Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

TRIBAL (TRIpod meets gimBAL)

Antony Newman

Antony Newman

Well-known member
Joined
Mar 7, 2012
Messages
1,805
Reaction score
6
Points
38
Location
London, UK.
TRIBAL SPIRIT

Note: This is not yet a living breathing product.
This is my attempt to create the best possible design for a mobile Gimbal.

+) The Spirit is the part that can operate autonomously from the Exoskeleton.

+) Its job is to keep the top platform 'level' when you are moving.
See post #18 for a moving image on a Stuart Platform in action.
Tribal%252520Spirit.png


TRIBAL X

The Exoskeleton part of this setup would transfer the weight to the ground - and will neatly hook up to the Spirit

Tribal%252520X.png


AJ
 
Last edited:
MUST HAVE Requirements

+) Rig must be as light as possible
+) Be able to smoothly point the camera
+) Be able to smoothly move the camera over an arc
+) Be able to focus with low latency image
+) Must be quick & easy to set up & intuitive to use
+) Camera controls accessible on body Or through wireless control
+) Rotation around optical centre possible
+) Ability to Stand with gimbal for a long time (without getting fatigued)
+) Ability to Walk with gimbal
+) Silent as possible
+) Gimbal doesn't 'get in the way' of the operator
+) Must have a +- 60 degree horizontal rotational capability ?
+) Must have a +- 60 degree vertical rotational capability ?
+) Support up to 12 lb (25 kg)

NICE TO HAVE requirements

+) Easy to balance
+) Compatible with a support system (to save our arms)
+) Ability to power camera through gimbal (less batteries)
+) Ability to Run with Gimbal (both feet leave ground for finite times)
+) Unlimited rotational capability
+) Support up to 24 lb (50kg) (eg when running)


ANY FEATURES the Ultimate Gimbal would possess?

+) Ability to damp out acceleration of camera (can someone with experience chip in on this...)
+) Ability to create a 'gimbal' path - and repeat the same path.
+) Ability to move camera direction when Not on your body (remote control)
+) (FAR SIGHTED USE CASE) Ability to move independently of operator.
+) (EVEN MORE FARSIGHTED) Ability to navigate terrain.

AJ
 
Why was this Game difficult?

FQSFG48FB0B260G.LARGE.gif



Manipulating something of very low weight

+) Your hands and arms shake because the force required to move your arms is much greater than that required to move the tweezers.
+) The lighter the camera ... the more shaky the footage.


Manipulating something Heavy

+) It doesn't change directly quickly
+) Very little shaking
+) You use most of your energy supporting the mass of the object rather than pointing it.

mqdefault.jpg


AJ
 
Last edited:
VIRTUAL FLOATING HEAVY CAMERA

+) What if the Camera + Lens was floating in front of you
+) What if the Camera + Lens felt like it has a mass of 22 lb (10 kg)

Under this bizarre 'hypothetical' setup:

i) The camera would not shake
ii) You would need to apply a LARGE force to move the camera (and give it velocity)
iii) You would need to apply a LARGE force to slow the camera velocity down (to stop it from moving)
iv) You would NOT have to support the camera weight - and so NO FATIGUE.

7438791054_957065ca30_b.jpg


AJ
 
Last edited:
VIRTUALISED CENTRE OF GRAVITY

If by magic the camera could be made to float - this would not be enough to get perfect pans.

This I understand from comments on the Letus & Ronin threads requires the centre of gravity to be at the optical centre of the Camera + Lens.


AJ
 
Last edited:
TRIBAL : ok .. What on earth is this?

IN A STATIC POSITION

If the Camera on top would initially not be moving (ie static).

If you pushed the camera upwards and then stopped pushing, you would give the camera a Velocity. This velocity would be maintained until the maximum extension of the legs was reached in the Sprit.

- - - - - - - - - - - - - - -

If you attempted to pan the camera - it would alter the lengths of its legs to cause the camera to Want to rotate about its optical centre. This will be because there will be a virtualised centre of gravity controlled in software.

- - - - - - - - - - - - - - -

The amount of force needed to start and stop the camera moving would be much greater than the real weight of the camera. This simulated 'weight' will be created by the software controlling the extension of the legs.


HANDLES THAT CONTROL THE SPRIT

I am still thinking about this (I think it depends on the display technology!)

These will measure the strain (ie force) that you are exerting on the top platform.
+) UP, DOWN, LEFT, RIGHT, ROTATION ON ALL AXIS.

This input could be used to by the software.

AJ
 
Last edited:
I really think you should draw something more, I can't even begin to understand what the hell you're trying to explain here, it feels like I'm listening to a drunken man (or worse) or maybe a genius who doesn't speak the common's language
 
It looks like you are thinking of a mechanism similar to this

In your design, Dolly (slider) shots would be limited the distance between one foot of the tripod and a line intersecting the points of the other two feet if the unit is not anchored.

