Search

[Resumen] Los robots de búsqueda y rescate juegan un papel fundamental en la exploración de entornos peligrosos y desastres naturales. Su empleo permite a los equipos de rescate trabajar de manera más segura y eficiente en entornos de alto riesgo. La implementación de sistemas de teleoperación mediante realidad virtual (RV) permite a los usuarios controlar estos robots de forma remota con un alto grado de precisión. En este artículo se analiza, diseña e implementa un sistema de telecontrol para la operación remota del torso robotizado de un robot tipo centauro (formado por 11 servomotores) con capacidad de manipulación dual, utilizando un sistema RV. El desarrollo se ha realizado en dos fases. En la primera, se ha llevado a cabo el modelado e implementación física del torso del robot acoplado a diferentes bases móviles (robot cuadrúpedo y de orugas). En la segunda se ha creado el entorno virtual de teleoperación mediante Unity3D y ha realizado la conexión de este con el sistema real mediante Unity-Matlab. Para llevar a cabo las pruebas, se han planteado una serie de retos en interiores y exteriores, que van desde la manipulación dual de escombros, asistencia a víctimas con equipos de primeros auxilios como botiquines u oxígeno. Los principales resultados han mostrado un incremento en la confianza del operador para llevar tareas de precisión y contacto, frente a métodos de telecontrol convencionales., [Abstract] Search and rescue robots have a critical role in exploring dangerous environments and natural disasters. Its use allows rescue teams to work more safely and efficiently in high-risk environments. Implementing teleoperation systems through virtual reality (VR) allows users to control these robots remotely with high precision. This article analyzes, designs and implements a remote control system for the remote operation of the robotic torso of a centaur-type robot (formed by 11 servomotors) with dual manipulation capacity using a VR system. The development has been carried out in two phases. First, the modelling and physical implementation of the robot’s torso coupled to different mobile bases (quadruped and tracked robot) have been carried out. In the second, the virtual teleoperation environment has been created using Unity3D, and the connection with the real system has been made using Unity-Matlab. Different challenges have been raised indoors and outdoors to carry out the tests phase, from the dual manipulation f debris to victims’ assistance with first aid equipment, such as first aid kits or oxygen. Compared to conventional remote control methods, the main results have shown an increase in the operator’s confidence to carry out precision and contact tasks.

Post-disaster environments resulting from catastrophic events, leave sequels such as victims trapped in debris, which are difficult to detect by rescuers in a first inspection. Technological advances in electronics and perception have allowed the development of versatile and powerful optical sensors capable of capturing light in spectrums that humans cannot. new deep learning techniques, such as convolutional neural networks (CNNs), has allowed the generation of network models capable of autonomously detecting specific image patterns according to previous training. This work introduces an autonomous victim detection system to be deployed by using search and rescue robots. The proposed system defines new indexes based on combining the multispectral bands (Blue, Green, Red, Nir, Red Edge) to obtain new multispectral images where relevant characteristics of victims and the environment are highlighted. CNNs have been used as a second phase for automatically detecting victims in these new multispectral images. A qualitative and quantitative analysis of new indexes proposed by the authors has been carried out to evaluate their efficiency in contrast to the state-of-the-art ones. A data set has been generated to train different CNN models based on the best obtained index to analyze their effectiveness in detecting victims. The results show an efficiency of 92% in automatically detecting victims when applying the best multispectral index to new data. This method has also been contrasted with others based on thermal and RGB imagery to detect victims, where it has been proven that it generates better results in situations of outdoor environments and different weather conditions.

Technological breakthroughs in recent years have led to a revolution in fields such as Machine Vision and Search and Rescue Robotics (SAR), thanks to the application and development of new and improved neural networks to vision models together with modern optical sensors that incorporate thermal cameras, capable of capturing data in post-disaster environments (PDE) with rustic conditions (low luminosity, suspended particles, obstructive materials). Due to the high risk posed by PDE because of the potential collapse of structures, electrical hazards, gas leakage, etc., primary intervention tasks such as victim identification are carried out by robotic teams, provided with specific sensors such as thermal, RGB cameras, and laser. The application of Convolutional Neural Networks (CNN) to computer vision is a breakthrough for detection algorithms. Conventional methods for victim identification in these environments use RGB image processing or trained dogs, but detection with RGB images is inefficient in the absence of light or presence of debris; on the other hand, developments with thermal images are limited to the field of surveillance. This paper's main contribution focuses on implementing a novel automatic method based on thermal image processing and CNN for victim identification in PDE, using a Robotic System that uses a quadruped robot for data capture and transmission to the central station. The robot's automatic data processing and control have been carried out through Robot Operating System (ROS). Several tests have been carried out in different environments to validate the proposed method, recreating PDE with varying conditions of light, from which the datasets have been generated for the training of three neural network models (Fast R-CNN, SSD, and YOLO). The method's efficiency has been tested against another method based on CNN and RGB images for the same task showing greater effectiveness in PDE main results show that the proposed method has an efficiency