Perception Computing-Aware Controller Synthesis for Autonomous Systems
Clara Hobbs1,a, Debayan Roy2, Parasara Sridhar Duggirala2,a, F. Donelson Smith2,b, Soheil Samii3, James H. Anderson1,c and Samarjit Chakraborty1,d
1University of North Carolina at Chapel Hill, USA
acghobbs@cs.unc.edu
bpsd@cs.unc.edu
csmithfd@cs.unc.edu
danderson@cs.unc.edu
esamarjit@cs.unc.edu
2Technical University of Munich, Germany
debayan.roy@tum.de
soheil.samii@gm.com
ABSTRACT
Feedback control loops are ubiquitous in any autonomous system. The design flow for any controller starts by determining a control strategy, while abstracting away all implementation details. However, when designing controllers for autonomous systems, there is significant computation associated with the perception modules. For example, this involves vision processing using deep neural networks on multicore CPU+accelerator platforms. Such computation can be organized in many different ways, with each choice resulting in very different sensor-toactuator delays and tradeoffs between cost, delay, and accuracy. Further, each of these choices requires the control strategy to be designed accordingly. It is not possible for a control designer to enumerate and account for all of these choices manually, or abstract them away as “implementation details” as done in traditional controller design. In this paper we outline this problem and discuss how automated controller-synthesis techniques could help in addressing it.
