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

2019/04/22 by Yoon, Insik, Anwar, Aqeel, Rakshit, Titash +1
#FOS: Computer and information sciences #Hardware Architecture (cs.AR) #Other Computer Science (cs.OH)

paper · doi:10.48550/arxiv.1905.06314

Abstract

In this paper we present an algorithm-hardware codesign 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 metaenvironments 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

Related

• Autonomous Navigation via Deep Reinforcement Learning for Resource Constraint Edge Nodes using Transfer Learning
• NAVREN-RL: Learning to fly in real environment via end-to-end deep reinforcement learning using monocular images
• Deep Learning and Reinforcement Learning for Autonomous Unmanned Aerial Systems: Roadmap for Theory to Deployment
• BERRY: Bit Error Robustness for Energy-Efficient Reinforcement Learning-Based Autonomous Systems
• Air Learning: A Deep Reinforcement Learning Gym for Autonomous Aerial Robot Visual Navigation