Encoder Navigation - OnBot
In the previous section you learned about how to use Elapsed Time to allow your robot to navigate the world around it autonomously. When starting out many of the robot actions can be accomplished by turning on a motor for a specific amount of time. Eventually, these time-based actions may not be accurate or repeatable enough. Environmental factors, such as the state of battery charge during operation and mechanisms wearing in through use, can all affect time-based actions. Fortunately, there is a way to give feedback to the robot about how it is operating by using sensors; devices that are used to collect information about the robot and the environment around it.
With Elapsed Time, in order to get the robot to move to a specific distance, you had to estimate the amount of time and the percentage of duty cycle needed to get from point a to point b. However, the REV motors come with built in encoders, which provide feedback in the form of ticks ( or counts) per revolution of the motor. The information provided by the encoders can be used to move the motor to a target position, or a target distance.
Moving the motors to a specific position, using the encoders, removes any potential inaccuracies or inconsistencies from using Elapsed Time. The focus of this section is to move the robot to a target position using encoders.
There are two articles in that go through the basics of Encoders. Using Encoders goes through the basics of the different types of motor modes, as well as a few application examples of using these modes in code. In this section we will focus on using RUN_TO_POSITION.
The other article, Encoders, focuses on the general functionality of an encoder.
It is recommended that you review both articles before moving on with this guide.
Basics of Programming with Encoders
Start by creating a basic op mode calledHelloRobot_EncoderAuton.
When creating an op mode a decision needs to be made on whether or not to set it to autonomous mode. For applications under 30 seconds, typically required for competitive game play changing the op mode type to autonomous is recommended. For applications over 30 seconds, setting the code to the autonomous op mode type will limit your autonomous code to 30 seconds of run time. If you plan on exceeding the 30 seconds built into the SDK, keeping the code as a teleoperated op mode type is recommended.
For information on how op modes work please visit the Introduction to Programming section.
For more information on how to change the op mode type check out the Test Bed - OnBot Java section.
The op mode structure below is simplified and only includes the necessary components needed to create the encoder based code.
As with all drivetrain related navigation, the directionality of one of the motors needs to be reversed in order for both motors to move in the same direction. Since the Class Bot V2 is still being used add the line rightmotor.setDirection(DcMotor.Direction.REVERSE); to the code beneath the rightmotor = hardwareMap.get(DcMotor.class, "rightmotor"); code line.
For more information on the directionality of motor check out the Basics of Programming Drivetrains section.
Recall from Using Encoders that using RUN_TO_POSITION mode requires a three step process. The first step is setting target position. To set target position add the lines leftmotor.setTargetPosition(1000); and rightmotor.setTargetPosition(1000); to the op mode after the waitForStart(); command. To get a target position that equates to a target distance requires so calculations, which will be covered later. For now, set target position to 1000 ticks.
The next step is to set both motors to the RUN_TO_POSITIONmode. Add the lines leftmotor.setMode(DcMotor.RunMode.RUN_TO_POSITION);and rightmotor.setMode(DcMotor.RunMode.RUN_TO_POSITION);to your code, beneath the setTargetPosition code lines.
The main focus of the three step process is to set a target, tell the robot to move to that target, and at what speed (or velocity) the robot should get to that target. Normally, the recommended next step is to calculate velocity and set a target velocity based on ticks. However, this requires quite a bit of math to find the appropriate velocity. For testing purposes, its more important to make sure that the main part of the code is working before getting too deep into the creation of the code. Since the setPower function was covered in previous sections and will communicate to the system what relative speed (or in this case duty cycle) is needed to get to the target, this can be used in the place of setVelocity for now.
Add the lines to set the power of both motors to 80% of duty cycle.
Now that all three RUN_TO_POSITION steps have been added to the code the code can be tested. However, if you want to wait for the motor to reach its target position before continuing in your program, you can use a while loop that checks if the motor is busy (not yet at its target). For this program lets edit the while (opModeIsActive()){}
Recall that, within a linear op mode, a while loop must always have the opModeIsActive() Boolean as a condition. This condition ensures that the while loop will terminate when the stop button is pressed.
