'Best' picture of your bot

[HarryBotter]

This contest will award prizes for the best robot picture submitted to this thread, as judged by the editors of Robot magazine. Judging will take place on November 15, 2006 and the winning picture will be featured in the upcoming issue of Robot magazine.

The winning submission(s) as determined by the editors will receive a one year subscription or extension to Robot Magazine.

[killerbotics]

I also have a high resolution of this!

"Titanium, the other white meat."

[killerbotics]

http://server.stuplich.net/~dick/bab/PR03861053029.jpg

At Battle at the Beach I.

[Kevin Berry]

I remember that fight. Thats the one where you spend two days psyching out Terry by adding & subtracting stand off devices, just to mess with his head!

Nice spark distribution on the pic, by the way.

[tswift]

:) Here is a link to my GardenBot robot. It currently can carry a payload of up to 175 lbs and tow a Scotts Fertilizer spreader around my lawn.

http://hometown.aol.com/mgatron/myhomepage/

Future modifications include adding a Vex based garden tool that will enable GardenBot to plant and water plants and tree seedlings for greenhouse, garden centers and tree farming.

[tswift]

:) My Hero 2007 Robot is currently controlled via Vex RC remote control, but I am in the process of upgrading it with a Parallax Servo Controller for autonomous operation. It currently uses 4 Vex motors for 4WD drive and 1 Vex motor to pan the robot's head an owl like 0-180 Degrees or to rotate it continuously for maximum sensor coverage for obstacle avoidance and navigation around the house. Currently it can dance and move about very fast under Vex RC remote control.

You can for yourself the progress I have made so far, if you check out photos of my Hero 2007 Robot using the following link:

http://www.vexforum.com/gallery/browseimages.php?c=8&userid=

[TheDuck]

:) Here is a link to my GardenBot robot. It currently can carry a payload of up to 175 lbs and tow a Scotts Fertilizer spreader around my lawn.

http://hometown.aol.com/mgatron/myhomepage/

Future modifications include adding a Vex based garden tool that will enable GardenBot to plant and water plants and tree seedlings for greenhouse, garden centers and tree farming.

Gardenbot! A robot with a practical use. That's fantastic! This is what robots are all about. Nice job.

[norris56]

Here is my "Follow Me" HexCrawler. Based on the CrustCrawler HexCrawler it's able to follow me around the house using a Thermal Sensor Array while avoiding common objects like walls and chairs using proximity IR sensors and a Ping.

[TheDuck]

A crawler! Very nice. The challenge of putting these together is certainly worth the reward of seeing them creeping along....and in this case, creeping after you!

Something I think we need to keep in mind is that this is the best picture, not the best bot (after all, how could rate bots without having some sort of objective). Don't be afraid to put your bots "in situ" and let us see them in action! Have some fun with it, too.

Keep 'em coming! Tell a friend! hahaha

[JonHylands]

Can you post more than one picture?

- Jon

[JonHylands]

Seeker II is probably one of the best looking robots I have done to date, although I suspect Seeker 3 will take that title once it is painted.

But this is a nice picture...

- Jon

[TheDuck]

Very nice picture, indeed! Not that it has any bearing on taking the picture but, what does it do? Can I guess it's searching for something? ;)

[JonHylands]

Hi Duck,

Seeker 2 is an autonomous mini-sumo robot. You can see a lot more pictures, videos, and details at http://www.huv.com/miniSumo/seeker2

Basically, it hunts for the other robot, and tries to push it out of the ring.

- Jon

[tatwood]

Congratulations to tswift, (aka Dan Ramirez), for posting photos of his Garden Bot, which is even now morphing into his Earth Bot. We credit Dan for creating a design that moves tools, supplies and soil around the house grounds—a bot capable of moving a significant payload, like any self-aware wheelbarrow, and avoiding obstacles, using Basic Stamp and PIC chips. He is upgrading the robot using tips and tricks from his BioSTAMP adventures [www.botmag.com/articles/bionic_stamp.shtml].

Second place runner-up goes to norris56, for a CrustCrawler hexapod that follows its warm-blooded master using thermal sensors, yet avoids walking into objects. The picture is superbly done, and we congratulate norris56 for taking the time to pursue this interesting application that reminds us of the behavior of pets.

As you can see, it’s not just the photo but it's partly the mission behind the photo that counts in our photo contests. Honorable mention goes to John Hylands' impressive sumo bot.

Whose home bot project will we see in the next round? We will find out on February 15, when the next iteration of our home robot photo project will conclude, and at which time the next round of winners will be announced! The lucky winners will receive not just a year’s subscription to Robot but also a lasting presence in our contest archives. Our storage media will be updated as technology evolves, and the history of these contests will endure.

In the meantime, jump in and post your robot photos and give us your thoughts--join the fun!

Tom Atwood, Editor-in-Chief
Toma@botmag.com

[wrighthobbies]

Hi All,

Here is Arty 2.0. Its a line racer that can hit over 1M/sec lap speeds.

http://www.wrighthobbies.net/bots/arty/arty2.0-1sm.jpg

Click Here for a larger version. (http://www.wrighthobbies.net/bots/arty/arty2.0-1.jpg)

And here is a video of Arty on a test track.

http://www.wrighthobbies.net/video/arty2-1.wmv

Arty is built using differentials from 1/18th scale HPI RC car kits. The differentials are attached to a custom polystyrene frame.

Arty uses articulated frame steering, bending in the middle to change direction. This give the bot a swimming motion on the track. A digital mini servo is mounted in the center to control the steering.

The bot is designed for speed; long, low and lightweight with flat track slicks. I take some teasing from the others in the club when I clean the track before Arty runs. Dust can make the tires lose traction.

To achieve high speeds, several design aspects have to be considered.


