This paper will cover a description of Autonomous Underwater Vehicles (AUV), Solar Powered AUVs, and Unmanned Aerial Vehicles. Which are all used to collect information remotely and report it back to the base station. The main focus of the paper is underwater communication in regards to AUVs. This is called acoustic communication. I will discuss mobile ad hoc networks (MANET) and the simulation tools I used to implement the these networks (Qualnet, Opnet, AUV Workbench).
KEYWORDS
Acoustic, networking, pressure, routing, manet
INTRODUCTION
The goal of my summer research with Dr. Boadi is to develop a network where mobile aerial, land, and sea robots/vehicles/devices can communicate with each other. My focus of the research is the networking of autonomous underwater vehicles (AUV). Our goal is to have network that is similar to the one depicted below:
We want to build a network like these to create a better surveillance system from different angles (land, water, and sky). The big picture is we want under water devices to be able to communicate with each other and surface devices; we want surface and land devices to be able to communicate with air and underwater devices and vice versa. Most importantly these devices have to be equipped with surveillance equipment. Since there is a lot of research done already concerning land and air networking and focus is the networking of underwater devices or AUVs.
To begin my research, I first had to know type of devices needed to be networked. Autonomous Underwater Vehicle (AUVs) are pre-programmed on shore then deployed in the water to collect data as they navigate the water own it's own. Unmanned Aerial Vehicles (UAVs) are remotely piloted or self-piloted aircraft that can carry cameras, sensors, communications equipment or other payloads. (http://www.fas.org/irp/program/collect/uav.htm ) UAVs are used in the following areas electronic attack (EA), strike missions, suppression and/or destruction of enemy air defense (SEAD/DEAD), network node or communications relay, combat search and rescue (CSAR), and derivations of these themes. (http://www.theuav.com/ )
My paper will focus mainly on AUVs but will include the different topics I needed to understand before starting my principle research. Initially I began my research learning the configuration of a mobile ad hoc network (MANET) because it is the foundation of how the underwater network will be configured. Next was studying underwater sensor networks. Once I had a better understanding of the different network, it was important to find the right tools to simulate these networks. I used Opnet and Qualnet initially but useless for underwater networks. Eventually we discovered AUV Workbench has the functionally to demonstrate our network. Once, I had all this knowledge, we thought it was important to see how others are using AUVs. As a result we went to visit Monterey Bay Aquarium Research Institute (MBARI) and the Scripps Institution of Oceanography (SIO) where they are actively using AUVs to do research.
MBARI uses two types of AUVs. The first class is the Dorado which was created for deep sea exploration. It was used for sea floor and mapping and recently used for the oil spill in the Gulf. In the Gulf it was used to collect sample of water at depths of 5,000 feet to help understand the nature of the oil spill. The Dorado class can stay in the water for 17 hours. The Dorado is depicted below:
The other class of AUVs is Tethys. The Tethys is useful for long range exploration in shallow waters (< 1,000 feet) and is more energy efficient thus it can be deployed two week at a time. Tethys is mainly used for chemical and biological sensing missions. Depicted below is the Tethys:
SIO auv is classified as a glider. Like the Tethys it very energy efficient and used in shallow depth. It was also used in the Gulf. Unlike the MBARI, these are used in a underwater acoustic network, MBARI only uses satellite communication. The AUV is depicted below:
Mobile ad hoc network (MANET)
The UAV network will be considered a Mobile ad hoc network (MANET). In this section will explain what a MANET is and discuss a couple routing protocols that are used.
Mobile ad hoc network are a self-configuring network of mobile devices connected by wireless links. This is important in regards to our research because a significant amount of our devices will be mobile (auv, uav) and we want to be able to network them. To make the ad hoc networks function as proficiently as possible, suitable on-demand routing protocols have to be integrated, to locate efficient paths from a source to a destination, taking mobility of the node into consideration. The Mobility influences ongoing transmissions, since a mobile node that receives and forwards packets may move out of range.[] As a result, links fail over time. For these instances a new path has to be established. Therefore, a quick path recovery method ought to be one of the major characteristics of a routing protocol. There are many protocols that can be used but I will only discuss Ad Hoc On-Demand Distance Vector (AODV) routing protocol and Fisheye.
