Proceedings Article10.1109/IC3.2018.00014
A Parameter Sharing Method for Reinforcement Learning Model between AirSim and UAVs
Shau Yin Tseng,Chin-Feng Lai,Ming-Shi Wang,Ching Ju Chen,Chia Yu Ho +4 more
- 06 Dec 2018
- pp 20-23
9
TL;DR: In a virtual environment using the Q - learning training unmanned aerial vehicle landing, then transplanted model of virtual environment in which to train good into real environment, makes the realistic environment of unmanned Aerial vehicle can use cheaper and quickly achieve the same task.
read more
Abstract: In recent years, unmanned aerial vehicle aerial photography has developed rapidly. Unmanned aerial vehicle can get a different perspective and allow us to do more difficult tasks. Controlling unmanned aerial vehicle requires a lot of manpower, so there are a number of studies that use reinforcement learning to make the unmanned aerial vehicle fly autonomously. It is an expensive and time-consuming task to use reinforcement learning and training unmanned aerial vehicle to accomplish specific tasks in a realistic environment. Therefore this study in a virtual environment using the Q - learning training unmanned aerial vehicle landing, then transplanted model of virtual environment in which to train good into real environment, makes the realistic environment of unmanned aerial vehicle can use cheaper and quickly achieve the same task.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Security in Internet of Drones: A Comprehensive Review
TL;DR: In this paper , a review has been done on recent UAV frameworks which have been designed and tested using the AirSim simulator, which is an open-source UAV simulator that has different features like ease of development, efficient motion capture, efficient obstacle detections and collision detections, use of different sensor models, and physics models.
29
A Neuro-inspired Approach to Intelligent Collision Avoidance and Navigation
Nikolaus Salvatore,Sami Mian,Collin Abidi,Alan D. George +3 more
- 11 Oct 2020
TL;DR: This work presents both spiking and conventional neural network architectures, trained using reinforcement learning, for high-speed collision avoidance using dynamic vision sensors, as supported by previous work making use of the AirSim simulator.
19
Optimal Control Techniques for Heterogeneous UAV Swarms
Sami Mian,John Hill,Zhi-Hong Mao +2 more
- 11 Oct 2020
TL;DR: This study develops Heterogeneous Decentralized Receding Horizon Control (HD-RHC) for swarm management in search & rescue missions, which builds upon existing multiagent UAV work, but adds the capacity to manage a fleet of heterogeneous, diverse robot platforms that are equipped for different mission capabilities.
6
Directly-trained Spiking Neural Networks for Deep Reinforcement Learning: Energy efficient implementation of event-based obstacle avoidance on a neuromorphic accelerator
Luca Zanatta,Alfio Di Mauro,Francesco Barchi,Andrea Bartolini,Luca Benini,Andrea Acquaviva +5 more
TL;DR: This study presents an energy-efficient implementation of a Reinforcement Learning algorithm using directly-trained Spiking Neural Networks (SNNs) for obstacle avoidance on a neuromorphic accelerator, outperforming a Convolutional Neural Network (CNN) with 6× less energy consumption.
6
A Quadcopters Flight Simulation Considering the Influence of Wind
Mayu Ida,Hiroki Nishikawa,Xiangbo Kong,Ittetsu Taniguchi,Hiroyuki Tomiyama +4 more
- 21 Oct 2020
TL;DR: In this article, a program considering the influence of wind and implementing the wind on AirSim is presented. But the experimental results show that flying time in a simulator using the proposed program is almost the same as the flying time calculated by the formula.
6
References
Design and use paradigms for Gazebo, an open-source multi-robot simulator
Nathan Koenig,Andrew Howard +1 more
- 01 Sep 2004
TL;DR: Gazebo is designed to fill this niche by creating a 3D dynamic multi-robot environment capable of recreating the complex worlds that would be encountered by the next generation of mobile robots.
3.8K
AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles
Shital Shah,Debadeepta Dey,Chris Lovett,Ashish Kapoor +3 more
- 15 May 2017
TL;DR: In this paper, the authors present a new simulator built on Unreal Engine that offers physically and visually realistic simulations for autonomous vehicles in real-world environments, including a physics engine that can operate at a high frequency for real-time hardware-in-the-loop (HITL) simulations with support for popular protocols (e.g., MavLink).
•Posted Content
Uncertainty-Aware Reinforcement Learning for Collision Avoidance
TL;DR: An uncertainty-aware model-based learning algorithm that estimates the probability of collision together with a statistical estimate of uncertainty is presented, and it is shown that the algorithm naturally chooses to proceed cautiously in unfamiliar environments, and increases the velocity of the robot in settings where it has high confidence.
379
•Posted Content
Autonomous UAV Navigation Using Reinforcement Learning
TL;DR: This paper provides a framework for using reinforcement learning to allow the UAV to navigate successfully in such environments, and conducted a simulation and real implementation to show how the Uavs can successfully learn to navigate through an unknown environment.
101
Autonomous UAV Navigation Using Reinforcement Learning
Mudassar Liaq,Yung-Cheol Byun +1 more
TL;DR: This thesis aims to overcome the limitations of the existing works for autonomous navigation for UAV’s by proposing a model for navigating a drone in an unknown environment without any human help or aid and presenting a model which is compatible with almost every off-the-shelf drone available in the market.