scholarly journals DeepPilot: A CNN for Autonomous Drone Racing

Sensors ◽  
2020 ◽  
Vol 20 (16) ◽  
pp. 4524
Author(s):  
Leticia Oyuki Rojas-Perez ◽  
Jose Martinez-Carranza
Keyword(s):  
Deep Learning ◽  
Flight Control ◽  
Rotational Speed ◽  
Angular Position ◽  
Body Frame ◽  
Mosaic Image ◽  
Vertical Speed ◽  
Translation Motion ◽  
Yaw Angle ◽  
Detection And Localization

Autonomous Drone Racing (ADR) was first proposed in IROS 2016. It called for the development of an autonomous drone capable of beating a human in a drone race. After almost five years, several teams have proposed different solutions with a common pipeline: gate detection; drone localization; and stable flight control. Recently, Deep Learning (DL) has been used for gate detection and localization of the drone regarding the gate. However, recent competitions such as the Game of Drones, held at NeurIPS 2019, called for solutions where DL played a more significant role. Motivated by the latter, in this work, we propose a CNN approach called DeepPilot that takes camera images as input and predicts flight commands as output. These flight commands represent: the angular position of the drone’s body frame in the roll and pitch angles, thus producing translation motion in those angles; rotational speed in the yaw angle; and vertical speed referred as altitude h. Values for these 4 flight commands, predicted by DeepPilot, are passed to the drone’s inner controller, thus enabling the drone to navigate autonomously through the gates in the racetrack. For this, we assume that the next gate becomes visible immediately after the current gate has been crossed. We present evaluations in simulated racetrack environments where DeepPilot is run several times successfully to prove repeatability. In average, DeepPilot runs at 25 frames per second (fps). We also present a thorough evaluation of what we called a temporal approach, which consists of creating a mosaic image, with consecutive camera frames, that is passed as input to the DeepPilot. We argue that this helps to learn the drone’s motion trend regarding the gate, thus acting as a local memory that leverages the prediction of the flight commands. Our results indicate that this purely DL-based artificial pilot is feasible to be used for the ADR challenge.

scholarly journals Orchard Mapping with Deep Learning Semantic Segmentation

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3813
Author(s):  
Athanasios Anagnostis ◽  
Aristotelis C. Tagarakis ◽  
Dimitrios Kateris ◽  
Vasileios Moysiadis ◽  
Claus Grøn Sørensen ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Semantic Segmentation ◽  
Automated Detection ◽  
Aerial Images ◽  
Training Dataset ◽  
Field Boundary ◽  
Different Seasons ◽  
Detection And Localization ◽  
Different Levels

This study aimed to propose an approach for orchard trees segmentation using aerial images based on a deep learning convolutional neural network variant, namely the U-net network. The purpose was the automated detection and localization of the canopy of orchard trees under various conditions (i.e., different seasons, different tree ages, different levels of weed coverage). The implemented dataset was composed of images from three different walnut orchards. The achieved variability of the dataset resulted in obtaining images that fell under seven different use cases. The best-trained model achieved 91%, 90%, and 87% accuracy for training, validation, and testing, respectively. The trained model was also tested on never-before-seen orthomosaic images or orchards based on two methods (oversampling and undersampling) in order to tackle issues with out-of-the-field boundary transparent pixels from the image. Even though the training dataset did not contain orthomosaic images, it achieved performance levels that reached up to 99%, demonstrating the robustness of the proposed approach.

Robot detection and localization based on deep learning

2017 ◽  
Author(s):  
Sha Luo ◽  
Huimin Lu ◽  
Junhao Xiao ◽  
Qinghua Yu ◽  
Zhiqiang Zheng
Keyword(s):  
Deep Learning ◽  
Detection And Localization

Real Time Loading Test Rig for Flight Control Actuators Under PHM Experimentation

2018 ◽  
Author(s):  
P. Chiavaroli ◽  
A. De Martin ◽  
G. Evangelista ◽  
G. Jacazio ◽  
M. Sorli
Keyword(s):  
Flight Control ◽  
Linear Models ◽  
Angular Position ◽  
Experimental Tests ◽  
Hydraulic Actuator ◽  
Loading Test ◽  
Pressure Transducers ◽  
Non Linear ◽  
Flight Controls ◽  
Servo Actuator

The article deals with the architecture, performance, and experimental tests of a test bench for servo-actuators used in flight controls. After the state of the art on the subject, the innovative architecture of the built bench is described, in which flight control actuator under test and load actuator are not in line but mounted perpendicularly. The model of the bench actuating systems is then presented, consisting of the servo-controlled hydraulic actuator, load cell, speed transducer, angular position transducer of the coupling and pressure transducers. For each of these components the nonlinear multi-physics mechatronic model is described, according to the adopted solutions. The adopted force control algorithm is discussed, showing the integrative compensation on the action line and proportional-derivative on the feedback, with speed feedforward. The experimental tests carried out on the bench under stalled conditions are also presented, whose results concerning time and frequency responses are compared with those obtained through the linearized and non-linear numerical model. Finally, the non-linear models of the flight control actuator under test, controlled in position, and of the loading servo-actuator of the bench are joined together, and the results of various simulations are described.

scholarly journals A new active asymmetry monitoring and control technique applied to critical aircraft flap control system failures

2019 ◽  
Vol 304 ◽  
pp. 04011
Author(s):  
Dario Belmonte ◽  
Matteo Davide Lorenzo Dalla Vedova ◽  
Gaetano Quattrocchi
Keyword(s):  
Control System ◽  
Flight Control ◽  
Angular Position ◽  
Control Technique ◽  
Monitoring And Control ◽  
Active Monitoring ◽  
Actuation System ◽  
Left And Right ◽  
And Control ◽  
Current Monitoring

Asymmetry limitation requirements between left and right wing flap surfaces play an important role in the design of the implementation of the secondary flight control system of modern airplanes. In fact, especially in the case of sudden breaking of one of the torsion bars of the flap transmission line, the huge asymmetries that can rapidly develop could compromise the lateral-directional controllability of the whole aircraft (up to cause catastrophic occurrences). Therefore, in order to guarantee the aircraft safety (especially during take-off and landing flight phase in which the effects of asymmetries could generate uncontrollable aircraft attitudes), it is mandatory to timely detect and neutralize these asymmetries. The current monitoring techniques generally evaluate the differential angular position between left and right surfaces and, in most the events, limit the Flaps Control System (FCS) asymmetries, but in severe fault conditions (e.g. under very high aerodynamic loads), unacceptable asymmetries could be generated, compromising the controllability of the aircraft. To this purpose, in this paper the authors propose a new active monitoring and control technique capable of detecting the increasing angular error between the different flap surfaces and that, after stopping the whole actuation system, acts on the portion of the actuation line still connected to the PDU to minimize the FCS asymmetries.

Sign in / Sign up

Export Citation Format

Share Document