Francisco Marques

An autonomous surface-aerial marsupial robotic team for riverine environmental monitoring: Benefiting from coordinated aerial, underwater, and surface level perception

This paper presents RIVERWATCH, an autonomous surface-aerial marsupial robotic team for riverine environmental monitoring. The robotic system is composed of an Autonomous Surface Vehicle (ASV) piggybacking a multirotor Unmanned Aerial Vehicle (UAV) with vertical takeoff and landing capabilities. The ASV provides the team with longrange transportation in all-weather conditions, whereas the UAV assures an augmented perception of the environment. The coordinated aerial, underwater, and surface level perception allows the team to assess navigation cost from the near field to the far field, which is key for safe navigation and environmental monitoring data gathering. The robotic system is validated on a set of field trials.

Kelpie: A ROS-Based Multi-robot Simulator for Water Surface and Aerial Vehicles

Testing and debugging real hardware is a time consuming task, in particular for the case of aquatic robots, for which it is necessary to transport and deploy the robots on the water. Performing waterborne and airborne field experiments with expensive hardware embedded in not yet fully functional prototypes is a highly risky endeavour. In this sense, physics-based 3D simulators are key for a fast paced and affordable development of such robotic systems. This paper contributes with a modular, open-source, and soon to be freely online available, ROS-based multi-robot simulator specially focused for aerial and water surface vehicles. This simulator is being developed as part of the RIVERWATCH experiment in the ECHORD european FP7 project. This experiment aims at demonstrating a multi-robot system for remote monitoring of riverine environments.

On the Design of a Robotic System Composed of an Unmanned Surface Vehicle and a Piggybacked VTOL

This paper presents the core ideas of the RIVERWATCH experiment and describes its hardware architecture. The RIVERWATCH experiment considers the use of autonomous surface vehicles piggybacking multi-rotor unmanned aerial vehicles for the automatic monitoring of riverine environments. While the surface vehicle benefits from the aerial vehicle to extend its field of view, the aerial vehicle benefits from the surface vehicle to ensure long-range mobility. This symbiotic relation between both robots is expected to enhance the robustness and long lasting of the ensemble. The hardware architecture includes a considerable set of state-of-the-art sensory modalities and it is abstracted from the perception and navigation algorithms by using the Robotics Operating System (ROS). A set of field trials shows the ability of the prototype to scan a closed water body. The datasets obtained from the field trials are freely available to the robotics community.

A cooperative multi-robot team for the surveillance of shipwreck survivors at sea

The sea as a very extensive area, renders difficult a pre-emptive and long-lasting search for shipwreck survivors. The operational cost for deploying manned teams with such proactive strategy is high and, thus, these teams are only reactively deployed when a disaster like a shipwreck has been communicated. To reduce the involved financial costs, unmanned robotic systems could be used instead as background surveillance teams patrolling the seas. In this sense, a robotic team for search and rescue (SAR) operations at sea is presented in this work. Composed of an Unmanned Surface Vehicle (USV) piggybacking a watertight Unmanned Aerial Vehicle (UAV) with vertical take-off and landing capabilities, the proposed cooperative system is capable of search, track and provide basic life support while reporting the position of human survivors to better prepared manned rescue teams. The USV provides long-range transportation of the UAV and basic survival kits for victims. The UAV assures an augmented perception of the environment due to its high vantage point.

Online self-reconfigurable robot navigation in heterogeneous environments

This paper presents a robot navigation system capable of online self-reconfiguration according to the needs imposed by the various contexts present in heterogeneous environments. The ability to cope with heterogeneous environments is key for a robust deployment of service robots in truly demanding scenarios. In the proposed system, flexibility is present at the several layers composing the robots navigation system. At the lowest layer, proper locomotion modes are selected according to the environments local context. At the highest layer, proper motion and path planning strategies are selected according to the environments global context. While local context is obtained directly from the robots sensory input, global context is inspected from semantic labels registered off-line on geo-referenced maps. The proposed system leverages on the well-known Robotics Operating System (ROS) framework for the implementation of the major navigation system components. The system was successfully validated over approximately 1 Km long experiments on INTROBOT, an all-terrain industrial-grade robot equipped with four independently steered wheels.

