What an FRC Robot Actually Is

Module 1: What an FRC Robot Actually Is

Series: FRC Technical Foundations Unit: 1 — Foundations Session length: ~3 hours (60 min instruction, 100 min studio, 20 min review) Prerequisites: None. This is the entry point.

What you should be able to do after this lesson

  • Describe a competition robot as a set of interacting subsystems rather than a single machine.
  • Identify the interface between two subsystems and explain why interfaces get frozen before designs do.
  • State what “done” means for a mechanism in measurable terms.
  • Work safely in a workshop without being told to.

The mistake almost everyone makes first

When students start in robotics, they think of a robot as an object. You build it, you program it, it works. That mental model survives about four weeks into a competition season, and then it breaks in a very specific way: something works perfectly on the bench, gets bolted to the robot, and stops working.

The intake that gripped every game piece in testing now jams, because it is mounted 15 mm lower than the prototype and the game piece catches the frame rail. The arm that lifted smoothly now stalls, because the drivetrain is pulling 180 A during acceleration and the battery sags to 9 V. The autonomous routine that ran flawlessly on a practice field now drifts, because the real carpet has a different friction coefficient.

None of these are failures of the subsystem. They are failures of the system. And that is the single most important idea in this entire series: an FRC robot is not a collection of parts that happen to be bolted together. It is an integrated electromechanical system in which every subsystem consumes something another subsystem produces — space, power, time, structural stiffness, CPU cycles, or driver attention.

The curriculum this series is adapted from states the goal well: develop students who can reason about a complete robot as an integrated electromechanical system, not merely assemble a kit or copy code. Everything in the next nine lessons serves that.

The six subsystems

Almost every FRC robot, in any season, decomposes into the same six pieces. The game changes what each one does; it rarely changes that they exist.

Structure. The frame, the bumpers, the superstructure that holds everything else. Its job is to hold the geometric relationships between everything else constant while the robot is being hit by other 125 lb robots.

Drivetrain. Gets the robot from one place on the field to another, reliably, at a speed you chose on purpose. Almost always the first thing built and the last thing you are allowed to break.

Manipulators. Whatever acquires, moves, holds, and releases game pieces. Intakes, arms, elevators, shooters. This is where the game-specific creativity lives.

Electrical. Battery, power distribution, motor controllers, the CAN bus, sensors, and all the wiring between them. It is the most invisible subsystem and the most common cause of a robot that dies mid-match.

Software. The code that turns driver intent and sensor data into motor commands. Includes the autonomous routines.

The human system. Drivers, pit crew, scouts, strategists. Real teams treat this as a subsystem with a design, a test plan, and practice hours, because it is.

Interfaces are the real design

Here is the idea I want you to carry out of this lesson more than any other.

When two subsystems meet, the meeting point is called an interface. An interface is not just where two brackets bolt together. A complete interface specification answers six questions:

  1. Mounting. What is the bolt pattern, and what surface does it land on?
  2. Envelope. How much space does this thing occupy, including its full range of motion?
  3. Power. How many motors, on what breakers, drawing what peak current?
  4. Sensing. What sensors does it have, plugged into what ports?
  5. Command. What does software need to be able to tell it to do? (“Go to a position” is a different interface from “run at a speed.”)
  6. Service. How do you get a broken part out of it in under five minutes, in a pit, with a robot on a cart?

 

Rule 1.1 — Freeze interfaces before you freeze designs. You can keep redesigning the inside of the intake all season. But the day you tell the electrical lead “the intake mounts here, uses two NEOs on 40 A breakers, and reports one through-bore encoder,” they can start building the harness. If that changes in week five, they build it twice.

This is why professional engineering organisations write interface control documents before anything is manufactured. It is also why the FTC curriculum this series adapts includes a non-negotiable integration gate: a subsystem may be installed on the competition robot only when it has (1) a defined interface, (2) a basic acceptance test, (3) a safe software state, (4) a service procedure, and (5) a named owner responsible for its documentation and spares.

I would keep that gate word for word. It is the single highest-leverage rule a rookie team can adopt.

“It works” is not a result

The other habit to build now, before you have anything to test, is refusing to accept the phrase “it works.”

