Curiosity Inspired Long Range Remote Vehicle - a project I want to get OCUK started on.

Associate
Joined
23 Mar 2006
Posts
1,739
Added my details to the spreadsheet.

I think donations should wait till a proper plan has been created with actual costs of things.
 
Soldato
OP
Joined
7 Aug 2004
Posts
10,987
Spread sheet is looking healthy now guys :)

So I’ve been thinking of a way to organise this and have come up with a plausible way of contributing for each person, my rough proposal is:

1) We all discuss, design, thrash out the ideas for the rover online, shared folders, drop box etc to all share work and designs (will take a few weeks)

2) After a few weeks of everyone’s input on the design (how everything will work with each other) we will have an agreed final product – my key proposal is – make the rover modular, in that each person’s expertise area is utilised to the best in that they will make their particular module of the vehicle (much like Nasa contracts out different parts of their space vehicles)


3) Once all completed, all ship the parts to OCUK Rover Tech HQ for construction of the final product (I’m am happy to invite all who are interested to come to HQ for a day of fun construction and a few beers of course!)

4) The next idea I came up with is vehicle testing stage one – ill mount the physically finished rover on a stand in a spare room, hook up all the required computers to run it and create a VPN/Remote desktop to the systems, in addition the ‘master server’ I will hook up a HD IP camera with full and clear view of the rover from various angles. This way all the software guys can log into the system 24 hours a day and fine tune the rovers operational software (create, debug, experiment)……also they can control the rover of course starting its wheels turning, steering abilities, lower/raise masts and cameras on the rover and so on…….basically a live, real life online access zone for EVERYONE to be able to access.

5) Ill be happy to do any physical ‘real world’ changes to the rover….such as if a programmer requests to move from mains powered electric sources to the onboard power systems, happy to provide a environmentally controlled env, i.e. a fancy way of saying ill put a large fan in the room to try out wind turbines or ill install high watt lighting to test the solar panals.

6) Once this has been done (will take as long as it takes), I will make an indoor test environment in my garage, i.e. sand and rocks all over the floor so again everyone can remote log in and test the rover, try out the parts they constructed, of course ill make the rover accasable via the net/wifi for speed and also give options of engaging UHF / mobile phone connections. (im also happy to get a couple of mobile smart phones of everyones request, android, win phones etc, ill also take out a 3 mobile unlimited data contract and also pay for that)

7) For fun im also happy to ‘mount’ the rover on a stand as mentioned in point 4, but in a chest freezer with controllable water spray nozzels, i.e. a harsh environment to test in

8) Stage 3 of testing is putting the rover ‘in the field’ (literally :p) in Snowdonia on private land (as mentioned I know the owner, it will be ok) – this will then be an operational long distance controlled rover if it all works.

9) We can either stop there or put it in a highly remote place in Snowdonia for a few days once we know it wont get stolen, I have some friends that regularly go into Snowdonia on a daily basis and will know of locations where no one really gets to.



How does all this sound? It gives an excellent testing environment through progressive stages of development and also allows everyone to get hands on, even if there 100’s of miles away.


EDIT: Also for the software (this is not my area of expertise so im sure you guys will think of a much better way of doing this!), but we could program it nice and simple perhaps, and enable minimal text msgs to be sent to the rover in addition to a 3G data connection. For example we could make commands such as

TR30,MF5

Which the rover could read as Turn Right 30 degrees Move Forward 5cm

Or

TL20,MB1

Which would make it turn left 20 degrees then move back 1cm

Or

MF3

Move forward 3 cm

How does that sound? Simple small commands, I suggest we make it with just one speed, not varying speeds, i.e. a low range motor with high torque and make it move say 1 mph at all times, so when it executes commands we know how fast it goes, or we could program in 2 speeds, L and H, low and high and code it,

What you guys think?
 
Last edited:
Caporegime
Joined
22 Nov 2005
Posts
45,167
what are you planning to do with the machine once all its roving has came to an end?

if your asking people to donate then I feel its a question we should know the answer to
 
Soldato
OP
Joined
7 Aug 2004
Posts
10,987
what are you planning to do with the machine once all its roving has came to an end?

if your asking people to donate then I feel its a question we should know the answer to

Defiantly, iv not really thought to be honest, we could post it around between everyone so they can have a play with it ?

Then after that..... we could start project 2 and use components from it to do another build?

Personally my 'dream' would be to approach Copenhagen sub orbitals and try and make a pitch to become there high altitude/space robotics division - I like that they are an open source development team. - who knows, 20 years they could build a rocket that could orbit/go to the moon and we could build the tech / satellites / rovers that it would take there - pie in the sky at the moment but got to have something lofty to aim for.
 
Soldato
Joined
20 Aug 2010
Posts
8,201
Last edited:
Associate
Joined
11 Nov 2003
Posts
1,696
Location
South Yorkshire
I think this is the best platform tbh, but really it all depends on what the developers want to code in.
http://www.st.com/internet/evalboard/product/250863.jsp
If machine vision is a requirement of this rover (for autonomous point-of-interest based navigation) then we'll need a platform that's far more powerful (unless that work is done off-board on a separate DSP or something).

