PIRT: A Runtime Framework to Enable Energy-Efficient Real-Time Robotic Applications on Heterogeneous Architectures

TL;DR

PIRT is a runtime framework that manages multiple accelerators on mobile and cloud systems to enable energy-efficient, real-time robotic applications with diverse tasks and stringent resource constraints.

Contribution

It introduces PIRT, the first framework to efficiently coordinate multiple accelerators for diverse robotic tasks on mobile and cloud platforms.

Findings

Robots achieve 25 FPS localization and 3 FPS obstacle detection.

Robots operate within an 11W power envelope at 5 mph.

PIRT improves performance and energy efficiency for robotic workloads.

Abstract

Enabling full robotic workloads with diverse behaviors on mobile systems with stringent resource and energy constraints remains a challenge. In recent years, attempts have been made to deploy single-accelerator-based computing platforms (such as GPU, DSP, or FPGA) to address this challenge, but with little success. The core problem is two-fold: firstly, different robotic tasks require different accelerators, and secondly, managing multiple accelerators simultaneously is overwhelming for developers. In this paper, we propose PIRT, the first robotic runtime framework to efficiently manage dynamic task executions on mobile systems with multiple accelerators as well as on the cloud to achieve better performance and energy savings. With PIRT, we enable a robot to simultaneously perform autonomous navigation with 25 FPS of localization, obstacle detection with 3 FPS, route planning, large map generation, and scene understanding, traveling at a max speed of 5 miles per hour, all within an 11W computing power envelope.