“It works” contains no information. Ask instead:

  • How many times did you run it? (If the answer is under ten, you have not tested it.)
  • What was the success rate?
  • How long did one cycle take?
  • Under what conditions did it fail?
  • What was the battery voltage during the test?

A mechanism that works 8 times out of 10 will fail roughly twice in a qualification match. Over ten matches, that is twenty failures. Teams routinely discover in week six that their 80%-reliable intake is the reason they are ranked 34th.

Rule 1.2 — Every technical claim needs evidence. A calculation, a CAD view, a wiring diagram, a code excerpt, a test log, a plot, or a repeatable demonstration. “I think it’s fine” is not one of these.

The shape of a season

FRC compresses everything. A game is announced at kickoff in early January, and teams have roughly six weeks before their first competition. That compression is why the habits above matter so much: there is no time to do things twice.

A rough shape that works:

  • Week 0–1: Read the manual properly. Decompose the game into scoring actions. Estimate cycle times. Decide what your robot will not do.
  • Week 1–2: Prototype the risky mechanisms. Build the drivetrain, because you already know you need one.
  • Week 2–4: Manufacture and assemble. Freeze interfaces early in this window.
  • Week 4–5: Integrate, wire properly, get software onto real hardware.
  • Week 5–6: Stop changing things. Drive practice, reliability testing, inspection readiness.

Most teams lose their season in week five, by continuing to add features when they should be practising. Protecting driver practice time is a design decision, not a scheduling one.

[ANECDOTE SLOT] — A short story here about a time you or a team you mentored kept changing a design too late into a build, and what it cost. Two or three sentences is enough; specific beats dramatic.

Safety, briefly and seriously

Safety is not a lecture you sit through once. It is a set of habits that make you faster, because injured people and damaged robots both cost time.

  • Eye protection whenever anything is being cut, drilled, or driven. No exceptions for “just one hole.”
  • Long hair tied back, no loose sleeves or lanyards near rotating machinery.
  • Nobody powers on a robot without saying out loud that they are doing it, and checking nobody’s hands are inside it.
  • Batteries are heavy, carry real current, and are never dropped or left where they can short.
  • Every tool goes back. A missing 7/16″ wrench in a pit costs you a match.

🔧 Exercise 1.1 — Robot Teardown Map

Time: 45 minutes. Equipment: Any complete robot — last season’s competition robot, a kitbot, or a training chassis. Paper, pens.

Working in pairs, without disassembling anything:

  1. Identify and label all six subsystems on the physical robot.
  2. Find and list five interfaces — places where two subsystems meet. For each, write down as many of the six interface questions as you can answer by looking.
  3. Find one interface that is clearly badly designed. Something you cannot service without removing an unrelated part, a wire routed through a moving mechanism, a bolt you cannot reach with any tool you own.
  4. Sketch the fix.

Evidence of learning: A labelled system diagram and one written interface specification, plus a one-paragraph critique of the bad interface.

🔧 Exercise 1.2 — The “It Works” Interrogation

Time: 30 minutes. Equipment: Any working mechanism and a stopwatch.

One student demonstrates a mechanism and says “it works.” Everyone else has three minutes to ask questions that turn that claim into data. Then actually measure it: run it twenty times, log every attempt, and compute a success rate and a mean cycle time.

Evidence of learning: A twenty-row test log with success rate and cycle time, plus at least one observed failure mode described in one sentence.

🔧 Exercise 1.3 — Tool and Safety Certification

Time: 25 minutes.

Each student identifies every hand tool in the workshop by name and correct use, locates the fire extinguisher, eye-wash, and first-aid kit, and demonstrates safe battery handling. Sign off individually.

Evidence of learning: A signed personal tool and safety checklist retained for the season.

Further reading

  • WPILib: Zero to Robot — the official on-ramp; skim the whole path now, work through it in Lesson 5.
  • FIRST Competition Manual — the authority for everything legal. Read it, do not rely on what someone told you.
  • Chief Delphi — the FRC community forum. Nearly every design question you have has been argued about here already.
  • Spectrum 3847 Resources — one of the best curated collections of team-produced FRC technical material.

Next Lesson

We start with the thing everything else bolts to. Lesson 2 — Structure That Doesn’t Move covers load paths, triangulation, bearings, and why a frame that flexes by two millimetres can ruin an autonomous routine.