Competition

Our robot came in second place in the competition. There were five groups and the group with the largest amount of points (earned per round) won the competition. Our robot performed excellently during most of the matches, but our major flaw that put us only in second place was that our robot still did not have enough traction to push certain robots.

Our group took recordings of matches with our robot against opposing robots. Here are a few examples of our robot's successes and failures.

Competition match. Our robot is the left one.

Competition match. Our robot is the lower left one.

Competition match. Ours is the left one.

Competition match. Ours is the left one.

Competition match. Ours is the right one.

Week 9 Update

Currently, we have the NXT sumo robot ready for the competition, which will be tomorrow, June 1st.

There were some more minor modifications in order to prepare the robot for it's weight requirements as well as it's ability to perform in the competition.

After further testing the robot in the ring, we noticed how the ultrasonic sensor apparently sensed an object in front of it even if there was nothing there. We then realized that it was sensing the ramp when it is deployed to the ground. To deal with this dilemma, we raised the ultrasonic sensor by one "unit" of a LEGO piece with the intention that it will no longer sense the ramp. The modification proved to be a success, as it now performs exactly as we intended to.

We also had to adjust the weight of the robot in order to meet the requirements of 1 kilogram and below.
Luckily, there were extra pieces on the robot that were not needed and were removed to reduce weight. As of now, our robot meets the required weight limit, as shown in the picture below.

Our current robot weighing in at 999.9 grams (0.2 grams away from exceeding the weight limit).

In addition, we have also had the opportunity to test our sumo robot against another group's robot. The video below shows the robots competing against each other.

 

 Our robot (left) competing against an opposing robot (right) in a practice match.

Week 8 Update

This week we mainly focused on calibrating the code and modifying the body in order to perfect our design. For example, we realized that our robot was a little bit too large from front to back, so we had to relocate the front light sensor in order to meet the size requirements. This change also altered the readings of the light sensor to a degree, however a simple recalibration of the values fixed the problem.

One other modification to the robot body is that added more weight towards the front of the robot. Before the light sensor was able to detect the outer edge, the skid we have in the back of the robot was knocked off the ring and was caught on the edge when the robot tried to move forward. This meant that we had too much weight focused towards the rear of the robot. To balance the weight, we placed more LEGO pieces towards the front of the robot. While doing this, we were also constantly measuring its total weight after each major addition in order to stay under the requirement of 1 kilogram.

With the competition happening next week, our group is confident that our sumo robot will be a major success. After testing it with a basic NXT robot (shown below), we feel that our robot has a sturdy design that will win us the competition.

Here is a video of our sumo robot (right) testing against the basic robot we built (left) to get a general idea of how our robot will fare in the upcoming competition.

Week 7 Update

After making some additional modifications to the design, we now have a complete NXT sumo robot.

One modification was the removal of the ultrasonic sensor located at the back of the robot. Previously, our group had trouble programming the robot with two ultrasonic sensors gathering information at the same time as seen in the video uploaded last week. We ultimately decided to remove the back ultrasonic sensor and have our robot rely solely on the forward sensor as well as the two light sensors.

We have also created and attached the ramp which is used to disrupt the opposing robot's maneuverability. Once the opposing robot is within a certain range determined by the ultrasonic sensor, the ramp will rise up with the intention of providing defensive and offensive protection.

The code has now been modified for the newly modeled robot. After testing the code, the results proved that the code is adequate enough to be able to compete.

The only major objective we have next week is to adjust any part of the code that needs to be adjusted in order to be prepared for the upcoming competition.

Below is a video of our robot running on the most current version of our code. We used a water bottle to get a basic idea of it's accuracy since there were no other NXT sumo robots to test with at the time.

Basic test of our robot's code.

The video below is humorous test run of an idea that occurred on the spot. We decided to upload this because we may create a new offensive tool based off this idea to aid us in the competition.

A random idea that the group came up with. Our idea was to trap the opposing robot in a box and push it out of the ring easily.

Week 6 Update

We have finished designing the main body of the sumo robot.We've also added the two ultrasonic sensors for detecting the opposing robot and two light sensors to detect if it is heading towards the edge of the ring.

In addition, we also tested the code we had created the previous week before. As of right now, the ultrasonic sensor on the front side of the robot is reacting exactly the way we want it; when it senses anything in the range it was coded with, it will keep moving forward until the front light sensor detects the edge of the ring. The back ultrasonic sensor, however, is currently not responding properly. We will edit the code further that will make the ultrasonic sensor perform the way we want it.

Below are pictures of the robot as well as a video of the current program.

Final Design of the robot. The front of the robot has an ultrasonic sensor aimed forward and a light sensor pointing down.

Back side of the robot. Includes an ultrasonic sensor pointing forward and a light sensor pointing down.

 

A test run to see how our code is translating into the robot. The back ultrasonic sensor was not working the way we intended it to.

Week 5 Update

        As it turns out, the treads we ordered from the website were discontinued. As of this moment, the design to incorporate treads is scrapped. Our new idea is to shorten the length between the two wheels (two powered by one motor) in order to better distribute the weight of the robot.

        After taking measurements of our robot, we learned that it exceeded the length limit it was allowed to be. As a result, we have decided to create a new design that will be more compact. We are also, at the same time, addressing the problem of the instability the wheels previously were experiencing.

        We have also developed a general code that should program the robot to do what we want. Since we don't really have a functioning model yet, however, we can't officially test the code at this time. We plan that this new, revised design will be functioning by next week and will be ready to test our program on.

Week 4 Update

We have a few modifications to our robot design.

First off, it was decided that we will use a normal light sensor to detect the outer ring of the arena as opposed to the color sensor.  At first, we tested the color sensor by creating a code that will show us the results of it directed at a white color and a black color. The results were not consistent and  therefore unreliable for our needs. We've performed these same tests with the light sensor. The results from this particular sensor gave us a distinct difference between the two colors. We tested this further by creating a simple code that told the robot to move indefinitely on a black surface until the light sensor detected the white edge. The light sensor successfully differentiated between the two colors and stop just as it sensed the difference.

Construction is still under way. Our wheels have trouble performing a zero point turn because there is not much weight keeping in on the ground. We have ordered extra LEGO parts online that will be used in out design.One of these parts include treads that we intend to incorporate into the design to going better traction.