36 Hits in 8.0 sec

A Model-free Deep Reinforcement Learning Approach To Maneuver A Quadrotor Despite Single Rotor Failure [article]

Paras Sharma, Prithvi Poddar, P.B. Sujit
2021 arXiv   pre-print
In this paper, we develop a model-free deep reinforcement learning approach for a quadrotor to recover from a single rotor failure.  ...  The approach is based on Soft-actor-critic that enables the vehicle to hover, land, and perform complex maneuvers.  ...  CONCLUSIONS In this paper, we modeled and evaluated a model free deep reinforcement learning based algorithm that uses Soft Actor-Critic methods to handle a single rotor failure in quadrotors.  ... 
arXiv:2109.10488v1 fatcat:2yxq6cyaejhk7enex6zxf3wzsm

Model Predictive Control for Micro Aerial Vehicles: A Survey [article]

Huan Nguyen, Mina Kamel, Kostas Alexis, Roland Siegwart
2020 arXiv   pre-print
Furthermore, an overview of recent research trends on the combined application of modern deep reinforcement learning techniques and model predictive control for multirotor vehicles is presented.  ...  learning methods have been utilized and if the controller refers to free-flight or other tasks such as physical interaction or load transportation.  ...  , and deep neural networks-based reinforcement learning approaches.  ... 
arXiv:2011.11104v1 fatcat:dil4kdnfcvfmxc7j6n3pqytlyi

Deep Reinforcement Learning-Based Adaptive Controller for Trajectory Tracking and Altitude Control of an Aerial Robot

Ali Barzegar, Deok-Jin Lee
2022 Applied Sciences  
The proposed controlling approach employs a reinforcement learning-based algorithm to actively estimate the controller parameters of the aerial robot.  ...  This research study presents a new adaptive attitude and altitude controller for an aerial robot.  ...  In addition to the aforementioned optimal control and active tuning approaches (for PIDs), Reinforcement Learning (RL) and Deep Reinforcement Learning (DRL) are other new approaches that have recently  ... 
doi:10.3390/app12094764 fatcat:hl7lsicz4zfjrd5m5pdtsdmdl4

A Review on Comparative Remarks, Performance Evaluation and Improvement Strategies of Quadrotor Controllers

Rupal Roy, Maidul Islam, Nafiz Sadman, M. A. Parvez Mahmud, Kishor Datta Gupta, Md Manjurul Ahsan
2021 Technologies  
This paper conducts a thorough analysis of the current literature on the effects of multiple controllers on quadrotors, focusing on two separate approaches: (i) controller hybridization and (ii) controller  ...  The quadrotor is an ideal platform for testing control strategies because of its non-linearity and under-actuated configuration, allowing researchers to evaluate and verify control strategies.  ...  Freddi et al. (2014) designed a quadrotor model in any failure case of a rotor by using feedback linearization.  ... 
doi:10.3390/technologies9020037 fatcat:ju7url6qo5bkhkses7a2raom7m

A Survey on Fault Diagnosis and Fault-Tolerant Control Methods for Unmanned Aerial Vehicles

George K. Fourlas, George C. Karras
2021 Machines  
Therefore, a fault-monitoring system must be specifically designed to supervise and debug each of these subsystems, so that any faults can be addressed before they lead to disastrous consequences.  ...  Typically, a UAV consists of three types of subsystems: actuators, main structure and sensors.  ...  A deep learning approach that utilized a hybrid Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) technique was developed in [71] , for the fault diagnosis of actuators on a six-rotor  ... 
doi:10.3390/machines9090197 fatcat:53feevbq25ggzdbe4kx3x7kdl4

Learning Deep Control Policies for Autonomous Aerial Vehicles with MPC-Guided Policy Search [article]

Tianhao Zhang, Gregory Kahn, Sergey Levine, Pieter Abbeel
2016 arXiv   pre-print
Reinforcement learning can in principle forego the need for explicit state estimation and acquire a policy that directly maps sensor readings to actions, but is difficult to apply to unstable systems that  ...  This data is used to train a deep neural network policy, which is allowed to access only the raw observations from the vehicle's onboard sensors.  ...  However, model-free RL is difficult to apply to unstable systems such as quadrotors, due to the possibility of catastrophic failure during training.  ... 
arXiv:1509.06791v2 fatcat:oy2gr3zsxbhazfbp7eqotih63e

Learning deep control policies for autonomous aerial vehicles with MPC-guided policy search

Tianhao Zhang, Gregory Kahn, Sergey Levine, Pieter Abbeel
2016 2016 IEEE International Conference on Robotics and Automation (ICRA)  
Reinforcement learning can in principle forego the need for explicit state estimation and acquire a policy that directly maps sensor readings to actions, but is difficult to apply to unstable systems that  ...  This data is used to train a deep neural network policy, which is allowed to access only the raw observations from the vehicle's onboard sensors.  ...  However, model-free RL is difficult to apply to unstable systems such as quadrotors, due to the possibility of catastrophic failure during training.  ... 
doi:10.1109/icra.2016.7487175 dblp:conf/icra/ZhangKLA16 fatcat:4mryusn4sbfxdpbifmo7gmguau

Laser-Based Reactive Navigation for Multirotor Aerial Robots using Deep Reinforcement Learning

