Transfer and Online Reinforcement Learning in STT-MRAM Based Embedded Systems for Autonomous Drones

In this paper we present an algorithm-hardware co-design for camera-based autonomous flight in small drones. We show that the large write-latency and write-energy for nonvolatile memory (NVM) based embedded systems makes them unsuitable for real-time reinforcement learning (RL). We address this by p...

Full description

Saved in:
Bibliographic Details
Main Authors: Yoon, Insik, Anwar, Aqeel, Rakshit, Titash, Raychowdhury, Arijit
Format: Conference Proceeding
Language:English
Subjects:
Automation
Cameras
Drones
Europe
Real-time systems
Reinforcement learning
System-on-chip
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper we present an algorithm-hardware co-design for camera-based autonomous flight in small drones. We show that the large write-latency and write-energy for nonvolatile memory (NVM) based embedded systems makes them unsuitable for real-time reinforcement learning (RL). We address this by performing transfer learning (TL) on meta-environments and RL on the last few layers of a deep convolutional network. While the NVM stores the meta-model from TL, an on-die SRAM stores the weights of the last few layers. Thus all the real-time updates via RL are carried out on the SRAM arrays. This provides us with a practical platform with comparable performance as end-to-end RL and 83.4% lower energy per image frame.
ISSN:1558-1101
DOI:10.23919/DATE.2019.8715066