[TheBigDog]

OK, this is the official thread of my rover building efforts. I am not going to be shy about asking for help with things. I have much to learn between now and having a working product.

Goal of Team Tuttle Park: Have a Rover with cameras that can be controlled via the internet.

I am doing a great deal of waffling about exactly where to start. I don't want to waste much money since I don't have too much to dedicate to the project.

Here are some of the things I have so far...

I am now looking at a tricycle style rover. Two powered wheels that do not turn, they will work like tank treads. The third wheel will use a servo to point it to work with the drive wheels. The idea is a very stable platform at fairly low speeds. This is not the rough and tumble RC car I had first envisioned, although it may get there eventually. The center of gravity will be kept low and to the center of the triangle of wheels, giving maximum stability even on uneven terrain.

Here are some of the parts I have picked to begin with...

Wheels and motors...
Tamiya Gear Head Motor & 130mm Pin Spike Tire Set

I want to have maximum battery. I am looking at li-ion poly batteries similar to the ones used on the mars rovers. These look like the best bang for the buck in giving me a power supply. If the 7.2v motors will run nice and slow for me at 3.6v then I will take six of these in parallel, or two blocks of three in series to give me 7.2v.
Polymer Li-Ion Cell: 3.7V 9000 mAh (86100100A) , 9A drain rate - PL-86100100A

That will give me about a kilo of battery plus wiring and insulation, and 660 grams of motors and wheels plus mounting brackets. The rear wheel and servo unit will be another 340 at least, giving me 2 kilos dedicated to driving, steering and power, leaving me 3 kilos for cameras, frame, computer, and radios. I really want to be able to do this in 5 kilos.

I am looking at this control unit...
Sabertooth Dual 10A Motor Driver

It will allow me to use computer control of the unit for precision. The idea is to use the cameras to evaluate the position of the rover on the moon, then give it a series of maneuvers to perform. It follows the set of instructions and then auto sends pictures from its new position. If it senses a problem with the maneuver it will stop and tell you about the problem. The operator will be able to change the warning threshholds to force a maneuver-you don't want to get stuck because a sensor tells you that you can't do something.

I am also conceding the use of a solar panel for self charging the rover. It won't take much to be useful, and helps insure a longer life for the rover.

Another thought I had was to have multiple battery packs and use a switching system. So one pack could be charging while the other was being used. This would also protect us in case one of the packs went bad. It becomes a matter of mass and priorities.

I would like to think of this rover as the mission fail-safe. Our other rovers can be more ambitious in aspects of their design, and what they hope to achieve.

The onboard computer, radio and cameras should be able to operate from the solar power alone if all of the batteries go dead and the rover becomes immobile. We would still be able to talk to it through the lander and get pictures from it during the lunar day.

I will post some sketches later.

Bill

[TheBigDog]

As I was drawing up my sketches I had another idea on the tricycle idea. Instead of powering two wheels I will power the one that I can point. The other two will just coast along. The steering wheel will have 180 degrees of motion. There will be a camera mounted to the steering wheel so we can see where it is pointing. I am also toying with the idea of having two drive wheels close together, like a tractor. Instead of using a servo to point them I will have them powered individually, and move them opposite one another to twist their orientation, then power them together to move the whole thing forward. I think my practical tests will help me to figure it all out. I will start with wood framing and move to metal as I finalize my design.

KISS

I have given myself a budget to play with. I am open to suggestions about how to spend it. I have a QuickCam that I can mount on the rover. Ultimately I will have at least two QuickCams and a 720P capable camcorder. I was actually toying with my idea of having cameras in stereo so we could take the images that they show and create a 3D image of the common parts of the view using simple trigonometry. But I am starting with just one.

The camcorder will be mounted on a pan/tilt device, and I will be able to control both the manual and digital zoom. It should have pretty low power needs considering that it will not need a video display. I will also have a tall boom, I want my comm antenna for talking to the lander to have the best range I can manage.

More later.

Bill

[Jay-qu]

It would be cool to be able to swap batteries in and out of the rover, but I think it might be to complex of an action to bother with. Im sure we could come up with a system that charges fast enough for our purposes. It may be an inconvenience, but once we are there I think the race will be mostly over

[TheBigDog]

Originally Posted by Jay-qu

It would be cool to be able to swap batteries in and out of the rover, but I think it might be to complex of an action to bother with. Im sure we could come up with a system that charges fast enough for our purposes. It may be an inconvenience, but once we are there I think the race will be mostly over

Not switch them in and out. Imagine having two batteries. You have an electrically controlled selector to use one batter or the other. I am not sure if the risk merits the extra mass, but it your battery goes bad it is priceless.

