Team 997 Spartan Robotics "Ultamate Ascent" Climber (vid)

[|DELTA|]

This is team 997's Climber Bot. It has two pairs of pistons, and a linear actuator to move the inner pair forward and backward. I does not currently have any other method of movement, although a modular drivetrain "attachment" is planned and in the works. There is also going to be a "disc dumper" attachment on the middle pair of pistons. This dumping phase of the climb is simulated as a dry run in the video. You may notice that there is a lot of rocking, and that the robot is bashed against the pyramid. This will be alleviated by another pair bumpers on the chamfered front of the 'bot. Any questions, comments, or critiques? Please feel free to post!

http://youtu.be/9lqxi2BZlF8

[swwrobotics]

It says that the video doesn't exist...

[dodar]

Speed needs to be greatly increased. Especially if you want to be a dumper as well.

[|DELTA|]

Hmmm. I will look into the video problem. And, as to the speed issue, I also believe that we are pushing it at 2:09, but that was with multiple mistakes from the driver, and also other unnecessary stops, such as the aforementioned bumps. Also. We are planning to have the entire climb done in an "autonomous mode". That way, we will be approaching 100% repeatability. Also, in this video, I am not certain that the tanks were precompressed prior tho the climb. You may also note that it looses pressure on the way up. That would be due to this.

[cmrnpizzo14]

Originally Posted by dodar

Speed needs to be greatly increased. Especially if you want to be a dumper as well.

I agree. You lose a lot of time just with the swinging. Is it possible that you could shape that end of your robot like a wedge so that you can immediately start pulling yourself up to the next bar? Hopefully the side of your robot would ride up over the rung.

[Austin2046]

Originally Posted by |DELTA|

Hmmm. I will look into the video problem. And, as to the speed issue, I also believe that we are pushing it at 2:09, but that was with multiple mistakes from the driver, and also other unnecessary stops, such as the aforementioned bumps.

you might want to add some sensors/switches and programming to eliminate the need for a driver to control each action. i think that'll speed up the time and eliminate errors.

[ENIAC]

Originally Posted by swwrobotics

It says that the video doesn't exist...

This should be the right link:
http://youtu.be/9lqxi2BZlF8

[dodar]

Originally Posted by |DELTA|

Hmmm. I will look into the video problem. And, as to the speed issue, I also believe that we are pushing it at 2:09, but that was with multiple mistakes from the driver, and also other unnecessary stops, such as the aforementioned bumps. Also. We are planning to have the entire climb done in an "autonomous mode". That way, we will be approaching 100% repeatability. Also, in this video, I am not certain that the tanks were precompressed prior tho the climb. You may also note that it looses pressure on the way up. That would be due to this.

Even still, you should be at level 3 in at most 90 seconds; at least, if you want to be able to score in the pyramid goal. And 90 seconds is a very long time.

[|DELTA|]

Originally Posted by ENIAC

This should be the right link:
http://youtu.be/9lqxi2BZlF8

Thanks, Mat.

[Arpan]

Could be very worth it to use some gas springs to accelerate retraction of the cylinders.

[mlantry]

were there wheels on your robot because to me it just looks like a big white block on the bottom of the robot

Edit i now saw the part about a modular drive

[Nemo]

Amazing job on this thing. Really cool design.

[Pendulum^-1]

Nice work. As others have mentioned, the swinging is a problem, in that it is slowing you down. Perhaps you could deploy a pool noodle as a forward protrusion (either a loop or a stick) to impact the departed bar on the return swings and have a dampening effect on the swinging. Should not take much...just something soft and flexible to touch the crossbar and take energy out of each swing of the robot.

[Akash Rastogi]

Originally Posted by |DELTA|

Hmmm. I will look into the video problem. And, as to the speed issue, I also believe that we are pushing it at 2:09, but that was with multiple mistakes from the driver, and also other unnecessary stops, such as the aforementioned bumps. Also. We are planning to have the entire climb done in an "autonomous mode". That way, we will be approaching 100% repeatability. Also, in this video, I am not certain that the tanks were precompressed prior tho the climb. You may also note that it looses pressure on the way up. That would be due to this.

With more storage tanks, more practice, and by charging your pneumatics beforehand I think you'll have some gain in speed but not too much. The step that I see which takes away some significant time is whatever mechanism you have to move your robot back and forth. For example, at the beginning of the hang, whatever you're using to pivot your robot up off the ground can probably be sped up. The actuation itself is not your issue, in my opinion. Putting a hard stop when you sway back and forth will also help significantly.

[Michael Corsetto]

Originally Posted by Arpan

Could be very worth it to use some gas springs to accelerate retraction of the cylinders.

I'll echo this. This will make your climb sooooo much quicker. Stick a light gas spring in line with your cylinders that will force the hooks up when you have no pressure retracting the cylinder. You will use half the air volume to climb, hopefully cutting your climb time in half!

Love the design, great work!

-Mike