Parallel Reinforcement Learning Simulation for Visual Quadrotor Navigation

Reinforcement learning (RL) is an agent-based approach for teaching robots to navigate within the physical world. Gathering data for RL is known to be a laborious task, and real-world experiments can be risky. Simulators facilitate the collection of training data in a quicker and more cost-effective manner. However, RL frequently requires a significant number of simulation steps for an agent to become skilful at simple tasks. This is a prevalent issue within the field of RL-based visual quadrotor navigation where state dimensions are typically very large and dynamic models are complex. Furthermore, rendering images and obtaining physical properties of the agent can be computationally expensive. To solve this, we present a simulation framework, built on AirSim, which provides efficient parallel training. Building on this framework, Ape-X is modified to incorporate parallel training of AirSim environments to make use of numerous networked computers. Through experiments we were able to achieve a reduction in training time from 3. 9 hours to 11 minutes, for a toy problem, using the aforementioned framework and a total of 74 agents and two networked computers. Further details including a github repo and videos about our project, PRL4AirSim, can be found at https://sites.google.com/view/prl4airsim/home

Authors

• Jack Saunders
• Sajad Saeedi 0001
• Wenbin Li 0002

Keywords

• Training
• Computers
• Visualization
• Navigation
• Computational modeling
• Atmospheric modeling
• Reinforcement learning
• Parallel Reinforcement Learning
• Training Time
• Digital Networks
• Reduce Training Time
• Parallel Training
• Neural Network
• Inertial Measurement Unit
• Multiple Agents
• Depth Camera
• Global Parameters
• Linear Velocity
• Deep Reinforcement Learning
• Multiple Environments
• Image Synthesis
• Machine Vision
• Client-side
• Physics Engine
• Distributed Architecture
• Experience Replay
• Unreal Engine
• Proximal Policy Optimization
• Replay Buffer
• Deep Q-learning
• Tens Of Hours
• API Calls
• Depth Images
• Number Of Agents
• Reward Function
• State Space