Edit the while loop to include the leftmotor.isBusy() and righmotor.isBusy()functions. This will check if the left motor and right motor are busy running to a target position. The while loop will stop when either motor reaches the target position.
Right now the while loop is waiting for the either motor to reach the target. There may be occasions when you want to wait for both motors to reach their target position, in this case the following loop can be used.
while (opModeIsActive() && (leftmotor.isBusy() || rightmotor.isBusy()))
Save and run the op mode two times in a row. Does the robot move as expected the second time?
Try turning the Control Hub off and then back on. How does the robot move?
In the Basic Encoder Concepts section, it is clarified that all encoder ports start at 0 ticks when the Control Hub is turned on. Since you did not turn off the Control Hub in between runs, the second time you ran the op mode the motors were already at, or around, the target position. When you run a code, you want to ensure that certain variables start in a known state. For the encoder ticks, this can be achieved by setting the mode to STOP_AND_RESET_ENCODER . Add this block to the op mode in the initialization section. Each time the op mode is initialized, the encoder ticks will be reset to zero.
For more information on the motor mode STOP_AND_RESET_ENCODERScheck out the STOP_AND_RESET_ENCODERS section of the Using Encoders guide.
Converting Encoder Ticks to a Distance
In the previous section, the basic structure needed to use RUN_TO_POSITIONwas created. The placement of leftmotor.setTargetPosition(1000); and rightmotor.setTargetPosition(1000); within the code, set the target position to 1000 ticks. What is the distance from the starting point of the robot and the point the robot moves to after running this code?
Rather than attempt to measure, or estimate, the distance the robot moves, the encoder ticks can be converted from amount of ticks per revolution of the encoder to how many encoder ticks it takes to move the robot a unit of distance, like a millimeter or inch. Knowing the amount of ticks per a unit of measure allows you to set a specific distance. For instance, if you work through the conversion process and find out that a drivetrain takes 700 ticks to move an inch, this can be used to find the total number of ticks need to move the robot 24 inches.
Reminder that the basis for this guide is the Class Bot V2. The REV DUO Build System is a metric system. Since part of the conversion process references the diameter of the wheels, this section will convert to ticks per mm.
For the conversion process the following information is needed:
Ticks per revolution of the encoder
Total gear reduction on the motor
Including gearboxes and motion transmission components like gears, sprockets and chain, or belts and pulleys
Circumference of the driven wheels
Ticks per Revolution
The amount of ticks per revolution of the encoder shaft is dependent on the motor and encoder. Manufacturers of motors with built-in encoders will have information on the amount of ticks per revolution. For HD Hex Motors the encoder counts 28 ticks per revolution of the motor shaft.
Visit the manufacturers website for your motor or encoders for more information on encoder counts. For HD Hex Motors or Core Hex Motors visit our Motor documentation.
Total Gear Reduction
Since ticks per revolution of the encoder shaft is before any gear reduction calculating the total gear reduction is needed. This includes the gearbox and any addition reduction from motion transmission components. To find the total gear reduction use the Compound Gearing formula.
For the Class Bot V2 there are two UltraPlanetary Cartridges, 4:1 and 5:1, and an additional gear reduction from the UltraPlanetary Output to the wheels, 72T:45T ratio.
The UltraPlanetary Cartridges use the nominal gear ratio as a descriptor. The actual gear ratios can be found in the UltraPlanetary Users Manual's Cartridge Details.
Using the compound gearing formula for the Class Bot V2 the total gear reduction is:
Unlike the the spur gears used to transfer motion to the wheels, the UltraPlanetary Gearbox Cartridges are planetary gear systems. To make calculations easier the gear ratios for the Cartridges are already reduced.
Circumference of the Wheel
The Class Bot V2 uses the 90mm Traction Wheels. 90mm is the diameter of the wheel. To get the appropriate circumference use the following formula
You can calculate this by hand, but for the purpose of this guide, this can be calculated within the code.