AODV
The best protocol for routing an unmanned aerial vehicle network, I believe is the ad hoc on-demand vector. This protocol beneficial when keeping your network connected. With the uncontrollable environment of the air (eg. Weather change, possible collisions) the protocol should be able to reroute the network in order to stay connected.
The Ad Hoc On-Demand Distance Vector routing protocol (AODV) is based on distance vector and also uses the destination sequence numbers to determine the freshness of the routes.[] It operates on the On-demand style. AODV needs hosts to know only routes that are active or in use. The benefit of AODV is that it attempts to reduce the amount of needed broadcasts. It constructs the routes on an on-demand basis, as opposed to keeping track of a entire record of routes for every destination. The usage of the AODV protocol for mobile ad hoc networking applications provided consistent results for large scale scenarios [ ].
Fisheye
Fisheye technique proposed by Kleinrock and Stevens to reduce the size of information required to represent graphical data. [] The fish's eye in high detail can acquire the pixels near the focal point. The detail diminishes as the space from the focal point widens. In routing, the fisheye methods translates to keeping an correct distance and path quality information about the close by neighbors of a node, with increasingly less detail as the distance increases.
AUV
The capabilities of the Autonomous Underwater Vehicle are just as important as figuring out how to network them. I've encountered mainly two types of AUVs, does for deep water and shallow water. Typically the deep sea AUV (Dorado) can stay in the ocean less than a day before they need to be recharge and are more expensive. Ideally shallow water AUV would be a better choice because they are energy efficient and wouldn't have to be constantly retrieved to be charge and are less expensive. This way there is abundance of AUV monitoring the waters. The drawback they can't monitor the deep ocean and all depths need to be observed for great surveillance. As of today AUV are mainly being used for oceanic research and little is known if it is being used for surveillance purposes.
Ocean bottom sensor nodes are used to enable applications for oceanographic data collection, pollution monitoring, offshore exploration and tactical surveillance applications.[1] Underwater sensors that are carried on Autonomous or Unmanned Underwater Vehicles are applicable "exploration of natural undersea resources and gathering of scientific data in collaborative monitoring missions". To better optimize these applications they should be able to communicate with each. This is possible by creating a wireless underwater acoustic network.
Wireless Underwater Acoustic Networking
Underwater Acoustic Sensor Networks (UW-ASN) consist of a variable number of sensors and vehicles that are deployed to perform collaborative monitoring tasks over a given area.[ ] To accomplish this goal, sensors and vehicles must organize themselves in an autonomous network that can adjust to the characteristics of the ocean setting. As I mentioned above once the UW-ASN is created multiple application can be optimized including the following:
Ocean Sampling Networks. Networks of sensors and AUVs can perform synoptic, cooperative adaptive sampling of the 3D coastal ocean environment.[]
Pollution Monitoring and other environmental monitoring (chemical, biological, etc.).
Distributed Tactical Surveillance. AUVs and immobile underwater sensors can work together to examine areas for targeting, reconnaissance, surveillance and intrusion detection systems.
There are several disadvantages when a UW-ASN is not enabled, some of these disadvantages include:
Real time monitoring is not possible. The collected data cannot be retrieved until the vehicles are recovered, which can be several months after deployment.
No interaction is possible between onshore control systems and the monitoring instruments.[ ] This delays any modification of the instruments once a change has occurred
, nor is it possible to reconfigure the system after particular events occur.
Failures may not be detected until the vehicle is recovered. Serve failure can lead to the overall failure of the mission that needs monitoring.
The amount of data that can be recorded during the monitoring mission by every sensor is limited by the capacity of the onboard storage devices (memories, hard disks, etc). []
Enabling a UW-ASN is not an easy task. The major challenges in the design of underwater acoustic networks are: i) limited battery power and inability to recharge batteries; ii) the available bandwidth is severely limited [2]; iii) channel characteristics, including long and variable propagation delays, multipath and fading problems []; iv) high bit error rates; v) failures due underwater sensors are prone to failures due to fouling.
Architecture
The following section will describe the communication architecture of underwater acoustic sensor networks. I will describe static two-dimensional UW-ASNs and static three-dimensional UW-ASNs and SEA Swarm.
Static two-dimensional underwater acoustic sensors network for ocean bottom surveillance. These are made up of sensor nodes that are secured to the bottom of the ocean floor. Typical applications may be environmental monitoring, or monitoring of underwater plates in tectonics [4]. Since 2D architecture isn't reverent to my research, you can refer to figure for a basic structure of this architecture.
