The Incision

Another productive couple of days for the Robodox. Since the last update, we've gotten a ton of work done: 

First of all, our fabrication division has been hard at work making parts for the chassis. We are ready to assemble BOTH ROBOTS this saturday. Great progress.

Our programming division has made great progress on the drive code, writing the main body and control systems in just one day. They are currently conducting tests on the Andymark Sheet metal chassis, henceforth referred to as the Mock Chassis. 

Our CAD and Design divisions got a ton of work done on their respective subsystems, as well:

Drive: The Drive CAD is just about finished, with only minor details and bolt holes still missing. We have decided to use Andymark 6" HiGrip Traction wheels due to our large quantity of them and their proven effectiveness in traversing the field. Our gearboxes bring our free speed to roughly twenty-one feet per second in high gear and nine feet per second in low gear theoretical. Bumper mockups were also added to the model to ensure the intake system has the proper clearances. 

Intake: On Wednesday, we considered which advantages, if any, our robot would have if we were able to outtake power cubes from both sides. This would allow far faster cycles from the exchange to the Power Cube zone, by cutting out the time to turn around. It would also allow for more precise maneuverability and speed in placing cubes on the Scale. However, after weighing the pros and cons of this strategy, we decided that, due to the relative complexity of the changes which would be required to existing systems, this strategy would not be worthwhile to pursue. 

After this decision, we considered wheel positions and integration with the elevator and chassis perimeter requirements. The intake mechanism we have planned will currently need to actuate upwards at some point to fit within the frame perimeter. We will ultimately be deciding the design of the intake this saturday before after lunch. 

Regarding the lift, there are several factors we have to decide upon before the design review meeting this weekend, as well. For example, the size of the bearings we chose has come before concerns, especially considering the loads they will bear during climbs. We have chosen to redesign several components of the elevator system such that the inner carriage can travel a larger distance in order to reach the scale and climb from the rung more effectively. 

More updates, as well as chassis renders and design pictures coming tomorrow!

-Michael

Calculations and Fabrications

Another day of break, another day to work. Today, we met in K2 from 9-7, and completed more of our central design work. We specifically looked at physics and motor calculations for the cascade elevator, as well as layout designs for the power cube intake. Tomorrow, we hope to translate these designs to 3D models in CAD and place our preliminary round of orders.

Now, for the physics. First, our requirements: we know from analysis of previous year's robots that the optimal speed for an FRC linear elevator is roughly five feet per second. Because this speed will be accomplished by the innermost carriage within the elevator, the primary stage will have to travel at a speed of two and a half feet per second, or 0.762 m/s. Our objective is for the robot to scale the scale at this velocity, carrying a load of approximately 70 kilograms. By taking into consideration tentative pulley dimensions, motor specifications at peak power, and knowledge of gear ratios, we were able to determine several important theoretical figures about our climbing mechanism. First of all, the required power needed to climb is calculated to be just over five hundred watts, requiring two 775 Pro motors. However, due to concerns with overheating by running 775 pro motors near peak power throughout the endgame, the team decided to design using four 775 pro motors instead of two. With these four motors. our robot would be able to lift the weight of one other robot attached to ours, easily, opening up options for our strategy team. The main limiting factor in that situation, however, would be our drivetrain acceleration, which could not run at the same time for fear of browning out the RoboRIO. In conclusion, our final decided gear ratios are 25:1, planned for versa planetary gearboxes with a two-motor adapter adapter. 

Next, we prototyped intake wheel arrangements for the power cube manipulator. After surpassing our initial time expectations for prototyping, we revisited the effort with specific objectives. It is only worthwhile to prototype a system if your team would gain something from doing so, and thus we decided continuing our efforts would give us insight into wheel distance for vertically aligned power cubes, as the fabric can deform slightly on one side. We decided to use a spring-tensioned, jointed pair of mirrored sectionally-sigmoidal-shaped arms (hard to explain, think integral sign) with three pairs of high friction wheels, spaced such that the tension is initially able to intake vertically-aligned power cubes, and expands to allow a secure grip on flat power cubes. We expect to calculate free-speeds of the intake before the end of the day, tomorrow, and cut out a rough design of the plates which will eventually mount to the competition robot.