Due to wear and manufacturing tolerances, the diameter of some wheels may be nominally different. For the most accurate results consider measuring your wheel to confirm that the diameter is accurate.
To summarize, for the Class Bot V2 the following information is true:
Ticks per revolution | 28 ticks |
Total gear reduction | 30.21 |
Circumference of the wheel |
|
Each of these pieces of information will be used to find the number of encoder ticks (or counts) per mm that the wheel moves. Rather than worry about calculating this information by hand, these values can be added to the code as constant variables. To do this create three variables:
COUNTS_PER_MOTOR_REVDRIVE_GEAR_REDUCTIONWHEEL_CIRCUMFERENCE_MM
The common naming convention for constant variables is known as CONSTANT_CASE, where the variable name is in all caps and words are separated by and underscore.
Add the variables to op mode class, where the hardware variables are defined. Defining the variables within the bounds of the class but outside of the op mode, allows them to be referenced in other methods of functions within the class. To ensure variables are referenceable they are set as static final double variables. Static allows references to the variables anywhere within the class and final dictates that these variables are constant and unchanged elsewhere within the code. Since these variables are not integers they are classified as double variables.
Now that these three variables have been defined, they can be used to calculate two other variables: the amount of encoder counts per rotation of the wheel and the number of counts per mm that the wheel moves.
To calculate counts per wheel revolution multipleCOUNTS_PER_MOTOR_REV by DRIVE_GEAR_REDUCTION Use the following formula:
Where,
=
COUNTS_PER_MOTOR_REV=
DRIVE_GEAR_REDUCTION=
COUNTS_PER_WHEEL_REV
Create the COUNTS_PER_WHEEL_REVvariable within the code. This will also be a static final double variable.
Once COUNTS_PER_WHEEL_REV is calculated, use it to calculate the counts per mm that the wheel moves. To do this divide theCOUNTS_PER_WHEEL_REV by the WHEEL_CIRCUMFERENCE_MM. Use the following formula.
Where,
=
COUNTS_PER_MOTOR_REV=
DRIVE_GEAR_REDUCTION=
WHEEL_CIRCUMFERENCE_MM=
COUNTS_PER_WHEEL_REV=
COUNTS_PER_MM
Create the COUNTS_PER_MMvariable within the code. This will also be a static final double variable.
COUNTS_PER_WHEEL_REVwill be created as a separate variable fromCOUNTS_PER_MM as it is used in calculating a target velocity.
Moving to a Target Distance
Now that you have created the constant variables needed to calculate the amount of ticks per mm moved, you can use this to set a target distance. For instance, if you would like to have the robot move forward two feet, converting from feet to millimeters and multiplying by the COUNTS_PER_MMwill give you the amount of counts (or ticks) needed to reach that distance.
Create two more variables called leftTarget andrightTarget. These variables can be fluctuated and edited in your code to tell the motors what positions to go to, rather than place them with the constant variables, create these variables within the op mode but above the waitForStart(); command.
The setTargetPosition();function takes in a integer (or int) data type as its parameter, rather than a double. Since both the leftTarget and rightTargetwill be used to set the target position, create both variables as int variables.
Right now the main distance factor is COUNTS_PER_MM , however you may want to go a distance that is in the imperial system, such as 2 feet (or 24 inches). The target distance in this case will need to be converted to mm. To convert from feet to millimeters use the following formula:
If you convert 2 feet to millimeters, it comes out the be 609.6 millimeters. For the purpose of this guide, lets go ahead an round this to be 610 millimeters. Multiply 610 millimeters by theCOUNTS_PER_MM variable to get the number of ticks needed to move the robot 2 feet. Since the intent is to have the robot move in a straight line, set both the leftTarget and rightTarget, to be equal to 610 * COUNTS_PER_MM
As previously mentioned the setTargetPosition(); function requires that its parameter must be an integer data type. The leftTargetand rightTargetvariables have been set to be integers, however theCOUNTS_PER_MM variable is a double. Since these are two different data types, a conversion of data types needs to be done.
