Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   What will be the next technical growth leap for the average team? (http://www.chiefdelphi.com/forums/showthread.php?t=154615)

frcguy 02-02-2017 21:15

What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by s_forbes (Post 1639546)
I want to be allowed to use hydraulic systems.

I like my field carpet without oil slicks, thank you very much.

PayneTrain 03-02-2017 07:33

Re: What will be the next technical growth leap for the average team?
 
I think in the next 4 years, lidar technology will get cheaper and more teams will pick it up, hopefully spurred on by a game that is designed to have a longer auto with less uniform targets.

I think the next big hardware leap would be Brushless motors. I think we're ready for that now, just need someone to make a proposal that satisfies the powers that be.

rich2202 03-02-2017 10:44

Re: What will be the next technical growth leap for the average team?
 
GPS. All you need is a few sources of signal.

The FMS system provides one signal source. Being able to lock onto a few more (cell phones?), and you could have a pretty good estimation of where you are on the field.

Working around R68 would be needed - no form of wireless communication to, from, or within the robot. Visible spectrum and non-RF sensors are allowed. With RF being 3 kHz to 300 GHz, you could try to make something outside of that range to send a time mark.

For this year's Airship, we are working on Doppler Radar, however, we don't see that as being useful for future games.

bobbysq 03-02-2017 11:06

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by rich2202 (Post 1639839)
GPS. All you need is a few sources of signal.

Or IR tracking like the HTC Vive, with the field having the tracking stations set up and having the sensor array in the KOP. It could be like the 2012 Kinect station, but less gimmicky, using technology that actually works, and with more applications in the field of robotics.

Jon Stratis 03-02-2017 11:10

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by bobbysq (Post 1639847)
Or IR tracking like the HTC Vive, with the field having the tracking stations set up and having the sensor array in the KOP. It could be like the 2012 Kinect station, but less gimmicky, using technology that actually works, and with more applications in the field of robotics.

I don't know, I wouldn't call this gimmicky... :D

Brian C 03-02-2017 11:19

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by Matt_Boehm_329 (Post 1639530)
Smaller, inexpensive water jet cutters that can cut 48x48x.25in aluminum to .001 in accuracy at 100 in/min for less than $1000 and a tech grant to get one in KoP :)


YESSssssssss!

mrnoble 03-02-2017 11:31

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by Brian C (Post 1639859)
YESSssssssss!

+1

mrnoble 03-02-2017 11:39

Re: What will be the next technical growth leap for the average team?
 
The three biggest advances in the last six years or so are:
  1. COTS frame and gusset options that allow significant customization
  2. COTS gearboxes that are compact, reliable and provide many options
  3. The smaller and more robust control system elements, including the RoboRio and motor controllers

Swerve-in-a-box for less than $100/module, requiring nothing other than attachment to standard VF to make a drivebase, would change mobility. I think that's coming.

Andrew Schreiber 03-02-2017 13:02

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by mrnoble (Post 1639866)
The three biggest advances in the last six years or so are:
  1. COTS frame and gusset options that allow significant customization
  2. COTS gearboxes that are compact, reliable and provide many options
  3. The smaller and more robust control system elements, including the RoboRio and motor controllers

Swerve-in-a-box for less than $100/module, requiring nothing other than attachment to standard VF to make a drivebase, would change mobility. I think that's coming.

I don't think you'll ever see sub $100. I mean, two VersaPlanetaries are like $150, add a wheel and some extra structure, I'd be happy with $200-250 range just because of how many parts they are.

Chris is me 03-02-2017 13:15

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by mrnoble (Post 1639866)
The three biggest advances in the last six years or so are:
  1. COTS frame and gusset options that allow significant customization
  2. COTS gearboxes that are compact, reliable and provide many options
  3. The smaller and more robust control system elements, including the RoboRio and motor controllers

Swerve-in-a-box for less than $100/module, requiring nothing other than attachment to standard VF to make a drivebase, would change mobility. I think that's coming.

If your team doesn't have the fundraising / organizational resources to afford the current COTS swerve modules, the team may not have the software / organizational resources to get a test bed done early enough and thoroughly programmed for effective use. Basically, I think we are already at the point where the limiting factor is software capability.

marshall 03-02-2017 13:40

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by Chris is me (Post 1639917)
If your team doesn't have the fundraising / organizational resources to afford the current COTS swerve modules, the team may not have the software / organizational resources to get a test bed done early enough and thoroughly programmed for effective use. Basically, I think we are already at the point where the limiting factor is software capability.

