RoboSavvy Forum

PS One controller modified to control RN-1 via Bluetooth

2007-09-30T17:31:55+01:00

Finaly found some time to work on my Bipeds.

I have experimented on some faster gaits for my Biolod. I conjured up a quick gait which involved rapid sidestepping. It clocked up the following time:

Average Time: 4.8 seconds
Distance: 1 metre
Robot: Standard Bioloid
Floor Surface: Wooden Flooring

That is a whopping improvement over the Robonovas top time - which makes it the new leader of this virtual race. I wonder if the Robonova could also go faster with a side stepping gait?

It would be great to hear of timings from any Kondo owners.

Statistics: Posted by BillB — Sun Sep 30, 2007 5:31 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-31T11:27:23+01:00

I suspect that there will be an error tollerence with my manual timing. I stopped the watch when the robot had definately (well visibly as least) reached the finish line rather that passing the finish line. This may account for the difference our times.

I wonder if there is a more scientific (and easy) way to measure the speeds, breaking light beams perhaps. By easy I mean I don't want to spend half a day soldering components together.

P.S. Nothing sinister meant by the - i was just ready for the next race .

Statistics: Posted by BillB — Fri Aug 31, 2007 11:27 am

PS One controller modified to control RN-1 via Bluetooth

2007-08-30T21:21:25+01:00

virtual race is open to all electromechanical biped humanoid - homosapien simulacrum's

I hope no one thinks i was being sarcastic.

Can any one with who's reading this beat BillB's time and hence challenge Bren.

I've only got my fastest RN time down to 5.4s on full charge.

PPS I have traced the delay problems i had with my "new" remote controller down to my bad programming of the PIC chip...the response now seems instantaneous, except off the analogue ports...

Statistics: Posted by NovaOne — Thu Aug 30, 2007 9:21 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-29T18:19:55+01:00

Oh, Sorry, Bren and BillB (please don't use your twisted evil face, it scares the begebers out of me)

Of course this virtual race is open to all electromechanical biped humanoid - homosapien simulacrum's who access this thread...did i forget anyone this time

Statistics: Posted by NovaOne — Wed Aug 29, 2007 6:19 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-29T09:57:03+01:00

The virtuial race is not over yet - need a time from Robo1( aka Bren). He won the race in the June 06 UK Robo1 (although I was racing with the Bioloid at the time).

Statistics: Posted by BillB — Wed Aug 29, 2007 9:57 am

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T23:10:16+01:00

I've just got my RN back together:

My best two times for the one metre walk using my original compass routine and the original fast walk routine was 22.7 seconds with one course correction and 26.1 with two corrections. (RN had 10 non-complete attempts)

So I tried the July 07 Run routine with no course corrections I got an average time for a completed run, of the two best runs (when RN did not fall over or veer off the side of the course/shelf) of 5.9 seconds (ie 5.7s and 6.1s)

Note: I have not changed the original July 07 code at all.

My main problem was the apparent time lag from pressing my remotes forward button and the RN responding? Not a good thing with races taking 5 seconds.

Thanks for the virtual race BillB you win this time, or should i say again.

I'll try to see if i can improve things before we start the Virtual race thread proper.

Chris

PS
Surface Used: Not very shiny laminate shelf
Timing Method: Manual Stopwatch
Distance: 1 Metre
Fastest Time: 5.7 Seconds
Robot: Modified Robonova 1 (but no mods that make the slightest difference...yet)

Statistics: Posted by NovaOne — Tue Aug 28, 2007 11:10 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T20:58:22+01:00

Just completed some initial speed trials. I raced a Standard RoboNova1 and a Bioloid 18 DOF Humanoid over a 1 metre straight.

First up was a Bioloid in the standard 18 DOF configuration using the standard Walking gait on the installation CD. It got off to a good start, veered slightly to the leftand finished in 12.0 seconds. In round 2 it faired better with a more respectable 11.8 seconds and on the third attempt 11.9 seconds. Average of 11.9 seconds.

Secondly it was the RoboNova using the Jul 07 fast walk gait from the Robonova website. It got of to a cracking start and was very fast off the mark. And completed the 1 metre dash in an astonishing 5.3 seconds then 5.1 seconds then 5.2 seconds. Giving an average of 5.2 seconds!!!!

Although the Robonova clearly won - the Bioloid gait that I used was far from designed for speed (looks like it was designed to be a graceful walk). So with a bit of gait design I would expect the Bioloid to be soon challenging the RN1 time.

Surface Used: Matt Varnish Wooden Parquet Floor
Timing Method: Manual Stopwatch
Distance: 1 Metre
Fastest Time: 5.1 Seconds
Robot: Unmodified Robonova 1

Statistics: Posted by BillB — Tue Aug 28, 2007 8:58 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T19:34:55+01:00

I will pop into the local DIY centre at the weekend and see what standard sizes/surfaces are available. I am keen that we use a design and materials that anyone can easily obtain and assemble.

