CSE seminar by Dr. Sparsh Mittal on Autonomous driving: constraints and hardware acceleration.
Title: Autonomous driving: constraints and hardware acceleration.
Speaker: Dr. Sparsh Mittal, Assistant Professor in the dept. of CSE, IIT Hyderabad
Room No: LH-2, Academic block
(This talk is mostly based on the paper “The Architectural Implications of Autonomous Driving: Constraints and Acceleration” by University of Michigan research group, published in ASPLOS 2018. )
Autonomous driving systems have attracted a signiicant amount of interest recently, and many industry leaders, such as Google, Uber, Tesla and Mobileye, have invested large amount of capital and engineering power on developing such systems. Building autonomous driving systems is particularly challenging due to stringent performance requirements in terms of both making the safe operational decisions and finishing processing at real-time.
This talk discusses the performance, predictability, storage, thermal and power constraints in designing autonomous driving systems. It then discusses the design of an end-to-end autonomous driving system using state-of-the-art algorithms to understand the design trade-offs for building such systems. The algorithms are accelerated over multicore CPU, GPU, FPGA and ASICs. Results show that compared to multicore CPUs, accelerators viz., GPUs, FPGAs and ASICs reduce the tail latency of the system by 169X, 10X, and 93X respectively.
Dr. Sparsh Mittal is a faculty member in CSE department.
Friday, March 23, 2018 - 10:00 to 11:00