Panning shots get trickier the further the camera is from the physical center of the rig, and if the camera is angled up or down from level.
 
Enrique,

Thanks for your feedback. I am starting from 'first principals' to design a lightweight Gimbal that might increase the precision of operator control, reduce fatigue, and minimise setup time. I don't have much in the way of knowledge in this area, which will likely mean I am missing out some 'pivotal design points. If I (or we!) succeed, we might just develope the blueprints for furture designs, or even new use cases.

Most people would say I am a bit on the crazy side (what sane person would reverse engineer 12,000 assembler subroutines of the Canon 5D Firmware, write a GUI ... and give it away for free?!).

http://magiclantern.wikia.com/wiki/2.0.4_AJ

And here's my first car ... I couldn't find one I liked ... and so I built my own.
(the guys on Pistonheads were less than amused that the hifi increased the car weight by 10%!)

2b.jpg


AJ
 
Last edited:
Todd,

Thanking you the feedback. Very useful.

+) I need to think about the options like 'using two struts' in post #1 to fix the positions in space of the top plate.

+) There is an option for a Ring piezo motor on the platform itself. This would enable pans left & right without 'tangling' the legs.

+) Agreed : Panning up and down would have a limited range - ie where the system could effectively support you.

AJ
 
Last edited:
I like the concept!To be as quiet and reactive as a gimbal (To smooth the XYZ position realtime), you must use strong direct drive linear motor, and to hold the weight (Around 8 kg by feet right ?) it will draws a lot of amps!

I've made a quick sketch inspired by Todd's link, using direct drive brushless motor with counterweight in red. The model in the middle is a version with more range and the last thing is folded.

Tribal_Concept_v1.png
 
Keeping the Spirit Level

The generic term (I think!) is called a Stuart platform

(A bit like the image below .. but rather than a (heavy) disc on the bottom, 3 x tubes would form a structurally rigid triangle as in the image in post 1)


http://en.m.wikipedia.org/wiki/Stewart_platform

Hexapod_general_Anim.gif




Lower eXoskeleton Section

To allow this to also be mobile (the tricker of the two use cases), I think we should consider a Virtual Platform that roughly the height of the operators hips - and slightly infront.

As a person walks this platform may bob up and down (low frequency).

The Upper sections will need to PUSH against this Virtual Platform, and would compensate for height and XYZ to completely stabilise and control the direction of the camera.

If we can create a system in which the Lower section is attached to an operator while they are walking. If (arbitrary) 75% of the Weight (Of Camera + Battery + Upper and Lower section) all transferred to the operators FEET ... and 25% of the weight is transferring to the hips, we should end up with a platform that feels 'super light' and does not fatigue.

AJ
 
Last edited:
Lower eXoskeleton Section

Something a bit like this ... but with two parallel struts from each hip that 'mate' with the Spirit (top section)

Tribal%252520X%252520pool.png


Pros

+) Crouching & Running possible with this ExoSkeleton

+) System weight can be transmitted to ground (though Hips .. not your Arms, Shoulder and Back!)

AJ
 
Last edited:
LOW FLY

+) To Fly Very low to the ground ... the Upper section (the Spirit) would turn upside down and can be reconnected to the Exoskeleton arms (two parallel struts that are attached to each of your hips).

+) The Handles (or joysticks, or spheres) would unlock and point the opposite direction.

+) Option (not fleshed out) to extend the length of the handles.

I don't have experience to know where the ideal floor position is ... out of harms way in front of you ... but out to far and this would make you want to fall forward. A possible compromise here could be a single CF pole that could go up 45 degrees Behind you with a weight at the end.

Thoughts anyone?

AJ
 
Last edited:
VISOR

Requirements:

+) Ability to navigate real world (transparent visor)
+) Ability to see what the Camera is Looking at in high enough resolution to Frame and Focus (1080p or higher)
+) Very low latency . The lower the better (to enable easy frame accurate focus - and not effect our inner ear balance)
+) Wireless HDMI input
+) Colour contrast / colour quality high enough enable accurate focus.
+) Camera image projected over a large field of view at a user definable Virtual Depth in front of them.
+) Remote control of Stuart Platform from second operator.

Ultimate solution:

+) To use head tracking as an option (combined with joystick input on Tribal Spirit) to control where camera is 'looking'
+) Configurable augmented reality (to assist operator)
+) Configurable Camera screen location (eg fixed in 3d space in front of them in a large 'window').
+) Even more exotic : Eye Tracking as an input
+) 4k image display
+) Wireless repeater to broadcast video image at a much greater range (latency less critical)
+) Comms input from director


Wireless HDMI DONGLE

Requirements:

+) HDMI 1080p input
+) Transmit < 1ms latency to receiver on Visor.
+) <5 Watts power
+) Transfer distance > 30ft

Ultimate solution:

+) 4k Input and transmission over a much greater distance

Visor.png


AJ
 
Last edited:
You must log in or register to reply here.
Back
Top