AAM-VDT: Vehicle Digital Twin for Tele-Operations in Advanced Air Mobility
Authors:
Tuan Anh Nguyen,
Taeho Kwag,
Vinh Pham,
Viet Nghia Nguyen,
Jeongseok Hyun,
Minseok Jang,
Jae-Woo Lee
Abstract:
This study advanced tele-operations in Advanced Air Mobility (AAM) through the creation of a Vehicle Digital Twin (VDT) system for eVTOL aircraft, tailored to enhance remote control safety and efficiency, especially for Beyond Visual Line of Sight (BVLOS) operations. By synergizing digital twin technology with immersive Virtual Reality (VR) interfaces, we notably elevate situational awareness and…
▽ More
This study advanced tele-operations in Advanced Air Mobility (AAM) through the creation of a Vehicle Digital Twin (VDT) system for eVTOL aircraft, tailored to enhance remote control safety and efficiency, especially for Beyond Visual Line of Sight (BVLOS) operations. By synergizing digital twin technology with immersive Virtual Reality (VR) interfaces, we notably elevate situational awareness and control precision for remote operators. Our VDT framework integrates immersive tele-operation with a high-fidelity aerodynamic database, essential for authentically simulating flight dynamics and control tactics. At the heart of our methodology lies an eVTOL's high-fidelity digital replica, placed within a simulated reality that accurately reflects physical laws, enabling operators to manage the aircraft via a master-slave dynamic, substantially outperforming traditional 2D interfaces. The architecture of the designed system ensures seamless interaction between the operator, the digital twin, and the actual aircraft, facilitating exact, instantaneous feedback. Experimental assessments, involving propulsion data gathering, simulation database fidelity verification, and tele-operation testing, verify the system's capability in precise control command transmission and maintaining the digital-physical eVTOL synchronization. Our findings underscore the VDT system's potential in augmenting AAM efficiency and safety, paving the way for broader digital twin application in autonomous aerial vehicles.
△ Less
Submitted 15 April, 2024;
originally announced April 2024.
LS-Net: Fast Single-Shot Line-Segment Detector
Authors:
Van Nhan Nguyen,
Robert Jenssen,
Davide Roverso
Abstract:
In low-altitude Unmanned Aerial Vehicle (UAV) flights, power lines are considered as one of the most threatening hazards and one of the most difficult obstacles to avoid. In recent years, many vision-based techniques have been proposed to detect power lines to facilitate self-driving UAVs and automatic obstacle avoidance. However, most of the proposed methods are typically based on a common three-…
▽ More
In low-altitude Unmanned Aerial Vehicle (UAV) flights, power lines are considered as one of the most threatening hazards and one of the most difficult obstacles to avoid. In recent years, many vision-based techniques have been proposed to detect power lines to facilitate self-driving UAVs and automatic obstacle avoidance. However, most of the proposed methods are typically based on a common three-step approach: (i) edge detection, (ii) the Hough transform, and (iii) spurious line elimination based on power line constrains. These approaches not only are slow and inaccurate but also require a huge amount of effort in post-processing to distinguish between power lines and spurious lines. In this paper, we introduce LS-Net, a fast single-shot line-segment detector, and apply it to power line detection. The LS-Net is by design fully convolutional and consists of three modules: (i) a fully convolutional feature extractor, (ii) a classifier, and (iii) a line segment regressor. Due to the unavailability of large datasets with annotations of power lines, we render synthetic images of power lines using the Physically Based Rendering (PBR) approach and propose a series of effective data augmentation techniques to generate more training data. With a customized version of the VGG-16 network as the backbone, the proposed approach outperforms existing state-of-the-art approaches. In addition, the LS-Net can detect power lines in near real-time (20.4 FPS). This suggests that our proposed approach has a promising role in automatic obstacle avoidance and as a valuable component of self-driving UAVs, especially for automatic autonomous power line inspection.
△ Less
Submitted 24 January, 2020; v1 submitted 19 December, 2019;
originally announced December 2019.
