- Duplicate citations
- Universidad de Las Palmas de Gran Canaria (ULPGC)
- World Robotic Sailing Championship - Wikipedia
- Kjell Ivar Øvergård
They provide an in-depth theoretical background for each sub-problem and define the theoretical lower-bound limits for their solutions. They conclude that even though localization precision cannot be improved, motion planning can be the focus of future research involving the cooperation of UAV-USV for SAR. An approach combining the Pelagi USV, a 4. The key idea behind the CADDY project is to use a UUV as a diver companion, taking photos and other tasks such as guiding the diver and bringing objects to the surface.
Carefully planned experiments led to a mean error of 1. The project trials were designed to detect information about a diver potentially in distress. Initial tests [ ] included localization, tracking, and diver activity detection, while the final validation [ ] presented the warning system for the diver in distress and showed that divers felt safe and comfortable using the system. Divers often are part of disaster scenarios as part of first-responders acting in extremely harsh environments such as the cave where the Thai boys were trapped in , and one diver lost his life trying to save them.
Table 5 presents the current work on disaster robotics involving USVs. Figure 3 presents a bubble chart for the visual representation of the number of works in Table 5 , classified by the operation and disaster management phases. The number of works is directly proportional to the size of a bubble, while the colors represent the maturity level for each operation.
Finally, this section presents a series of technological and non-technological guidelines which have an important role in the DM with USVs. Visual representation for the number of works in Table 5. This section presents and discusses other technological hardware and software recommendations and guidelines for USV research focusing on Disaster Robotics. Connectivity : Communication is fundamental for the operation and to guarantee that relevant sensor data reaches the operation center on land [ ].
Still, communication problems with USVs often occur, since connectivity depends on the environment, weather, and wave conditions. Furthermore, the availability of broadband connectivity is limited in remote areas of the ocean [ ] where only satellite communications with limited bandwidth are available. Even in shallow water and obstructed regions, in situations which cannot be handled by the USV alone e. Therefore, communication problems must not be underestimated.
Robust delay-tolerant protocols and equipment should always be considered for DM missions.
However, as the human interaction with the USV increases, the bandwidth requirements also must increase to provide real-time video and responsive control, among other bandwidth demanding requirements involving USV sensors data [ ]. Finally, in disaster scenarios where the fixed infrastructure may be compromised, IEEE Localization : Bad weather, being near the coast or man-made buildings e. Therefore, localization strategies should be versatile in case of GPS problems.
Such features will be helpful for several operations, including SAR, and detection of hazards in the surroundings of USVs. Information Sharing : The operator must not be the only person with access to real-time information from the USV, including cameras. Information access to responders and experts must be straightforward. Enough bandwidth and an adequate user interface are required to allow multiple simultaneous users.
For instance, disaster damage to man-made structures may be above and below the water level and sometimes only accessible with UUVs—e. UUVs can take advantage of USV localization and underwater mapping information [ ], communications as well as communication capacity. Finally, UAVs can offer a view from the disaster site which is not possible for USVs [ 61 , 79 , ], paramount for detecting victims [ ] or hazards through an upper field of view.
For example, Zhang et al. This way, it is also possible to rapidly carry UUVs to disaster sites and perform underwater tasks as needed. Similarly, USV towing capabilities could be used to tow containment booms [ , ] or vessels in distress [ , ] to help contain further environmental disasters e. It may also be possible to connect multiple USVs to form an autonomous containment line to pollutants or a blockade to alien vessels [ 95 ].
Thrusters : There are objects such as plastic objects, and plants, such as eelgrass, which may damage underwater propellers [ , ].
Furthermore, propellers should not pose a hazard to humans in disaster sites e. Therefore, USV design should consider proper propeller casing to preserve its integrity and prevent injuries to others in need e. Cruz and Alves [ ] argue that sailboats can be effective for both monitoring and disaster response due to the lack of propellers and the potential for power savings. However, the absence and excess of winds may limit the use of sailboats in real disaster response scenarios.
On the other hand, Scerri et al. Bathymetry : Even though bathymetry instruments are essential tools to address many problems, they are prone to errors which depend on sensor limitations and the environment e. Performing bathymetry surveys during high tides is a good strategy as it is possible to place the USV as close to the shore as possible to assess regions near the water with risk of collapse to improve the bathymetry results.
