AME 3623: Project 5: Rate Gyroscopes and Damping Control

Key to successful control in embedded systems is the ability to integrate information about the external state of your system into the control decisions that are being made. In this project, we will finally close a loop between sensing and actuation. We will add one more component to your circuit: an inertial measurement unit (IMU). The IMU contains three distinct sensors (with three degrees of freedom each): an accelerometer, a rate gyroscope and a magnetometer. The gyroscope will be used in this project to dampen rotational disturbances. In project 6, we will use all nine degrees-of-freedom to estimate the orientation of the hovercraft.

All components of the project are due by Thursday, March 29th at 1:30 pm.
Groups are the same as for project 1.
Discussion within groups is fine.
Discussion across groups may not be about the specifics of the solution (general programming/circuit issues are fine to discuss).

At the end of this project, you should be able to:

extract rotation rate information from an inertial measurement unit, and
use the sensory data to dampen out rotational disturbances.

Component 0: Library Installation

In your Arduino environment: in the sketch menu, select Manage Libraries / Add .Zip Library
Navigate to and select ImuUtils.zip. Click Ok
In your Arduino environment: in the sketch menu, select Manage Libraries / Add .Zip Library
Navigate to and select QuaternionFilters.zip. Click Ok
Likewise for MPU9250.zip

At the top of your project, include the following lines:

#include "ImuUtils.h"
#include "PeriodicAction.h"

The new library provides the following functions:

imu_setup() initializes the inertial measurement unit. This function assumes that your hovercraft is not in motion when it is called.
imu_update() updates the state of the inertial measurement unit. This function expects to be called regularly (at 100-200 Hz) and updates the following variables:
- IMU.gx, IMU.gy, IMU.gz: rotation rate about the sensor axes in deg/sec
- IMU.ax, IMU.ay, IMU.az: linear acceleration along the sensor axes in g
- IMU.roll, IMU.pitch, IMU.yaw: orientation in degrees (note that we will need to calibrate our magnetometer for this to work properly)
Note that IMU is a global variable defined in ImuUtils.

Component 1: Hardware and Circuit

In order for your hovercraft to lift off the floor, it must be balanced. Move your circuit board into a position such that your craft is balanced front-to-back and left-to-right. We will be providing battery mounts for this purpose.

The IMU is a MPU-9250. Mount the Inertial Measurement (IMU) on the mast that we provide. The advantage of the mast is that your sensor will be less subject to magnetic interference from the hovercraft motors and wires, as well as sources buried in the floor.

Connect the IMU to your circuit. The pins are:

BLACK Ground
+3.3V power: connect to the 3.3V supplied by the Teensy
SDA: connect to SDA0 of the Teensy (Arduino pin 18)
SCL: connect to SCL0 of the Teensy (Arduino pin 19)

Component 2: Software

Copy the following function from project 1:
- void display_heading_velocity(int16_t velocity)
Copy the functions that you developed in project 4 into your current project.
Implement a new PeriodicAction (called sensor_task) that executes once per 5 ms The associated sensor_step() function should:
- Call imu_update().
- Display the rotation velocity of the craft.
Implement the following function:
- void set_hovercraft_acceleration(float ddx, float ddy, float ddtheta)
  This function accepts as input a desired acceleration of the hovercraft, translates this desired acceleration into an appropriate thrust level for each of the lateral fans, and changes the thrust state of these fans.
Implement a new finite state machine in fsm_step(). This state machine will:
1. Ramp up the central fan to a ~70% duty cycle in 5 seconds.
2. Wait in this state for 30 seconds.
3. Ramp down the central fan to zero.
Implement a new PeriodicAction (called control_task that executes once per 10 ms. The associated control_step() function will implement a damping rotation controller. Specifically:
```
float ddx = 0.0;
float ddy = 0.0;
float ddtheta = -Kv * IMU.gz;   // gz = gyro about the Z axis
set_hovercraft_acceleration(ddx, ddy, ddtheta);
       
```

Project Template

Below is your starting point for your project.

#include "PeriodicAction.h"
#include "ImuUtils.h"

// Promise that we will implement this function later
void control_step();

// Create a task that will be executed once per 10 ms
PeriodicAction control_task(10, control_step);

// Promise that we will implement this function later
void sensor_step();

// Create a task that will be executed once per 5 ms
PeriodicAction sensor_task(5, sensor_step);

void setup() {
  // Initialize the USB serial connection
  Serial.begin(115200);

  // Initialize the IMU
  imu_setup();

  // Add your project-specific setup code here

}

/* 
 * Take one step in querying the sensors
 */
void sensor_step()
{
  imu_update();
}



/**
 * No changes necessary to the loop (except to add new tasks)
 */
void loop() {
  // Check to see if it is time to execute the various tasks
  sensor_task.step();
  fsm_task.step();
  control_task.step();
}

Component 3: Testing

Perform initial testing while holding the lip of the Hovercraft with two hands (I recommend not sticking fingers in the fans).
Start with a conservative Kv (i.e., a small value).
Slowly increase the gain until the craft resists the rotations.
At this point, you can place the craft on the floor (or a table, if you are careful).
If the craft tends to oscillate while on the ground, then decrease your Kv.

What to Hand In

All components of the project are due by Thursday, March 29th at 1:30 pm

Demonstration/Code Review: All group members must be present. Given time, this can be done during class. The demonstration must be completed by Tuesday April 4th.
Check in the following to your project 5 area of your Dropbox:
- Documented code: See the project 1 specification for detailed documentation requirements.
Personal report: Catme will request that you fill out a survey about the project. These must be completed by Tuesday April 4th in order to receive credit for the project.

Grading

Personal programming credit:

Each person must accumulate at least three personal programming credits over the course of the semester.
To receive credit, you must be the primary designer, implementer and debugger of the component. This does not mean that your other group members should not be looking over your shoulder. But: you must do the "driving."

Group grade distribution:

35%: Project implementation
30%: Demonstration of working project (to either of the TA or the instructor)
35%: Documentation

Group Grading Rubric

Grades for individuals will be based on the group grade, but weighted by the assessed contributions of the group members to the non-personal programming items.

andrewhfagg -- gmail.com

Last modified: Wed Mar 28 15:09:50 2018