Archive for the ‘Grand Challenge 2004’ Category

CajunBot to visit Governor Blanco and legislature

Sunday, May 30th, 2004

CajunBot is scheduled to visit Governor Kathleen Blanco and the Louisiana legislature on June 2nd. CajunBot fans and supporters are invited to witness the event. Please plan to be in the Capitol at 1pm. CajunBot will make a grand entrance driving through the scenic lawn (walkways) in front of the majestic State Capitol. She will then climb the steps of the Capitol and greet Governor Blanco on the landing. Team CajunBot and major sponsors will follow CajunBot. Family and friends of the team members will be present in the guests. After meeting the Governor, the team will proceed to the Capitol to visit with the legislature.

Arun Lakhotia

Danny Majors moving to Florida

Sunday, May 2nd, 2004

Danny Majors, leader of CajunBot Chase Team and our marketing guru, will be moving to Florida in the next few days. From the time when Danny first said ‘Think about it’ for having a CajunBot float in the Mardi Gras parade, Danny has been our marketing guru. He initiated this journal site; plastered the trailer with CajunBot banner and images; and initiated the idea of 4×4 bumper stickers (soon to become reality). While Danny is moving to Florida, he is not moving away from the team. He has promised to be back for the next Grand Challenge. Our best wishes to Danny on his move to Florida.

Arun Lakhotia

Thanks Terry Sampson, cajunbot.com webmaster, for his support

Sunday, May 2nd, 2004

Terry Sampson, our webmaster for cajunbot.com, is moving on to other projects. Terry has been a great webmaster. He designed and developed the website on a very short notice, and kept it up to date. I love the picture gallery he put together, telling a story through pictures. Our many thanks and best wishes to Terry.