Statistics: Posted by BillB — Tue Aug 28, 2007 7:34 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T17:52:52+01:00

That leaves us no choice then, we/someone needs to build a running track.....or would a 3.5 metre roll of smooth black vinyl be ok, if it stayed flat when rolled out?

This could be placed on the floor, although we wouldn't get to see any "interesting final leaping strategies"

Statistics: Posted by NovaOne — Tue Aug 28, 2007 5:52 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T16:50:51+01:00

The problem with the Formula 1 (Rat Race) was not so much the surface - but the barrier walls.

The barrier walls work to the advantage of the robot ‘rats’ by bouncing them back in the right direction. But for Bipeds it was a disadvantage. Once a biped lightly touched the wall it added friction to the movement of a foot whilst the other foot moved freely. The net effect was that the robot rapidly turned away from the goal and into the wall.

Statistics: Posted by BillB — Tue Aug 28, 2007 4:50 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T16:19:43+01:00

The competition surface could be standardised to a surface that is cheap & easily available anywhere in the world.

A Laminated/Melamine shelf could be a good standard. Perhaps even with a line painted down the middle to support any guidance systems (by borrowing a trick or two from the Micro Mouse robots).

I agree, I think you guys competed on the schools "Formula 1 (Rat Race)" course....this is the spec:

Track details - See Figure 2
Overall, each lane on the track is 7.2 metres long by 0.4 metres wide. The lane surface will be painted matt
black. A 50 mm wide white tape is laid up in the middle of the lane. The start box occupies the first 375mm
of the track with the start line being defined by 50 mm wide strips of white tape. The track will be laid as
flat as possible, but Rats should be able to cope with a step of up to 3 mm across the track where the boards
are joined, and be able to cope with white lines of up to 3 layers of masking tape.
A start beam will detect jump-starts. The start beam will be aligned with the far edge of the start line.
The finish line is defined by 50 mm wide strips of white tape. The finish line timing beam will be aligned
with the near edge of this line.
The braking area comprises the final 1.2 m of the track. A block of polystyrene foam will be placed at the
end of the braking area. If the foam is displaced by a braking Rat, the 5 second penalty will apply. A catch
net will be provided behind the polystyrene to arrest out-of-control Rats, but some Rat damage may occur.
A 300 mm high gantry will straddle the track at the finish line.

The surface wasn't ideal...It wasn't shinny, I think we should supply our own track though?

A sumo/dance ring and some standard steps would be good as well. (I think i might make some standard steps)

Statistics: Posted by NovaOne — Tue Aug 28, 2007 4:19 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T16:00:12+01:00

The competition surface could be standardised to a surface that is cheap & easily available anwwhere in the world.

A Laminated/Melamine shelf could be a good standard. Perhaps even with a line painted down the middle to support any guidance systems (by borrowing a trick or two from the Micro Mouse robots).

Using a shelf would also be easy to time - simply stop the clock when the robot drops off the end (would encourage some interesting final leaping strategies).

Statistics: Posted by BillB — Tue Aug 28, 2007 4:00 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T15:05:07+01:00

Give me a day, to put my RN back together and I will post my "running" speed.

We also need to agree on a surface.
I have a smooth laminate shelf 300mm wide 1.4 metres long, so my maximum timed length for now is going to have to be 1 metre

Chris

PS Why can't we try a simple 2 stage "relay" race this year, say we each run 1.5 metres?
If you are using a remote, don't mind running the 1st stage and can get the baton into my RN gripper (held at arms length). I could put a photo diode proximity on the gripper to automatically close it and start off my compass based autonomous 2nd stage.
NB
If you put magnets on you robots hands, the baton could have a short steel section or insert to keep the mass down. The max diameter for my little grippers is 15mm.

There's only one problem....We need a pair to compete against?

Statistics: Posted by NovaOne — Tue Aug 28, 2007 3:05 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T14:21:30+01:00

Great idea NovaOne.

Perhaps we should capture times for a standard distance perhaps 2 metres.

It would be interesting if people could also post the length of the leg (if it is not standard). The ratio of the leg to speed would be an interesting unit of comparison.

P.S. I look forward to a race in November. Perhaps in a few years time perhaps we can look forward to a relay race with magnetic batons.

Statistics: Posted by BillB — Tue Aug 28, 2007 2:21 pm

PS One controller modified to control RN-1 via Bluetooth

2007-08-28T13:05:21+01:00

Hi

I'm up for that, If you start a thread in the general section It should be good. least we will know what the other peoples speeds are.

As my bots in bits will not know that the speed is until about 2 weeks.

bren

Statistics: Posted by Robo1 — Tue Aug 28, 2007 1:05 pm