Mass is not your friend - The heavier the bot, the more difficult it is to accelerate and decelerate and to stay on track in the corners. Try reducing the weight as much as possible. Li-Ion/Li-Poly batteries are one way to help achieve this.
Varying speed - Just as with RC cars, you will need to slow down in the corners and speed up in the straight aways. This is handled by the microcontroller.
Minimize inertial effect - How you steer your bot will affect it's overall speed. Differential steering turns the entire mass of your bot. Balance the mass over the wheel base to improve the turning speed or use a method of separating the mass from the steering mechanism. Steve Hassenplug's robots minimize inertial impact by towing the majority of the mass behind the steering mechanism - http://www.teamhassenplug.org/. He is able to achieve greater than 1M/sec robots using Legos.

[HarryBotter]

Thanks Eddie, a very innovative approach to high speed line following. It wasn't that long ago that these kind of speeds were unachieveable.

Howard Gordon and I have been wondering about the applicability of vision to racing 1/18 scale RC racers autonomously in the RoboRacer category of this forum. I know that John Orlando has been using vision in line following and wonder what kind of results he's having. It seems that vision should allow for better path planning but am unsure if that works in practice.

Do you have any thoughts on the subject?

Cheers, Harry

[wrighthobbies]

Thanks Eddie, a very innovative approach to high speed line following. It wasn't that long ago that these kind of speeds were unachieveable.

Howard Gordon and I have been wondering about the applicability of vision to racing 1/18 scale RC racers autonomously in the RoboRacer category of this forum. I know that John Orlando has been using vision in line following and wonder what kind of results he's having. It seems that vision should allow for better path planning but am unsure if that works in practice.

Do you have any thoughts on the subject?

Cheers, Harry

Harry,

Mel Gehrs has achieved some impressive speeds with his vision bot. He's not up to 1M/sec (yet) but he's just about unbeatable on the advanced line following course. He is using John Orlando's camera and a Rabbit controller.

He is in the process of convertin to Li-Ion batteries to reduce the weight and perhaps increase the speed even further.

I believe the challenge with vision at high speeds is the data rate from the camera. At 1M/sec, things happen quickly. If a camera has a framerate of 10 fps, the bot moves 10cm before the next update.

Perhaps 2 cameras could be used, one looking close and one looking further out and between the 2, decisions could be made at a faster rate. Plus, cameras work better when they are above the track, not close to it. That makes bots top heavy and harder to control at higher speeds.

[HarryBotter]

I believe the challenge with vision at high speeds is the data rate from the camera. At 1M/sec, things happen quickly. If a camera has a framerate of 10 fps, the bot moves 10cm before the next update.

That kind of resolution could certainly pose a problem with a line follower but may have less impact on the RoboRacer where we would just have to stay on the track and avoid fellow racers while going really fast. :)

I may have it wrong but I think the AVRcam has a framerate of 30 fps which also helps a bit.

I think vision could add some significant advantages if some sort of course planning could be implemented, even for line following.

As an example, say at an upcoming curve the outside track barrier has a 12" wide block of a contrasting color at a point perpendicular to the straightaway centerline. By tracking the width of that color blob the robot could compute how far it is from the contrasting color, and start slowing and turning at the appropriate times. This example also requires different colored track boundaries, something suggested earlier.

Or using a similar principle in a line follower, if the camera could indicate where the line is some distance ahead of the bot while it is centered on the line it may start making earlier adjustments that may speed up a change in direction.

Might any of this work?

Harry

[wrighthobbies]

Hey Harry,

That kind of resolution could certainly pose a problem with a line follower but may have less impact on the RoboRacer where we would just have to stay on the track and avoid fellow racers while going really fast. :)

I may have it wrong but I think the AVRcam has a framerate of 30 fps which also helps a bit.

I think vision could add some significant advantages if some sort of course planning could be implemented, even for line following.

As an example, say at an upcoming curve the outside track barrier has a 12" wide block of a contrasting color at a point perpendicular to the straightaway centerline. By tracking the width of that color blob the robot could compute how far it is from the contrasting color, and start slowing and turning at the appropriate times. This example also requires different colored track boundaries, something suggested earlier.

Or using a similar principle in a line follower, if the camera could indicate where the line is some distance ahead of the bot while it is centered on the line it may start making earlier adjustments that may speed up a change in direction.

Might any of this work?

Harry

All of this may work. Mel could explain is method better than I, but I believe he tracks the line as several objects, both far and near. That gives him the aiblity to see further out.

A combination of a camera and IR or sonar sensors could work for racing. Are you thinking of an oval track with walls and the robot racecars running together? Side and front obstacle detectors and a camera should provide enough sensory input for something like that. The big question is how much speed would it allow for.

[HarryBotter]

could work for racing. Are you thinking of an oval track with walls and the robot racecars running together? Side and front obstacle detectors and a camera should provide enough sensory input for something like that. The big question is how much speed would it allow for.

Hi Eddy,

We're exploring this in the RoboRacer category of the forum. And yes, the big question is how to maneuver at speed. :)

Our eventual goal is to be able to compete with RC racers using these autonomous robots. A more immediate goal would be to complete one lap...at any speed. :)

BTW, Tom and I are very excited about your possible participation in our forum and will set up a new category of line following robot development.

Cheers, Harry

[tswift]

:) See Gilbert III exploring Antarctica! My backyard after a recent snowstorm, sporting a Nighthawk wireless video camera.

http://www.vexforum.com/gallery/showimage.php?i=362&original=1&c=member&imageuser=392

[tswift]

:) The prototype Bionic Arm can easily lift 5 pounds using a single Vex motor.

http://www.vexforum.com/gallery/showimage.php?i=404&catid=member&imageuser=392

[engineer]

137RN-! after robogames 2007