Legacy Forum: Preserving Nearly 20 Years of Community History - A Time Capsule of Discussions, Memories, and Shared Experiences.

Walking in a Straight Line

Anything that doesn't fit our other forums goes here.
5 postsPage 1 of 1
5 postsPage 1 of 1

Walking in a Straight Line

Post by BillB » Thu Jul 19, 2007 4:49 pm

Post by BillB
Thu Jul 19, 2007 4:49 pm

After competing in the UK RoboOne 5m dash it became apparent how difficult it is to keep humanoid robots walking in a nice straight line. Each time mine viered off course I had to stop and do a time expensive turn on the spot to back to the correct position then start walking agin. I had to do this many times along the dragstrip.

I would be interested to hear from anyone who has sucessfully managed to keep their humanoid robots going in a straight line and what strategies they took. Using gyros come to mind, or perhaps physically weight balancing the bots, or even setting up a user control system so that the robots walking gait can manually be nudged/corrected mid stride.
After competing in the UK RoboOne 5m dash it became apparent how difficult it is to keep humanoid robots walking in a nice straight line. Each time mine viered off course I had to stop and do a time expensive turn on the spot to back to the correct position then start walking agin. I had to do this many times along the dragstrip.

I would be interested to hear from anyone who has sucessfully managed to keep their humanoid robots going in a straight line and what strategies they took. Using gyros come to mind, or perhaps physically weight balancing the bots, or even setting up a user control system so that the robots walking gait can manually be nudged/corrected mid stride.
BillB
Savvy Roboteer
Savvy Roboteer
User avatar
Posts: 232
Joined: Sun Aug 06, 2006 1:00 am
Location: Hampshire, UK

Post by Robo1 » Thu Jul 19, 2007 6:44 pm

Post by Robo1
Thu Jul 19, 2007 6:44 pm

Hi Bill

After that event I stripped my bot and have ordered some new parts (I will bet you next time). I'm going to use a compass http://www.sparkfun.com/commerce/product_info.php?products_id=7915. The theory is when I press the move forward button it will take the compass bearing then when it is moving forward I will make the gait change slightly to keep the bot going in that direction.

The two methods I'm going to try are making one leg take a shorter step Or some form of shifting the weight (momentum) around.

But for some one like your self, can't you use the analogue input from the xbox controller and make the gait change due to the input. You could also use this for the navigation too.

Bren
Hi Bill

After that event I stripped my bot and have ordered some new parts (I will bet you next time). I'm going to use a compass http://www.sparkfun.com/commerce/product_info.php?products_id=7915. The theory is when I press the move forward button it will take the compass bearing then when it is moving forward I will make the gait change slightly to keep the bot going in that direction.

The two methods I'm going to try are making one leg take a shorter step Or some form of shifting the weight (momentum) around.

But for some one like your self, can't you use the analogue input from the xbox controller and make the gait change due to the input. You could also use this for the navigation too.

Bren
Robo1
Savvy Roboteer
Savvy Roboteer
Posts: 501
Joined: Fri Jun 30, 2006 1:00 am
Location: UK - Bristol

Post by BillB » Thu Jul 19, 2007 8:57 pm

Post by BillB
Thu Jul 19, 2007 8:57 pm

Bren,

Look forward to seeing your compass setup in action.

I will probably go for the analogue joystick option - the theory is that the analogue XBOX controller direction will dynamicaly change the offset of the Bioloid hip joints, and the distance from the analogue joystick centre could determine whether I use a slow or fast walk (or even running?) gait.

Well that is the theory anyway :?

Bill
Bren,

Look forward to seeing your compass setup in action.

I will probably go for the analogue joystick option - the theory is that the analogue XBOX controller direction will dynamicaly change the offset of the Bioloid hip joints, and the distance from the analogue joystick centre could determine whether I use a slow or fast walk (or even running?) gait.

Well that is the theory anyway :?

Bill
BillB
Savvy Roboteer
Savvy Roboteer
User avatar
Posts: 232
Joined: Sun Aug 06, 2006 1:00 am
Location: Hampshire, UK

Post by NovaOne » Thu Jul 19, 2007 9:49 pm

Post by NovaOne
Thu Jul 19, 2007 9:49 pm

I am also working on the compass option using a I2C Devantech CMPS03.

Other than EEPROMs Ive never connected other devices to the I2C bus on my RN yet.

Oh and I also need to master basic reliable walking without falling over. :cry:

I keep you posted this weekend :arrow:
I am also working on the compass option using a I2C Devantech CMPS03.

Other than EEPROMs Ive never connected other devices to the I2C bus on my RN yet.

Oh and I also need to master basic reliable walking without falling over. :cry:

I keep you posted this weekend :arrow:
NovaOne
Savvy Roboteer
Savvy Roboteer
Posts: 405
Joined: Thu Jul 05, 2007 7:30 am

Post by NovaOne » Sun Jul 22, 2007 4:54 pm

Post by NovaOne
Sun Jul 22, 2007 4:54 pm

I think the I2C on MR C-3024 are for alternative program EEPROMs only:?:

I believe this is the same problem as KurtE had over a year ago :oops:

I have switched to using 2 spare digital i/o ports for SCA and SCL, which should work according to page 93 of the RN manual.

I will post my program in the RoboNova section this evening.

Chris
I think the I2C on MR C-3024 are for alternative program EEPROMs only:?:

I believe this is the same problem as KurtE had over a year ago :oops:

I have switched to using 2 spare digital i/o ports for SCA and SCL, which should work according to page 93 of the RN manual.

I will post my program in the RoboNova section this evening.

Chris
NovaOne
Savvy Roboteer
Savvy Roboteer
Posts: 405
Joined: Thu Jul 05, 2007 7:30 am


5 postsPage 1 of 1
5 postsPage 1 of 1