You don't just click and drag the "Swerve drive go roundy roundy" VIs from the palette onto the canvas?

Lil' Lavery 03-02-2017 13:58

Re: What will be the next technical growth leap for the average team?
 
The next big step is going to be the continued development of consumer-grade machine tools. Additive manufacturing capability is already percolating down to the hobbyist and prosumer levels, and teams have adopted that technology already. As we see similar developments in affordable CNC mills/routers/lathes/combo machines, I suspect we'll see significant strides in machinery capabilities of teams who work out of classrooms and garages.

cadandcookies 03-02-2017 14:10

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by s_forbes (Post 1639560)
It would be a lot of fun. I like fun!

I think we might have different definitions of the word fun. Personally my definition does not include unnecessary field delays (we get enough necessary ones as is). I could definitely see this happening though, provided someone can come up with a system that might be less susceptible than the current pneumatics system to leaks. I can't say I know enough about hydraulics to know how likely this is to happen, but the visceral "I don't want fluid all over my field and someone else's robot" reaction is, I think, going to stick around until someone proves there's a workable solution.

Quote:

Originally Posted by Jared Russell (Post 1639687)
...as well as a few capabilities that have not yet been widely featured on an FRC field, but might in the next couple of years. In most cases here, I think the real opportunity for innovation is from the commoditization of robotics technologies from consumer products and early-stage robotics industries ramping up capability while driving down cost:
  • 2D/3D Scanning LIDAR
  • Solid-state LIDAR
  • Torque control
  • Robot localization / SLAM - Precise drift-free robot pose estimation by recognizing landmarks on the field
  • Fast, accurate vision tracking of less structured targets
  • Brushless motors
  • Hydraulics and hybrid motor/fluid actuators
  • Low-cost, low-backlash gearing

Love this list. LIDAR is definitely going to be a much more real option in the near future. We're definitely in the bumpy/expensive part of adoption now, but this is certainly a thing where consumer products will help drive prices down to the point where we can get modules designed for competitive robotics.

I think localization is going to get there sooner rather than later. I know a few teams that have been experimenting with it in FTC using the image targets and Vuforia, and I could imagine a similar scheme being used in FRC. FRC/competition robotics fields already have a number of features that make them much better for this than "real-world" scenarios, so it's really a matter of either the commoditization effect that you mentioned or FIRST making things a little easier for teams. As a side note, I would love to see a challenge eventually, once localization becomes more mainstream, where mapping is a significant challenge in a longer autonomous period. Perhaps a semi-randomized field, not quite like Stronghold, but similar.

Something else that might become more of a mainstream over the next few years is roller-sized mecanum / "vectored intake" wheels. We aren't quite at the point where they're commodity for FRC (given most of the teams that use them like this are using 4" versions), but I think that will get here soon too. Not quite as far out as anything you've mentioned, but I also don't think they're really highly accessible to average teams yet. I suppose that isn't a leap so much as a step forward, though.

Andrew Schreiber 03-02-2017 14:22

Re: What will be the next technical growth leap for the average team?
 
Quote:

Originally Posted by marshall (Post 1639931)
You don't just click and drag the "Swerve drive go roundy roundy" VIs from the palette onto the canvas?

Code:

import swerve from _future_
But, honestly, with the closed loop functions on the SRX and the VersaPlanetary Encoder stage feedback controls have gotten better. The actual math behind swerves has been published pretty thoroughly. It'll take some loop tuning but at this point I think a lot of the hard parts of swerve have been removed.

There's still UX issues with them but, tbh, there's similar UX issues with most drive systems in FIRST. That'd be a great place to see some innovation - the UX of the operator control. CheezyDrive (SplitArcade) seems to be a good solution for tank drives in that it discourages the Drive, Turn, Drive, Turn approach that is brutal to drive systems. I'd really like to see some dev on similar things for mecanum to address the fact that sideways motion is expensive and as you are driving it auto rotates you drive wheels into position.

Taylor 03-02-2017 14:34

Re: What will be the next technical growth leap for the average team?
 
The next advancement will not be technological, it will be cultural.
With permanent full FRC fields becoming more and more prolific, access to events, both official and unofficial, will increase dramatically. This will create demand for mass media coverage, switching from "aren't these cute kids and their cute robots cute?" to actual game reviews, results, and prognostication.


All times are GMT -5. The time now is 02:12.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi