UAV-Based Situational Awareness System Using Deep Learning

Geraldes, Rúben; Gonçalves, Artur; Lai, Tin; Villerabel, Mathias; Deng, Wenlong; Salta, Ana; Nakayama, Kotaro; Matsuo, Yutaka; Prendinger, Helmut

doi:10.1109/access.2019.2938249

Cited by 61 publications

(41 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our contribution is to develop the interface of the drone management system. The algorithmic details for some components of the system have been studied in related works (e.g., the algorithm for automated image analysis [13] and the algorithm for conflict detection and resolution [10]- [12]), whereas the overall design of the immersive interface is discussed in this paper for the first time.…”

Section: Design Methodologymentioning

confidence: 99%

“…This can prohibit a real-time analysis of image sensor data with state-of-the-art techniques (i.e., deep convolutional neural networks). Thus, we use a cost-efficient algorithm [13] that has a lower accuracy but is especially suited for a real-time analysis on UAV hardware.…”

Section: Automation In the Drone Management Systemmentioning

confidence: 99%

“…With respect to software, research has advanced, among others, the decision-making logic for UAV routing [4], coordination mechanisms [8], [9], and collision detection/resolution (e.g., [10]- [12]). Other works have focused on situational awareness when detecting and locating objects [13]. There have also been field tests [14].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Interface Design for Human-Machine Collaborations in Drone Management

et al. 2021

Self Cite

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) operate with large degrees of autonomy when executing tasks from a ground-based controller. However, questions regarding the human-machine intersection remain open. In this work, we design an (immersive) interface for drone management. Our interface supports operators in mission execution of one or several (automated) UAVs within a fleet. This allows operators to assign tasks to UAVs, while safe operations are ensured through in-flight conflict detection and prevention.Our interface fulfills both functional (e.g., reliability) and non-functional requirements (e.g., perceived cognitive load during use). A key design challenge stems from the latency in information exchanges, which introduces information asymmetries between the humans and UAVs. We overcome previous information asymmetries by proposing a set of design principles, whereby we partially relegate the informationprocessing and decision-making capabilities to the UAVs. We then validate the effectiveness of our interface design experimentally, which succeeds in reducing the perceived cognitive load while improving the task performance. Altogether, this work has implications for designing interfaces in human-machine collaboration, so that humans can effectively control, interact, or collaborate with automated machines, such as UAVs.INDEX TERMS unmanned aerial vehicle (UAV), user interfaces, system analysis and design, drone management, in-flight conflict detection and resolution (CDR), field test

show abstract

Section: Design Methodologymentioning

confidence: 99%

Section: Automation In the Drone Management Systemmentioning

confidence: 99%

See 1 more Smart Citation

Interface Design for Human-Machine Collaborations in Drone Management

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Modern technical advances in next-generation network and communication infrastructure enable reliable management and organization by utilizing mobile computing platforms, e.g., autonomous unmanned aerial vehicles (UAVs) [1][2][3][4][5][6][7][8][9][10][11][12][13]. Even though autonomous UAVs are considered major components in next-generation network design and implementation, it has several research challenges [14,15].…”

Section: Introductionmentioning

confidence: 99%

Coordinated Multi-Agent Deep Reinforcement Learning for Energy-Aware UAV-Based Big-Data Platforms

et al. 2021

View full text Add to dashboard Cite

This paper proposes a novel coordinated multi-agent deep reinforcement learning (MADRL) algorithm for energy sharing among multiple unmanned aerial vehicles (UAVs) in order to conduct big-data processing in a distributed manner. For realizing UAV-assisted aerial surveillance or flexible mobile cellular services, robust wireless charging mechanisms are essential for delivering energy sources from charging towers (i.e., charging infrastructure) to their associated UAVs for seamless operations of autonomous UAVs in the sky. In order to actively and intelligently manage the energy resources in charging towers, a MADRL-based coordinated energy management system is desired and proposed for energy resource sharing among charging towers. When the required energy for charging UAVs is not enough in charging towers, the energy purchase from utility company (i.e., energy source provider in local energy market) is desired, which takes high costs. Therefore, the main objective of our proposed coordinated MADRL-based energy sharing learning algorithm is minimizing energy purchase from external utility companies to minimize system-operational costs. Finally, our performance evaluation results verify that the proposed coordinated MADRL-based algorithm achieves desired performance improvements.

show abstract

“…Besides, they have distinct advantages compared to other types, such as higher maneuverability and hovering capabilities, robustness, vertical takeoff and landing, ability to be easily equipped with different types of sensors, and low maintenance and purchase cost [7]. Therefore, there has been an increasing trend of adopting multirotor UAVs for military and non-military applications [8] such as providing security and conveying protection for mobile targets, capturing live videos and pictures for sport events [9], information collection for better situational awareness [10], and wildlife monitoring [11]. In most scenarios, multirotor UAVs maneuver ahead and at a standoff distance from the target to capture the real-time airborne information required by their mission.…”

Section: Introductionmentioning

confidence: 99%

A Dynamic Artificial Potential Field (D-APF) UAV Path Planning Technique for Following Ground Moving Targets

Jayaweera

Hanoun

2020

IEEE Access

View full text Add to dashboard Cite

Path planning is a vital and challenging component in the support of Unmanned Aerial Vehicles (UAVs) and their deployment in autonomous missions, such as following ground moving target. Few attempts are reported in the literature on multirotor UAV path planning techniques for following ground moving targets despite the great improvement in their control dynamics, flying behaviors and hardware specifications. These attempts suffer several drawbacks including their hardware dependency, high computational requirements, inability to handle obstacles and dynamic environments in addition to their low performance regarding the moving target speed variations. In this paper, a novel dynamic Artificial Potential Field (D-APF) path planning technique is developed for multirotor UAVs for following ground moving targets. The UAV produced path is a smooth and flyable path suitable to dynamic environments with obstacles and can handle different motion profiles for the ground moving target including change in speed and direction. Additionally, the proposed path planning technique effectively supports UAVs following ground moving targets while maneuvering ahead and at a standoff distance from the target. It is hardware-independent where it can be used on most types of multirotor UAVs with an autopilot flight controller and basic sensors for distance measurements. The developed path planning technique is tested and validated against existing general potential field techniques for different simulation scenarios in ROS and gazebo-supported PX4-SITL. Simulation results show that the proposed D-APF is better suited for UAV path planning for following moving ground targets compared to existing general APFs. In addition, it outperforms the general APFs as it is more suitable for UAVs flying in environments with dynamic and unknown obstacles. INDEX TERMS Unmanned Aerial Vehicles, path planning, Artificial Potential Field, ground moving targets.

show abstract

UAV-Based Situational Awareness System Using Deep Learning

Cited by 61 publications

References 33 publications

Interface Design for Human-Machine Collaborations in Drone Management

Interface Design for Human-Machine Collaborations in Drone Management

Coordinated Multi-Agent Deep Reinforcement Learning for Energy-Aware UAV-Based Big-Data Platforms

A Dynamic Artificial Potential Field (D-APF) UAV Path Planning Technique for Following Ground Moving Targets

Contact Info

Product

Resources

About