Learning Transferable Policies for Monocular Reactive MAV Control

Abstract

The ability to transfer knowledge gained in previous tasks into new contexts is one of the most important mechanisms of human learning. Despite this, adapting autonomous behavior to be reused in partially similar settings is still an open problem in current robotics research. In this paper, we take a small step in this direction and propose a generic framework for learning transferable motion policies. Our goal is to solve a learning problem in a target domain by utilizing the training data in a different but related source domain. We present this in the context of an autonomous MAV flight using monocular reactive control, and demonstrate the efficacy of our proposed approach through extensive real-world flight experiments in outdoor cluttered environments.

International Symposium on Experimental Robotics (ISER 2016)

References in corpus (4)

• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• End-to-End Training of Deep Visuomotor Policies
• Domain Adaptive Neural Networks for Object Recognition
• Robust Monocular Flight in Cluttered Outdoor Environments

Cited by in corpus (10)

• MagNet: a Two-Pronged Defense against Adversarial Examples
• DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
• Unifying Map and Landmark Based Representations for Visual Navigation
• End-to-end Learning of Driving Models from Large-scale Video Datasets
• Memory Augmented Control Networks
• Generalization through Simulation: Integrating Simulated and Real Data into Deep Reinforcement Learning for Vision-Based Autonomous Flight
• Domain Adaptation Using Adversarial Learning for Autonomous Navigation
• Adversarial Imitation Learning from Incomplete Demonstrations
• Learning Transferable UAV for Forest Visual Perception
• Transferable Representation Learning in Vision-and-Language Navigation