Hot off their trip to Einstein, Israeli team BumbleB 3339 gives a peek behind the curtain.
After reading Recycle Rush rules we noticed that the robot who will control the bins (Recycling Containers) will control the game. We decided that our robot main goal is to be the “king of the bins.” In order to control the bins, we had to be able to control not only the bins that are in our side of the field but also the bins that are placed on the STEP. Our strategy was that the robot will be primarily feeded with totes from human player station. The reason to this decision was consistency – we thought we will be able to be more consistent in making points from the human player than the landfill.
This year we have decided to take the CAD team into new levels. The aim was to design the robot on the computer before we even start to build anything.
The CAD team worked very hard during the first 3 weeks of the season and completed the design of all major systems in the robot. Unfortunately, not all of the systems were designed in 3D due to lack of time and manpower. But, we gave enough breathing time to start build the robot. At the end of these 3 weeks the CAD team made a design meeting that the 3D model of the robot was presented to the whole team in order to make sure everyone understand the robot design based on the concept and principles the team decided on at design meetings.
The Robot Systems
- Drivetrain – We decided that the drivetrain will be mecanum driven, this year the field was full with obstacles, we needed an holonomic drivetrain. Our inspiration was team 118 with their swerve drive in 2008. We dreamed of being able to cross the field like 118 did in their robot reveal video. We chose mecanum drive because we have not done any holonomic drivetrain ever and mecanum seemed the easiest and most reliable for us. After we decided we are going on mecanum we had to learn how to use it. As simple as it sounds we went to Chief Delphi in order to consult with experienced people. We asked: Which mecanum is better Vex or AndyMark? How to build a good mecanum drive? How much gearbox reduction is needed? We decided to choose Vex wheels although there were bad critics on them from 2014 because Vex stated they have improved them and because they were much cheaper. We were advised to place the wheels in a square shape and to keep the center of gravity as close to the center of the square (see tilt mechanism). On gearbox reduction it was recommended to use 10.5:1 reduction. We chose Vex Versaplanetary with 10:1 reduction in order to save some weight – planetary gearbox is lighter and Vex had a single reduction option of 10:1 which saved more weight. The wheels were direct driven from the gearboxes. We took some risk with the shaft of the gearboxes that payed off, the shaft was short that we could only hold the wheel from one side with a hub but it worked well and saved some more weight.
- Tilt Mechanism – We needed to keep the center of gravity closer to the center of the robot in order to maintain a good driving ability. We thought to tilt the stack backwards with 2 pistons to control the angle of the stack since our design was to hold the stack outside the robot and not inside like most teams.
- Aligning Arms – These arms were designed to align totes to the center of the robot, it designed for totes that enter the field through the chute door. The arms were driven by a single snow blower motor with a timing belt.
- Track – both tote arms and container claw were moving on the same track since the recycling container must be on top of the totes there won’t be any clash. The track is made from hollow aluminum angles we found in one of our sponsors warehouse. We chose this material because it was lightweight and strong enough to support itself without any more reinforcements.
- Tote Arms – These arms were designed to be able to collect totes and stack them up to 6 at a time. They were also driven by a snow blower motor with a timing belt. The arms were designed to flip recycling containers to a standing position. The arms are moving on the track with a CIM motor connected to a worm gearbox (WormBox). We decided to use worm gearbox in order to get a steady height of the stack without worrying the stack will fall.
- Container Claw –The claw was designed to pick up a standing recycling container and hold it on top of the stack, or to cap it on existing stacks. It is controlled with a single piston connected to a mechanism like a double four bar mechanism. At first it was designed to be as simple as we could – only a metal arch with the shape of the container that was supposed to hold the container from its frame. It wasn’t as reliable as we hoped to we added the piston to it after week 0 competition. The claw is moving on the track with a MiniCIM motor connected to a worm gearbox (WormBox).
- Can Burglars – These were our secret weapon in our design concept that will break the game. In order to be the “king of the bins” we knew we should take 2 recycling containers in the first 2 seconds of the game. We designed it with 2 long arms that moves by two pistons. The pistons are longer than the way the arms had to go from the robot to the recycling containers in order to apply a strong downward force that will give us advantage when entering a battle with another robot (we actually won every can battle we had in Carson). The arms had a pivot point near the end of the arms in order to enable a quick detach of the recycling containers without worrying of getting stuck in them. During Israel Regional we could steal the cans in about 750 milliseconds which was faster than we wanted. In order to compete in the championship we knew we had to make them faster so we made some physical modification by moving pivot point of the pistons, and tried to optimize our code. We managed to minimize our stealing time to about 450 milliseconds.
Our team won a regional for the first time in Israel Regional. We won the Excellence in Engineering award. We were the first ever Israeli team to win a division (Carson) and advancing to Einstein. Our robot had the ability to make 3 full stacks worth 126 points.