R2C2 Chasing CHiP the Dog

Next task on Robotics Specialization Course was to calibrate the motors, IMU and camera and then to build a differential drive controller to follow an April Tag. But unfortunately the Raspberry Pi v2 camera module I had turned out to be defective. Probably it got damaged while sitting in my car for few days under scorching sun
Fortunately Ruvinda Dambarage, one of my colleagues had a v1 camera that I could borrow.
After working on this late night at workplace all was good. But what could be more cooler than using another robot to test your robot?
So here my WowWee CHiP robot Dog giving me a helping hand at April Tag handling while I do the shooting required for the assignment.

[Published with Open Live Writer on Windows 10]

R2C2 - Raspberry Pi + ROS Robot for Coursera Robotics Course (Part 1 - Assembly)

As part of carrier oriented personal goals for 2016/17 I have being taking the Robotics Specialization course at Coursera offered by Penn university. In this 6 course specialization you have to complete a capstone project related to robotics. There are two tracks you can select for this. Either a MATLAB based simulation project or if you are more adventurous, you can choose to build a real autonomous robot from basic hobbyist robot components you can easily source from online vendors.
Well even though you only need to complete either of the tracks to pass the course, since doing the simulation track is more straightforward and less exotic why not doing both? So here I go ahead with doing both tracks.
First step was to acquire the required components. As a guideline course moderators have provided a wish list at Adafruit for sourcing the components. But I have choose to do a little bit of my own research and use alternative components that I can easily find either locally or for a lower price from ebay.
So here is the components I used.

• Raspberry Pi 3 Model B – ebay (US $36.96) 
• Raspberry Pi Camera Board v2 – ebay (US $20.99)
• 3-wheel Smart Robot Car Chassis with Speed encoder and Battery Box – Local Shop (LKR 1100.00 - Approx. US $7.25)
• I2C TB6612 Stepper Motor PCA9685 Servo Driver Shield V2 Fr Arduino – ebay (US $5.91)
• 10DOF LSM303D + L3GD20 + BMP180 Pressure Field + Gyroscope Module Sensors – ebay (US $6.97) 
• Adjustable Pi Camera Mount – ebay (US $1.89)

When the chassis is observed, even though it had some holes to mount components, none were aligning with the Raspberry Pi. So I got some holes drilled up on to mount the Raspberry Pi before assembling the chassis.

I also flashed the Raspberry Pi image provided at course webpage and configured for remote access since the way I have chosen to place the Raspberry Pi prevents easy access to the HDMI port once the Motor shield is mounted.
After assembling the chassis I had mounted the Raspberry Pi using some plastic mounts and screws. I also mounted the Camera mount base and the IMU using a single screw.

Then I screwed the Raspberry Pi camera to the camera mount and attached it to the camera mount base. And attached the camera cable to Raspberry Pi.

Since I purchased a Arduino version of the Motor shield instead of Raspberry Pi version (purely to save some bucks since Arduino version was much cheaper)  I had to mount it separately instead mounting directly on top of the Raspberry Pi. Then connected the wires from motors to the motor hat.

Then I used a Level converted and some jumper wires to connect Raspberry Pi, Motor Hat and the IMU. Here both IMU and Motor Hat runs in 5V logic levels while Raspberry Pi is Running at 3.3V logic levels.

Note: Actual components used are slightly different than in the above diagram.

Unlike the Adafruit chassis kit, the one I used did not have a top shelf. So I got an ad-hock platform 3D printed. Unfortunately sitting in my car under scorching sun, it got bent down in the middle.

But nevertheless it serves the purpose for mounting the battery packs. I used double sided tapes to secure the battery packs on top of the platform.

Once all done and modifying the controller template to issue a 0.3m/s forward and 0.5rad/s angular velocity as instructed in the course, I could verify that all were connected in order.