Fabrication is going better than ever thanks to the dedication and time given by our wonderful coaches and mentors. This season, we are on track for a three-hundred hour time commitment, which will be instrumental in testing and prototyping, in addition to the motion of building a second robot. 

Here's to another productive work day ahead!

Decisions, decisions

We just show you how, but you make the decisions. When you have this much power, you have to make big decisions.
— Bob Ross

The 'Dox have completed a successful second day of build season. Now that the game has began to sink in a little bit, and the large influx of people attending our kickoff has diminished, we were able to accomplish our objectives in a reasonably straightforward manner. Today, we reached three major milestones.

Before lunch, subsystem leads worked with their respective design groups on potential designs to determine pros and cons of different systems before we decided which to prototype. However, we realized soon afterwards that, as a result of dividing into different subsystems, our derived and driven requirements were not communicated effectively to other subsystems. To rectify this, we held a subsystem lead meeting to flush out several unknowns. 

The first unknown we discussed regarded strategy. After researching ideas from other teams, we considered the possibility of climbing while carrying one or more additional robots. We determined that there were essentially three different methods to climb while leaving opportunities for other robots. Firstly, a climbing mechanism could only contact a small proportion of the rung, leaving space for other robots to climb. We determined that it would be unreasonable to expect an alliance of three robots to have two which could both fit on the rung, and soon discounted this option. Secondly, a robot could deploy ramps or other rigid plates, allowing other robots to drive onto itself, before lifting all or two of them above the twelve inch climb height. Due to our ambitious requirements as driven from our prior day's strategic decisions, we felt as though we would not have sufficient student experience to lead a separate system capable of lifting two other robots from a drive perspective. The third option we considered was to mount a copy of the rung onto our robot, allowing other teams on our alliance to climb on a uniform standard. Despite depending on the mechanisms of other robots to secure a ranking point, we decided to utilize this strategy due to the relative simplicity of this method compared to the other two, as well as providing the option to utilize the full rung on the field. 

The second decision we faced was to select a mechanism to deliver power cubes to the scale. Prior brainstorming conducted on kickoff narrowed our linear lift to two general categories; a single-jointed arm, and a multi-stage linear elevator. In our design review meeting after lunch, we compared and contrasted the qualities of both mechanisms. A single jointed arm is considerably less complex than a multi-stage elevator, but comes with several risks which were difficult to ignore, such as the possibility of violating the maximum volume during the game, or unpredictably changing the robot's center of gravity while lifted to nearly seven feet tall, allowing other robots to potentially knock us over. A multi-stage elevator would be effective due to its increased repeatability in software, rigidity in climbing, ease of allowing other robots to climb from our own, and predictable constant center of gravity. This will allow our programming team to develop more robust and dependable code to control the major subsystems of this robot, and we ultimately decided to pursue this design.

The third and final significant design decision we faced today was among the last facing our drivetrain. At the culmination of the kickoff meeting, we were undecided as to the benefits of using pneumatics on this robot, due to weight concerns and the relatively low-defense nature we predict for this game. Initially, we were unable to see other mechanisms which might utilize them, however, for the benefits pneumatics could provide for augmenting the intake and outtake systems of the cube manipulator, in addition to serving as a static lock for the elevator, we decided ultimately to implement pneumatics into our robot. 

As you can see, today was extremely productive, and we left K2 with a clear idea of what our robot will eventually look like and how it will perform in-game. Tomorrow, we will present our decisions to the team as a whole, opening our discussion to a wider audience to find flaws before they may occur. Additionally, we will continue fabricating tomorrow, including beginning the structural members of the chassis. 

 

Power Up!

Kickoff: the day in which the next season's FIRST Robotics Competition game is released, allowing a team to begin the arduous and extraordinary process of game and rule analysis, strategy, design, fabrication, electronics, and programming. Today marks the first day of this six week period, for many students the last one we'll ever experience. 

Today was packed- nearly sixty individuals crowded into K2 this morning at around 6:30 PST to decide our 2018 Robot name and attend kickoff. The team decided to name this year's robot Emma, the Embryologist. 

