Web-Based Real-Time Alarm and Teleoperation System for Autonomous Navigation Failures Using ROS 1 and ROS 2 DOI Creative Commons
Nabih Pico,

Giovanny Mite,

Daniel S. Moran

и другие.

Actuators, Год журнала: 2025, Номер 14(4), С. 164 - 164

Опубликована: Март 26, 2025

This paper presents an alarm system and teleoperation control framework, comparing ROS 1 2 within a local network to mitigate the risk of robots failing reach their goals during autonomous navigation. Such failures can occur when robot moves through irregular terrain, becomes stuck on small steps, or approaches walls obstacles without maintaining safe distance. These issues may arise due combination technical, environmental, operational factors, including inaccurate sensor data, blind spots, localization errors, infeasible path planning, inability adapt unexpected obstacles. The integrates web-based graphical interface developed using frontend frameworks joystick for real-time monitoring robot’s localization, velocity, proximity is equipped with RGB-D tracking cameras, 2D LiDAR, odometry sensors, providing detailed environmental data. provides sensory feedback visual alerts web vibration walls, faces potential collisions objects, loses stability. evaluated in both simulation (Gazebo) real-world experiments, where latency measured performance assessed 2. results demonstrate that systems operate effectively real time, ensuring safety enabling timely operator intervention. offers lower LiDAR inputs, making it advantageous over 1. However, camera higher, suggesting need optimizations image data processing. Additionally, platform supports integration additional sensors applications based user requirements.

Язык: Английский

Human and environmental feature-driven neural network for path-constrained robot navigation using deep reinforcement learning DOI
Nabih Pico, Estrella Montero,

Alisher Amirbek

и другие.

Engineering Science and Technology an International Journal, Год журнала: 2025, Номер 64, С. 101993 - 101993

Опубликована: Фев. 25, 2025

Язык: Английский

Процитировано

1
Memory-driven deep-reinforcement learning for autonomous robot navigation in partially observable environments DOI Creative Commons
Estrella Montero, Nabih Pico, Mitra Ghergherehchi

и другие.

Engineering Science and Technology an International Journal, Год журнала: 2025, Номер 62, С. 101942 - 101942

Опубликована: Янв. 24, 2025

Язык: Английский

Процитировано

1
High-Quality Text-to-Image Generation Using High-Detail Feature-Preserving Network DOI Creative Commons
Wei‐Yen Hsu, Jing Lin

Applied Sciences, Год журнала: 2025, Номер 15(2), С. 706 - 706

Опубликована: Янв. 13, 2025

Multistage text-to-image generation algorithms have shown remarkable success. However, the images produced often lack detail and suffer from feature loss. This is because these methods mainly focus on extracting features text, using only conventional residual blocks for post-extraction processing. results in loss of features, greatly reducing quality generated necessitating more resources calculation, which will severely limit use application optical devices such as cameras smartphones. To address issues, novel High-Detail Feature-Preserving Network (HDFpNet) proposed to effectively generate high-quality, near-realistic text descriptions. The initial (iT2IG) module used maps avoid Next, fast excitation-and-squeeze extraction (FESFE) recursively high-detail feature-preserving with lower computational costs through three steps: channel excitation (CE), (FFE), squeeze (CS). Finally, attention (CA) mechanism further enriches details. Compared state art, experimental obtained CUB-Bird MS-COCO datasets demonstrate that HDFpNet achieves better performance visual presentation, especially regarding preservation.

Язык: Английский

Процитировано

0
Web-Based Real-Time Alarm and Teleoperation System for Autonomous Navigation Failures Using ROS 1 and ROS 2 DOI Creative Commons
Nabih Pico,

Giovanny Mite,

Daniel S. Moran

и другие.

Actuators, Год журнала: 2025, Номер 14(4), С. 164 - 164

Опубликована: Март 26, 2025

This paper presents an alarm system and teleoperation control framework, comparing ROS 1 2 within a local network to mitigate the risk of robots failing reach their goals during autonomous navigation. Such failures can occur when robot moves through irregular terrain, becomes stuck on small steps, or approaches walls obstacles without maintaining safe distance. These issues may arise due combination technical, environmental, operational factors, including inaccurate sensor data, blind spots, localization errors, infeasible path planning, inability adapt unexpected obstacles. The integrates web-based graphical interface developed using frontend frameworks joystick for real-time monitoring robot’s localization, velocity, proximity is equipped with RGB-D tracking cameras, 2D LiDAR, odometry sensors, providing detailed environmental data. provides sensory feedback visual alerts web vibration walls, faces potential collisions objects, loses stability. evaluated in both simulation (Gazebo) real-world experiments, where latency measured performance assessed 2. results demonstrate that systems operate effectively real time, ensuring safety enabling timely operator intervention. offers lower LiDAR inputs, making it advantageous over 1. However, camera higher, suggesting need optimizations image data processing. Additionally, platform supports integration additional sensors applications based user requirements.

Язык: Английский

Процитировано

0