In this case theCOUNTS_PER_MMneeds to be converted to an integer. This is as simple as adding the line (int) in front of the double variable. However, you need to be cautious of potential rounding errors. Since COUNTS_PER_MMis part of an equation it is recommended that you convert to an integer after the result of the equation is found. The example of how to do this is exhibited below.
Edit the setTargetPosition(); lines so that both motors are set to the appropriate target position. To do this add the leftTargetand rightTarget variables to their respective motor.
Try running the code and observing the behavior of the robot. Consider some of the following
Is the robot moving forward by two feet?
Does the robot seem to be moving in straight line?
Is the code running without error?
Setting Velocity
Velocity is a closed loop control within the SDK that uses the encoder counts to determine the approximate power/speed the motors need to go in order to meet the set velocity. When working with encoder setting a velocity is recommended over setting a power level, as it offers a higher level of control.
To set a velocity, its important to understand the maximum velocity in RPM your motor is capable of. For the Class Bot V2 the motors are capable of a maximum RPM of 300. With a drivetrain, you are likely to get better control by setting velocity lower than the maximum. In this case, lets set the velocity to 175 RPM
Recall that setVelocity is measure in ticks per second.
Create a new double variable called TPS. Add TPS the to the op mode under where rightTarget is defined.
Since RPM is the amount of revolutions per minute a conversion needs to be made from RPM to ticks per second. To do this divide the RPM by 60, to get the amount of rotations per second. Rotations per second can the be multiplied by COUNTS_PER_WHEEL_REV, to get the amount of ticks per second.
Exchange the setPower();functions for setVelocity();. AddTPS as the parameter for setVelocity();.
Try to build the code. Do you get errors?
With the current state of the code you are likely to get errors similar to the ones pictured below:
This is because the setVelocity();function is a function of theDcMotorEx Interface. The DcMotorEx Interface is an extension of the DcMotor Interface, that provides enhanced motor functionality, such as access to closed loop control functions. To use setVelocity(); the motor variables need to be changed to DcMotorEx . To do this both the private variable creation of the motors, and the hardware mapping need to be changed to DcMotorEx.
Since DcMotorEx is an extension of DcMotor, DcMotor specific functions can be used by variables defined as DcMotorEx.
Once you have made these changes the basic, drive two feet code is done! The code below is the finalized version of the code. In this the other hardware components and telemetry have been added.
Turning the Drivetrain Using RUN_TO_POSITION
In the Robot Navigation - OnBot Java section, the mechanism of setPower(); was discussed. setPower();dictates what direction and speed a motor moves in. On a drivetrain this dictates whether the robot moves in forward, reverse, or turns.
In RUN_TO_POSITIONmode the encoder counts ( or setTargetPosition();) are used instead of setPower(); to dictate directionality of the motor. If a target position value is greater than the current position of the encoder, the motor moves forward. If the target position value is less than the current position of the encoder, the motor moves backwards
Since speed an directionality impacts how a robot turns, setTargetPostion(); and setVelocity();need to be edited to get the robot to turn. Consider the following code:
The rightTarget has been changed to be a negative target position. Assuming that the encoder starts at zero due to STOP_AND_RESET_ENCODER this causes the robot to turn to the right. Velocity remains the same for both motors. If you try running this code, you may notice that the robot pivots along its center of rotation. To get a wider turn changing the velocity so that the right motor is running at a lower velocity than the left motor. Adjust the velocity and target position as needed to get the turn you need.
For more information on how direction and speed impact the movement of a robot please refer to the explanation of setPower(); in the Robot Navigation section.
The following code walks through adding a turn to the program, after the robot moves forward for 2 feet. After the robot reaches the 2 foot goal, there is a call to STOP_AND_RESET_ENCODERS this will reduce the need to calculate what position to go to after a position has been reached.
Last updated