The game is FIRST Power Up, an arcade-themed challenge in which robots attempt to gain control of their alliance's Switch, a central Scale, and their opposing alliance's Switch in order to incrementally score points during a match. Robots can gain control of their alliance's switch by placing Power Cubes, thirteen inch square milk crates, onto each side. Power cubes can also be given to an alliance's Human Player on their respective alliance station wall by entering the cube into the Exchange Zone, where it is entered into that alliance's Vault to score points and earn Power Ups. At the end of the match, robots may climb onto the central scale to score additional points, and may also hang from/climb on other team's robots. 

Today, we began with an intensive rule analysis to ensure all members know the game thoroughly. We conducted a game simulation, as well-after tracing a 1:1 scale model of the Arcade onto asphalt, humans pretended to be robots in order to learn how to play the game. This allowed our strategists to develop insight into the role the Vault, Scales, and Switches will play during the match, and how Power Ups could be utilized to most effectively score points.

We calculated the absolute maximum score an alliance could earn in one match for comparison purposes (in green) , and a projected week one score bracket (in orange): https://i.imgur.com/aFuD1yC.jpg

After the team fully understood the game, we worked to develop a general robot requirements list. In order to be highly competitive, our robot must:

  • Score Power Cubes onto the Switch
  • Score Power Cubes into the Exchange and Vault
  • Climb to earn ranking points
  • Score Power Cubes onto the Scale
  • Intake Power Cubes from the floor

We determined that leaving out any of these objectives could be problematic if no robots on our alliance are able to complete these tasks. For example, if no robot on our alliance can manipulate the scale, then as long as the other alliance has at least one robot which can, the match is essentially decided based on the 125 points the scale could possibly award (considering the usage of a Force Power Up)

Using these fundamental robot requirements, we were able to begin setting derived and driven requirements for our other systems, currently divided into three rough areas.

  1. Linear Lifting Structure: concerned with housing the Cube Manipulator and Climb subassemblies. Must be able to:
    1. tentatively reach a maximum height of seven feet,
    2. structurally support the weight of the robot in tension
    3. structurally support the weight of the gamepiece in addition to it's own structure in compression.
    4. compact to a minimum height of 55 inches
    5. vary in height continuously
  2. Climb: concerned with lifting the robot at the end of the match to climb the scale. Must be able to:
    1. lift our robot one foot above the platform zone
    2. hook onto the rung or another robot with similar lifting style
    3. complete a climb in seven seconds or less
  3. Cube Manipulator: concerned with manipulating the Power Cube game elements. Must be able to:
    1. intake Power Cubes from the floor
    2. mount to the Linear Lifting Structure
    3. outtake Power Cubes to the Switch
    4. outtake Power Cubes to the Scale
    5. outtake Power Cubes to the Exchange
    6. handle cubes from all directions

We decided to use our standard 6WD WCD drivetrain for this game due to the relatively flat and familiar surfaces. However, CAD analysis of the drivetrain ground clearance profile suggests that the bottom of the chassis would have less than one eighth inch of clearance from the platform zone ramp at the closest point. For this reason, we decided to use six inch wheels to refrain from concerns which might arise from driving onto the platform zone at an angle. We have not yet decided an objective floor speed or if we plan to use pneumatics in this robot for a shifting gearbox.

Tomorrow, our objectives are to take the rules test, resume discussion, research existing mechanisms, begin fabricating game elements, and begin prototyping for potential designs.

Pre-Kickoff Meetings

Hello, and welcome to the first Robodox blog post for the 2018 season! We've had an extremely busy week of meetings and organization, and are excited to attend kickoff in less than nine hours. 

Here are our objectives for the 2018 season:

  1. Bring home our first blue banner
  2. Practice driving for two weeks prior to bag & tag
  3. Build a second robot
  4. Improve documentation

Accomplishing all of these will require maximum effort on our part.  Due to our large student experience base and the large number of seniors on the team this year, we have chosen to build two robots--one for competition and one for drive/programming practice--pushing our fabrication and design capabilities to the limit. 

To assist with this, we have finished CAD models of off-season robot chassis models, available on the CADs page of this website. These CAD models are special, however; they parametrically update in response to user-driven variables, such as width and length. The result is that the chassis CAD is projected for completion by the end of the first weekend at the latest. Students can begin fabricating parts as soon as the drivetrain is decided, and we will hopefully hit the ground running to begin a successful season. 

Good luck to all, and try to get some sleep!

-Michael McNabb

2018 Chief Engineer