Carlos Sampedro, Hriday Bavle, Alejandro Rodriguez-Ramos, Paloma de la Puente, Pascual Campoy
2018 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
In this paper, we present a fast reactive navigation algorithm using Deep Reinforcement Learning applied to multirotor aerial robots.  ...  Taking as input the 2D-laser range measurements and the relative position of the aerial robot with respect to the desired goal, the proposed algorithm is successfully trained in a Gazebo-based simulation  ...  ACKNOWLEDGMENT The authors would like to thank the UPM and the MON-CLOA Campus of International Excellence for funding the predoctoral contract of the corresponding author.  ... 
doi:10.1109/iros.2018.8593706 dblp:conf/iros/SampedroBRPC18 fatcat:zaywz6tdp5gqpk5jf7jcmjhire

Final Program

2020 2020 International Conference on Unmanned Aircraft Systems (ICUAS)  
and a great pleasure to welcome you to this year's conference.  ...  ., and in my capacity as the President of the Association, it is a privilege, a great pleasure and an honor to welcome you to the 2020 International Conference on Unmanned Aircraft Systems (ICUAS'20).  ...  This approach searches for obstacle-free, low computational cost, smooth, and dynamically feasible paths by analyzing a point cloud of the target environment, using a modified connect RRT-based path planning  ... 
doi:10.1109/icuas48674.2020.9214039 fatcat:7jr6chhfija47kgtwoxqmfmmoe

Drone Deep Reinforcement Learning: A Review

Ahmad Taher Azar, Anis Koubaa, Nada Ali Mohamed, Habiba A. Ibrahim, Zahra Fathy Ibrahim, Muhammad Kazim, Adel Ammar, Bilel Benjdira, Alaa M. Khamis, Ibrahim A. Hameed, Gabriella Casalino
2021 Electronics  
In this paper, we described the state of the art of one subset of these algorithms: the deep reinforcement learning (DRL) techniques.  ...  To name a few; infrastructure inspection, traffic patrolling, remote sensing, mapping, surveillance, rescuing humans and animals, environment monitoring, and Intelligence, Surveillance, Target Acquisition  ...  Acknowledgments: We would like to show our gratitude to Prince Sultan University, Riyadh, Kingdom of Saudi Arabia.  ... 
doi:10.3390/electronics10090999 doaj:57ededb7d1a0445eaf34975cb6625c1f fatcat:kya3fbblszd27i4exlybnji4ni

Human-in-the-Loop Methods for Data-Driven and Reinforcement Learning Systems [article]

Vinicius G. Goecks
2020 arXiv   pre-print
Recent successes combine reinforcement learning algorithms and deep neural networks, despite reinforcement learning not being widely applied to robotics and real world scenarios.  ...  This can be attributed to the fact that current state-of-the-art, end-to-end reinforcement learning approaches still require thousands or millions of data samples to converge to a satisfactory policy and  ...  model-free approach.  ... 
arXiv:2008.13221v1 fatcat:aofoenmwcvckvagbttrkskevty

A Review on IoT Deep Learning UAV Systems for Autonomous Obstacle Detection and Collision Avoidance

Paula Fraga-Lamas, Lucía Ramos, Víctor Mondéjar-Guerra, Tiago M. Fernández-Caramés
2019 Remote Sensing  
In this regard, Deep Learning (DL) techniques have arisen as a promising alternative for improving real-time obstacle detection and collision avoidance for highly autonomous UAVs.  ...  Advances in Unmanned Aerial Vehicles (UAVs), also known as drones, offer unprecedented opportunities to boost a wide array of large-scale Internet of Things (IoT) applications.  ...  and videos from a DL-UAV), but it is also more expensive and heavier than alternatives like single-rotor UAVs.  ... 
doi:10.3390/rs11182144 fatcat:54xs26xnvzf7rfa5b64tuzkz44

A Survey on Swarming With Micro Air Vehicles: Fundamental Challenges and Constraints

Mario Coppola, Kimberly N. McGuire, Christophe De Wagter, Guido C. H. E. de Croon
2020 Frontiers in Robotics and AI  
Once the operations of the single MAV are sufficiently secured for a task, the subsequent challenge is to allow the MAVs to sense one another within a neighborhood of interest.  ...  Robustness is often hailed as a pillar of swarm robotics, and a minimum level of local reliability is needed for it to propagate to the global level.  ...  In this approach, a clustering algorithm is used to learn a model of the robots' expected behavior.  ... 
doi:10.3389/frobt.2020.00018 pmid:33501187 pmcid:PMC7806031 fatcat:p3zp5y3r65cn7nilxzu6gedxaq

Survey on Coverage Path Planning with Unmanned Aerial Vehicles

Tauã Cabreira, Lisane Brisolara, Paulo R. Ferreira Jr.
2019 Drones  
The surveyed coverage approaches are classified according to a classical taxonomy, such as no decomposition, exact cellular decomposition, and approximate cellular decomposition.  ...  This paper aims to explore and analyze the existing studies in the literature related to the different approaches employed in coverage path planning problems, especially those using UAVs.  ...  The first step is to build a 3D terrain model using control points in order to obtain an analytical model.  ... 
doi:10.3390/drones3010004 fatcat:j3nsrywfnjcy3aw3zrug6xzmyu

Application Specific Drone Simulators: Recent Advances and Challenges

Aakif Mairaj, Asif I. Baba, Ahmad Y. Javaid
2019 Simulation modelling practice and theory  
A performance evaluation through relevant drone simulator becomes indispensable procedure to test features, configurations, and designs to demonstrate superiority to comparative schemes and suitability  ...  However, incidents such as fatal system failures, malicious attacks, and disastrous misuses have raised concerns in the recent past.  ...  , and reinforcement learning algorithms for various autonomous drones.  ... 
doi:10.1016/j.simpat.2019.01.004 fatcat:oy4rssrl5fagtixx5wrr747apm
« Previous Showing results 1 — 15 out of 36 results