I conveniently named the robot as R2C2 which stands for Raspberry Pi ROS (Robot) for Coursera (Robotics) Course. It is also partly inspired from R2D2 (even though former is a droid unlike this which is a differential drive robot)
[Published with Open Live Writer on Windows 10]

Kinect Services for RDS 2008 R3

Now with the release of Kinect SDK for Windows, Microsoft is putting a lot of bang on the next generation human computer interaction technology to the mainstream PC. This time it’s none other than the Kinect Services for Microsoft Robotics Developer Studio - a Windows-based environment for hobbyist, academic and commercial developers to create robotics applications for a variety of hardware platforms.
Similar to Kinect SDK for Windows 7 Kinect Services for RDS 2008 R3 is available at Microsoft Research site for free download. According to the site, “Kinect Services for RDS provides sample services that use the Kinect for Windows SDK to allow access to the Depth and RGB data from a Kinect sensor. In addition to a service for a real Kinect, there is also a service for a simulated Kinect that works with the RDS simulator.”
Kinect Services for RDS 2008 R3 is vailable for download at:
http://research.microsoft.com/en-us/downloads/f8cda115-e9ec-44d1-abcd-3dfdd09d2e77/
There’s already a tutorial of powering the Kinect using iRobot’s internal battery is available at http://www.ros.org/wiki/kinect/Tutorials/Adding%20a%20Kinect%20to%20an%20iRobot%20Create

Now that’s one more reason to buy a Kinect Sensor.
Reference & Image Courtesy:
http://www.eetimes.com/electronics-news/4217801/Robots-get-Kinect-s--eyes-and-ears-
[Posted using Windows Live Writer 2011 on Windows 7 64bit]

Nao Developer Program

:-)

Aldebaran Robotics, a start-up headquartered in Paris has released it’s medium-sized humanoid robot for the developer preview. It’s invitation based only accepting 200 developers worldwide.

It seems however there is a price tag for the participation since the package also includes the humanoid robot itself. One-time entry fee is 3,600€ (including taxes). While Nao probably won’t get any cheaper soon even with a public release for which it is not ready yet it is still bit steep price for me without a sponsor. :(

More details and invitation requests at. http://www.aldebaran-robotics.com/en/node/1674

[Posted with Windows Live Writer on Windows 7]

Day 3 – Fat and Greasy battles…

:-)

Today is the day 3 of the Imagine Cup 2009 world finals. The Big big battle day.

All the competitors had completed their coding. (And by the we learned that Gauna, the fellow competitor who was sick and had been working from his room had to fly back to his country. Anyway his robot will play on behalf of him)

There were battles organized between each player. (So altogether 15 battles) Each battle would consists of 3 matches which each consists of 3 rounds of 60 seconds in length. If you beat the opponent in more number of rounds you win the match. If you beat the opponent in more number of matches you win the battle. A win in a battle give 3 points,  tie is given 1 point for each and loose is 0 points.

Apart from that each participant were given a bonus mark of up to 3 depending on their achievement with the real robots. Byron and Gauna were awarded 3 points each and all the other were given 1 point for this task.

So at the end the leader board was as follows.      

Competitor Name	Country	Battles	Wins	Ties	Defeats	Score
Pablo Gauna (Gauna)	Argentina	5	1	0	4	+6
Byron Knoll (Yokozuna)	Caneda	5	4	0	1	+15
Fuming Lin (ZeroLin)	China	5	3	0	2	+10
Imrich Zivcak (Goldie)	Czech Republic	5	1	0	4	+4
Lukas Perutka (Srotbot)	Czech Republic	5	4	0	1	+13
Pramuditha Aravinda (QUT0268)	Sri Lanka	5	2	0	3	+7

We also had a fun session with the robots after the end of the competition.

Hmm so that’s it. But anyway I was presented with the robot that I was playing with considering my achievements so far. Wow… I like you Robbie…

Competition Update:

Our embedded team and Software design team has been ended up in top 12. Here I have been to 4th place. Now we are keeping our faith regarding the Mash Up team to have a big big prize whose results will be released on world festival on 7th.