Arun Lakhotia

Recollection of the big day

Sunday, April 18th, 2004

Here are my recollections of the Big Day, March 13, 2004, the DARPA Grand Challenge day. Having completed the QID, we had already achieved the minimum objective of the team – pass the QID. That had been our ‘next step’ since December. We knew of some of the deficiencies of our design, such as, we had enough gas to run for 10 hours, but that may not have been enough to complete the course. The Mechanical Engineering team was not happy about the design, i.e., not carrying enough gas. We had told them, ‘if CajunBot ran for 10 hours we’d be going wild.’ And they would retort, ‘if we did not win because of gas, you’d blame us’. But that is engineering trade-off. There is no point having a chain in which some links are super strong, when we know it has links that are weak. The chain in our case consisted of vehicles, mechanical add-ons, electronics, computers, and software. Our electronics boxes were hobby grade, what one would design as a first version, tied together with tie-wraps. It would have been a miracle for CajunBot’s electronics to stand the vibrations for 10 hours. The computers were off-the-shelf, nothing special. They could potentially crash bringing everything to a halt. The final piece, software, was put in place just about 48 hours earlier. There was not enough time to get it rugged anyway. There are a million of reasons why it could have failed. We went into the final race with our fingers crossed. We had the satisfaction of meeting our minimal objective for the challenge. For the race, the minimal objective we had established was to get out of the range of the cameras, say about a mile. The evening before the race, we had the opportunity to test CajunBot get out of the chute, and go about 200 yards. After a bit of tuning, it had performed flawlessly. Morning of the race we were reasonably confident that CajunBot will start-off well. Anything after the first few hundred yards would have been lagniappe. Having studied the terrain we were very confident that she may go the two mile, or so, loop around Slash X. We were more worried about how she will perform once she got into the mountains, for she had never actually climbed any mountains before. The day started very early for us, around 2 AM. We had to be at Slash X around 3:30 to get the route definition file. Our hotel in Hesperia was around an hour away. And we had to factor in the possibility of emergency on the road, such as a flat tire. Even though it was really early, we were surprisingly awake. Slash X was already humming. There was activity all over the place. It had the atmosphere of a carnival in the desert, minus the music. May be a desert camp is a better analogy. Soon after 4am, there were helicopters buzzing around. There was this large military helicopter, probably a black hawk, came down blowing dust all over the place and spilled a group of servicemen in uniforms, who walked, rather rushed, straight to start. Soon after a white helicopter appeared. It had a huge bulge on the left side. Took us sometime to figure that was a camera. Two more helicopters showed up, not as sophisticated as the white one. They had their doors open and cameramen hanging out. There was also media running all over the ground. There was excitement all over the place. Only two team members were permitted to accompany CajunBot to the start area. That means all all the other team members were turned into bystanders. Two of my kids had arrived in California to see the race. I wanted them to experience first hand this historic event. We made our way to the spectators area. This place was already full. The Caltech crowd appeared to be the largest. It was a chilly morning, and there were three Caltech supporters, all boys, with their chests bare and painted with their teams name. There were supporters holding banners. The place had the feel of an athletic event, except for the cheerleaders. I know how to navigate through crowds, after all I have traveled in the crowded trains of Mumbai. Very soon my kids and I were right in the front, against the wire fence, overlooking the chute. The teams were lining up behind the chutes. The evening before we had already rehearsed the process. Most teams consisted of academics, students, and computer geeks, people who are hard to tame to follow process. Yet, the flow was very smooth. Guess the helicopters, the uniformed personnel, and just significance of the occasion created an ambiance where these people were willing to follow orders. The event started with some introductions, national anthem, and more talking. There was a lot of important stuff being said, but by then I had tuned out. I was busy watching time. Sandstorm was to leave at 6:45 (if I recall correctly), the next bot in exact 5 minutes after that, and so on. This was the first time we were just spectators. During the QID we were too busy getting our own pieces together. Now all we could do was see, muscle space with other spectators, and take pictures. There was too much cacophony to talk, or to call Pablo and Firas, who were to responsible to bring CajunBot to the chute and let go. So here I was, squeeze against a wire fence, along with my kids, reduced to a spectator. Which incidentally, was not bad either. Since there is nothing much more I could do, I was also relaxed, a little detached (sort of Hindu thinking kicking in). The countdown started, the green flag was waived, and Sandstorm took off. It was just a maginificient take off. Breathtaking. This is the first time I had seen it running. It was impressive. Sandstorm completed the loop around Slash X in no time and headed into the mountations. The team and the university has sure earned its place. We were seventh in the list. By time CajunBot came to the chute, five of the bots were already out of the race. Sandstorm was the only one still running. Virginia Tech’s bot broke down at the front gate, Team Ensco’s had tossed over, another bot had run into some problems further down, and DARPA had shut the course until it was cleared. Here we were, CajunBot in the chute, with five bots out of the race, and our start delayed because of some jam on the track. It was becoming cleared why DARPA called it a grand challenge. My kids had started singing the ‘are we there yet’ tune. The helicopters and the media were not exciting anymore. They were waiting for CajunBot to start, and we did not know when it will start. Besides, the sun had also gone past its cute stage in the morning, and was getting nasty. My eyes were glued on Pablo and Firas. I was trying to read their expressions and body language. Trying to know what they were doing. Whether everything was alright. The green flag waived, the commentator announced CajunBot, her siren came on, the amber lights blinking. Man, I said, this is the flashiest, bot. CajunBot’s siren was louder than other vehicles and so was her flashing light. We had taken each of DARPA’s requirements seriously. The rules called for warning sound of ‘at least’ 85dB sounds at 10 feet. Our siren was rated for 100db at 10 ft. We had taken the rules literally. The sirens stopped, as they were supposed to. CajunBot lurched forward, breaking into motion. What a relief. She was moving. She moved towards the start gate. Slowed down, and corrected itself. Another relief. She’s seeing and steering. Every little indication of her capabilities was relieving. It meant all the links in the chains were holding together. The cameras were pointing right at the start gate. CajunBot went past those cameras. Another sigh, She did not break down at the gate. But then CajunBot kept going straight. Turn, turn, I muttered under my breath. Why aren’t you turning. She kept going straight. I could see some of my team members on the other side of the track. It appeared she was heading to them. What is it, I was thinking, this is not the right time to say hi. She kept going into the wire fence, into the bushes, and against the cement barrier. I was waiting, thinking, she will turn. She always does. For some reason I did not hear her cry, the siren, the loud siren had gone off. But I did not hear the siren. I ran towards the area where she was stuck. Scott was already there. He was always everywhere. I asked him what happened. He said, “The sires are on, that means the kill switch is pressed.” Now I could hear the cry. Its killed itself. Damn, the kill switch. The requirements said make it accessible, and we did. We took every requirements literally. The media was all over the place. We were in the spotlight. There were bulbs flashing and photographers scrambling to get picture from a vantage point. What is with with CajunBot, she attracts all the media attention. This is one time we could do without the attention. There was nothing more to do. We pulled CajunBot out of where she was trapped and pushed her for a little while. We were beginning to regain composure. No, we will not push her back. There is nothing wrong with her, just the kill switch was trpped. She can go one her own. Josh got the joystick back into operation, powered up CajunBot, and drove her back to our tent. As we walked we were thinking, “we wish this and we wish that.” But no one ever said “who did this and who did that.” There was no finger pointing, or blame to stick. That’s what made this team so special. Whenever there were emergencies, and there was one every day, the usual reaction was: “This is where we are. Where do we go from here.” This time the answer was obvious: Time to go eat. We were all running on empty. There was a Mexican restaurant we had eyed on the way. It looked like an interesting place. We packed our bags, put CajunBot in the trailer, and headed to the Mexican restaurant.