Static 3D UW-ASNs for ocean surveillance consist of networks of sensors whose depth can be controlled, and may be used for surveillance applications or monitoring of ocean phenomena (water streams, pollution, etc). []
A SEA Swarm (Sensor Equipped Aquatic Swarm) is a sensor cloud that drifts with water currents and enables 4D
(Space and time) monitoring of local underwater events such as contaminants, marine life and intruders.[]
3D ARCHITECTURE
Three dimensional architecture would ideal for enabling a UW-ASN but protocols that I have researched seem highly complicated to implement and optimize. Most 3D architecture appears to use layered protocols which are still in process of being optimized. For every layer (physical, network, data, etc) it has to be changed in order to be used underwater. There is new research about cross-layer protocol in order to cut back on power and redundancy. The lack of a good protocol makes 3D architecture unsuitable for our network.
Three dimensional underwater networks are used to detect and observe phenomena that cannot be adequately observed by means of ocean bottom sensor nodes, i.e., to perform cooperative sampling of the 3D ocean environment.[] In these underwater networks, the floating sensor nodes observes the phenomenon at different depths. One possible solution would be to attach each uw-sensor node to a surface buoy, by means of wires whose length can be regulated so as to adjust the depth of each sensor node []. This allows easy deployment but the drawback to this is, if too many buoys are deployed it can get in the way of navigating ships, or can be detected by enemies and deactivate in cases military usage. The buoys can be easy tampered with or destroyed. Due to this, another approach can be to secure sensor devices to the floor of the ocean. This architecture is depicted in figure two. As mentioned each sensor is secured to the ocean floor and equipped with a floating buoy that can be inflated by a pump.[] The purpose of the buoy is to move the sensor toward the surface. The sensors have a electronically controlled engine that can adjust the length of the wire connected to the anchor thus changing the depth of the sensors. To implement this architecture there are challenges that need to be addressed, including:
• Sensing coverage. Sensors have to regulate their depth in order in order to obtain sampling of the phenomenon at all depths.
• Communication coverage. Information will have to be relayed to the surface station by multi-hops by sensors. Thus, network devices should coordinate their depths in such a way that the network topology is always connected, i.e., at least one path from every sensor to the surface station always exists.[]
Hereafter we analyze the factors that influence acoustic communications in order to state the challenges posed by the underwater channel for underwater sensor networking.
4D Architecture
In SEA swarm architecture, the swarm is guided at the surface by floating sonobuoys that gather the information from undersea sensors using acoustic modems and report it in real-time by the use of radio to a monitoring center. Unlike traditional tethered sensors, a large number of underwater mobile sensor nodes are dropped to the venue of interest to form a SEA Swarm (Sensor Equipped Aquatic Swarm) that moves as a group with water current [] The goal of this architecture is to use a proficient anycast routing algorithm for dependable underwater sensor data relaying to any of the surface sonobuoys, see figure 3. There are numerous key advantages of SEA swarm architecture. First, mobile sensors offer 4D (space and time) monitoring, hence developing dynamic monitoring coverage. Second, the massive amount of sensors helps supply more control on redundancy and granularity. Third, floating sensors increase system re-configurability because they can control their depth; moreover they resurface once depleted of energy and can thus be recovered and reused.[]
Although there are many advantages in enabling a SEA Swarm but it can be challenging due to restricted resources (bandwidth and energy) and mobile nodes. An underwater acoustic channel has limited bandwidth and propagation latency five times greater than the radio channel. Such severe limitations in communication bandwidth coupled with high latency and limited energy make the network vulnerable to congestion due to packet collisions.[] With such conditions, it is important to reduce the amount of packet transmissions for these reasons: reducing congestion and reducing power consumption. To enable such architecture, Depth Based Routing (DBR) can be used, which can be referred as hydraulic pressured based routing. When using DBR, packet forwarding decisions are locally made based on the measured pressure level (or depth) at each node such that a packet is greedily forwarded to the node with lowest pressure among the neighbors.[] The challenge with using DBR is undependable acoustic channel and the occurrence of voids (no nearby node).