[Note: This post will be updated with photos and videos once processed)

[Posted using Windows Live writer on Windows 7 RC]

Day 2 – 24 Hours hectic programming…

:-)
Today is the start of the competition day. Hmm 24 hours of continuous time to compete. Had to rush trough competition check in before repeat competitor briefing since it was really late when we arrived yesterday.
We were given a new version of the Sumo challenger to code. And were explained that the competition will be hosted in simulation environment for the fairness for all the competitors and in addition we were given real physical robots to play with and to create a robot that can do a particular action to a vision stimulus. That would be a mandatory challenge that would give few more bonus points.
Real robots ready to play.

I’m coding and playing with the real robots.

Even this is a 24 ours continuous competition, competitors were allowed to have their time. It was even allowed to go out of the competition hall and have a rest at your hotel room or even work from there.

The most apparent change in the new challenge kit is that it mimics the real robots behavior. now robots do rarely go over other.

So Let’s see back on next day. Big Big battle day!

MRDS 2008 R2 is released

:-)

A new version of Robotics Developer Studio (RDS), called the R2 release, has been released for download. RDS 2008 R2 includes the new DSS Log Analyzer, several additional sensors in simulation, Software Quality Metrics, and various other bug fixes and documentation enhancements.

Express version of the product can be downloaded from the product site at http://www.microsoft.com/robotics/#GetIt There is also an upgrade kit for the Standard edition is available. It is also said that the Academic version is also rolled out to the MSDNAA but couldn’t see it to checkout in either Imagine Cup MSDNAA account or ACM MSDNAA account. Neither could be found in the (Admin)page containing full list of .sdc files. Hopes it would be available soon in either of those places.

[Posted Using Windows Live Writer on Windows 7 Beta]

How the things went on Mars…

:-)
Follow up to the Post:Landed on Mars. Chase for the water begins…
You all know that I have being selected as a world finalist in this year Imagine Cup Robotics and Algorithms and probably would be glad to see my Robot’s working at Endurance Crater, Mars.
Well I have being thinking to post this video a long time but had to always postpone due to various reasons. So Enjoy it now!

[Posted Using Windows Live Writer on Windows Vista]

Landed on Mars. Chase for the water begins…

:-)

Today the list of competitors advanced for the Imagine Cup Robotics and algorithms were published. Hmm It seems 237 are on chase for the water on Mars.

Top 20 Countries are as follows…

TW : 39
CN : 31
FR : 23
HU : 19
US : 16
KR : 10
BR : 9
DE : 8
IN : 8
RU : 6
PH : 5
PL : 5
EG : 5
CZ : 4
AU : 4
JP : 3
DZ : 3
CH : 3
LK : 3
AR : 3

So… Now It’s a matter of water or death.

[Posted using Windows Live Writer on Windows 7 Beta]

Imagine Cup Robotics Round 2 mission released… Wild goose chasing at Endurance crater, Mars…

:-)

Follow up to the post In to the Mars – Neck breaking Opportunity.

On last Friday (27th) challenge kit for the Robotics and algorithms round 2 was released. As it was expected this is re-launch of Mars challenge that was held as part of Robochamps league. But this time the simulation environment is more responsive and the task is bit more spiced up with collecting the data from rocks using a spectrometer.

The sensors we have access includes color Navigation camera, Panoramic camera installed on the ‘head’ of the rover and two B&W hazard avoidance cameras on front and rear of the rover. Spectrometer which is in fact physically a touch sensor and a comlink which can be used to obtain the Pan, roll and pitch of the Rover as well as GPS location and some mission details (Positions of navigation areas)

The drives available includes Horizontal and vertical rotation of head, A Six wheel differential drive, articulated arm on which the spectrometer is attached on the end.

The objective is to navigate trough the given 6 points. Analyze a rock from each area and reach the heat shield. The navigation includes descending into the crater as well similar to the real Opportunity mission at Endurance crater.

