remote controlled boat

Here is one example of an "industrial" UAV servo. There are numerous types of radio control-ready powerful servos available. You can find many more by doing searches on google and also by going to www.rcgroups.com they have numerous threads on unmanned air vehicles(not toys) ect.
http://www.rogersonaircraftequipmentgroup.com/1.html

There are numerous manufacturers with integrated control systems for UAV's that would easily be adaptable to your project.

another example: http://store.diydrones.com/SearchResults.asp?Cat=20

 

Yea I Understand that but the University is using this project just for the concept of making a boat autonomous and semi remote. It is more about achieving the concept than the practicality. Once the concept can be established then it can be improved. The project is to be used for deep sea rescue...So basically the boat would be remote control for about 50m and then boat becomes autonomous and locate the person in distress via gps etc

 

This is right up my alley...

Here is our R&D boat:
http://www.deepdowncorp.com/docs/EW Videos/StaSys Operational_NEW.wmv

We took an old 12m Trojan Express and added a Glendinning remote control system to it. You can see in the video that we removed the entire helm and replaced it with 3 touchscreens/computers and instead of a set of throttles and a wheel, we just have a joystick.

What is even more amazing (and fun) is that we have a wireless controller that resembles a Playstation controller. We control the entire boat with that, from the horn and anchor to the throttles and rudder. It's a lot of fun seeing the look on people's faces when they see we have no helm and that I'm standing up on the bow driving the boat with a game controller.

If you watch the video you can see the glendinnings on a wooden board (white boxes with little silver moving levers at the beginning). We don't have the video of our wireless joystick up on the site unfortunately.

 

You may not recognize the importance of some of the comments you're getting. If you can learn how to define the requirements for your system, you'll find that's a much more useful skill than what you'll learn from the details of the design.

For an autonomous vehicle, fault tolerance and availability of the control system are some of the most important system requirements. You need to define what an acceptable probability of loss of control is and design to that. You'll find it will drive the mechanical design far more than you think. This goes to every part of the system - actuators, power, sensors, computation, communication.

You need to define system requirements for the controls, like what kind of bandwidth the system should have and what the maximum rates need to be. You'll need to know how much force is required.

You may think defining requirements is a waste of time, but in fact it is the fastest way to produce a working system.

 

Tom's advice is spot on. We've built and operated several semi-autonomous remotely-controlled vessels for open-water model testing purposes. The largest, at 55' LOA, was a control challenge because it was twin waterjet propelled and required remote reversing bucket control in addition to all the controls of throttle, propulsion engine clutch in/out, heading (organic autopilot), stabilization system operation, etc.

For safety reasons, we had to build in a very robust set of failsafe features that provided a remote 'emergency all-stop' capability and an organic emergency all-stop capability that would bring the model to a stop quickly in the event the remote wireless communication link was lost for any reason. The effective range of the wireless ethernet setup we use is about 1000 meters or so. Further to that, additional auto-shutdown features had to be built in to the control system at the sub-system level to handle various failure modes that might occur within the vessel control package itself.

Be very certain of the local rules and regulations too. We 'get away' with our operations of unmanned autonomous craft (although we would probably not fare well if challenged by the local USCG constabulary..) , but our colleagues at the nearby Navy autonomous vehicle test facility are not so 'lucky'; they are expressely forbidden to turn any of their autonamous or remotely-controlled craft 'loose' without a human being on board. When they are taking any PR vidoe of the craft in operation, the person on board has to 'hide' in the bilges somewhere out of camera view.

The rules regarding safe operation of a vessel and standing a proper watch are not waived for autonomous craft. Cause any damage with it and you will certainly ..very certainly..be found at fault and liable. We condut our testing in an area that is well clear of any other marine traffic...and yet there have still been instances where we 'e-stopped' a model just to be certain that nothing bad could occur when a boat happened to transit through our testing area. Part of our established set of operational testing guidelines...

 

4186326,

Building an autonomous / FPV / RC vessel is easier than it sounds and probably harder than you think. In principal the project may fail on the most minute detail (we had a leaking gland that caused havoc), it is up to your own judgment to make the vessel as fail safe as possible.

If you after a proof of concept, than things are simpler as by definition the vessel would be deployed in a controlled environment. The following suppliers have been used to build many of our prototypes:

High current servo controller:
http://www.pololu.com/catalog/product/1393

Radio Trx for telemetry:
http://www.digi.com/products/wireless/xtend.jsp#overview

Microprocessor for control logic:
http://www.picaxe.co.uk

Actuating motors:
Car wiper motor, local industrial / military surplus outlet, Bosch, mobility scooters, UAV grade Servos

Hope I helped.