Acoustic Propagation
Underwater acoustic communications are mainly influenced by path loss, noise, multi-path, Doppler spread, and high and variable propagation delay.[] Every one of these factors decide the temporal and spatial changeability of the acoustic channel, and make the existing bandwidth of the Under-Water Acoustic channel (UW-A) restricted and considerably reliant on both range and frequency. The table below explains the relationship between range and bandwidth:
Solar Powered Vehicles
With regards to battered powered AUVs, three concerns remain as prime limitations; energy, navigation over extended times and distances, and communication of a user with the remote platform on a relatively real-time basis.[] There is belief Solar powered AUVs can eventually eliminate all three of these limitations.
Solar energy systems permit the stamina of AUVs to be amplified considerably or allow them to stay in the water longer before they have to be recovered because they don't have to be brought to be charged. The capability to carry out long endurance remote monitoring without the call for support ships and the cheap costs of obtaining that data with use satellite communication make the advancement of Solar powered AUVs an significant objective for today's ocean community.
As I've mentioned before there is a lot research in the regards of unmanned aerial vehicle this include solar powered UAVs. Solar engineered UAV will be very useful in our research. They already have a strong history and there is strong knowledge of their capabilities. It will make it easy for long term surveillance and networking. Unlike the UAVs, obviously the AUV doesn't have great access to the sun so will we have to find a solar powered AUV that can be charged fast so it can go back to its task. Having these type of AUVs will the reduce cost of retrieving the AUV frequently.
Limitations
Solar energy system have a small amount of energy and must be able capable to utilize available energy. This confines the payload to low powered devices. The sensor must be able to stay calibrated for a long period of time or be able to be remotely calibrated while in the water. Biofouling of both the scientific sensors and solar panels used to collect energy is a problem.[]
QUALNET V OPNET
Initially, we used Qualnet and Opnet to try to use to simulate our hybrid network. Although, they are good for two dimensional architecture, they didn't offer the tools to simulate three-dimensional architecture that include underwater networks. Qualnet offers 3D simulation. Below I will discuss the disadvantage and advantage of Qualnet and Opnet.
Opnet
OPNET is a well-established and highly professional product. The graphical interface clarifies most of the usual operations, whereas the creation of new models constantly includes the definition of a finite state machine. I thought this fact was difficult at the beginning, but I found that finite state machines undeniably an effective tool for modeling. In OPNET, differently from Qualnet, it results rather easy to describe an application that circumvent part of the protocol stack. This part can be rather significant to speed up and/or to reduce the level of redundant detail throughout the simulations.
Qualnet
Qualnet has several nice features. It comes with a rich suite of models, unlike Opnet it can scale up to thousands of nodes, and its exceedingly modular design makes it easy to modify and/or extend the basic functionalities. Analysis and visualization tools are simple but adequate for broad studies. QualNet is based on GloMoSim but it dramatically expands its capabilities in terms of contributed models and protocols, graphical tools for experiment planning, analysis and visualization, as well as, in terms of available documentation and technical support.[] I found Qualnet to be better than Opnet because it offered graphical and mathematical tools for experiment building, good documentation, possibility of parallel and/or distributed implementations, and possibility to specify a realistic 3D model of the environment. But neither is useful for underwater network simulations but that is most useful is AUVWorkbench.
AUV Workbench
Key features of AUV workbench are:[]
•Open source, Java, XML, X3D graphics
•Mission planning
•Robot mission execution
•Hydrodynamics response
•Sonar modeling
•3D visualization
•Compressed radio frequency (RF) and acoustic communications
AUV Workbench has nightly builds of the AUV Workbench codebase that allows automatic updating of workbench software, keeping mission planners equipped with the latest technology to conduct full-scale mission planning, rehearsal, execution and playback for analysis of multiple robots.[] Below is a screenshot of a AUV workbench demo.
FUTURE RESEARCH
Since UW-ASN is such a new field there is a constant need for improvement. There future research for more efficient protocols and architecture. With regards with my research I have to keep watch of the progress of AUVs engineering. I have to do more research on the power consumption of different devices so the network can be the most power efficient. I will also, research more interface software to handle the data and the remote interaction with the vehicles.
CONCLUSION
There is still a lot of progress to be made to get to our final goal. We need a secure device to work as a medium between the AUVs and UAVs, and there needs to be stronger protocols to handle such communication. From my summer research I have the foundation to build upon our goal. To start implementing our goal, I found out that we need to implement an acoustic underwater network using AUVs and other underwater devices that utilizes 3D architecture or SEA swarm architecture and I can simulate this using AUV Workbench. This way there can be more efficient underwater surveillance.