The navigation points are obtainable from the comlink interface however the position that the heat shield dropped is bit of random and cannot be known for certain. So it ultimately becomes a wild goose chase to find out the heat shield. Even though it is said to be near the last selected point in practice it appeared to be near means almost half across the map.

With few days of experiments and coding I was able to come up with a solution that could navigate all the points successfully and analyze the required data. However I would certainly have to improve this solution with better algorithms and more reliability and or adaptability over the time. But no problem I have plenty of time left even till the challenge officially start on 20th of April.

More details.

http://imaginecup.com/Competition/mycompetitionportal.aspx?competitionId=22

http://www.robochamps.com/rc/info/Downloads/challenges.aspx

[Posted using Windows Live Writer on Windows 7 Beta]

In to the Mars – Neck breaking Opportunity.

:-)

Since the Imagine Cup Robotics & Algorithm competition Round 1 deadline is extended till April 1st I had decided to go and give a try on the Robochamps previous challenges. So I decided to give a try in the Mars Challenge which seems to be very interesting.

The sample kit is provided only with a VPL solution. However it is not a big deal converting a VPL diagram to C# code I first converted the VPL diagram into a C# solution and started developing an entry.

The environment and the robot recreates a scenario experienced by Opportunity, one of NASA’s real rovers in the Endurance Crater. There are 6 Navigation points that we should visit and finally we must reach the heat shield to complete the mission.

The obstacles are the rocks that are scattered over the terrain and safely maneuvering into the crater. The sensors available are A Navigation camera, Panoramic camera and two hazard avoidance cameras in front and rear. There is also a GPS data from the Communication link. So as an start I designed a solution which uses only the GPS data and travels the navigation points in sequence. The problem with this is the rower may hit a rock or stuck in the crater forever.

Following is the video footage of test run. Well fortunately there wasn’t any rocks on the path and even though the rover rolled into the crater in a neck breaking way finally it landed on correctly.

Hmm I think I should be bit more careful with a multi million dollar scientific robot like this. Well let’s see how we can improve this more later..

[Posted using Windows Live Writer on Windows 7 Beta]

Fat and Bloody (Greasy) – iRobots playing Sumo…

Follow up to the post Coaching iRobots to play Sumo. B-)

:-)

On the 31st Jan the submission SDK of the Imagine Cup Robotics & Algorithms challenge was published. And… this time the opponent – RCChallenger is no more a newbie in the game and really offensive. This causes fast battles between the robots and in most cases one goes other or both topples over. So I had to spend about 2 days testing some new methods to take care of my robot toppling over. Even though almost all the methods that I tried out were incapable of handling the speed the things happenes in a battle, I could turn the odds to my side.

So at the end I decided to go for a submission run hoping that the luck will be on my side. Fortunately during 1st round I could recover from a toppling over situation and push the opponent out. The impact caused a bumper in the opponent to malfunction and on 2nd round it goes out itself. With the 3rd round also going well it was a 3-0 win.

Following is the playback of the match from the recorded file. My robot is the blue colored tagged one.

Any sumo masters to take up the challenge???

[Posted using Windows Live Writer on Windows 7 Beta]

Coaching iRobots to play Sumo. B-)

:-)

Since the Algorithms and Project Hoshimi Invitationals has been removed from Imagine Cup -the world’s premier student technology competition, a new theme, named Robotics & Algorithms which is based on Microsoft Robotics Developer Studio products has been introduced. The task during Round 1 is to program a simulated robot entity simulated based on iRobot Create to successfully defeat an opponent in a Sumo ring.Even though it was not clear at first what to do. A clear step-by step guide has been published on Apps Community for MRDS. After meddling trough the code and changing few parameters and algorithms I was able to successfully modify the sample project provided with challenge kit to compete in a more reliable fashion.

Wow it's a win by 3-0. Well It was just luck in 1st round that my robot win. And during next rounds opponents goes out of the ring by itself :-P Anybody dare for a Sumo round?