tackles the outstanding problem of memory protection in G
 PUs.\n\nAnalyzing and Improving Fault Tolerance of Learning-Based Navigati
 on Systems\n\nWan\n\nLearning-based navigation systems are widely used in 
 autonomous applications, e.g., robotics, unmanned vehicles and drones. Spe
 cialized hardware accelerators have been proposed for high-performance and
  energy-efficiency for such navigational tasks. However, transient and per
 manent faults are increas...\n\n---------------------\nNeural Pruning Sear
 ch for Real-Time Object Detection of Autonomous Vehicles\n\nZhao\n\nObject
  detection plays an important role in self-driving cars. However, mobile s
 ystems on self-driving cars with limited resources lead to difficulties fo
 r object detection. To facilitate this, we propose a compiler-aware neural
  pruning search framework  to achieve high-speed inference on autonomous..
 .\n\n---------------------\ngGuard: Enabling Leakage-Resilient Memory Isol
 ation in GPU-accelerated Autonomous Embedded Systems\n\nYadlapalli\n\nGrap
 hics processing units (GPUs)  are being widely used as co-processors for p
 erformance acceleration in many autonomous embedded systems such as roboti
 cs and  autonomous vehicles. However, current GPU hardware and systems sof
 tware, including GPU device drivers, compilers, and operating systems, do 
 ...\n\n---------------------\nRoboRun: A Robot Runtime to Exploit Spatial 
 Heterogeneity\n\nBoroujerdian\n\nThe limited onboard energy of autonomous 
 mobile robots demands efficient computing solutions. A shortcoming of stat
 e-of-the-art solutions is that they ignore the robot's environment heterog
 eneity making static, worst-case assumptions. As this heterogeneity impact
 s the computing payload, an optimal s...\n\n\nTopic: Autonomous Systems\n\
 nVirtual Program: Presented In-Person
END:VEVENT
END:VCALENDAR