UAV Emergency Landing Site Selection System using Machine Vision

Faheem, Rao Muhammad; Aziz, Sumair; Khalid, Adnan; Bashir, Mudassar; Yasin, Amanullah

doi:10.21174/jomi.v1i1.24

Cited by 11 publications

(4 citation statements)

References 8 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This research assesses the efficacy and constraints of different approaches in quantifying roads from aerial imagery, therefore offering valuable insights into their practical implementations, including urban planning and navigation. The author of [3] described how the system enables unmanned aerial vehicles (UAVs) to identify suitable emergency landing sites autonomously. The inclusion of machine vision in the system enables automated decision-making capabilities during emergency situations, thereby enhancing UAV safety.…”

Section: Literature Surveymentioning

confidence: 99%

Autonomous UAV forced landing site prediction using machine learning

Gracelin Hepsiba J,

Purnima RG,

Shinthu Bargavi V

et al. 2024

Int. J. Sci. Res. Arch.

View full text Add to dashboard Cite

Unmanned aerial vehicles, or UAVs, are being used in an increasing range of applications, including surveillance, search and rescue, and environmental tracking. However, unanticipated engine issues, engine failures, and breakdown of the flying surface may necessitate forced landings, putting the UAV and its surroundings in danger. If there are any obstacles in the way of the UAV's ability to land safely, such as buildings or trees, it must be able to return to its emergency landing place. Thus, in these emergency scenarios, automated technology that can identify safe landing places rapidly. This paper presents an innovative approach that adds feature extraction, including HOG, HSV, LBP, and SFIT. GMM, SVM and kernels that use machine learning techniques to instinctively select the proper UAV-forced landing places. Through the use of machine learning and feature extraction techniques, we raised our accuracy by 40% over the baseline. The proposed system integrates data from several sources, including topography maps, satellite images, and board sensors. The machine learning algorithms predict possible landing sites. Annotated datasets with factors including topographic height, land cover type, slope, and proximity to obstacles are used to train these algorithms. especially artificial neural networks, or ANNs.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Autonomous UAV forced landing site prediction using machine learning

Gracelin Hepsiba J,

Purnima RG,

Shinthu Bargavi V

et al. 2024

Int. J. Sci. Res. Arch.

View full text Add to dashboard Cite

show abstract

“…Similarly, artificial neural networks were used to predict traffic accidents from visual or motion data (Gwak et al , 2018; Halim et al , 2016). The literature also contains few examples of using machine vision for selecting UAV landing sites (Aziz et al , 2016; Rao et al , 2016). To the authors' best knowledge, there has been no systematic study on objective prediction of impending drone accidents from operator's physiological data in (near-) real time.…”

Section: Literature Reviewmentioning

confidence: 99%

A feedforward neural network for drone accident prediction from physiological signals

Sakib

Chaspari

Behzadan

2021

SASBE

View full text Add to dashboard Cite

PurposeAs drones are rapidly transforming tasks such as mapping and surveying, safety inspection and progress monitoring, human operators continue to play a critical role in ensuring safe drone missions in compliance with safety regulations and standard operating procedures. Research shows that operator's stress and fatigue are leading causes of drone accidents. Building upon the authors’ past work, this study presents a systematic approach to predicting impending drone accidents using data that capture the drone operator's physiological state preceding the accident.Design/methodology/approachThe authors collect physiological data from 25 participants in real-world and virtual reality flight experiments to design a feedforward neural network (FNN) with back propagation. Four time series signals, namely electrodermal activity (EDA), skin temperature (ST), electrocardiogram (ECG) and heart rate (HR), are selected, filtered for noise and used to extract 92 time- and frequency-domain features. The FNN is trained with data from a window of length t = 3…8 s to predict accidents in the next p = 3…8 s.FindingsAnalysis of model performance in all 36 combinations of analysis window (t) and prediction horizon (p) combinations reveals that the FNN trained with 8 s of physiological signal (i.e. t = 8) to predict drone accidents in the next 6 s (i.e. p = 6) achieved the highest F1-score of 0.81 and AP of 0.71 after feature selection and data balancing.Originality/valueThe safety and integrity of collaborative human–machine systems (e.g. remotely operated drones) rely on not only the attributes of the human operator or the machinery but also how one perceives the other and adopts to the evolving nature of the operational environment. This study is a first systematic attempt at objective prediction of potential drone accident events from operator's physiological data in (near-) real time. Findings will lay the foundation for creating automated intervention systems for drone operations, ultimately leading to safer jobsites.

show abstract

“…In case of emergency situations, mainly engine failure, engine fire, flight instruments failure, or control surface damage or failure, continuing to fly becomes either impossible or can poses a serious threat to the safety of the flight. In such circumstances, a forced or emergency landing on a suitable surface such as a flat field becomes a must especially if it is not possible to return safely to the runway [5]. In [6], an emergency landing controller is proposed for an Unmanned Aerial Vehicle by segmenting the emergency landing period into four sub-levels known as slipping guiding, straight line down, exponential pulling up, and shallow sliding.…”

Section: A Fault/failure Tolerant Systems For Flight Controlmentioning

confidence: 99%

An Intelligent Autopilot System that learns flight emergency procedures by imitating human pilots

Baomar

Bentley

2016

2016 IEEE Symposium Series on Computational Intelligence (SSCI)

View full text Add to dashboard Cite

We propose an extension to the capabilities of the Intelligent Autopilot System (IAS) from our previous work, to be able to learn handling emergencies by observing and imitating human pilots. The IAS is a potential solution to the current problem of Automatic Flight Control Systems of being unable to handle flight uncertainties, and the need to construct control models manually. A robust Learning by Imitation approach is proposed which uses human pilots to demonstrate the task to be learned in a flight simulator while training datasets are captured from these demonstrations. The datasets are then used by Artificial Neural Networks to generate control models automatically. The control models imitate the skills of the human pilot when handling flight emergencies including engine(s) failure or fire, Rejected Take Off (RTO), and emergency landing, while a flight manager program decides which ANNs to be fired given the current condition. Experiments show that, even after being presented with limited examples, the IAS is able to handle such flight emergencies with high accuracy.

show abstract

UAV Emergency Landing Site Selection System using Machine Vision

Cited by 11 publications

References 8 publications

Autonomous UAV forced landing site prediction using machine learning

Autonomous UAV forced landing site prediction using machine learning

A feedforward neural network for drone accident prediction from physiological signals

An Intelligent Autopilot System that learns flight emergency procedures by imitating human pilots

Contact Info

Product

Resources

About