Furthermore, calm waters are always the best scenario of choice for a survey, since environmental disturbances such as waves and wind may affect instruments. Sensor Payload and Threat Detection : Appropriate sensor positioning must be considered during the design and testing of the USV, to avoid problems while in operation: as seen in [ 62 ], the sensor payload must be robust to withstand the water force, especially in the case of underwater sensors since it may damage or knock them out of alignment.
Furthermore, a USV designed for DM should ideally be equipped to detect different types of threats, such as nuclear, biological, chemical and even explosive detectors: for example, the CBRNE sensor system, which integrates Chemical, Biological, Radiation, Nuclear, and Explosive sensors [ ].
Load Capacity : The USV must be able to carry all sensors, batteries, and extra weight, but if the vehicle is expected to work in shallow waters or ebbs, the weight must not be excessive to allow for mobility [ 90 ]. Real Scenario Testing : USVs must be thoroughly stressed and tested in real-world situations before their actual deployment, risking complete mission failure. Disaster sites can be dangerous both to humans and USVs. Currently, exercises with the navy and disaster missions [ , ] simulated in competitions [ , , ] are ways to perform such evaluations and operational validation.
Universidad de Las Palmas de Gran Canaria (ULPGC)
Schneider et al. However, in some cases such as those involving extreme hazards, e. This section presents research and management issues that impact the performance of USV deployment in real disaster scenarios. Ideally, they should be easy to use and deploy.
World Robotic Sailing Championship - Wikipedia
However, a major problem for response teams is that disaster events are sporadic. Thus, there may be a long period between training disaster response teams to the USV use. Therefore, training exercises and competitions are fundamental [ , , ] to test, prepare, and maintain emergency response personnel readiness, mitigating the long periods where response teams and technology are idle. Transparency : Many disaster-related problems can raise concerns from the population regarding effectiveness to respond to disasters.
USVs and unmanned systems in general , combined with social media, can be valuable assets for real-time disclosure of risk management information. For instance, unmanned systems strategically positioned along the coast can be used to forecast and automatically warn affected populations about extreme weather or HAB beforehand, using social networks.
Two main robotic platforms are being used in the project: USVs for flooding prevention and response, and UAVs for mapping areas prone to landslides. However, description of the later is out of the scope of this paper. This section presents the current and near future contributions of the authors in the field of USVs, with focus to disaster applications. These contributions are organized as: boat prototypes Section 7. It can withstand open seas and strong winds and support long-term missions autonomously [ ]. Its main applications will focus on monitoring tasks, which include environmental protection and border surveillance, as well as natural disaster mitigation through long-term extreme weather forecasting.
Lutra boats are approximately 1. We have made modifications for both software and payload of the boats. For instance, we have adapted the Ardupilot autopilot boards to perform basic waypoint navigation, return to launch, and maneuvers compensation by software, integrated with Robot Operating System ROS , a robotics middleware.
We have also improved the communication capabilities between the boats: in addition to the Wi-Fi connectivity, we integrated long-range radios for basic telemetry and inter-boat communication. We have proposed a new payload system for the N-Boat II [ ], including environmental monitoring sensors and real-time and online communications. The new N-Boat II sailboat architecture is vital for autonomous long-term missions since it can stream available sensor data over the Internet to DM and environmental monitoring agencies.
The idea is to prepare the N-Boat II to be used as an early warning system for environmental and natural disasters. This payload is designed to run Convolutional Neural Networks CNN , in applications for real-time obstacle detection and other image processing tasks. Finally, the boats can also carry bathymetry sensors and a system to collect water samples remotely. In the future, we intend to include an ADCP sensor to measure water flow—an important parameter for hydrologic studies associated with extreme weather.
During a flood, the USV might encounter strong water current, winds, water vortexes, and floating debris that might jeopardize the mission. Before testing with real boats, the software designers must have tools to test the USVs safely, in simulation scenarios. Simulators intend to prepare the USV and the rescue team for the actual mission.
Kjell Ivar Øvergård
Presently, there are few open source robotics simulators and fewer which simulate aquatic environments with enough accuracy to simulate a flooding disaster. One of our research goals is to design a robotic simulator to support accurate wind, wave, and hydrologic models to mimic a flooding scenario [ ]. Also, several different USV models e. Some of these USV models are being calibrated with the actual boats described in Section 7.