Open questions will appear on this page for discussion. ย After consensus is reached (or after a long time with no discussion), the question will be moved to the resolved category along with its discussion.

Open Questions:

EQ2013-06-11a: On-board sensor/controller/actuator communications hardware: wired or wireless?

Resolved Questions:


5 comments on “Electrical
  1. Jon Riley says:

    There are likely to be readers with varied knowledge when it comes to these matters of communications and sea water, so it may help to be more gentle without intending to offend those with the knowledge and experience.

    Although wireless LAN (802.11) works fine over water in the air, (provided the antenna is above the water), unfortunately radio waves don’t work very well underwater. Radio frequency signals are as good as absorbed by the water and are not an effective option for communicating with a submerged vessel.
    The oceans permit very low-frequency signals to roll through tremendous distances, but they are not useful to us since they have such low bit rates as to be useful only to communicate 1 or 0.

    If your goal is an autonomous sub, then you will need to give the sub enough ‘smarts’ to operate in the environment untethered. However, as batteries deplete, bilges flood, missions complete etc, then the sub must be equipped with the basic capability to return to base or at least return to surface. Naturally, there are many ways to achieve these basic requirements and that is where innovation and original creativity are applied.

    Communications back to a ‘mother ship’ (whilst submerged) is constrained by the environment. If you are ok to regard a tethered remotely operated vehicle as autonomous, then you could be ok to use a suitable cable (e.g. coax, unshielded twisted pair UTP or optic-fibre). In each case the useful length is determined by the mass of the tethering cable and the bandwidth you want. Optic-fibre provides high-bandwith and amazing distance. It also provides the option for multiple fibres and real-time video etc, but it comes at a relatively high cost. As the length increases, the engineer has to also consider the cable drag, buoyancy and how that alters the auv’s manoeuvrability and missions.

    If your definition of autonomous means un-tethered, then your sub needs to have the ability to operate without communications back to the command and control centre and to operate it’s mission with some level of autonomy.

    By my reckoning, just because a sub has a link back to base doesn’t mean that it isn’t autonomous. Autonomy is more readily evident if the communications with base is lost.
    If you start your planning with autonomy in mind, then treat communications back to base as a luxury. Depending on your mission, you may wish to keep the best communications channels open whenever possible. This may mean sending video and audio streams back to base. Streaming video and audio does require modest bandwidth from the communications channel/path. Even compressed video (e.g. skype or lync etc) requires hundreds of kilobits per second. Streaming audio could be also hundreds of kilobits per second. The minimum & maximum depends on the quality you want. Streaming data back from a sub may require hundreds of kilobits of bandwidth.

    Providing useful control input (to set course, depth and waypoints etc), and receiving some useful telemetry data may require a whole lot less bandwidth than streaming video and audio.

    If your needs are lower bandwidth then other communications options are possibly worthy of exploration. Sound frequencies (from say 100Hz to 20kHz) travel huge distances through water and travel approximately four times faster than in air. That means audio signals are a useful method of communicating underwater.

    Many marine creatures use sound to communicate underwater. If we are to study their environment and operate our systems with them, then we might do well to be sensitive to the damage we could cause to the hearing or behaviour of those creatures before we deafen them with high-volume sirens/horns! We wouldn’t want to see poor creatures washing up on beaches due to our insensitivity.

    It is possible to modulate data using audio frequencies, but the frequencies impose physical limitations on the total bandwidth.
    If you are old enough to recall the dial-up modems most folks used to connect to the Internet in the 1980’s and early 1990’s then you will appreciate the usable bandwidths of maybe 28kbit/s when modulated signals are carried via 2400Hz.
    Even so, low-bandwidth signals that can be carried over long distance are very useful in communicating with subs.
    If all you need to communicate is a command, and all you need to receive is confirmation then maybe acoustic communication is sufficient. After all, you may store video and audio from the mission on-board on HDD or SD for later retrieval and analysis.
    Acoustic signalling can be very helpful to navigating autonomous subs. For example, synchronised acoustic beacons within audio range can be used for navigation. Placing three or more beacons on the bottom at known locations allows an autonomous sub to use them to navigate by triangulation (geometry). The distance from each beacon can be precisely calculated (by measuring the time elapsed to receive the signal from each). When I said synchronous, it means they share/use a common clock. That doesn’t mean they need to ping at the same instant. Provided the sub knows when they will ping, then it can calculate the distance from each beacon and so place itself spatially in an envelope.