Arun Lakhotia

Team CajunBot celebrates with a BBQ

Thursday, March 25th, 2004

Thanks to Patrick and Rita Landry, Team CajunBot will celebrate its journey with a BBQ. Patrick and Rita will host the team at their residence. Now that we have had a break of 10 days, this event will start off the next phase of the project. We have gained considerable momentum in the past few months. We will get our heads together to see how to leverage that momentum.

Arun Lakhotia

CajunBot delayed – Will arrive on Monday

Wednesday, March 17th, 2004

Due to some mechanical problems with the trailer, CajunBot will be delayed. We are expecting her to arrive on Monday, March 22, at 1:30pm.

Arun Lakhotia

Correction: CajunBot arriving Thursday, 1:30 at Dupre Library

Wednesday, March 17th, 2004

CajunBot will arrive at Dupre Library on Thursday at 1:30pm. She will be on display until the rest of the day.

Arun Lakhotia

Correction: CajunBot arriving Thursday, 1:30pm

Wednesday, March 17th, 2004

CajunBot will be arriving on Thursday, 1:30pm at Abdalla Hall. Earlier I had incorrectly announced arrival for Wednesday.

Arun Lakhotia

Team CajunBot back home, CajunBot on her way

Tuesday, March 16th, 2004

Team CajunBot arrived at Lafayette airport as scheduled. They were pleasantly surprised to be greeted by Dr. Steve Landry, Vice President Academic Affairs, UL Lafayette and Channel 10 TV. CajunBot is on its way, being driven by CajunBot Chase Team – Watson ‘Trouble’ Arnold, aDam Dupre, and Danny Majors. They are scheduled to arrive in Lafayette on Wednesday. I will post their expeted time of arrival soon.

Arun Lakhotia

Team CajunBot thanks C&C Technologies

Monday, March 15th, 2004

Its been an incredible ride, one that would not have been possible without C&C Technologies (www.cctechnol.com). Early in September 2003 when we were still toying with the idea of participating in the Grand Challenge, C&C provided us with a C-NAV differential GPS system, the blue dome on CajunBot. This system has 10cm accuracy, a great improvement over 1m accuracy of garden-variety GPS. In February, C&C went a step further with the POS/MV Inertial Navigation System. This system added two more GPS to CajunBot, the round white plates at the front and back. Using the three GPS, we could CajunBot can accurately sense its location, heading (direction), roll, pitch, and speed over ground. The equipment provided by C&C is valued at around $130K. But their most valuable contribution went beyond that. C&C also permitted Pablo Mejia and Curtis Calloway, engineers with C&C, to work on this project. Pablo brought with him enormous experience in navigating autonomous submarines and boats. Within a week of joining us, Pablo assumed the key role, developing software to get data from the sensors, to control the actuators, and to drive CajunBot in desired direction. The success of CajunBot could not have been possible without participation of C&C and Pablo Mejia.

Arun Lakhotia