Gaze-directed telemetry in high latency wireless communications: The case of robot teleoperation

The proposed telemetry system consists of a graphical user interface with reconfigurable multiple widgets whose transmission is prioritised based on the users gaze. The proposed approach encompasses a novel framework for telemetry of remote machines, based on ROS (Robot Operating System). It includes a modular ensemble of GUI, gaze device interoperability, and a ROS sensory topic modulator, which alternates different strategies to optimise all displayed information transmission quality, in a way that best suits the user. The approach was validated on a teleoperation scenario having multiple users control a robot in a designed course.

Saliency-based cooperative landing of a multirotor aerial vehicle on an autonomous surface vehicle

This paper presents a method for vision-based landing of a multirotor unmanned aerial vehicle (UAV) on an autonomous surface vehicle (ASV) equipped with a helipad. The method includes a mechanism for helipad behavioural search when outside the UAVs field of view, a learning saliency-based mechanism for visual tracking the helipad, and a cooperative strategy for the final vision-based landing phase. Learning how to track the helipad from above occurs during takeoff and cooperation results from having the ASV tracking the UAV for assisting its landing. A set of experimental results with both simulated and physical robots show the feasibility of the presented method.

Sediment Sampling in Estuarine Mudflats with an Aerial-Ground Robotic Team

This paper presents a robotic team suited for bottom sediment sampling and retrieval in mudflats, targeting environmental monitoring tasks. The robotic team encompasses a four-wheel-steering ground vehicle, equipped with a drilling tool designed to be able to retain wet soil, and a multi-rotor aerial vehicle for dynamic aerial imagery acquisition. On-demand aerial imagery, properly fused on an aerial mosaic, is used by remote human operators for specifying the robotic mission and supervising its execution. This is crucial for the success of an environmental monitoring study, as often it depends on human expertise to ensure the statistical significance and accuracy of the sampling procedures. Although the literature is rich on environmental monitoring sampling procedures, in mudflats, there is a gap as regards including robotic elements. This paper closes this gap by also proposing a preliminary experimental protocol tailored to exploit the capabilities offered by the robotic system. Field trials in the south bank of the river Tagus’ estuary show the ability of the robotic system to successfully extract and transport bottom sediment samples for offline analysis. The results also show the efficiency of the extraction and the benefits when compared to (conventional) human-based sampling.

An Aerial-Ground Robotic Team for Systematic Soil and Biota Sampling in Estuarine Mudflats

This paper presents an aerial-ground field robotic team, designed to collect and transport soil and biota samples in estuarine mudflats. The robotic system has been devised so that its sampling and storage capabilities are suited for radionuclides and heavy metals environmental monitoring. Automating these time-consuming and physically demanding tasks is expected to positively impact both their scope and frequency. The success of an environmental monitoring study heavily depends on the statistical significance and accuracy of the sampling procedures, which most often require frequent human intervention. The bird’s-eye view provided by the aerial vehicle aims at supporting remote mission specification and execution monitoring. This paper also proposes a preliminary experimental protocol tailored to exploit the capabilities offered by the robotic system. Preliminary field trials in real estuarine mudflats show the ability of the robotic system to successfully extract and transport soil samples for offline analysis.

Laser-Based Obstacle Detection at Railway Level Crossings

This paper presents a system for obstacle detection in railway level crossings from 3D point clouds acquired with tilting 2D laser scanners. Although large obstacles in railway level crossings are detectable with current solutions, the detection of small obstacles remains an open problem. By relying on a tilting laser scanner, the proposed system is able to acquire highly dense and accurate point clouds, enabling the detection of small obstacles, like rocks laying near the rail. During an offline training phase, the system learns a background model of the level crossing from a set of point clouds. Then, online, obstacles are detected as occupied space contrasting with the background model. To reduce the need for manual on-site calibration, the system automatically estimates the pose of the level crossing and railway with respect to the laser scanner. Experimental results show the ability of the system to successfully perform on a set of 41 point clouds acquired in an operational one-lane level crossing.