    Sub position can be determined via several methods. Unfortunately, GPS doesn’t work underwater either.

    To start, let us assume that you will know you pressure (depth) and distance to the sea floor (depth sounder). Using a digital 3-axis gyroscope (not much bigger than a postage stamp) you should be able to accurately determine your heading within 5 degrees, determine pitch and roll too. Using a accelerometer of similar size you should be able to work out whether you are rising, falling, drifting in current etc. The speed through the water may be callibrated on the revolutions of the props. If you want to use an original propulsion method then you will need to work out how to relay your sub’s speed through the water back to the bridge. ๐Ÿ™‚

    Consider also modulated light (e.g. lasers) briefly as another possible communications option for short distances.
    The reader will be very likely the user of an infra-red remote control for the reader’s television? It is reasonable that such a solution would also work underwater for short distances (subject to the power of the luminance being modulated).
    If you have done any scuba-diving you will be likely aware that visibility underwater can vary widely in different locations. In some cases of it may be difficult to see a torch placed directly on ones face-mask due to water visibility being less than a typical coffee with milk. Other locations or seasons the visibility can be 30 metres or more.
    Naturally, modulated light communications is line-of-sight and directional… so your sub and ‘mother ship’ both need to know which direction precisely to point their beams respectively. (Not so easy if you are in doubt and both are moving).

    Furthermore, there are many marine creatures also that depend on their vision (and their healthy retinas) in low light conditions for their survival. If you don’t respect the marine creatures and want to use modulated lasers underwater, then you should not be exploring their environment.

    We’ve covered a lot in one post. Perhaps it would help to sort further posts on communications generally into the tethered and untethered varieties?

    There is a lot to discuss in the untethered, but the tethered comms is fairly straight-forward if you cabling is robust, impervious to water at the junctions, low mass (without being buoyant) and flexible enough to support your manoeuvrability and spooling requirements. If you are in the untethered camp then you will probably realise that there are many scientists and engineers that have been here before you. Probably a lot drowned in the early days, but there are many navies with submarines in the oceans today that depend on vital submarine communications for their operations and survival.

    We didn’t yet cover the option yet of dragging a comms buoy. Although it is possible to drag a buoy with an antenna to communicate back to base (send video, audio, telemetry etc), it does limit the depth and movement of the auv.

    It is more probable that a releasable buoy may be deployed to rise to the surface to signal the sub’s location if/when the sub suffers an unrecoverable fault. A communications buoy that can be released from the sub is more a tertiary engineering consideration for an unmanned auv. It is more likely that the sub may emit a periodic acoustic chirp or ping if it hasn’t depleted power (or reached hull-crush depth) but is incapable of surfacing as a method of signalling it’s location and condition. If well-engineered with a suitable alternative (secondary) buoyancy option, then catastrophic failure of the buoyancy system is really only a contingency. For this reason, deployable comms beacons or towable comms buoys are less likely to be seen on aovs unless they are designed for exploring only in shallow coastal areas, shoals and estuaries.
    I hope this is useful to those who aspire to their own projects and need the basic information to start. I regret not providing some links to the authoritative sources on some of these matters, but maybe in future. ๐Ÿ™‚

    I respect all of you who attempt to explore the marine environment. Two-thirds of the surface of the Earth is submerged and we have only explored small areas of it. Exploring the oceans safely and efficiently requires developing smart auvs.

    I hope all your auv experiments return safely to your lab at the end of your missions. ๐Ÿ™‚
    Kind regards, Jon

  2. Cecil Johnson says:

    i think wireless wouldnt be the energy efficient way to go . i would go for a semi autonomous system rather than relying on remote control

  3. Mike Godin says:

    Sorry, I should have been more clear in the original question statement (it has now been clarified). While the typical approach is to go with wired, wireless does have some advantages (i.e., fewer wires, fewer physical ports on the main controller, ability to traverse a cm or so of water), and of course some disadvantages (higher power, likely higher sensor & actuator cost).

  4. Phil Marcelino says:

    Given that the one of the design goals is to be left out overnight, I would think that we would want a wireless solution. That implies a much larger cost. I worked on a system that used the iridium network, I know this isn’t an option for this project, but given that the device will be left out a mechanism to be able to recover it will be needed.

    This I think will have to be a wireless system.

    • Mike Godin says:

      Good news on the IOIO front. The developers have generated an Android driver that allows running the IOIO in device mode on an Android device or USB hub. This should simplify adding multiple IOIOs to the system.

Leave a Reply