Bill

[Jay-qu]

But using 2 batteries in parallel would be effectively the same thing, no?

[TheBigDog]

Yes and no. Having them in parallel give you the life of both batteries combined. but it one of them goes bad then it spoils the other. If you have your whole power supply as one connected unit then it is only as effective as the worst cell in the bunch. The idea I have is to isolate them so if one begins to have degraded performance, or shorts out, then you can isolate it and still have decent performance.

Bill

[Jay-qu]

I dont know if it would spoil the other, the current would be lower - does that have an effect on the circuitry?

[TheBigDog]

OK. I have settled on the trike design. Low center of gravity will make it nice and stable. One powered wheel that will also steer. Two wheels coast and support mass.

There will be two fixed cameras associated with the steering. One will be attached to the column that the steerable wheel is on, so it will always be pointed where the rover is steering. The second will be fixed to the frame, always pointing where the rover is pointing. In the field of view of each camera is a protractor that is attached to the opposite part of the rover (steering camera sees frame mounted protractor, frame camera sees steering protractor). In this way we can see the in either camera the angle that the steering is currently set to without having to use telemetry on a sensor. We can also derive the location of objects in the common field of view in fairly accurate 3D. If either camera fails we still have the navigational control that we want, but we lose some of the extras we get from having both working.

The steering will be set with a lever to fix it at an angle and then drive for so long to execute an arcing turn of so many degrees. We then set the angle again and make the next maneuver. Every move will be well calculated and deliberate. I am going to start with the basic frame, steering, motive power, and the cameras as I have described and build from there.

My partner on Team Tuttle Park is my son John. Maybe we will have something he can enter in the school science fair this year.

Bill

[freeztar]

Originally Posted by TheBigDog

OK. I have settled on the trike design. Low center of gravity will make it nice and stable. One powered wheel that will also steer. Two wheels coast and support mass.

Imho, an obvious flaw in the trike design you describe is that only one wheel (out of three) controls all movement. If it gets stuck....kaput.
Individual wheel motors would allow for more traction and manueverability.
It also introduces some mobility redundancy (of course at the cost of more motor weight and battery reqts).

In the field of view of each camera is a protractor that is attached to the opposite part of the rover (steering camera sees frame mounted protractor, frame camera sees steering protractor). In this way we can see the in either camera the angle that the steering is currently set to without having to use telemetry on a sensor. We can also derive the location of objects in the common field of view in fairly accurate 3D. If either camera fails we still have the navigational control that we want, but we lose some of the extras we get from having both working.

I think this will work well coupled with your idea of a "blind sensing" system that you described in the original "Hypo-xPrize" thread.

The steering will be set with a lever to fix it at an angle and then drive for so long to execute an arcing turn of so many degrees. We then set the angle again and make the next maneuver. Every move will be well calculated and deliberate. I am going to start with the basic frame, steering, motive power, and the cameras as I have described and build from there.

I still think motor redundancy will add value. I would not take a 3-wheeler with 1-wheel drive out into fine desert sands, but I might consider it if the vehicle was 4-wheel drive (especially if it was AWD).

My partner on Team Tuttle Park is my son John. Maybe we will have something he can enter in the school science fair this year.

Bill

Cool, hey John.

[TheBigDog]

Originally Posted by Freeze

Imho, an obvious flaw in the trike design you describe is that only one wheel (out of three) controls all movement. If it gets stuck....kaput.
Individual wheel motors would allow for more traction and manueverability.
It also introduces some mobility redundancy (of course at the cost of more motor weight and battery reqts).

This is a two edged sword. If you have 4wd and one of the four fails your whole vehicle can be disabled. So while you have more redundancy, you also have a higher probability of seeing a failure.

As far as getting stuck... We will have to take measures to insure that it is capable of unsticking itself. That may involve having an auxilary drive system for the rear wheels, very low gear, very low power, that is only used in case of emergency, but not part of the primary drive system. If we can do it with a precious few grams it may well be worth it. If we can make the thing pretty stuck proof under 1G, then .17G should be a breeze.

This is also the primary reason that I want to have multiple small rovers, so we can experiment with the engineering and not have all of our eggs in one basket when we can afford not to.

Bill