D. Tardioli, L. Cano, A. R. Mosteo
arXiv:2404.09688

Navigation of UAVs in challenging environments like tunnels or mines, where it is not possible to use GNSS methods to self-localize, illumination may be uneven or nonexistent, and wall features are likely to be scarce, is a complex task, especially if the navigation has to be done at high speed. In this paper we propose a novel proof-of-concept navigation technique for UAVs based on the use of LiDAR information through the joint use of geometric and machine-learning algorithms. The perceived information is processed by a deep neural network to establish the yaw of the UAV with respect to the tunnel’s longitudinal axis, in order to adjust the direction of navigation. Additionally, a geometric method is used to compute the safest location inside the tunnel (i.e. the one that maximizes the distance to the closest obstacle). This information proves to be sufficient for simple yet effective navigation in straight and curved tunnels.

D Tardioli, D Sicignano, L Riazuelo, A Romeo, JL Villarroel, L Montano
Journal of Field Robotics 33 (6), 765-801

Safety, security, and rescue robotics are crucial in emergencies like mining accidents or tunnel collapses, where robot teams can perform cooperative exploration, intervention, or logistics. Deploying multirobot teams in confined environments presents challenges such as task and motion planning, localization, mapping, safe navigation, coordination, and communication. Robots must move autonomously while maintaining connectivity for collaboration and operator oversight. This work introduces a system integrating deployment planning, semantic feature recognition, multirobot navigation, localization, mapping, and real-time communications, tested in two scenarios: the complex Spanish Santa Marta mine and the simpler Spanish-French Somport tunnel, each highlighting different operational challenges.

C. Rizzo, D. Tardioli, D. Sicignano, L. Riazuelo, J. L. Villarroel and L. Montano Journal of Field Robotics 32 (12), 1381-1397

Deploying a multi-robot team in confined environments poses multiple challenges that involve task and motion plan- ning, localization and mapping, safe navigation, coordination of robots and also communications among all of them. In recent years, increasing attention has been paid to these challenges by the robotics community, but many problems remain unresolved. In this paper we address a technique for planning the deployment of a robot team in so-called fading envi- ronments, such as tunnels or galleries, where signal propagation presents specific characteristics. In order to maintain constant connectivity and high signal quality in the communication network formed by the robots and the base station, the robot deployment is driven by real-time signal measurements. First, an analysis of the signal propagation to obtain the general characteristic parameters of the signals in this kind of environment is carried out. Second, a technique which uses these parameters to drive the deployment is developed. A general strategy for this kind of environment in which the signals exhibit similar behavior is implemented. A complete system involving all of the above-mentioned robotics tasks has been developed. Finally, the system has been evaluated by means of simulation and in a real scenario.