Today we built a 4WD chassis using 80/20 motors and Andymark Compliant Wheels. We constructed this drive-train to test if it was the safest and most efficient way to conquer the crater rim. By experimenting with the drive-trains, we can narrow down a different type(s) of a harvesting system we want to use. We are still debating whether using a zip tie harvester that extends and reaches into the crater or a zip tie harvester that requires us to go inside the crater.
The drive train we created is shown below. It includes custom cut metal bars, cardboard, and compliant wheels.
Today, we decided to set up the field for our school and understand the proper configuration of the field. We made sure the entire team learned how to set up the field and understand where the robot will be and the tasks it had to complete.
We also took down notes for how to score in both Autonomous and TeleOp. To start the JPL design process, we had to fully understand the game first.
Here are our notes:
We also observed the texture of the rim and tried to find wheels that have enough traction to go over the rim. Afterward, we went over the business plan and prepared to present it to the robotics program.
Kristoffer Kaufmann’s take on the Business Meeting:
“All of the teams had very different Business Plan, which shouldn’t have surprised me, but I ignorantly assumed that most of the cost would be general team costs, such as Sign-up and Regional costs. While the Sign-up costs were the most expensive item on their lists, they weren’t the majority, meaning that the variation in costs was quite large. I think that the constructive criticism provided to each team really helped, and I hope that we will all be better prepared at the second one.”
Today we visited Plymouth Elementary here in Monrovia, CA and showed off our Mecanum chassis to the kids. We also allowed many of the kids to drive the robots around so they could see what it’s like and know how enjoyable it is. Crowds of children rushed over when we brought out the robots. Loki, another robots team at Monrovia High School also brought their Mecanum chassis as well. While we had people supervising the children, we explained to the parents the benefits of First and encouraged them to get their kids involved in First. We introduced them to joining First Lego League and going to the First website to find out more information. We also explained that Monrovia middle school and Monrovia high school had great teams that their children could eventually be involved in to learn new things.
After finding a mechanism that we became particularly fond of last meeting, we began to design other mechanisms that complements the Radial Dump Truck. We concluded that using two of the arms would allow us to score into both the silver and the gold sides without having to move over to the other side, fulfilling one of our initial requirements of our robot. Although, in order to do this efficiently we need to figure out a design for a harvester that could separate gold and silver minerals. We began to create sketches of separating mechanisms to feed into the two arms.
Today we continued to prototype different ways we can harvest the minerals efficiently and safely. We searched for a mechanism that could both collect and separate minerals in order to be able to score both gold and silver. We tried to create a zip tie harvester with a a flap that changes the path of the mineral and sends it either left or right. We also experimented with a mechanism that uses two vertical zip tie stands that either spin left or right, sending the mineral in the intended direction.
Today we officially began experimental prototyping and figuring out the little details that we did not foresee as necessary in the beginning of brainstorming. During the down selecting process, concerns were expressed on the radial dump truck mechanism. We were afraid that by using the a dump truck that was one long arm would create too much momentum and make the mechanism inaccurate. After creating a prototype of the arm, we found that the speed of the minerals did not really matter as long as they are launched at the right angle.
Below are measurements for the Radial dump truck that we recorded to plan out possible configurations for the mechanism to fit with the rest of the designs in an 18″ square cube.
Radial Dump Truck(length): 18 inches
Radial Dump Truck (width): 3 inches
After realizing the potential behind the mechanism, we began to document the measurements at which the arm would have to be at in order for it to function properly.
After testing the arm several times, we found that the ideal location for the dumb truck to be at would be 17″ high and around 2″ away from the lander.
After today’s meeting, our entire team has become fond of the design. We plan on continuing to further design a robot off of this idea.
Last meeting we began to cut down the amount of viable options for our mechanisms. We separated each mechanism into 4 different categories. The different categories were Drive trains, Depositors, Harvesters, and Lifters. Out of each brainstorming idea, we got rid of all the designs with the exception of the two to three we found were best. The criteria we followed in order to find the best ideas were whether or not they were manufacturer-able, efficient, and were compatible with our game strategy. Below are the outcomes of our selective eliminations.
6 wheel drop center
Radial Dump Truck
Dump Truck on a linear slide
Hanger on a linear slide
Hook on a wheel
Zip tie harvesters
Tennis/ Half tennis harvesters
Although we had already done this, we made sure to explain to everyone who had missed the crucial down-selecting meeting in hopes of having the most amount of educated hands on deck. We also began doing light prototyping with cardboard. This was an experimental phase of our design process where we began to see the smaller details that must be figured out in order to properly design our mechanisms.
