Autonomous

There are generally 2 modes in FTC: Autonomous and TeleOp. As programmers, it’s our job to code both.

Autonomous is a 30 second game mode that involves the robot moving all by itself. In autonomous, points are widely skewed; everything is worth more, and the total points available for scoring range from 100-200 points.

It’s difficult to accomplish this feat as almost everything can go wrong. Generally, autonomous is coded in a step by step manner.

  1. Configure all your hardware. You should be using a Linear OpMode (you can erase the variables and contents of the methods in Linear OpModes that come with the SDK to get started). Example:
    motor_a = hardwareMap.dcMotor.get("left");
  2. Get the robot to move. Whether this entails encoders, time, or anything, the basis of autonomous lies on movement. Example:
    motor_a.setPower(1); motor_b.setPower(2);
  3. Get your sensors working, No matter which sensors you use, make sure you can get the data you need from them reliably. Example:
    rateOfRotation = gyro.getRotation();
  4. Make judgements. Figure out what values mean what – now that you can poll for data, what data should signal you to do what? Example:
    if(irSeeker.signalDetected()) beaconFound = true;
  5. Piece the movement together with the judgements and TEST. Testing is the most important part. Example: On the Lancers, we have many spreadsheets full of test data, running at least 10 trials for each form of autonomous and testing each possible scenario.
  6. If something doesn’t work, take a different approach. Example: In the Cascade Effect 2014-15 season, the IR Beacons used were circular; they returned different data. In addition, the IR sensors found it difficult to differentiate between diffeent positions. For this reason, we weighted the IR’s judgement as 25% reliable, and introduced a sonar that could judge 75% reliably. We then used the weighted data to figure out what position was most probable.
Skip to toolbar