OP, are there plans to get a forum set up for this kind of discussion? A lot of this will get lost if it's constrained to a single thread.
 
Soldato
OP
Joined
7 Aug 2004
Posts
10,987
If machine vision is a requirement of this rover (for autonomous point-of-interest based navigation) then we'll need a platform that's far more powerful (unless that work is done off-board on a separate DSP or something).

OP, are there plans to get a forum set up for this kind of discussion? A lot of this will get lost if it's constrained to a single thread.

Good point, I guess we could setup an off site forum, but id like to keep it on OCUK if possible, shall I contact a mod and put forth the idea of a sub forum for this?
 
Soldato
Joined
20 Aug 2010
Posts
8,201
If machine vision is a requirement of this rover (for autonomous point-of-interest based navigation) then we'll need a platform that's far more powerful (unless that work is done off-board on a separate DSP or something).

Yeah, that stuff would eat the cycles. Maybe it could be at the ground station though instead of being onboard?
 
Soldato
Joined
28 Oct 2006
Posts
12,456
Location
Sufferlandria
Good point, I guess we could setup an off site forum, but id like to keep it on OCUK if possible, shall I contact a mod and put forth the idea of a sub forum for this?

I think an off-site forum would be best and you could keep this thread (or start a new one) as a build log to keep everyone updated with progress without the masses of technical discussion that will be needed.
 
Soldato
OP
Joined
7 Aug 2004
Posts
10,987
I think an off-site forum would be best and you could keep this thread (or start a new one) as a build log to keep everyone updated with progress without the masses of technical discussion that will be needed.

Cool will let a few others chirp in with whats the best idea/forum to use here then sort something out.

Completely randomly I thought of using iPad 3 batteries as the main power hold, as far as I can tell they are high capacity for a small volume, wide and flat but a sandwich or 2 of them on the base of the rover chassis would spread its weight nicely
 
Associate
Joined
17 Feb 2006
Posts
898
Location
Fleet
EDIT: Also for the software (this is not my area of expertise so im sure you guys will think of a much better way of doing this!), but we could program it nice and simple perhaps, and enable minimal text msgs to be sent to the rover in addition to a 3G data connection. For example we could make commands such as

TR30,MF5

Which the rover could read as Turn Right 30 degrees Move Forward 5cm

Or

TL20,MB1

Which would make it turn left 20 degrees then move back 1cm

Or

MF3

Move forward 3 cm

How does that sound? Simple small commands, I suggest we make it with just one speed, not varying speeds, i.e. a low range motor with high torque and make it move say 1 mph at all times, so when it executes commands we know how fast it goes, or we could program in 2 speeds, L and H, low and high and code it,

What you guys think?

We will have to design the command messages carefully to allow extensibility, but also to make sure we transmit commands using the least possible amount of data. I dont think we can use string based commands as they have the potential to be very wasteful in terms of data transferred and processing (manipulating strings is very slow compared to integers). We should probably just use TLV encoding for the commands and associated data eg.

Command size=1byte (enough for 256 commands)
Length size=2bytes (enough to specify a value length of 65535 bytes)
Value=variable size

eg:

command enumerations
turn_right =1
move_forward=2

TR30,MF5

So in your example in binary the commands would be:-
Code:
T        L                V
00000001 0000000000000001 00011110
00000010 0000000000000001 00000101

This would be very quick to process and also easy to debug. Im sure the other software guys will have some more ideas though :)
 
Last edited:
Soldato
Joined
28 Oct 2006
Posts
12,456
Location
Sufferlandria
If your sending the commands by text message, wont they be parsed as strings anyway?

If it's going to be sent over some sort of UHF from a base station, you could have the base station computer interpret the string commands and transmit them in whatever format anyway.
 
Associate
Joined
9 Dec 2008
Posts
387
Edit: I realized after posting this that it might cause an annoying distraction to the serious discussion going on here. Moved it to the space flight thread...
 
Last edited:
Man of Honour
Joined
13 Nov 2009
Posts
11,595
Location
Northampton
I'm not entirely sure how much torque a stepper motor would produce but if we're talking small movements such as a few centimeters at a time these would give a very accurate resolution giving us the best possible control, as I feel it would be quite crucial given the environment.

I think tracks would also be the way forward as it will be a lot easier to get implement steering control rather than a tradition rack and pinion type system.

Although they could be geared to give us the best possible torque output as stepper motors from what I understand can move a fair bit faster than 1mph(0.447m/s)
 
Last edited:
Soldato
Joined
19 Dec 2006
Posts
9,260
Location
Saudi Arabia né Donegal

I think you left out

0) Mission brief and regional research :p

There should definitely be a move towards either getting some forum space here or some of our own (or better yet a temp forum here until we can get our own up and running) so we can start nailing stuff down. I've taken the liberty of spamming 5UB with just such a request to see if it's possible, I hope that's okay?
 
Back
Top Bottom