Today we met to apply the our prioritized key features to our brainstormed ideas. We worked feature by feature, applying our priorities to each and choosing one or two ideas to start prototyping.
We ruled out suspension and treads because we didn’t see a need to enter in to the crater. We’re going to prototype a six wheel drop center and mecanum drivetrain to see how each behaves when partially on the crater, and we’re going to leave the door open for DARPA wheels — Chase is interested in trying them out — because we think they would be fun to build and not detrimental to our game strategy.
Our battle cry this year is “make new mistakes,” and claws on articulated arms are a known old problem in FTC, so we eliminated them from all lists. (They also appear in Land & Latch and Harvest.) We spent a short amount of time testing with throwing elements in to the lander to see if that was a viable strategy. It didn’t work reliably even with a human throwing from 6″ away, so we eliminated both flickers and flywheel shooters. Air cannons seemed difficult to build, as did an expanding conveyor belt that would allow us to score from the crater. This left us with dumping elements either with a linear slide or radially-actuated arms. We will prototype both.
After eliminating the known-difficult builds — claw on an arm, suction cups and vacuum harvesters — we were left with three general ideas: zip tie harvesters — with and without an arm to extend them in to the crater — various tennis-inspired harvesters and a “thavma tube,” which is a 4″ diameter tube with rubber bands on one side that will allow elements in when forced down on them but will not let elements out based on gravity alone.
We decided that the Thavma Tube, while effective, would probably be relatively slow at collecting. Fast harvesting is one of our highest priorities, so we eliminated this option.
Unfortunately, our team members who championed the tennis-style harvesters during our brainstorming session could not be at today’s meeting. We left those ideas open as possibilities in case they want to prototype them.
We’ve build zip tie harvesters before and we know that they’re both effective and very fast at harvesting these types of element. A zip tie harvester is going to be our main prototyping approach for harvesting.
Land and Latch
Again, the claw was eliminated from consideration due to its “old mistake” status. Anti-gravity and drones would both require FTC-illegal components. Our favorite idea is simply putting a hanger on a linear slide and powering it both up and down. We’ve built linear slides for several games before and have all the equipment to build another one. The only new challenge for this year is powering the slide both up and down.
Architecture and Strategy
After deciding on our prototyping targets, we started in on sketching out some ideas for a full robot. For harvesting, Emilio championed a zip tie harvester no top of a linear slide that would extend diagonally in to the crater and scoop up elements.
This would provide us with fast harvesting — a key requirement — without requiring us to venture in to the crater.
We also talked about a few depositing ideas. One with a linear slide to raise the harvesting mechanism up to the lander so it can shoot out the back of the robot, and one for radial dumpers on servo arms that the harvester could deposit in to.
For the linear slide approach, a problem we identified is that we already need one linear slide to Land and Latch, and it will need to be high torque in order to lift the robot. We want to score quickly, so we need high speed in our depositing mechanism.
Mr. Porter explained a simple way to estimate force required to deposit minerals with a radial arm. We estimated that it would take roughly 120 inch-oz to deposit a mineral with a 12″ arm, and FTC-legal servos can provide about 180 inch-oz of power. So potentially we could use that approach, and we could add a counterweight to the other end of the arm if the servo is not powerful enough.
As an assignment for the weekend, we asked Edward to determine how much power would be required to lift a 42lb. robot 6″ in 5 seconds. This should tell us how many motors we need for our Land and Latch system.
In order to use our Key Priorities to select features we would need to actually designate the priorities, we decided to it would be easiest if we ranked them. To do this, we used a technique recommended by Mr. Porter called “planning poker,” in which each team member ranks each requirement on a scale of 1 to 5 both in terms of difficulty and importance. We share our rankings with each other at the same time, and when there are different ideas about what the rankings should be, we discuss them and re-rank.
The idea behind this technique is that everybody on the team may have different information about each requirement. Something that seems difficult to one team member who knows little about it may seem easy to somebody who has done it before. This way we ensure that we get input from everybody on the team and arrive at reasonable estimates.
As a result of this process, our highest priorities are:
That the robot not way too much to lift
That the robot is reliable
That the robot is able to pick up elements very quickly
That the phone camera is able to see the vision targets
Our next highest priorities are
The robot’s center of mass is low and centered
The robot is very easy to drive
The robot is serviceable
Our medium importance priorities are
The robot is fast, because there is relatively little driving this year
The robot can score both types of mineral, because in Los Angeles only scoring silver minerals should be enough to win Regionals